table of contents Sierra Nevada Forest Plan Amendment – Part 4.4

4.4. Forest Service Sensitive Species (Focus species)4.4. species) OverviewOverviewOverview Forest Service Policy (FSM 2670.32) states that all programs and activities will be reviewed as part of the NEPA process to determine the potential effect of such proposed activities on sensitive species. Further, policy states that the impacts of such activities must be avoided or minimized and that any permitted activities must not result in a loss of viability or create significant trends toward Federal listing. The objective of this policy is to conserve species so that they do not become endangered or threatened because of Forest Service actions. Additionally, their habitats must remain well distributed throughout their geographic range on National Forest System lands (FSM 2670.22).

The programmatic Biological Evaluation (BE) for Sensitive Species in the Pacific Southwest Region that are known or suspected to occur within the Sierra Nevada Forest Plan Amendment planning area has been integrated directly into this Final Environmental Impact Statement (in Parts 4.4 and 4.6 of this chapter). Standards for conducting a BE for endangered, threatened, proposed and sensitive species (FSM 2672.42) were carefully applied and the results are included in this section and Part 4.6.

Within this part and Part 4.6 of Chapter 3, the species are identified, their habitat relationships are discussed and known ranges are described for each species. In addition, a discussion of effects, conservation measures and standards and guidelines are included for each species. The discussions for each taxon concludes with a finding that the preferred alternative in combination with land use designations and standards and guidelines for the Sierra Nevada Forest Plan Amendment FEIS are not likely to result in a loss of viability or a trend towards Federal listing.

The documentation in this section constitutes a programmatic BE for Forest Service Sensitive species. Forest Service policy and Sensitive Species standards and guidelines in the Sierra Nevada Forest Plan Amendment FEIS directs that BEs for these species will also be prepared for all proposed actions that are implemented at the project level. The individual sensitive species determinations can be found in the letter entitled: Biological Evaluation determinations for Forest Service Species found within the Sierra Nevada Forest Plan Amendment Planning Area dated December 20, 2000.

A determination of “may affect individuals but is not likely to contribute to the need for Federal listing” was made on all Sensitive species within the planning area except for 10. Of those 10 species, 7 determinations were “no effect” while the remaining 3 determinations were “may affect and likely to contribute to the need for Federal listing.” The three species having the latter determination include the mountain yellow-legged frog, foothill yellow-legged frog and the Yosemite toad.

FEIS Volume 3, Chapter 3, part 4.4, page 1 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 4.4.1. Mammals 4.4.1.1. PACIFIC FISHER (Martes pennanti) I. Affected Environment A. Background Habitat The fisher is among the most habitat-specific mammals in North America, and changes in the quality, quantity, and distribution of available habitat can affect their distributional range in California (Buskirk and Powell 1994). Forest type is probably not as important to fishers as the vegetative and structural aspects that lead to abundant prey populations and reduce fisher vulnerability to predation (Powell 1993). In general, fishers use forest or woodland landscape mosaics that include conifer-dominated stands, and avoid entering open areas that have no overstory or shrub cover (Buskirk and Powell 1994). They select forests that have low and closed canopies. Late-successional coniferous or mixed forests provide the most suitable fisher habitat because they provide abundant potential den sites and preferred prey species (Allen 1987).

Fishers use large areas of primarily coniferous forests with fairly dense canopies and large trees, snags, and down logs. A vegetated understory and large woody debris appear important for their prey species. The following California Wildlife Habitat Relationships (CWHR) types are important to fishers: generally structure classes 4M, 4D, 5M, 5D and 6 (stands with trees 11” diameter at breast height or greater and greater than 40% cover) in ponderosa pine, montane hardwood-conifer, mixed conifer, montane riparian, aspen, red fir, Jeffrey pine, lodgepole pine, subalpine conifer, and eastside pine (Timossi 1990). Part XXX of this chapter describes the CWHR model.

At a landscape scale, patches of preferred habitat and the location of open areas with respect to these patches may be critical to the distribution and abundance of fishers in an area (Buskirk and Powell 1994). Fishers will probably use patches of preferred habitat that are interconnected by other forest types, whereas they will not likely use patches of habitat that are separated by sufficiently large open areas (Buskirk and Powell 1994). Riparian corridors (Heinemeyer and Jones 1994) and forested saddles between major drainages (Buck 1983) may provide important dispersal habitat or landscape linkages for the species. Riparian areas are important to fishers because they provide important rest site elements, such as broken tops, snags, and coarse woody debris (Seglund 1995).

Powell and Zielinski (1994) suggested that habitat suitable for resting and denning sites may be more limiting and that these habitats should be given more weight than foraging habitats when planning habitat management. Key fisher habitat features for resting and denning sites are displayed in Table 4.4.1.1a.

FEIS Volume 3, Chapter 3, part 4.4, page 2 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Table 4.4.1.1a. Key habitat features for fisher resting and denning sites.

Subregion Mean Den Tree Mean Rest Site Mean Rest Site Mean Den Mean Rest Site dbh (in.) Tree dbh (in.) Basal Area Canopy Canopy Closure Closure Conifer Oak Conifer Oak Square ft /acre Percent Percent Northern and b b b b b b b a 31 24 30 19 260 80 88 Central Sierra c d d d d b d d Southern Sierra 49 27 44 26 273 94 93

A For purposes of this table, the Northern and Central Sierra Subregion consist of the Plumas, Lassen, Tahoe, Lake Tahoe Basin, and Eldorado National Forests. Fishers are not known to occur on the Modoc, Inyo or Humboldt-Toiyabe. B Data are derived from Truex et al. (1998) Fisher meta-analysis-Tables 4, 6 and 7. Since no studies of fisher have been completed (due to apparent absence of the species from these subregions) in the Northern or Central Subregions, data from Dr. Rick Golightly's Eastern Klamath Study [theses of Dark (1997) and Seglund (1995)] on Weaverville RD of the Shasta-Trinity NF in an elevation range of 1,900 to 4,800 feet were used to best approximate habitat fisher might use if present in these subregions. C For habitat composition purposes, the Southern Sierra Subregion is considered to consist of the Sequoia, Sierra, and Stanislaus National Forests. D Data are from the Southern Sierra Study (Zielinski, pers. obs.), which falls 80% on the Sequoia National Forest, and 20% distributed between the Mountain Home State Demonstration Forest, the Tule River Indian Reservation, and several private inholdings. The elevation range is from approximately 2,500 to 9,500 feet.

Rest site structures used by fishers include: hollow logs; tree cavities; rocks; snags; ground burrows; fallen trees; canopy of live trees, commonly in witches brooms; and slash and brush piles (Heinemeyer and Jones 1994). In California, trees are the most commonly used rest sites. Buck (1983) reported that rest site trees were greater than 25 feet tall, with diameters of 19 to 67 inches. In the southern Sierra Nevada, oak and white fir were the tree species most frequently used for resting (Truex et al 1998). Down logs greater than 30-inch maximum diameter accounted for approximately 85 percent of all logs used as rest sites (Truex et al 1998), indicating the importance of large woody debris in the forest habitat.

Reproduction and Associated Habitat Natal dens, where kits are born, are most commonly in tree cavities at heights of greater than 20 feet, while maternal dens, where kits are raised, may be in cavities closer to the ground so active kits can avoid injury in the event of a fall from the den (Lewis and Stinson 1998). Most natal and maternal dens are in large conifers (white fir in southern Sierra, Douglas fir in Eastern Klamath) or oaks (black oak in southern Sierra), which may be live or in snag form (Truex et al. 1998). Only eight fisher natal and maternal dens are known in the Sierra Nevada. Protection of these reproductive sites is essential to prevent degradation of habitats used by reproductive females and to minimize disturbance of females during the reproductive period. In the absence of empirical data on female microhabitat use, an area equivalent to approximately 10% of the average home range or 700 acres would be adequate to protect these important areas. Table 4.4.1.1b shows average fisher home range sizes in California.

While protection of den sites is essential, it is important to note that location of den sites is difficult and time consuming. Project-level surveys are unlikely to locate new den sites. Radio- collared females monitored as part of demographic studies would yield the best information on the location of natal and maternal dens and microhabitat use. In lieu of demographic studies, larger, home range sized areas could be established around detection locations. Detection locations likely represent a location used by the individual within the home range so by protecting a large area around the detection, one is likely to protect habitat used by fishers.

FEIS Volume 3, Chapter 3, part 4.4, page 3 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 When the detection is of a female fisher, it is likely that one or more denning structures may be contained in the protected area.

Table 4.4.1.1b. Average home range sizes for fishers in California. Fisher Subregion MEAN MALE MEAN FEMALE (National Forest) Home Range (acres) Home Range (acres) Source Southern (SEQ, SIE, STAN) 9,855a 1,644a Zielinski and others (1997) Central & Northern (ELD, TAH, b b Dark (1997), Seglund 12,500 6,200 b LTB, PLU, LAS) (1995) Modoc-Inyo-Humboldt-Toiyabe N/A N/A Fisher not known to occur All California Means 11,178 3,922 Arithmetic Mean a Mean of two home range estimating techniques: 95% minimum convex polygon, and adaptive kernel. b Data from the Eastern Klamath study area (Dark 1997 and Seglund 1995) may approximate fisher if they occur in the north and central Sierra Subregions, due to similarity of habitats

Landscape Level Habitat Use The presence of large conifers and hardwoods is a highly significant predictor of fisher occurrence (Carroll et al. 1999). There are two possible reasons for the importance of large hardwoods to fishers: (1) cavities, which are frequently used as resting and denning sites, are more common in hardwoods than in conifers, and (2) large hardwoods produce mast (acorns), which may in turn stimulate higher prey densities (Powell and Zielinski 1994).

Density of overhead cover is another predictor of fisher occurrence (Carroll et al. 1999). Landscapes with high levels of overhead cover may protect fishers from predation, reduce the amount of energy fishers expend when traveling between foraging sites, provide more favorable microclimates, and increase prey numbers or prey vulnerability (Buskirk and Powell 1994, Powell and Zielinski 1994). Fishers also use habitat where shrubs contribute to “overhead” canopy (Buck et al. 1994, Dark 1997, Seglund 1995).

Roads may affect fisher distribution; areas with more roads may have increased fisher mortality due to road kill (Heinemeyer and Jones 1994). In addition, roads are associated with habitat alteration and fragmentation, and roads provide access for trappers who target other species, but might incidentally trap a fisher (Carroll et al. 1999).

Current and Historic Conditions Range and Distribution The fisher (Martes pennanti) is found in North America, from the mountainous areas in the southern Yukon and Labrador Provinces in Canada southward to central California and Wyoming, the Great Lakes and Appalachian regions, and New England (Nowak and Paradiso 1983). In California, the subspecies (M. p. pacifica) occurs in the northern Coast Ranges and Klamath Province at elevations of 82 to 3,280 feet (Golightly et al. 1997), and occurs sympatrically with the marten in the southern Sierra Nevada at elevations of 5,000 to 8,500 feet in mixed conifer forests (Zielinski et al. 1996). The fisher historically occurred in the Lassen, Plumas, Tahoe, Lake Tahoe Basin, Eldorado, Stanislaus, Sierra, and Sequoia National Forests, but was not known to occur in the Modoc, Inyo or Humboldt-Toiyabe National Forests. Fishers in the Sierra Nevada currently appear to occupy less than half of their historic range (Zielinski et al. 1997). Recent surveys indicate that fisher are absent from their former range for a distance of almost 240 miles in the central and northern Sierra, from Yosemite National Park northward (Zielinski et al.1995). This gap in distribution may be

FEIS Volume 3, Chapter 3, part 4.4, page 4 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 effectively isolating the existing southern Sierra Nevada population in the Sequoia National Forest and a portion of the Sierra National Forest, from the remainder of the fisher’s range in California, Oregon, and Washington.

Status and Trends Fisher populations are presently at low numbers, or absent throughout most of their historic range in Montana, Idaho, Washington, Oregon, and California (Heinemeyer and Jones 1994). In recent decades, a scarcity of sightings in Washington, Oregon, and the northern Sierra Nevada may indicate fisher extirpation from much of this area (Aubry and Raley 1999, Carroll et al. 1999, Zielinski et al. 1996). The Sierra Nevada and northwestern California populations may be the only naturally-occurring, known breeding populations of fishers in the Pacific region from southern British Columbia to California (Zielinski et al. 1997). Moreover, mortality rates of adult female fishers in the southern Sierra population appear to be high (Truex et al. 1998). No empirical population estimates are available for California, but fishers are considered rare. Because fishers occur at lower elevations than martens, they are more likely to be directly affected by human activities

The USDI Fish and Wildlife Service (FWS) was petitioned to list the fisher under the Endangered Species Act in 1990 and 1994. In both cases, the FWS reported that there was insufficient information to make a determination. A third petition was submitted to FWS on November 27, 2000.

Habitat Risk Factors Protecting biologically productive low- to mid- elevation forests may ensure long-term viability of fisher populations. Processes and patterns at the regional scale, such as metapopulation dynamics, land use, and vegetation patterns, may be important determinants of fisher distribution. To conserve fishers in the Sierra Nevada will require the retention or restoration of sufficient habitat and habitat connectivity throughout the planning area. Given the apparent reluctance of fishers to cross open areas (Earle 1978), and the more limited mobility of terrestrial mammals relative to birds, it is more difficult for fishers to locate and occupy distant, but suitable, habitat. The loss of structurally complex forest (SNEP 1996) and the loss and fragmentation of suitable habitat by roads and residential development has likely played a significant role in both the loss of fishers from the central and northern Sierra Nevada and its failure to recolonize these areas. Fishers prefer continuous or nearly continuous forests, and the decline in Pacific Northwest fisher populations was attributed to extensive logging (Powell 1993).

Fishers are long-lived, have low reproductive rates, large home ranges (for carnivores of their size) and exist in low densities throughout their range. This implies that fishers are highly prone to localized extirpation, colonizing ability is somewhat limited, and that populations are slow to recover from deleterious impacts. Isolated populations are therefore unlikely to persist. Habitat connectivity is a key to maintaining fisher within a landscape. Activities under Forest Service control that result in habitat fragmentation or population isolation pose a risk to the persistence of fishers. Timber harvest, fuels reduction treatments, road construction may result in the loss of habitat connectivity resulting in a negative impact on fisher distribution and abundance. Where these activities occur on non-Forest Service lands, these risks would also apply.

FEIS Volume 3, Chapter 3, part 4.4, page 5 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Key habitats are structurally complex late-successional coniferous forests (Freel 1991, Buskirk and Powell 1994), generally CWHR types 4M, 4D, 5M, 5D, 6 in stands at least 80 acres in size (Freel 1991), with the following attributes:

• Overhead cover (based on Freel 1991, Buskirk and Powell 1994) • Presence of large diameter snags (Allen 1987, Chapel et al. 1992, Freel 1991, Buskirk and Powell 1994) distributed across the landscape. • Large diameter (at least 15 inch dbh by 15 feet long) down logs (Chapel et al. 1992, Freel 1991, Buskirk and Powell 1994) distributed across the landscape. • Large diameter (greater than 24 inch dbh) live conifer and oak trees with decadence such as broken tops or cavities (Chapel et al. 1992, Freel 1991). • Complex structure near the ground (e.g., down logs, large down branches, root masses, live branches) (Buskirk and Powell 1994). • Multi-layered vegetation (vertical within-stand diversity) (Chapel et al. 1992, Freel 1991)

Any of these elements may be affected by Forest Service management activities or by similar activities on non-Forest Service lands. Reduction of any of these elements could pose a risk to fishers. In addition, selection of natal (birthing) and maternal (kit raising) dens is highly specific. Habitat components must exist in the proper juxtaposition within specific habitats in order to provide a secure environment for birth and rearing of fisher kits. All known natal and maternal dens in the western United States have been in large diameter coarse woody debris, snags, or cavities of large diameter live conifers or oaks (Powell and Zielinski 1994, Zielinski et al. 1995, Truex et al. 1998).

Risk factors outside the control of the Forest Service include rural or recreational development that may fragment habitat, increases in road density and traffic levels, and increases in human access to fisher habitat. Non-habitat based risk factors outside the control of the Forest Service include disease and climate change. Fishers are susceptible to both canine and feline distemper. Studies in urban-wildland interfaces suggest a correlation between the prevalence of disease in wild populations and contact with domestic animals (Riley in prep). The fisher is at the southern-most extent of its range in the Sierra Nevada. A species occupying the periphery of its biogeographic range, occurring at mid to upper elevations is at the greatest risk from climate change.

II. Environmental Consequences A. Measures Used to Assess Environmental Consequences As described in the preceding “affected environment” section, a variety of factors influence the fisher and its habitat. These factors are listed here, along with measures that were used to assess each alternative’s effects on the fisher.

1. Protection and recruitment of large old trees (conifer and hardwood) Measure: large trees.

2. Retention of dense forest canopy Measure: canopy closure

FEIS Volume 3, Chapter 3, part 4.4, page 6 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 3. Retention and recruitment of large snags Measure: large snags

4. Retention and recruitment of large down wood Measure: coarse (large) woody debris

5. Intermix of California black and canyon live oak in suitable coniferous habitats Measure: intermix of California black and canyon live oak in suitable coniferous habitats.

6. Management of human presence and associated activities Measure: recreation Measure: roads

7. Distribution and abundance of fishers. Measure: survey requirements and status and trend

8. Management of reproductive sites and protected areas. Measure: protected areas for fishers

9. Quality and quantity of habitat. Measure: abundance of old forest conditions

10. Quality, quantity and distribution of habitat of prey species Measure: prey habitat.

B. Assumptions and Limitations Information on the spatial distribution of management activities at the landscape or regional level is limited. Although we can map the location of current stand conditions, without a spatially explicit schedule of treatments, it is not possible to predict the distribution of stand conditions on future landscapes. Each alternative’s projected trends for acres in different size classes, as derived from the simulation modeling (Appendix B), are based on a series of assumptions, each associated with varying levels of uncertainty. The level of uncertainty surrounding the projections increases as we move father into the future. Consequently, references to vegetation projections will be limited to description of trend and the relative ranking of the alternatives over the first fifty years (see section on Landscape Dynamics for more information). Projections of habitat suitability, derived from CWHR models, incorporate the vegetation projections and interpret habitat type, tree size, and vegetative cover to predict habitat suitability for reproduction, cover or foraging habitat for wildlife. Because it is an “expert system,” based on broad categorizations that synthesizes a large number of environmental variables into a suitability score, the scores are somewhat insensitive to subtle changes in habitat that may nevertheless affect habitat use by wildlife. Like the vegetation projections, predicted changes in habitat suitability are limited to the projected changes over fifty years. As a result of the nonspatial nature of the model projections, much of this evaluation focuses on effects to fisher habitat elements at the stand scale based on interpretation of the Standards and Guidelines. However, the landscape scale is critical to understanding the effects on fisher populations and habitat (e.g., Carroll et al. 1999, Truex et al. in prep.). The absence of knowledge about the spatial arrangements of treatments and effects at the landscape scale, especially in regard to habitat

FEIS Volume 3, Chapter 3, part 4.4, page 7 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 connectivity, is a handicap to this assessment. Lacking this information diminishes confidence in evaluating the effects of all the action alternatives on fishers and their habitat.

Large Trees Large trees are less common in forests today than earlier in the century (McKelvey and Johnson 1992). Large trees can be lost due to natural processes or as a result of management activities. Thick bark may protect live trees from light to moderate intensity fires. If fuels treatment programs reduce the rate of severe fire events that kill large live trees, they could have greater benefits than risks to large live trees. Due to the uncertainty of fire effects on large trees, it is important to recognize that fire effects could either be severe or negligible.

Canopy Cover Natural processes (for example wildfire) or management activities (prescribed fire and mechanical vegetation treatments) can reduce canopy cover. Questions remain about the degree to which management activities, particularly prescribed fire, may reduce canopy cover. The degree to which prescribed fires reduce canopy closure and the time required for treated stands to return to pre-treatment canopy closure levels have not been scientifically studied.

Canopy densities associated with the occurrence of fishers typically have been determined through field studies in which closure is measured using a densiometer. Project planning typically uses information from remote sensing such as aerial photography and satellite imagery. The relationship between densiometer measurements, taken from below the canopy, and the values for canopy closure derived from remote sensing imagery has not be quantified.

Alternatives that affect a large portion of the landscape on a frequent basis and propose to reduce stand densities represent greater risk to forest canopy. Furthermore, alternatives that have a strong and enforceable monitoring and adaptive component represent less risk to forest canopy levels.

Snags Large snags are probably less common in forests today than earlier in the century. Large snags can be lost due to natural processes and as a result of management activities. Large snag replacement rates depend on how rapidly (or slowly) trees grow, die, and decay. Large snags can be lost as a result of management, especially when large trees that could become snags are harvested. While fires create some snags and consume other snags, net rates of snag loss or gain are unknown. The effects of prescribed fire on large snags in the Sierra are unknown. Until research suggests otherwise, it is assumed that prescribed fire will produce a net loss of large snags similar to the loss of snags from wildfire. Although the empirical data are lacking, for the purposes of this analysis, it is assumed that substantial investment in prescribed fire will yield lower probabilities of stand replacing fires. Alternatives with the greatest amounts of treated acres would be most likely to decrease the number and distribution of both large and small snags unless special mitigations are added to protect large snags from loss (e.g., lining snags or raking debris away from base).

Coarse Woody Debris The largest classes of coarse woody debris are recruited from large snags, so much of the discussion regarding snags relates to large coarse woody debris. Given that large snags are probably less common in forests today than earlier in the century, the availability of coarse

FEIS Volume 3, Chapter 3, part 4.4, page 8 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 woody debris, especially large logs, is also likely limited. This seems counter-intuitive, given that fuel loading has increased, but the bulk of the fuel loading occurs in the form of small woody debris (fine fuels) and small trees. Large logs are not easily replaced, because they originate with large snags, and snag renewal rates depend on how rapidly (or slowly) trees grow, die, and decay. Small woody debris are most vulnerable to fire, whether wildfire or prescribed fire. Coarse woody debris are believed to be lost at a somewhat lower rate than small debris, with logs being one of the most vulnerable elements. However, empirical data are lacking regarding exact rates of loss.

Intermix of California black and canyon live oak in suitable coniferous habitats Large diameter black oaks and canyon live oaks compose almost half of the rest sites used by fishers in the southern Sierra Nevada (Zielinski et al. in prep.). To maintain oaks in the mixed conifer forest types used by fisher, four factors should be taken into consideration: 1) retention of large live California black and canyon live oak when implementing management activities, 2) restoration of oaks where they have been eliminated or reduced in abundance, 3) ensuring regeneration of existing oak trees, and 4) recruitment (retention) of large live oaks so that they may age and fulfill additional ecological roles as snags and down logs.

Recreation Recreational activities can affect wildlife species; however, this relationship is poorly understood (Knight and Gutzwiller 1995). Recreational activities can alter wildlife behavior, cause wildlife displacement from preferred habitat, and decrease reproductive success and individual vigor. Peak recreation levels often coincide with the most critical phases of the species’ life cycle, such as during breeding and reproduction. Flight from human presence and interruption of behavior increase energetic costs experienced by an individual. Because of physiological constraints, these costs are greater for smaller animals. Non-motorized recreation is often assumed to have lower impacts than motorized recreation, but this may not be the case. For example, bighorn sheep respond more strongly to hikers than to road traffic, helicopters, or fixed-wing aircraft (MacArthur et al. 1982). However, it is unclear how results from studies on other species might apply to forest carnivores.

Roads Roads can impact fishers in the following ways: (1) vehicles can kill animals and potentially increase mortality rates; (2) roads can fragment habitat and affect the ability of animals to use otherwise suitable habitat on opposing sides of the road; (3) roads, and the presence of vehicles and humans, can cause wildlife to modify their behavior in the vicinity of roads; and (4) roads allow human access to wildlife habitat and can increase the direct impacts of human activities. Predicted habitat for wide-ranging carnivores in the Rocky Mountains was associated with low road densities (Mace et al. 1999, Carroll et al. in press). There may be a threshold value for road density, above which the habitat cannot sustain certain wildlife species (see Mech et al. 1988). Studies have not yet specifically addressed the effects of roads on fisher populations. The regular reports of road-killed fishers in California suggest that reduced road density and road speeds would benefit fishers. Given the current low density of fishers in the Sierra Nevada, the loss of even a small number of individuals through road-related mortality could significantly impact the population.

FEIS Volume 3, Chapter 3, part 4.4, page 9 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Fisher Distribution and Abundance As a method of understanding geographic distribution, project-level surveys are not as rigorous as probability-based surveys (such as random or systematic placement of survey units). Surveys that precede management activities may establish the presence of a fisher and may prevent degradation of occupied habitat. However, since they are not part of a regulated sampling framework, only limited inferences can be drawn from these kinds of surveys. Demographic studies are the most rigorous method to assess population status. Monitoring habitat occupation and condition can be used as a surrogate for monitoring populations but unless there is empirical evidence indicating the relationship between habitat conditions and population trends, the results of habitat monitoring may be difficult to interpret.

Protected areas provide administrative oversight in the areas where fishers currently occur. Buffers around natal and maternal den sites protect the habitat of importance to reproductive females. However, protection of fisher den sites is insufficient as a sole means for protecting fisher populations. In addition, den sites are difficult to locate and it is unknown how frequently den sites are reused. A complementary approach, the creation of buffers around sits where fishers are detected during surveys or other activities (detection buffers), are particularly important for the area outside the current range.

Fisher Diet Although the fisher is reported to be a specialist in late seral mixed conifer-hardwood forests, recent analysis of the diet of fishers in the southern Sierra Nevada portray an opportunistic predator with a diverse diet (Zielinski et al. 1999). Like many carnivores, fishers exploit prey that is patchily distributed in time and space. Consequently, fishers may switch prey in response to changing prey densities (Kuehn 1989). Many of the prey species found in the diet of fishers occur primarily in large tree and dense canopy conifer and oak woodland habitats, chaparral, and deciduous riparian areas. Habitat for species associated with deciduous riparian and chaparral areas was not modeled for the Forest Plan Amendment due to insufficient information on these habitat types (Appendix B). Analysis of changes in habitat utility values (Appendix B) for prey species indicates that different species would benefit under different alternatives. Considering the habitat utility values for the suite of prey species used by fishers, the alternatives were indistinguishable and analysis of individual alternatives was not conducted.

C. Effects of Alternatives Large Live Trees All alternatives would retain very large conifers (50” or greater diameter at breast height) and hardwoods at sufficient levels as to present low risk to this habitat element over time (see Figure 3.1h in Landscape Dynamics section). Within the fifty-year period, it is difficult to separate the alternatives based on vegetation projections. All of the alternatives propose using prescribed fire to varying degrees to reduce the risk of catastrophic fire that could kill large, live trees.

Alternative 8 represents the lowest risk to large trees for fishers because (1) it would retain and recruit large trees through its retention standards for specific levels of canopy cover, basal area, and large trees in suitable California spotted owl habitat and (2) it contains provisions for conducting specific investigations to develop habitat management guidelines for the fisher. Alternatives 6 and Modified 8 could pose some risk to large trees through emphasis on the use

FEIS Volume 3, Chapter 3, part 4.4, page 10 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 of prescribed fire but both alternatives assume a reduced risk of stand-replacing wildfire that is also a threat to large trees.

Alternatives 2, 5, 4 and 7 would represent the greatest risk to large live trees. Alternative 2 would pose a higher risk to large, live trees due to the low level of fuels reduction treatments and the risk of stand-replacing fires. Both Alternative 2 and 5 are projected to have a slight increase over the current condition in acres burned. Alternative 4 has a more aggressive fuels reduction program that could affect large tree recruitment in DFPZs and strategically placed area treatments (SPLATs). In addition, Alternatives 4 and 7 have the highest projected levels of large tree (30” or greater diameter at breast height) removal over the first fifty years.

Dense Forest Canopy Alternatives 5 and 8 would provide the highest levels of dense forest canopy. Alternatives 6 and 7 would result in the greatest predicted risk to known fisher locations by reducing forest canopy from that associated with denning habitat to a level associated with travel and foraging. Alternative 4 would present risks to dense canopy forest through the establishment and maintenance of DFPZs and SPLATs. Modified 8 would generally retain a minimum of 50% canopy closure in westside owl habitats. However, fuels treatments in the General Forest may reduce canopy closure up to 20% to achieve fuels objectives. On the Sequoia National Forest, 66% of the average fisher home range was in 60% or greater canopy closure (Zielinski et al. in prep). These fuels treatments could effectively reduce the quality of habitat available for fishers, particularly for resting and denning, to a level that is sufficient only for foraging or dispersal.

Region-wide, vegetation projections suggest a loss of CWHR classes 4M, 4D and 6 which are stands with medium to large trees (11 inches and greater diameter at breast height) and moderate to dense canopy closure (40% or greater canopy closure). However, there is a concomitant increase in CWHR classes 5M and 5D (24” or greater diameter at breast height) suggesting that 4M and 4D are growing into 5M and 5D and some CWHR 6 (multi-storied with trees 24” or greater diameter at breast height) is moving into other stages as a result of natural disturbances or management activities (see Figure 3.1l in Landscape Dynamics section). When pooled, the structure types preferred by fishers show little appreciable change either by alternative or over the first fifty years. Because the projections do not model the spatial arrangement of suitable habitat, they are of limited value for forecasting which alternative will most likely restore habitat connectivity across the Sierra Nevada.

Snags and Coarse Woody Debris Given that 21 percent of the known natal dens in California occurred in large snags, and 30 percent of the rest sites in the southern Sierra Nevada were in large snags, snags play a key role in the persistence of fisher populations. Reductions in snag levels may indirectly result in decreased rates of reproduction and increased rates of mortality for this species. Figure 3.1j in the Landscape Dynamics section shows the projected trend in snags greater than 15” diameter at breast height. All alternatives show increased numbers of snags over current management. All action alternatives meet the desired condition of five snags per acre by the 50-year mark. Alternatives 2, 5, and 8 show the highest levels of snags. Table 4.4.1.1c shows relative risks to coarse woody debris, which would be recruited from snags. Alternatives 2, 5, 7, and 8 represent the lowest risk to coarse woody debris due to retention standards for both coarse woody debris and snags, use of mechanical treatments and minimizing impacts to these

FEIS Volume 3, Chapter 3, part 4.4, page 11 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 elements when using prescribed fire. Alternatives 6 and Modified 8 emphasize the use of prescribed fire that will generate snags and coarse woody debris but will also consume an unknown proportion of large snags.

Table 4.4.1.1c. Level of risk to coarse woody debris by alternative. Alt 1 Alt 2 Alt 3 Alt 4 Alt 5 Alt 6 Alt 7 Alt 8 Mod 8 Low X X Moderate to Low X X Moderate X X Moderate to High X X X High

Oak/Conifer Intermix Alternatives 2, 5, 6, and 8 and Modified 8 would reestablish the greatest component of large oaks in conifer stands because these alternatives provide direction for retaining large hardwoods, regenerating existing sites, and recruiting oaks. Of these Modified 8 would provide for the greatest recruitment of oaks into larger sizes by retaining large oaks beginning at 12 inches rather than at 15 inches. Alternatives 3, 4, 6, and 8 would retain individual large (greater than or equal to 15 inches) hardwoods during fuels treatments and other management activities. Alternative 7 would only retain black oaks greater than 24” diameter at breast height. It is unlikely that this would provide adequate retention and recruitment of hardwoods in habitats used by fishers.

Recreation All of the action alternatives (Alternatives 2 through 8) establish limited operating periods (LOPs) to protect den sites, limiting the recreation-related impacts during the birthing and kit- rearing periods. Alternative 5 requires the evaluation of all recreation sites within suitable forest carnivore habitat and would restrict new developments within 5 miles of a forest carnivore detection. Alternative 2 would restrict OHV use in den sites all year, conferring a high level of protection from motorized recreation for a small but important area, reproductive sites.

Roads Alternative 5 gives high priority to reducing road densities across landscapes. This alternative would also prohibit new road construction in unroaded areas larger than 5,000 acres and manage ecologically significant unroaded areas between 1,000 and 5,000 acres (a total of approximately 492,000 acres) as reserves. Alternative 5 would provide the greatest benefits to fishers in terms of reducing impacts from roads and related human disturbances. Alternative 3 would prohibit permanent road construction in unroaded areas larger than 5,000 acres. In other areas, new road construction would be offset by reducing existing roads with the goal of achieving a net reduction in road density at the watershed scale. Alternative 4 has the greatest potential for new road construction, and it does not provide for concurrent reduction in existing roads in other areas. Modified 8 provides for seasonal and multi-year closures, and decommissioning of existing roads and would consider den sites and disturbance to wildlife in planning new construction of roads.

FEIS Volume 3, Chapter 3, part 4.4, page 12 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Survey Requirements Alternatives 5, 6, 7 would require surveys outside of the southern Sierra fisher conservation area in suitable habitat, which could provide new information about the distribution of the fisher. Alternatives 2 and 8 would require project level surveys only in the southern Sierra fisher conservation area. Modified 8 would not require project level surveys but would conduct systematic surveys to determine fisher status and distribution. Although systematic surveys are essential to the goals of long-term monitoring, and to the development of new regional habitat models, they may not be a sufficient strategy to conserve fishers in the near term. Fishers are at risk of extirpation in the Sierra Nevada and project -level surveys provide information at a finer scale and can alert managers to the potential alteration of occupied habitat. Project-level surveys provide immediate conservation benefit to a species like the fisher, in which the loss of single individuals, especially females, would be detrimental to recovery. These surveys may be necessary until systematic surveys are fully implemented or until the short-term need to identify occupied habitat has passed. An intermediate approach is outlined in Modified 8: if a fisher is detected outside of the southern Sierra fisher conservation area, finer scale surveys would be implemented surrounding the detection over an appropriate landscape area. Alternatives 1, 3, and 4 would have the least benefit for the fisher in regard to survey requirements.

Protected Areas All of the action alternatives would protect existing and new maternal or natal fisher den sites. Alternative 5 provides for the protection of habitat surrounding new fisher detections by designating an area of 5200 acres around these locations and managing them according to old forest emphasis standards. Alternatives 6 and 8 also establish home range sized buffers around existing and future detections (7500 acres) and manage them for fisher habitat objectives. Alternative 4 also establishes detection buffers but specifies no specific management of the areas. Alternative 5 lacks the southern Sierra fisher conservation area, which would provide special oversight within the currently occupied range of fishers. Alternative 2 and 8 would establish the southern Sierra fisher conservation area and den site buffers; however, only Alternative 8 protects existing and newly discovered fisher detections within and outside the southern Sierra fisher conservation area. Modified 8 provides the southern Sierra fisher conservation area and buffers at existing and new fisher den sites but does not establish detection buffers. Under Modified 8, if a fisher is detected outside the southern Sierra fisher conservation area, an area the size of a home range would be evaluated and managed to retain suitable habitat for fishers. Within the southern Sierra fisher conservation area, within each planning watershed (5,000-10,000 acres), at least 60% of the area would be retained in stands with trees CWHR size 4 (11-24 inches) or greater and at least 50% canopy closure. In addition, efforts would be made to minimize impacts to important legacy structures such as snags and large diameter logs during vegetation treatments. Under Modified 8, treatments in fisher den site buffers would be avoided except where site-specific analysis requires treatment to protect human health and safety. Treatments would be restricted to mechanical removal of fuels except where mechanical treatments are not feasible. As all known fisher den sites occur in either the Urban Defense or Threat zones, there is the potential for den site buffers to be treated to meet fuels reduction objectives. This poses some risk to important reproductive habitat. Alternative 8 provides the greatest protection benefits for fisher persistence in the Sierra Nevada by protecting reproductive sites, habitat surrounding known and future detection locations, and providing special consideration within the species current range.

FEIS Volume 3, Chapter 3, part 4.4, page 13 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Quality and quantity of habitat Fisher habitat suitability is predicted to increase under all alternatives an average of 12.1 percent, ranging from a high of 18.6 percent under Alternative 4 to a low of 4.0 percent under Alternative 2. The differences in the projections may, in part, reflect assumptions about fuels treatments and wildfire in the modeling. In addition, implementation of the standards and guidelines, which add constraints on treatments and are generally not included in the modeling, could alter the outcomes produces by the alternatives.

Alternatives 2, 5, 6, 8 and modified 8 would maintain the greatest amount of the landscape in old forest conditions. Alternative 2 establishes approximately 4.8 million acres in Biodiversity Reserves that maximize old forest condition. Old Forest Emphasis Areas are established under Alternative 6 (approximately 1.6 million acres), and Alternative 8 and Modified 8 (approximately 2.3 million acres). Both Alternatives 2, 6 and Modified 8 specifically require project-level consideration of landscape linkages for wildlife habitat and movement. Modified 8 requires landscape–level (30,000-50,000 acres) consideration of habitat connectivity during landscape analysis. In addition, landscape linkages and bottlenecks will be mapped and explicitly addressed in the conservation assessment for fisher. Alternative 5 designates approximately 1.7 million acres in Old Forest Emphasis Areas and requires the delineation of a network of areas connected by habitat suitable for species movement. Alternatives 2, 5, 6 and Modified 8 pose the least risk to connectivity of fisher habitat. Of these alternatives, Alternative 5 represents the least risk as it explicitly considers connectivity when designing the network of Old Forest Emphasis Areas.

Prey Habitat Because the species found in fisher diets occur in a wide variety of habitats, the alternatives would have differential effects on prey species. However, when viewed as a whole, effects to the suite of species would be negligible under all alternatives. The behavioral plasticity of the fisher should be adequate to compensate for modest shifts in the relative abundance and composition of available prey species, assuming that habitat adequate for foraging fishers is conserved. Consequently, fishers would have equal benefits from this measure under any alternative.

Summary of Consequences to Fisher Management and protection of fishers and their habitat would vary among the proposed alternatives. All alternatives would generally improve upon existing management (Alternative 1) for fishers. In the Sierra Nevada, fishers occur predominantly in national forests.

To evaluate the alternatives for overall benefit to fisher persistence, the measures evaluated above were grouped into three categories: (1) vegetation structure and composition, (2) recreation and roads, and (3) survey requirements and site protection. Vegetation composition and structure reflects how well the alternative maintains or enhances the habitat elements important to fishers. The recreation and roads category considers how human presence, recreational development, and roads under each alternative could impact fishers. Under the survey requirements and site protection category, the alternatives are evaluated based on requirements for surveys and level of protection provided to den sites and detection locations.

FEIS Volume 3, Chapter 3, part 4.4, page 14 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Vegetation Composition and Structure In the vegetation structure and composition category, Alternatives 5 and 8 represent the greatest benefit to fisher persistence and offer the best opportunity to increase fisher abundance and distribution in the Sierra Nevada. Both alternatives would protect fisher habitat through their provisions for retaining and recruiting large trees, snags and coarse woody debris; retaining dense forest canopy; and promoting hardwoods on conifer sites. In addition, Alternative 5 explicitly considers connectivity when designing Old Forest Emphasis Areas. Alternative 2 would provide habitat protections similar to Alternatives 5 and 8. However, because Alternative 2 relies primarily on fire suppression to manage the threat of severe wildfires, the risk of catastrophic fire would be high under this alternative. Although similar to Alternative 8 in large tree protections, Modified 8 poses a greater risk to canopy closure by allowing reductions of up to 20% in certain areas during fuels treatments. Modified 8 would provide habitat considerations at the planning watershed scale within the southern Sierra fisher conservation area, but does not directly address the habitat needs of dispersing fishers outside this area except when a detection is recorded. Although connectivity is not addressed up front as is Alternative 5, Modified 8 considers habitat connectivity when evaluating potential projects and during landscape analysis. In addition, Modified 8 requires the mapping and consideration of habitat bottlenecks within the fisher conservation assessment. Consequently, Modified 8 represents somewhat more risk to fisher habitat. Alternative 3, while similar in some regards to Alternative 5 and 8, would have less beneficial impacts on fishers in terms of dead and down wood and hardwoods on conifer sites. Under Alternative 6, canopy closure in denning areas could be reduced to 40 percent. Therefore, Alternatives 3 and 6 are viewed as less beneficial for fishers and fisher habitat over time.

Human disturbance All of the action alternatives would protect fisher den sites from human disturbance; however, none of the alternatives would reduce road-related risks to the same extent as Alternative 5. Alternative 5 would reduce potential recreation-related impacts in close proximity to fisher locations and would reduce the impacts of roads and related human disturbance by reducing road density and protecting unroaded areas.

Survey Requirements and Site Protection Alternatives 2, 5, 6, 7 and 8 require surveys of suitable habitat for fisher presence. Systematic surveys, which would supply valuable information on fisher distribution and population trends, are addressed only in Modified 8, which doesn’t address project-level surveys like the other alternatives. Instead, Modified 8 takes a different approach, effectively invoking project-level surveys in the event of a fisher detection outside the southern Sierra fisher conservation area. Alternatives 2, 8 and Modified 8 would provide the greatest protection for the area where the fisher currently occurs via establishment of the southern Sierra fisher conservation area. However, neither Alternative 2 nor 8 requires surveys outside this area, increasing risk to fishers occurring outside the southern Sierra fisher conservation area. Modified 8 does not require project level surveys except as noted above, increasing the risk of degrading occupied habitat. Although Alternative 5 would not establish the southern Sierra fisher conservation area, it shares other similar standards with Alternatives 2 and 8 for protection of den sites and existing and future detection locations. Although Modified 8 provides den site buffers, there is some risk that the buffers would be subject to vegetation treatments. Alternative 8 provides the best protections for known locations by establishing the Southern Sierra Fisher Conservation Area, den site buffers when dens are located, and establishing home range size buffers around

FEIS Volume 3, Chapter 3, part 4.4, page 15 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 existing and future fisher detections. Modified 8 shares requirements for the establishment of the southern Sierra fisher conservation area and den site buffers, but would not establish detection buffers per se. Lastly, Alternatives 2, 8 and Modified 8 initiate demographic studies that would help reduce uncertainty around the status and trend of fishers in the southern Sierra Nevada. This information would also refine conservation actions for fishers by enabling management to target those elements most at risk.

Alternatives 3 and 4 would provide lower benefits to fisher abundance and distribution as these alternatives would decrease the availability of habitat elements important to fishers; they provide minimal protection from the impacts of recreation and roads; and these alternatives would not provide increased protection of known fisher areas and would not require surveys for fisher presence.

Environment and Population Outcomes for the Fisher Environmental and population outcomes were derived by professional opinion to estimate environmental and population conditions that would exist in 50 years for the fisher under each alternative. Assigning these outcomes, while inherently subjective, is based on a reasoned thought process and the best available information. The environmental outcome addresses the capability of the environment on Forest Service lands to support population abundance and distribution. The population outcome addresses environmental conditions on all lands within the bioregion and other risk factors that may affect population abundance and distribution.

Table 4.4.1.1d. Outcome ratings for the fisher. Environment outcomes evaluate the estimated environmental conditions on Forest Service lands after 50 years under each alternative. Population outcomes evaluate the estimated population conditions based on the environment outcome and other risk factors.

Outcome Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Environment C- D+ C C- D+ B- C C- B- C+ Population D D- C- C- D C C- D+ C C-

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Rationale for Outcome Ratings The vast majority of fisher habitat, both current and historic, occurs on Forest Service lands (more than 90% and more than 75%, respectively). Trapping, which is believed to have been primarily responsible for the historical decline in the fisher population, was stopped in 1945. Fishers apparently remain at low population density in the Sierra Nevada, with an apparent gap in the

FEIS Volume 3, Chapter 3, part 4.4, page 16 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 population between the Shasta-Trinity National Forest and the southern portion of the Sierra Nevada. The reasons for lack of increase and expansion of the fisher population are unclear, but may include the lack of appropriate habitat elements, including large trees and snags, the lack of connectivity among patches of remaining habitat, the fragmenting effect of major highways, and human disturbance associated with the presence of smaller roads. The current environmental conditions for fishers were estimated to be patchy with some disjunct areas of suitable habitat effectively isolated (Outcome C-). The current population status was estimated to be somewhat more disjunct with little chance of interaction across gaps in the distribution (Outcome D).

For the purposes of this assessment, the assumptions and criteria described in the analysis of environmental consequences above generally apply. Given the uncertainties inherent in the model projections and the lack of spatial location of stand conditions in the future, the effects of implementation of the Standards and Guidelines on distribution and quality of habitat were also considered. Because of the uncertainty regarding the location and implementation of vegetation treatments, alternatives that provided protection at several spatial scales ranked higher in these outcomes. Within the fifty-year period of these outcomes, there is greater potential to realize a downward trend than to make appreciable change in a positive direction. This is due to the slow growth rate of trees and recruitment of legacy structures important to fishers and the fact that fishers are long-lived with relatively low reproductive rates.

Environmental Outcome Alternatives 5, and 8 were judged to result in the greatest improvement of conditions for the fisher. Positive features of Alternative 5 include: (1) allocation of significant area to unroaded areas, (2) the establishment of den site buffers, (3) a requirement generally prohibiting significant new recreational developments within 5 miles of carnivore detections, and (4) protection of habitat conditions around fisher locations, and (5) the requirement to survey for threatened, endangered, and sensitive species. Positive features of Alternative 8 include: (1) establishment of the southern Sierra fisher conservation area, (2) partial protection of reproductive habitat in den site buffers, (3) protection of habitat conditions around detection locations, and (4) a requirement to maintain all suitable California spotted owl habitat. Both of these alternatives provide protection for known locations through den site and detection buffers, and some consideration of landscape level habitat (Alternative 8—southern Sierra fisher conservation area) and habitat connectivity (Alternative 5—consideration of landscape connectivity when designing areas of old forest emphasis).

Alternatives 1 and 4 were judged to result in continued deterioration of conditions for the fisher. Alternatives 2, 3, 6, 7 and Modified 8 were judged to be intermediate with slight improvement under some alternatives. Alternative 2, while similar in some regards to Alternatives 5 and 8, was judged to be less beneficial for fisher habitat due to low level of fuels reduction treatment, reliance on fire suppression, and the increased risk of stand-replacing wildfire events. Modified 8 also provides similar protections to Alternatives 5 and 8, such as establishing the southern Sierra fisher conservation area and buffers around den sites. In addition, Modified 8 provides some consideration of fisher habitat at the watershed scale within the southern Sierra fisher conservation area and during landscape analysis. However, Modified 8 may allow degradation of conditions in 4M and 4D stands used by fishers, and it does not establish habitat protections around detection locations or require project level surveys except around a verified detection outside the southern Sierra fisher conservation area. Alternatives 3, and 7 were judged to be intermediate with some deterioration in conditions for

FEIS Volume 3, Chapter 3, part 4.4, page 17 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 fishers. These alternatives do provide for establishment of den sites buffers, but these areas provide for only partial retention of appropriate den conditions.

Population Outcome Three additional assumptions were applied in consideration of the population outcomes: 1) the amount of treatments (e.g. harvest) on private lands in the Sierra Nevada is stable or increasing; 2) human populations are increasing and will lead to increased urbanization; and 3) the popularity of recreational activities is increasing. Since the vast majority of fisher habitat occurs on Forest Service lands, the majority of predicted effects on fisher population distribution and abundance would also be expected to occur there. These criteria do have some impact, however, in the central and northern Sierra Nevada in areas of mixed ownership where there could be significant reduction in landscape quality and connectivity for fishers. Of principal concern in judging population outcomes were: 1) environmental outcome for the alternative, 2) provisions for expansion of fishers into unoccupied portions of the historic range.

Alternatives 5, and 8 were judged to provide the most benefit to fisher population distribution and abundance by protecting the current and future location of fishers through detection buffers and, in Alternative 8, the southern Sierra fisher conservation area. Alternative 5 protects the most area in roadless condition and would provide consideration of forest carnivores when planning new recreation developments, which could provide some protection from increasing levels of development and recreation. Alternative 5 also requires surveys in suitable habitat and would establish detection buffers when new locations are discovered. This provides the best opportunity for expansion of fishers into unoccupied areas of their historic range. Alternative 8 requires surveys only in the southern Sierra fisher conservation area. Alternative 8 provides a cautious approach to treatments in fisher habitat and would protect against habitat degradation or loss in the short term while the effects of treatments on important habitat elements are being assessed.

Modified 8 would also provide benefit to fisher population and distribution and abundance by providing some protection for den sites and the current range of the fisher, and proactive measures to encourage recovery to its historic range. This alternative provides less protection for fisher locations in the near term as it does not require fine-scale surveys for presence unless there is a detection of a fisher outside the southern Sierra fisher conservation area.

Alternatives 1 and 4 were estimated to continue the degradation of fisher distribution and abundance. Although habitat projections for Alternative 4 are high, there is a high level of treatment under this alternative with few protections for fisher locations. If treatment levels on private lands are steady or increasing, this alternative could have significant impacts in areas of mixed ownership for abundance and connectivity of habitats for fishers.

Alternatives 2, 3, 6, and 7 are intermediate in projected outcomes for fisher populations. Alternatives 2 and 3 set aside large amounts of area as reserves and roadless areas that could provide some protection against development but lack the other protections afforded under Alternatives 5 and 8. Alternative 6 requires surveys, and protects habitat around existing and future fisher locations. Alternative 7 also requires surveys but fails to provide protection of positive detections.

FEIS Volume 3, Chapter 3, part 4.4, page 18 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 4.4.1.2. MARTEN I. Affected Environment A. Background Habitat At a regional level, martens are affected by climate, land use patterns, and metapopulation dynamics (Zielinski et al. 1997). On a landscape scale, patches of preferred habitat and the distribution of open areas with respect to these patches may be critical to the distribution and abundance of martens (Buskirk and Powell 1994). Martens have not been found in landscapes with greater than 25 percent of the area in openings, even where suitable habitat connectivity exists (Chapin et al. 1998, Hargis et al. 1999, Potvin et al. 2000).

Martens prefer coniferous forest habitat with large diameter trees and snags, large down logs, moderate-to-high canopy closure, and an interspersion of riparian areas and meadows. Important habitat attributes are: vegetative diversity, with predominately mature forest; snags; dispersal cover; and large woody debris (Allen 1987). Martens selected stands with 40 to 60 percent canopy closure for both resting and foraging and avoided stands with less than 30 percent canopy closure (Spencer et al. 1983).

Martens generally avoid habitats that lack overhead cover, presumably because these areas do not provide protection from avian predators (Allen 1982, Bissonette et al 1988, Buskirk and Powell 1994, Spencer et al. 1983). In Yosemite National Park, martens avoided areas lacking overhead cover and preferred areas with 100 percent overhead cover, especially when resting (Hargis and McCullough 1984). Preliminary results of studies in the southern Sierra Nevada indicate marten rest sites are associated with closed canopy, multi-layered conditions (Zielinski et al. 1995). Various studies in the Sierra indicate that martens have a strong preference for forest-meadow edges, and riparian forests appear to be important foraging habitats (Spencer et al. 1983, Martin 1987).

It appears that the key to forest carnivore habitat quality is the structural diversity of forest vegetation (Allen 1987). Small open areas and regenerating stands (or plantations) are used by marten as foraging habitat, but these openings are of optimum value when they occupy a small percent of the landscape and occur adjacent to mature forest stands meeting requirements for denning or resting habitat. It is suggested that small dispersed tree harvest units within a forested matrix should have less impact on marten populations than large contiguous clearcuts and, in some instances, may prove beneficial (Thompson and Harestad 1994). In contrast, loss of interior forest habitat may outweigh size of opening concerns, and progressive cutting from a single patch could retain the most interior forest habitat (Hargis et al. 1999)

Important forest types include red fir, lodgepole pine, subalpine conifer, mixed conifer-fir, Jeffrey pine, and eastside pine (Zeiner et al. 1990). The following California Wildlife Habitat Relationships (CWHR) habitat stages are moderately to highly important for the marten: 4M, 4D, 5M, 5D and 6. (The CWHR model is described in Part 4.1.3 of this chapter.) Martens are closely associated with relatively mesic, late successional coniferous forests, although they may occur in other vegetation types. Complex physical structures (large snags, large down woody material, and debris piles), especially near the ground, appear to provide protection from predators, prey sources, access to subnivean (below snow) spaces, and protective thermal microenvironments, especially in the winter (Buskirk and Powell 1994, Spencer et al. 1983,

FEIS Volume 3, Chapter 3, part 4.4, page 19 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Thompson and Harestad 1994). Sites used for subnivean entry have: (1) greater percent cover of coarse woody debris, (2) greater total volume of coarse woody debris, (3) greater numbers of log layers, (4) greater volume of undecayed and moderately decayed logs, (5) less volume of very decayed logs, and (6) fewer small root masses than surrounding forest stands (Corn and Raphael 1992). Hence, large coarse woody debris (snags, down logs, large branches and root masses) is an important winter habitat component. Key habitat structure includes low branches of live trees, tree boles in various stages of decay, large coarse woody debris, presence of red squirrel middens, a shrub layer to the canopy, and large diameter trees and snags.

Table 4.4.1.2a. Westside suitable habitat in the marten core elevation range (5,500 to 10,000 feet). Martes a Westside Habitats Habitat Element Travel/Forage Denning/Resting a a Canopy Cover >=40% >=70% a a Largest Live Conifers >=24"dbh, >=6/acre =24"dbh, >=9/acre b Live Tree Basal Area -- 163-326 sq ft/acre >=350 sq ft/acre1 c c Largest Snags Avg 2.5/acre >=24"dbh Avg 5.0/acre >=24"dbh Coarse Woody Debris Largest logs ( >15 ft long) for 5-10 Largest logs ( >15 ft long) for 5-10 c c tons/acre in Decay Classes 1-3. tons/acre in Decay Classes 1-2. a Freel (1991) b Spencer et al. (1983). Fig 3. c Compilation of information from data tables, and Freel (1991).

Empirical data on marten use of forested habitat on the eastside of the Sierra Nevada are sparse. Marten in these habitats appear to focus on microhabitat elements available in greater proportion than westside areas, such as rock piles, to meet habitat needs. Marten were found resting on barren ground in association with rock piles and scree slopes (Kucera 1996). At the Sagehen study site on the Tahoe National Forest, Spencer et al. (1983) noted that marten activity generally occurred within 190 feet of a meadow, and was rarely noted more than 1267 feet from meadows. Riparian lodgepole pine was strongly preferred in the Sagehen Creek Basin, as were sites with Douglas squirrel feeding sign (middens) (Spencer et al. 1983). Table 4.4.1.2b displays key components of eastside marten habitat.

FEIS Volume 3, Chapter 3, part 4.4, page 20 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Table 4.4.1.2b. Eastside suitable habitat for the marten. Denning/Resting Habitat Travel/Forage Habitat Habitat Component Inyo NFa East Tahoe NFb Northern Utahc Educated Extrapolationd Percent Canopy Cover Average 21 >40 45 >40 (Range) (8-34) Not Provided Not Provided --- Large Live Conifers Average Size (dbh) 35” 24” Not Provided >24” Number per Acre 17-45 Not Provided Not Provided >6 Live Tree Basal Area (ft2/ac) Average 181 Not Provided 150 150 (Range) (93-269) (163-326) Not Provided --- Large Snags Average Size (dbh) 39” 40” Not Provided 30” Number per Acre 3.0 Not Provided Not Provided --- Avg Basal Area (ft2/ac) Not Provided 49 53 --- Coarse Woody Debris Avg Size (large end 46” “Largest” Not Provided >20” diameter) Length & Decay >15’ Long, De- Not Provided Not Provided >15’ long, Decay Class 1-2 Class(es) cay Class 1-2 Number per Acre >4.0 Not Provided Not Provided --- Tons per Acre Not Provided Not Provided 6.83 5-10 a Source: Kucera (1996) tables 7, 11, 14, and 16. b Source: Spencer et al. (1983) and Spencer (1987). c Source: Hargis et al. (1999). d Data specific to travel/foraging habitat are not available.

Home Range Marten home ranges are very large relative to their body size by mammalian standards. Consistent with observations that home range size expands as habitat quality declines, home range sizes for eastside martens (Inyo and Humboldt-Toiyabe National Forest data displayed in Table 4.4.1.2c) are at least three times larger than that known for westside martens (Buskirk and Ruggiero 1994, Buskirk and Zielinski 1997). However, it is not clear how martens use these habitats.

Table 4.4.1.2c. Marten home range sizes in the Sierra Nevada. Subregion Mean Male Mean Female HR (Acres) HR (Acre) Source a a Southern (Seq, Sie, Stan) 807 254 Zielinski and others (1997) b Central/Northern (Eld, Tah, LTB, Plu, Las, Mod) 960 801 Simon (1980) Spencer (1981) Inyo, Humboldt-Toiyabe 2,749 1,155 Kucera (1996) All California Means 1,505 737 Arithmetic Mean a Mean of two home range estimating techniques: 95% minimum convex polygon, and adaptive kernel. b Data from the Tahoe studies of Simon and Spencer best approximate marten in the north and central Sierra Subregions, due to similarity of habitats and lack of local research data.

Breeding Biology and Breeding Habitat. Marten natal dens are typically found in cavities in large trees, snags, stumps, logs, burrows, caves, rocks, or crevices in rocky areas. The dens are lined with vegetation and occur in structurally complex, late successional forests (Buskirk and Ruggiero 1994). Canopy cover and the number of large old trees in these patches exceed levels available in the surrounding suitable habitat. The availability of habitat suitable for natal dens may limit reproductive

FEIS Volume 3, Chapter 3, part 4.4, page 21 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 success and population recruitment; this has direct repercussions on future population size (Buskirk and Ruggiero 1994). Establishment of a buffer, equivalent to approximately 10% of an average marten home range (100 acres) at verified marten den sites provides protection for important habitat used by reproductive females. Dispersal distances of 24 to 60 miles have been reported for martens. In highly fragmented habitats, the ability to disperse may be a factor that limits breeding.

Marten diet The marten diet in the Sierra Nevada changes seasonally but is predominately microtine rodents, tree squirrels (Tamiasciurus douglasii and Glaucomys sabrinus), snowshoe hares (Lepus americanus) and – especially in the summer – ground squirrels (Eutamias sp. and Spermophilus lateralis) (Zielinski et al. 1983, Martin 1987). If one item were to be singled out as critical to the diet of martens in the Sierra, it would be voles (Microtus sp.), which contribute to the diet to varying degrees during all seasons of the year (Zielinski et al. 1983). The Douglas squirrel is very important prey in the winter whereas ground-dwelling sciurids occur exclusively in the diet during the snow-free period of the year.

Current and Historic Conditions Range and Distribution The marten occurs from the southern Rockies in New Mexico northward to the treeline in Canada and Alaska, and from the southern Sierra Nevada eastward to Newfoundland in Canada. In Canada and Alaska, martens have a vast and continuous distribution. In the contiguous western United States, martens are limited to mountain ranges within a narrow band of coniferous forest habitats (Buskirk and Ruggiero 1994).

In California, the marten was historically distributed throughout the Sierra Nevada, California Cascades, and the Coast ranges, from the Oregon border southward to Sonoma County. Martens are currently distributed throughout the Sierra Nevada and Cascades (Buskirk and Zielinski 1997). The species’ core elevation range is from 5,500 to 10,000 feet, and they are most often found in the Sierra Nevada above 7,200 feet. The bulk of the marten's Sierra Nevada and southern Cascade (in California) distribution occurs on national forest lands. Verified marten detections (either by track or photo) exist for all Sierra Nevada national forests, although negative survey results occurred at numerous locations in central Plumas and southern Tulare Counties (Kucera et al. 1995).

Status and Trend In most western States and Canadian Provinces where it occurs, the marten is managed as a furbearer (Buskirk and Ruggiero 1994). Although the marten is classified as a furbearer in California, there has been no open trapping season for marten in California since 1954; cessation of trapping coincided with a period when timber harvest and human population in California were dramatically increasing (Zielinski et al. 1997).

Marten distribution in North America has undergone regional contractions and expansions (Buskirk and Ruggiero 1994). Recent studies and sightings indicate that martens are relatively well distributed in a pattern similar to their historical distribution in the Sierra Nevada (Kucera et al. 1995). The decline in marten population size and range during the early part of the twentieth century has been attributed to habitat modifications, with trapping and predator control as contributing factors (Bennett and Samson 1984). Three factors make martens

FEIS Volume 3, Chapter 3, part 4.4, page 22 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 vulnerable to local extirpation and extinction: (1) low reproductive potential; (2) an affinity for overhead cover and avoidance of extensive open areas, especially in winter; and (3) very large home ranges.

Although marten are wide-ranging carnivores and likely occur in suitable habitats across the Sierra Nevada, martens are considered rare compared to many other species of mid-sized carnivores. Empirical population estimates are not available for marten in California. Based upon the elevational ranges suitable for marten in the Sierra Nevada, the overwhelming majority of potential marten habitat in California is contained within the eleven National Forests covered by this document.

Habitat Risk Factors The marten is among the most habitat-specific mammals in North America, and changes in the quality, quantity, and distribution of available habitat could affect their distributional range in California (Buskirk and Powell 1994). Further, martens are predisposed to impacts from human activities because they require mesic, structurally complex forest habitats. In northern Utah, martens responded negatively to low levels of habitat fragmentation, when the average distance between openings was less than 317 feet (Hargis et al. 1999). To ensure Sierra-wide marten population persistence over long time periods, landscape level habitat management for multiple marten populations could be important (Schneider and Yodzis 1994). Andren (1994) suggested that as landscapes become fragmented, there is a negatively synergistic combination of increasing isolation and decreasing patch size of suitable habitat that compounds the results of simple habitat loss. For some species, this may result in a decrease of greater magnitude than can be explained solely by the loss of suitable habitat. Marten may be a species that demonstrates this pattern of exponential population declines at relatively low levels of fragmentation (Bissonette et al. 1997).

Risks to marten habitat under Forest Service control include those activities that cause the removal of overhead cover, removal of large diameter trees and coarse woody debris, and the conversion of mesic to xeric sites with associated changes in prey communities (Campbell 1979). Although overhead cover is regenerated via successional processes in the ecosystem, the removal of coarse woody debris can only be ameliorated by artificial additions to the system or by the growth and decadence of new large diameter trees (Buskirk and Ruggiero 1994).

Martens utilize meadows and riparian habitats in close proximity to conifer forest. Elements of importance in these areas are availability of prey species and availability of cover. Grazing and fire suppression can be regarded as the most significant threats to these elements. Grazing reduces the amount of shrub and herbaceous cover available for prey species such as voles. Livestock degrade meadow and riparian areas and cause soil compaction that can also effect prey species. The history of fire suppression in the Sierra Nevada has allowed tree cover to encroach on meadow and riparian areas reducing herbaceous cover for prey and effectively reducing meadow size.

In addition to prey availability, marten require cover. Meadows and riparian areas in a matrix of conifer forest would be preferred. Alternatives that retain coarse woody debris such as logs, either in piles or as single large elements, would be more beneficial for marten than those that

FEIS Volume 3, Chapter 3, part 4.4, page 23 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 do not. Downed logs provide habitat for prey species and provide marten access to subnivean areas in winter; large logs with cavities provide rest sites for marten.

There is no information available to evaluate the effects of roads on marten population persistence, though martens are killed by vehicles (Zielinski, pers. comm.). Rather, use of roads is of primary concern, particularly the access they provide trappers, houndsmen and snowmobiles (Duncan Furbearer Interagency Workgroup, 1989).

Non-habitat Risk Factors Habitat alteration, loss of key habitat elements, changes in the availability of prey, roads and recreation development are all habitat risk factors under Forest Service control. These same risk factors apply to non-Forest Service lands as well. Non-habitat risk factors outside the control of the Forest Service include development and climate change. Rural or recreational development in boreal and subalpine areas may fragment habitat and increase road density and traffic levels posing a risk to martens. Like the fisher, the marten is at the southernmost extent of its biogeographic distribution in the Sierra Nevada. Species at the southernmost extent of their range, occupying mid-to upper elevations, and occupying the southern portion of the bioregion are at greatest risk from climate change.

II. Environmental Consequences A. Measures or Factors Used to Assess Environmental Consequences As described in the preceding “affected environment” section, a variety of factors influence the marten population and its habitat. These factors are listed here, along with measures that were used to assess each alternative’s effects on the marten.

1. Protection and recruitment of large old trees Measure: large trees.

2. Retention of dense forest canopy Measure: canopy closure

3. Retention and recruitment of large snags Measure: large snags

4. Retention and recruitment of large down wood Measure: coarse (large) woody debris

5. Presence of meadows and riparian habitat in proximity to conifer forests Measure: meadows and riparian habitat.

6. Human Presence Measure: recreation Measure: roads

7. Distribution and abundance of martens Measure: survey requirements and status and trend

FEIS Volume 3, Chapter 3, part 4.4, page 24 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 8. Management of reproductive sites and protected areas Measure: protected areas for martens

9. Quality and quantity of habitat Measure: abundance of old forest conditions

10. Quality, quantity, and distribution of prey species habitat Measure: meadow and riparian habitat condition

B. Assumptions and Limitations Information on the spatial distribution of management activities at the landscape or regional level is limited. Although we can map the location of current stand conditions, without a spatially explicit schedule of treatments, it is not possible to predict the distribution of stand conditions on future landscapes. Projections of habitat suitability, derived from CWHR models, incorporate the vegetation projections and interpret habitat type, tree size, and vegetative cover to predict habitat suitability for reproduction, cover or foraging habitat for wildlife. Because it is an “expert system”, based on broad categorizations that synthesizes a large number of environmental variables into a suitability score, the scores are somewhat insensitive to subtle changes in habitat that may nevertheless affect habitat use by wildlife. As a result of the nonspatial nature of the model projections, much of this evaluation focuses on effects to fisher habitat elements at the stand scale based on interpretation of the Standards and Guidelines.

The red fir zone forms the core of marten occurrence in the Sierra Nevada. Most activities proposed by this Forest Plan Amendment would be conducted below the red fir zone. Limited prescribed fire use and vegetation management activities would be conducted in this zone because it generally has lower fire hazard and risk ratings. Thus, potential impacts from vegetation and fuels management activities on martens would be limited. Where martens occur below the red fir zone on the westside of the Sierra Nevada, the assessment presented in the fisher section would generally apply. Considerations of impacts to eastside marten habitat are provided below.

High elevation habitats tend to exhibit less ecological resilience; hence, these habitats take much longer to recover from disturbance than more productive middle elevation sites. Alternatives that directly impact high elevation habitats would be considered more risky for the marten. Conversely, alternatives that focus treatments in lower elevation vegetation types would be less risky for the marten. Alternatives that have a strong and enforceable monitoring and adaptive management component are considered less risky than those that do not.

Large Trees Large trees are less common in forests today than earlier in the century (McKelvey and Johnson 1992). Large trees can be lost due to natural processes or as a result of management activities. Thick bark may protect live trees from light to moderate intensity fires. If fuel treatment programs reduce the rate of severe fire events that kill large live trees, they could have greater benefits than risks to large live trees. Due to the uncertainty of fire effects on large trees, it is important to recognize that fire effects could either be severe or negligible.

FEIS Volume 3, Chapter 3, part 4.4, page 25 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Canopy Cover Natural processes (for example wildfire) or management activities (prescribed fire and mechanical vegetation treatments) can reduce canopy cover. Questions remain about the degree to which management activities, particularly prescribed fire, may reduce canopy cover. The degree to which prescribed fires reduce canopy closure and the time required for treated stands to return to pre-treatment canopy closure levels have not been scientifically studied. Alternatives that affect a large portion of the landscape on a frequent basis and propose to reduce stand densities represent greater risk to forest canopy. Furthermore, alternatives that have a strong and enforceable monitoring and adaptive component represent less risk to forest canopy levels.

Snags Large snags can be lost as a result of management, especially when large trees that could become snags are harvested. While fires create some snags and consume other snags, net rates of snag loss or gain are unknown. The effects of prescribed fire on large snags in the Sierra are unknown. Until research suggests otherwise, it is assumed that prescribed fire will produce a net loss of large snags similar to the loss of snags from wildfire. Although the empirical data are lacking, for the purposes of this analysis, it is assumed that substantial investment in prescribed fire will yield lower probabilities of stand replacing fires. Alternatives with the greatest amounts of treated acres would be those most likely to decrease the number and distribution of both large and small snags.

Coarse Woody Debris Coarse woody debris is recruited from large snags, so much of the discussion regarding snags relates to coarse woody debris. Given that large snags are probably less common in forests today than earlier in the century, the availability of coarse woody debris, especially large logs, is also likely limited. This seems counter-intuitive, given that fuel loading has increased, but the bulk of the fuel loading occurs in the form of small woody debris (fine fuels) and small trees. Large logs are not easily replaced, because they originate with large snags, and snag renewal rates depend on how rapidly (or slowly) trees grow, die, and decay. Small woody debris are most vulnerable to fire, whether wildfire or prescribed fire. Coarse woody debris are believed to be lost at a somewhat lower rate than small debris, with logs being one of the most vulnerable elements. However, empirical data regarding exact rates of loss are lacking.

Meadows and Riparian Habitat Evaluation of the treatment of meadows and riparian areas in the alternatives is difficult in the absence of a unified approach to these ecologically important features. The effects of grazing, roads, recreation and fire were considered independently. Lack of spatially explicit information for grazing allotments, pack stations and Important Bird Areas necessitates some caution in interpreting any cost/benefit that may result from their overlap with marten distribution. Trade-offs exist between allowing fire into riparian and meadow areas to reduce ladder fuels, the risk of catastrophic fire, and forest encroachment on meadows, versus the retention of snags and logs in these areas.

Recreation Recreational activities can affect wildlife species; however, this relationship is poorly understood (Knight and Gutzwiller 1995). Recreational activities can alter wildlife behavior, cause wildlife displacement from preferred habitat, and decrease reproductive success and

FEIS Volume 3, Chapter 3, part 4.4, page 26 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 individual vigor. Peak recreation levels often coincide with the most critical phases of the species’ life cycle, such as during breeding and reproduction. Effects of recreation on martens have not been studied and it is unclear how results from studies on other species might apply to forest carnivores.

Roads Roads can impact martens in the following ways: (1) vehicles can kill animals and potentially increase mortality rates; (2) roads can fragment habitat and affect the ability of animals to use otherwise suitable habitat on opposing sides of the road; (3) roads, and the presence of vehicles and humans, can cause wildlife to modify their behavior in the vicinity of roads; and (4) roads allow human access to wildlife habitat and can increase the direct impacts of human activities. There may be a threshold value for road density, above which the habitat cannot sustain certain wildlife species. Studies have not yet specifically addressed the effects of roads on marten populations.

Distribution of the Marten Martens are believed to currently occupy much of their historical range in the Sierra Nevada; however, this species may not be common anywhere. Because of their expected shorter dispersal distances relative to fishers, marten populations could be at risk through habitat alteration and the hazards associated with small, isolated populations. To address the risk relative to marten distribution and abundance, more information is needed to understand the species’ range, population size, and growth rate. Two measures are used to assess how well each alternative meets these information needs: survey requirements and demographic studies.

Systematic surveys at a broad scale may fail to detect individuals for a variety of reasons not necessarily related to species absence. However, these surveys can provide a better picture of the species distribution and habitat associations across the region than more limited surveys. As a method of understanding geographic distribution, project-level surveys are not as rigorous as probability-based surveys (such as random or systematic placement of survey units). Surveys that precede management activities may establish the presence of a marten. Since they are not part of a regulated sampling framework, only limited inferences can be drawn from these kinds of surveys. Project-level surveys do provide information at a finer spatial scale than systematic surveys and informs managers when an area scheduled for treatment is being used by marten.

Large protected areas for forest carnivores under this Forest Plan Amendment are centered on the currently known distribution of fishers (e.g. Southern Sierra Fisher Conservation Area). This area represents only a fraction of the distribution of martens in the Sierra Nevada. Buffers around natal and maternal den sites protect the habitat of importance to reproductive females. However, protection of den sites is probably insufficient as a sole means for protecting marten populations. In addition, den sites are difficult to locate and it is unknown how frequently den sites are reused. Detection buffers are approximately the size of home ranges and protect habitat surrounding existing and new occurrences.

Marten Diet The assessment of the marten diet is based on the representation, via frequency of occurrence, in the scats of animals that were research subjects at one location in the Sierra Nevada: the crest of the Sierra and the near eastside in the Tahoe National Forest. The marten diet described here may not represent what martens eat elsewhere within the planning area.

FEIS Volume 3, Chapter 3, part 4.4, page 27 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Uncertainty There are several sources of uncertainty that may affect this assessment of martens. First, there is lack of information regarding the explicit spatial location of proposed management activities. Without this information, it is difficult to accurately assess the effect of proposed activities on marten habitat abundance, distribution, and connectivity. Second, there is uncertainty about the effects of proposed activities on important marten habitat elements. For example, information for predicting effects of prescribed fire on large trees, dead and down wood, and canopy closure is unavailable. In addition, means to evaluate treatment risks to these habitat elements relative to benefits conferred by reducing the risk of severe fire effects are lacking. These issues are of less concern for the marten than the fisher, however, as marten habitat occurs at higher elevations where fire is a less common threat to habitat. At these elevations, prescribed fire and silvicultural treatments to reduce fire hazard would be less common compared to the level of these activities in the mixed conifer zone. There is also uncertainty inherent in the theoretical vegetation projections modeled for this assessment.

C. Effects of Alternatives on the Marten Large Live Trees All alternatives would retain very large trees (50” or greater diameter at breast height) at sufficient levels as to represent low risk to this habitat element over time (see Figure 3.1h, in Landscape Dynamics section). Alternative 8 represents the lowest risk to large trees because (1) it would retain and recruit large trees through its stand retention standards for canopy cover, basal, area, and large trees in suitable California spotted owl habitat and (2) it contains provisions for conducting specific investigations to develop habitat management guidelines for the marten. Alternatives 6 and Modified 8 could pose some risk to large trees through emphasis on the use of prescribed fire but both alternatives are projected to have reduced risk of stand-replacing wildfire that is also a threat to large trees. Alternatives 2, 5, 4 and 7 would represent the greatest risk to large live trees. Alternative 2 would pose a higher risk to large, live trees due to the low level of fuels reduction treatments and the risk of stand-replacing fires. Both Alternative 2 and 5 are projected to have a slight increase over the current condition in acres burned. Alternative 4 has a more aggressive fuels reduction program that could affect large tree recruitment in DFPZs and strategically placed area treatments (SPLATs). In addition, Alternatives 4 and 7 have the highest projected level of large tree (30” or greater diameter at breast height) removal over the first fifty years. These risks are most significant where marten occur in lower elevation westside habitats and in eastside habitats.

Dense Forest Canopy Alternatives 5 and 8 would provide the highest levels of dense forest canopy. Alternatives 6 and 7 could reduce forest canopy from levels associated with denning habitat to levels associated with travel and foraging. Alternatives 6 and 8 would reduce canopy cover-related risk to westside martens but increase risk to eastside marten populations. Of the two, Alternative 8 poses a lower level of risk due to its canopy closure standards for suitable California spotted owl habitat on the Westside and its provisions for maintaining canopy cover in California spotted owl home ranges and northern goshawk post-fledging areas on the eastside. Modified 8 would allow the reduction of canopy closure to 30%. Data for marten on the Inyo National Forest indicate that 27 of 76 (36 percent) marten rest sites were in areas with 40% or greater canopy closure (Kucera 1997). Treatments in marten habitat on the eastside could effectively move stands from denning and resting habitat into travel and foraging quality habitat. Alternative 4 poses a significant risk to dense canopy forest because it would establish

FEIS Volume 3, Chapter 3, part 4.4, page 28 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 and maintain both defensible fuels profile zones (DFPZs) and strategically placed area fuels treatments (SPLATs). However, red fir and true fir zones would presumably be lower priority for these treatments.

Region-wide, vegetation projections suggest a loss of CWHR classes 4M, 4D and 6 which are stands with medium to large trees (11 inches and greater diameter at breast height) and moderate to dense canopy closure (40% or greater canopy closure). However, there is a concomitant increase in CWHR classes 5M and 5D (24” or greater diameter at breast height) suggesting that 4M and 4D are growing into 5M and 5D and some CWHR 6 (multi-storied with trees 24” or greater diameter at breast height) is moving into other stages as a result of natural disturbances or management activities (See Figure 3.1l, in Landscape Dynamics section). When pooled, the structure types preferred by martens show little appreciable change either by alternative or over the first fifty years. Because the projections do not model the spatial arrangement of suitable habitat, they are of limited value for forecasting which alternative will most likely ensure habitat connectivity across the Sierra Nevada.

Snags Snag levels may indirectly affect marten reproduction and mortality rates. Figure 3.1j (in the Landscape Dynamics section) shows the projected trend in snags greater than 15” diameter at breast height. All alternatives show improvement over current management. All action alternatives meet the desired condition of 5 snags per acre by the fifty-year mark. Alternatives 2, 5, and 8 show the highest levels of snags.

Coarse Woody Debris Alternatives 5 and 8 represent the lowest level of risk for coarse woody debris. Alternatives 2, 5, 7, and 8 represent the lowest risk to coarse woody debris due to retention standards for both coarse woody debris and snags, use of mechanical treatments and minimizing impacts to these elements when using prescribed fire. Alternatives 4 and 7 provide less snag protection, which could lead to lower levels of recruitment of coarse woody debris over time. Alternative 4 has the highest level of fuels treatment and poses the greatest risk to coarse woody debris recruitment. Alternative 2 poses a long-term risk to coarse woody debris because it relies primarily on fire suppression as a fire control method, thereby increasing the risk of severe wildfire effects over time. Alternatives 6 and Modified 8 emphasize the use of prescribed fire that will generate snags and coarse woody debris but will also consume an unknown proportion of large snags.

Meadow and Riparian Habitats Alternatives 2, 5, 6, 7, and 8 appear to represent the least risk to the marten when one considers meadow and riparian habitats. Alternative 5 does not establish important bird areas (IBAs); however, this alternative provides for greater stubble height (which is favorable for marten prey) and contains many of the provisions to reduce grazing impacts found in the other four alternatives. Although Alternative 7 does not establish IBAs, it shares many attributes with Alternatives 2, 5, 6 and 8 and emphasizes mechanical treatment over prescribed fire, resulting in reduced risk to snags and coarse woody debris in riparian areas used by martens. Alternatives 2, 6, and 8 establish IBAs, which could indirectly benefit the marten by providing conditions for meadow-dependent prey species, assuming that there is elevational overlap between IBAs and marten locations. Alternative 8 would maintain or enhance structural complexity in forest stands adjacent to wet meadows and riparian areas which would enhance

FEIS Volume 3, Chapter 3, part 4.4, page 29 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 marten foraging opportunities. Modified 8 should also provide foraging opportunities for marten by enhancing prey habitat in meadow and riparian areas by restricting treatments and use of fire in these areas.

Recreation Alternatives 2, 3, 5, 6 and 8 provide LOPs for den site protection, limiting the impact of recreation for all or part of the year. This strategy confers a high level of protection from motorized recreation for a small but important area, reproductive sites but, employed alone as in Alternatives 2 and 3, would allow impacts from recreation at other times and in other areas of suitable habitat. Alternative 5 requires the evaluation of all recreation sites within suitable forest carnivore habitat and would restrict new developments within 5 miles of a forest carnivore detection. Alternatives 1, 4, and 7 fail to consider the potential effects of recreation on martens and represent the greatest risk to martens from recreation.

Roads Alternative 5 would emphasize reducing road densities across landscapes and would prohibit road construction in unroaded areas. Of all the alternatives, Alternative 5 would have the least road-related risks to martens. Alternative 3 would prohibit permanent road construction in unroaded areas larger than 5,000 acres. New road construction in other areas would be offset by reducing existing roads with the goal of achieving a net reduction in road density at the watershed scale. Alternative 4 has the greatest potential for new road construction, and therefore has the most road-related risks to martens. Modified 8 provides for seasonal and multi-year closures, and decommissioning of existing roads and would consider den sites and disturbance to wildlife in planning new construction of roads.

Survey Requirements Only Alternatives 5 and Modified 8 require surveys outside the southern Sierra fisher conservation area; such surveys could provide new information about marten distribution in the Sierra Nevada. The survey requirements in Alternative 5 would contribute the most toward conserving martens and their habitat. Modified 8 would rely on broad-scale, systematic surveys to protect martens and marten habitat. Alternatives 2 and 8 stipulates project level surveys for fishers that may also result in marten detections in the southern Sierra fisher conservation area. Alternatives 1, 3, 4, 6, and 7 have fewer survey requirements.

Trend in Population Size Although martens appear to be more abundant and well distributed than fishers, uncertainty remains around the status and trend of marten populations. Obtaining demographic information on martens would address this uncertainty and would improve conservation actions for martens. Alternatives 2 and 8 would provide this information, and could ultimately improve management of marten habitat. The remaining alternatives (Alternatives 1, 3, 4, 5, 6, 7, and Modified 8) do not have direction for obtaining demographic information so population status and trend would remain uncertain.

Protected Areas Alternatives 2, 8 and Modified 8 would protect martens where they co-occur with the fisher in the southern Sierra fisher conservation area. Under Alternative 5, marten detections would be considered when old forest emphasis areas were delineated; however, protected areas would

FEIS Volume 3, Chapter 3, part 4.4, page 30 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 not be established for detected martens. All of the action alternatives would establish protected areas around maternal and natal den sites.

Alternative 2, 8 and Modified 8 establish both the southern Sierra fisher conservation area and den site buffers for marten (100 acres), protecting overlapping habitat for martens and fishers. Alternative 3, 4, 5, 6, and 7 would protect marten den sites, but these alternatives do not have the southern Sierra fisher conservation area. Alternatives 4, 6 and 8 would establish a buffer of 1500 acres around existing and future marten detections.

Quality and quantity of habitat Marten habitat suitability is predicted to increase modestly under all alternatives an average of 4.8 percent, ranging from a high of 10.0 percent under Alternative 4 to a low of –1.0 percent under Alternative 2. The differences in the projections may, in part, reflect assumptions about fuels treatments and wildfire in the modeling. In addition, implementation of the standards and guidelines, which add constraints on treatments and are generally not included in the modeling, could alter the outcomes produced by the alternatives. Lastly, montane habitats used by martens mature slowly and might be expected to exhibit less change over a fifty year period from growth alone than lower elevation, westside habitats.

Alternatives 2, 5, 6 and 8 would maintain the greatest amount of the landscape in old forest conditions. Alternative 2 establishes approximately 4.8 million acres in Biodiversity Reserves that maximize old forest condition. Old Forest Emphasis Areas are established under both Alternative 6 (approximately 1.6 million acres) and Alternative 8 and Modified 8 (approximately 2.3 million acres). Both Alternatives 2, 6 and Modified 8 specifically require project-level consideration of landscape linkages for wildlife habitat and movement. Modified 8 requires landscape–level (30,000-50,000 acres) consideration of habitat connectivity during landscape analysis. In addition, landscape linkages and bottlenecks will be mapped and explicitly addressed in the conservation assessment for fisher which may benefit martens where they co-occur with fishers. Alternative 5 designates approximately 1.7 million acres in Old Forest Emphasis Areas and requires the delineation of a network of areas connected by habitat suitable for species movement. Alternatives 2, 5 and 6 pose the least risk to connectivity of marten habitat. Of the three, Alternative 5 represents the least risk as it explicitly considers connectivity when designing the network of Old Forest Emphasis Areas.

Prey species Because of the importance of microtine rodents in the marten diet, the quality of meadow habitat (especially meadows surrounded by mature lodgepole and red fir forests) and riparian forests with a grass understory influence the quality of marten habitat. The consequences that apply to the measure, “Meadows and Riparian Habitat” above, reflect the consequences to vole habitat, vole populations, and hence marten foraging success. Using meadow and riparian quality as the measure, Alternatives 2, 5, 6, 7, and 8 and Modified 8 would have the greatest benefits to marten foraging habitat. Alternative 5, in particular, provides for greater stubble height and contains provisions for reducing the impacts of grazing on vole habitat.

The CWHR utility values for each alternative integrate the CWHR categories of forest type, density, and tree size class, their relative value to different species, and a forest growth simulator to forecast future habitat value for Sierra Nevada vertebrate species. (Refer to Part 4.1.2 of this chapter for a discussion of the CWHR model and habitat utility values.) The

FEIS Volume 3, Chapter 3, part 4.4, page 31 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 habitat utility values for the most frequent marten prey species in the Sierra Nevada, after 50 years under each alternative, are listed in the Table 4.4.1.2d below.

Table 4.4.1.2d. Projected percent change in habitat utility values for the most frequent marten prey species in the Sierra Nevada, after each alternative has been implemented for 50 years. Alt 1 Alt 2 Alt 3 Alt 4 Alt 5 Alt 6 Alt 7 Alt 8 Mod 8 Mean Red-backed vole 13.3 -0.3 6.0 10.7 6.6 5.4 2.5 0.6 10.7 5.6 Montane vole 177.6 171.5 171.8 173.2 173.6 171.4 171.0 171.5 178.3 172.7 Long-tailed vole -5.4 -13.1 -16.9 -10.7 -129 -16.0 -12.1 -13.2 -5.1 -12.5 Lodge chipmunk -5.5 -15.2 -1.4 -3.3 -13.0 -10.6 -7.3 -13.3 -3.6 -9.8 GM ground squirrel -0.5 -6.8 -7.3 -3.7 -5.4 -7.0 -4.8 -5.8 0.9 -5.2 Douglas squirrel 13.0 5.7 19.3 20.6 8.9 18.3 15.4 9.1 13.3 13.8 Flying squirrel 12.0 4.2 17.3 18.5 7.5 16.1 12.8 7.5 11.9 12.0 Snowshoe hare 458.3 381.6 248.7 324.1 397.7 252.1 289.1 317.0 441.9 333.6

Given the assumptions inherent in this predictive tool, all alternatives have the same direction of change in habitat for each species. For example, all alternatives lead to predicted increases in the habitat value for Douglas squirrels, and all alternatives lead to predicted decreases in future habitat value for long-tailed voles.

Table 4.4.1.2d does not include all the vertebrate prey species of martens, but it does represent their most dominant prey. Decreases in habitat utility values (from a mean of –12.5 percent for the long-tailed vole to a mean of –5.2 percent for the golden mantled ground squirrel) are forecast for three of the species, and increases are forecast for the other five (from a mean high of 333.6 percent increase for the snowshoe hare to a mean high of 5.6 percent for the red- backed vole). Habitat for the montane vole, perhaps the most constant dietary item across the seasons, is forecast to increase under all alternatives, with the greatest increase projected under Modified 8. Douglas squirrels are one of the largest and most common winter prey items and probably help martens survive the severe Sierra winters. Habitat for this species is forecast to increase under all alternatives, with the largest increase predicted under Alternative 3 and 4.

Summary of the Effects of the Alternatives Management and protection of martens and their habitat would vary among the alternatives. Martens currently appear to be well distributed throughout their historic range in the Sierra Nevada. They can be affected, however, by forest management due to their association with old forest conditions and their sensitivity to forest fragmentation. The alternatives were evaluated against the standard of maintaining the current distribution of martens in the Sierra Nevada bioregion and preventing changes in habitat or changes in management that would lead to a decrease in distribution or abundance.

To evaluate the alternatives for overall risk to marten persistence, the measures evaluated in the preceding sections were grouped into three categories: (1) vegetation structure and composition (including forests and meadows), (2) roads, and (3) survey requirements and site protection. The vegetation composition and structure category reflects how well an alternative would maintain or enhance the habitat elements important to martens. The human disturbance category reflects how an alternative would impact martens in terms of human presence, including recreation and roads. The survey requirements and site protection category describes how an alternative would survey for martens and protect den sites and detections.

FEIS Volume 3, Chapter 3, part 4.4, page 32 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Vegetation Composition and Structure Based on the vegetation composition and structure measures, Alternative 5 has the lowest risk to marten persistence and continued distribution throughout the Sierra Nevada bioregion. All alternatives would retain large live trees at sufficient levels as to represent low risk to this habitat element over time. Alternative 5 has a lower level of risk than Alternatives 6, 8 or Modified 8 in two regards: (1) it has less risk to canopy closure in eastside conifer habitats and (2) it has higher projected snag levels. Alternatives 5 and 8 have equivalent risk in terms of the coarse woody debris element; Alternatives 5 and 8 pose less risk to the coarse woody debris element than Alternatives 6 and Modified 8, which would implement more fire hazard reduction treatments although Modified 8 requires consideration of large snags and logs during vegetation treatments. Alternative 8 would apply stand structural standards for retaining large trees, basal area, and canopy cover in suitable California spotted owl habitat that would benefit martens. In addition, Alternative 8 has provisions for conducting specific investigations to develop habitat management guidelines for the marten. Alternative 2 provides management direction for protecting marten habitat that is similar to direction in Alternative 8. Alternative 2 has a greater risk to martens compared to Alternatives 5 and 8 because it would primarily rely on fire suppression over much of the landscape, resulting in an increased risk of stand- replacing fire. Compared to Alternatives 5 and 8, Alternative 3 poses a greater risk to dead and down wood. Risks are also associated with Alternative 6, which would allow canopy closure in denning areas to be reduced to 40 percent. Alternative 5 would most effectively protect the integrity of meadow ecosystems: meadow edges provide important habitat for one of the marten’s primary prey, voles. In the category of vegetation structure and composition, Alternative 5, more than any other alternative, promotes the maintenance of marten habitat features.

Human Disturbance Alternative 5 requires that new recreational developments (for example ski areas) be evaluated for their compatibility with marten needs when they are proposed in suitable marten habitat. Alternative 5 would also reduce the impact of roads and related human disturbance by protecting unroaded areas. All of the action alternatives would provide some level of protection from human disturbance near den sites, but none of the alternatives reduces the risk of roads to the same extent as Alternative 5.

Survey Requirements and Site Protection Only Alternatives 2, 5, and 8 require surveys of suitable habitat for marten presence. Alternative 5 would require surveys outside the southern Sierra fisher management area. Alternative 8 provides the best protections for known occurrence by establishing both den site buffers when dens are located and establishing home range size buffers around existing and future marten detections. Although martens appear to be more abundant and well distributed than are fishers, uncertainty remains around the status and trend of marten populations. Obtaining demographic information on martens would address this uncertainty and would inform management decisions regarding conservation actions for martens. Alternatives 2 and 8 would provide this information and could ultimately reduce the risk to martens from management activities. The remaining alternatives fail to reduce this uncertainty and could lead to increased risk for martens throughout the Sierra Nevada. Systematic surveys, which would supply valuable information on marten distribution and population trends region-wide, are required only under Modified 8, and alternative which does not establish buffers around marten detections.

FEIS Volume 3, Chapter 3, part 4.4, page 33 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Environment and Population Outcomes for the Marten Environmental and population outcomes were derived by professional opinion to estimate environmental and population conditions that would exist in 50 years for the marten under each alternative. Assigning these outcomes, while inherently subjective, is based on a reasoned thought process and the best available information. The environmental outcome addresses the capability of the environment on Forest Service lands to support population abundance and distribution. The population outcome addresses environmental conditions on all lands within the bioregion and other risk factors that may affect population abundance and distribution.

Table 4.4.1.2e. Outcome ratings for the marten. Environment outcomes evaluate the estimated environmental conditions on Forest Service lands after 50 years under each alternative. Population outcomes evaluate the estimated population conditions based on the environment outcome and other risk factors.

Outcome Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Environment B B- B B B- B+ B B B+ B Population B B- B B B- B+ B+ B B+ B

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Rationale for Outcome Ratings Approximately 80 percent of marten habitat occurs on National Forest lands. The remainder of marten habitat occurs on private land (16%) and other public lands (4%). Martens are closely associated with the complex structure found in older conifer forests, including large trees, snags, coarse woody debris, and high canopy closure. They generally occur in low densities, occupy large home ranges, are sensitive to fragmentation and large forest openings, and prey on microtine rodents, especially in and adjacent to meadows. Relative to other forest carnivores, martens are not believed to be as sensitive to human activities and disturbances. The current status of marten habitat was judged to be broadly distributed across the historic range of the species with some temporary gaps (Outcome B). Martens were estimated to be broadly distributed across their range in the Sierra Nevada with some potential for temporary gaps (Outcome B).

For the purposes of this assessment, the assumptions and criteria described in the analysis of environmental consequences above generally apply. Given the uncertainties inherent in the model projections and the lack of spatial location of stand conditions in the future, the effects of implementation of the Standards and Guidelines on distribution and quality of habitat were also considered. Because of the uncertainty regarding the location and implementation of vegetation

FEIS Volume 3, Chapter 3, part 4.4, page 34 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 treatments, alternatives that provided protection at several spatial scales ranked higher in these outcomes.

Environmental Outcome Alternatives 5 and 8 were judged as improving conditions for the marten in 50 years. Key factors contributing to these improved conditions included canopy closure retention on both westside and eastside marten habitats and den site and detection buffers for marten locations. Alternatives 2, 3, 6, 7 and Modified 8 were likely to result in little if any change in marten habitats across the planning area, although each alternative may have had some positive features.

Alternatives 1 and 4 had some likelihood of reducing the distribution or abundance of suitable marten environments. These alternatives could establish some permanent gaps in distribution that could isolate some subpopulations, primarily because they lack explicit features for maintaining or improving forest canopy closure or complex structures in the substantial acreage allocated to general forest. However, even in these alternatives, marten habitats would likely continue to be present in sufficient abundance and distribution in 50 years to provide for population interaction throughout the species’ range in the planning area.

Population Outcomes Three additional assumptions were applied in consideration of the population outcomes: 1) the amount of treatments (e.g. harvest) on private lands in the Sierra Nevada is stable or increasing; 2) human populations are increasing and will lead to increased urbanization; and 3) the popularity of recreational activities is increasing. Since the vast majority of marten habitat occurs on Forest Service lands, the majority of predicted effects on marten population distribution and abundance would also be expected to occur there. These criteria do have some impact, however, in the central and northern Sierra Nevada in areas of mixed ownership where there could be significant reduction in landscape quality and connectivity for martens. Of principal concern in judging population outcomes were: 1) environmental outcome for the alternative, 2) provisions for maintaining connectivity of marten populations.

Alternatives 5, 6 and 8 would provide the most benefit to marten populations by protecting den sites and current and future detection locations. Features of these alternatives that resulted in this rating include: 1) survey requirements; 2) protection of den sites; and 3) protection of known and future detection locations. In addition, Alternative 5 protects the most area in roadless condition and would provide consideration of forest carnivores when planning new recreation development which could provide some protection for martens from increasing levels of recreation. Modified 8 would provide slightly less benefit to martens as it does not provide detection buffers and could allow the reduction of canopy closure in eastside habitats to 30% potentially reducing landscape connectivity for martens. However, marten are expected to retain a broad distribution within the planning area under Modified 8, and Alternatives 2, 3, and 7.

Alternative 1 and 4 were judged to have some likelihood to result in slight decreases in marten abundance and distribution. These alternatives lack explicit features for retaining canopy closure and complex stand structure, and provide fewer protections than the other alternatives. Even under these alternatives, marten were expected to remain broadly distributed with small gaps throughout the planning area.

FEIS Volume 3, Chapter 3, part 4.4, page 35 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 4.4.1.3. SIERRA NEVADA RED FOX I. Affected Environment A. Background Habitat Little is known about the habits of the Sierra Nevada red fox. It appears that the Sierra Nevada red fox prefers red fir and lodgepole pine forests in the subalpine zone and alpine fell fields of the Sierra Nevada (Ingles 1965). This species does not appear to require dense canopy closure (Duncan Furbearer Interagency Workgroup 1989); however, it uses forested areas in proximity to meadows, riparian areas, and brush fields (Zeiner et al. 1990). Sierra Nevada red foxes use openings to a greater extent than either martens or fishers, possibly due to their larger size. Grinnell et al. (1937) considered the Sierra Nevada red fox to be restricted to the highest forested peaks and ridges, foraging above timberline in fall and even mid-winter. Forested habitats are used for reproduction and cover (Zeiner et al. 1990). Dense vegetation and rocky areas provide additional cover. Young may be reared in cavities or spaces within rock piles and talus slopes (Zeiner et al. 1990).

Many Sierra Nevada red fox sightings are at high elevations. Recent sightings of red fox (of unknown subspecies) on the Lassen National Forest have occurred in mixed conifer and red fir forests above 5,000 feet. In the northern Sierra Nevada, roughly equal numbers of sightings have been recorded in fir and mixed conifer, with additional sightings in mixed pine and lodgepole pine. In the southern Sierra Nevada, reports are predominantly from mixed conifer forests with additional sightings in lodgepole pine and red fir (Schempf and White 1977). Sightings have been reported in open Jeffrey pine and sagebrush stands as well (Bogard photo documentation, Duncan Interagency Furbearer Workgroup 1989). The Sierra Nevada red fox may move seasonally from higher elevations in the summer to mid-elevations during the winter (Zeiner et al. 1990). Winter habitat requirements are unknown. Other fox subspecies in Canada and the eastern United States are territorial and use large areas with average home ranges varying from 1,700 to 4,000 acres (Duncan Interagency Furbearer Workgroup 1989). However, the Sierra Nevada subspecies may require larger home ranges than eastern subspecies because the prey base is presumed to be more limited in the Sierras (Duncan Interagency Furbearer Workgroup 1989).

Diet of the Sierra Nevada Red Fox The Sierra Nevada red fox preys upon a variety of small to medium sized mammals, birds, other vertebrates, carrion and fruits (Zeiner 1990). In the western states, the diet of Sierra Nevada red fox is composed of microtines, mice, chipmunks, woodrats, squirrels, and lagomorphs (Ingles 1965) with pika and snowshoe hare apparently being important in the Sierra region (Duncan Furbearer Interagency Workgroup 1989).

Current and Historic Conditions Historically, the Sierra Nevada red fox was continuously distributed at high elevations in the Sierra Nevada from Tulare County northward to Sierra County. This species also occurred in the vicinity of Mt. Shasta and Lassen Peak westward to the Trinity Mountains of Trinity County (Grinnell et al. 1937). Although the Sierra Nevada red fox seems to range from 4,000 to 12,000 feet in elevation, they are seldom sighted below 5,000 feet, and most often above 7,000 feet. The Sierra Nevada red fox historically occurred at low densities, averaging perhaps one per square mile, and few sightings were ever recorded (Grinnell et al. 1937). Low prey

FEIS Volume 3, Chapter 3, part 4.4, page 36 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 availability and competition for a limited prey resource might be important factors that limit population densities (Martin 1989).

Status and Trend The current distribution and population status of the Sierra Nevada red fox is uncertain (California Department of Fish and Game 1990). California is home to both the indigenous Sierra Nevada red fox and an introduced population of non-native red foxes that currently occurs in lowland areas in the state (Burkett and Lewis 1992, Lewis et al. 1993). The relatively low number of recent Sierra Nevada red fox sightings suggests a small, possibly declining population (California Department of Fish and Game 1990). In addition, there is no way to visually distinguish the native and non-native foxes (Lewis et al. 1993).

Sierra Nevada red foxes have been seen most recently around Lassen National Park. The Sierra Nevada red fox survey in 1997 and 1998 reported one male and one female Sierra Nevada red fox in the Lassen National Forest. The Sierra Nevada red fox may be on the extreme edge of its range, existing in marginal habitat (Martin 1989). However, the species is potentially distributed across all of the Sierra Nevada national forests. National forest lands account for about 70 percent of the species’ range, with other public lands accounting for 10 percent. Private lands, such as industrial timber lands, account for 20 percent.

Although the Sierra Nevada red fox was widely distributed in the Sierra Nevada, it is unlikely it was ever common. There is also some question as to whether the Sierra Nevada red fox is rare or just rarely seen. It is likely, however, that the Sierra Nevada red fox population declined as a result of trapping, grazing, poisoning, and human presence in the early part of the twentieth century (Aubry 1997, Grinnell et al. 1937). More recently, fire exclusion has allowed conifers to encroach in meadow and riparian habitats, and livestock grazing has reduced the quality of meadows and riparian areas.

Habitat Risk Factors To conserve the Sierra Nevada red fox will require the retention of sufficient habitat and habitat connectivity throughout its range. The Sierra Nevada red fox requires a composite of habitat types including open forest, meadows and fell fields. At present, there is no reliable mapping effort that can describe with reasonable accuracy the amount or location of meadows and meadow complexes and their proximity to forest conditions suitable for red fox. Consequently, there is no means to insure that habitat suitable for Sierra Nevada red fox is not degraded.

The Sierra Nevada red fox forages in open areas, including meadows, and may use log or rock structures adjacent to meadows as denning habitat. Conifer encroachment in meadows and riparian areas has effectively reduced the availability of these habitats for meadow-associated species. Intensive sheep grazing in alpine meadows is considered to be the highest risk to the Sierra Nevada red fox (Grinnell et al. 1937). Grazing reduces the quality of meadow and riparian habitat, reduces prey habitat and abundance, and increases the level of human presence associated with livestock activities. Low prey availability leads to large territory size, low-density distribution, low reproductive success, and low survival rates.

Non-Habitat Risk Factors Grinnell et al. (1937) described the Sierra Nevada red fox is sensitive to human presence. Like the wolverine, it is likely that Sierra Nevada red foxes are negatively affected by the current rise in

FEIS Volume 3, Chapter 3, part 4.4, page 37 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 popularity of winter recreation, especially the use of newer, more powerful snowmobiles. Non- motorized recreation may affect Sierra Nevada red foxes as well. Dispersed recreation, such as hiker and livestock trail use, pack stations, and alpine campgrounds, can increase human presence in remote high country areas favored by this species and may concentrate use in meadow areas. Concentrated recreation, such as at ski resorts and snow parks, increases road density, traffic, and human access to high elevation habitats.

Based upon historic descriptions of habitat and behavior, any actions taken to minimize new and open roads, to limit human encroachment into the higher elevations, and to improve conditions of high elevation meadows will likely benefit this fox subspecies.

Non-Forest Service Risk Factors Any similar activities occurring on non-Forest Service lands would present the same risks as described above. Recreational activities in the high-country of National Parks can be viewed as a threat to the Sierra Nevada red fox. The southern Sierra Nevada parks, in particular, receive more recreational use than any other region in the world (Duane 1996). Although the parks receive a high level of backcountry use during summer, they do not experience as high a level of snowmobile use as the National Forests. The national parks also provide a diversity of high quality habitats from old forests to open canopy stands interspersed with meadows. A number of recorded Sierra Nevada red fox sightings are concentrated near Yosemite and Sequoia-Kings Canyon National Parks (Schempf and White 1977).

Where rural housing developments and areas of concentrated recreation occur at high elevations, they are a threat to Sierra Nevada red foxes. Housing, ski resorts, and snow parks can lead to increased road density, increased traffic on existing roads, and increased human presence in surrounding areas. Road construction and increased human settlement in the Sierra Nevada provide access to areas previously unavailable and may facilitate the dispersal of the non-native red fox into Sierra Nevada red fox habitats. The introduced red fox is highly adaptable and capable of dispersing long distances (Zeiner 1990, Lewis et al. 1993). Conversion of late seral stage forests to early seral favors coyotes and the non-native, introduced red fox. These species are potential competitors with Sierra Nevada red fox for prey resources. Competition may be occurring with the introduced red fox for prey, den sites, and habitat (Lewis et al. 1993). Contact with the non-native fox may also result in interbreeding and disease transmission or increased mortality from rabies outbreaks (Lewis et al. 1993).

II. Environmental Consequences A. Measures or Factors Used to Assess Environmental Consequences As described in the preceding “affected environment” section, a variety of factors influence the Sierra Nevada red fox population and its habitat. These factors are listed here, along with measures that were used to assess each alternative’s effects on the Sierra Nevada red fox.

1. Retention of dense forest canopy Measure: canopy closure

2. Presence of meadows and riparian habitat in proximity to conifer forests Measure: protection of meadows and riparian habitat.

FEIS Volume 3, Chapter 3, part 4.4, page 38 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 3. Occurrence of Sierra Nevada red fox Measure: survey requirements

4. Human Presence Measure: recreation use Measure: roads

5. Management of reproductive sites and protected areas. Measure: protected areas for Sierra Nevada red foxes

B. Assumptions and Limitations: Dense canopy cover is not believed to be an important habitat element for the Sierra Nevada red fox. The provisions in the alternatives for protecting forest canopy for the marten and wolverine are assumed to be adequate for the Sierra Nevada red fox as well. The Sierra Nevada red fox would benefit from landscapes having a variety of seral stages and habitat types.

Sierra Nevada red fox utilize open forest, meadows and riparian habitats in close proximity to conifer forest. Elements of importance in these areas are availability of prey species and availability of cover. Grazing and fire suppression can be regarded as the most significant threats to these elements. Grazing reduces the amount of shrub and herbaceous cover available for prey species such as voles, squirrels, woodrats, and lagomorphs. Livestock degrade meadow and riparian areas and cause soil compaction that can also effect prey species. The history of fire suppression in the Sierra Nevada has allowed tree cover to encroach on meadow and riparian areas reducing herbaceous cover for prey and effectively reducing meadow size. Meadows and riparian areas in a matrix of conifer forest would be preferred. There are trade-offs between allowing fire into riparian areas to reduce fuel levels and retaining snags and logs in these areas. Similarly, there are trade-offs between using prescribed fire to limit forest encroachment in meadows and retaining snags and logs in these areas. Meadow-related environmental consequences for the Sierra Nevada red fox are similar to those described in the previous section for the marten.

Unlike martens, Sierra Nevada red foxes are believed to be sensitive to human presence and disturbance (Grinnell et al. 1937). Human disturbance can have both direct and indirect effects on Sierra Nevada red fox populations. Human activities can cause direct mortality through vehicle- related mortality. Most of the potential habitat for the Sierra Nevada red fox occurs in the highest elevation forests and the alpine regions where current road densities are lower than in lower elevation forests. However, in the northern Sierra Nevada roads exist at high elevations. Human activities also may have indirect effects on Sierra Nevada red fox populations and distribution by altering behavior, reducing reproductive success and decreasing individual survival. Development of winter recreation areas such as ski resorts and snow parks may require new road construction and increase vehicular traffic levels and human presence in high country areas. Increased human presence at resorts and vacation housing increases the potential for negative human-wildlife interactions and direct human- caused mortality of Sierra Nevada red fox.

Little is known about the current distribution and abundance of the Sierra Nevada red fox population in the Sierra Nevada. To assess the risk of various management strategies to Sierra Nevada red fox persistence, scientists and managers must learn more about the distribution of this species and its habitats.

FEIS Volume 3, Chapter 3, part 4.4, page 39 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

C. Effects of Alternatives on the Sierra Nevada Red Fox Forest Canopy All alternatives would retain dense canopy forests. Excessive amounts of dense canopy forest could be detrimental to the Sierra Nevada red fox. Alternative 4 would create the greatest number of canopy openings through strategically placed area treatments (SPLATs) and defensible fuels profile zones (DFPZs). However, these strategies would not be targeted at the high elevation habitat used by the Sierra Nevada red fox. Alternative 7 would provide a diversity of canopy conditions that might be beneficial for Sierra Nevada red fox.

Meadows and Riparian Areas Alternatives 2, 5, 6, 7, 8 and Modified 8 appear to provide the greatest benefit to the Sierra Nevada red fox in terms of maintaining and enhancing meadow and riparian habitats. While Alternative 5 would not incorporate important bird areas (IBAs), it would provide for greater stubble heights and contains many of the provisions for reducing grazing impacts found in the other four alternatives. Although Alternative 7 would not establish IBAs, it shares many attributes with Alternatives 2, 5, 6, and 8. In addition, Alternative 7 emphasizes mechanical treatments over prescribed fire, resulting in reduced risk to snags and coarse woody debris used by the prey of the Sierra Nevada red fox. By establishing IBAs, Alternatives 2, 6 and 8 could indirectly benefit the Sierra Nevada red fox by providing habitat for meadow-dependent prey species, assuming an overlap between IBAs and Sierra Nevada red fox locations. Alternative 8 would maintain or enhance structural complexity in forest stands adjacent to wet meadows and riparian areas which would enhance Sierra Nevada red fox foraging opportunities. Modified 8 should also provide foraging opportunities for Sierra Nevada red fox by enhancing prey habitat in meadow and riparian areas by restricting treatments and use of fire in these areas.

Recreation Alternatives 3 and 5 would restrict human disturbance in unroaded areas. Alternative 5 would prohibit off highway vehicle (OHV) use in unroaded areas. Alternative 3 directs managers to evaluate existing OHV routes in unroaded areas to determine their compatibility with other resource objectives. Alternatives 2, 3, 4, 6, 8, and Modified 8 could restrict activities within a five-mile area around verified sightings or detections of Sierra Nevada red fox. These alternatives therefore pose a lower risk to Sierra Nevada red foxes and their habitat. Alternative 5 would reduce recreational impacts to Sierra Nevada red foxes and increase the amount of habitat for the species by generally prohibiting new recreational developments within 5 miles of Sierra Nevada red fox detections and evaluating existing recreation sites in suitable habitat. Alternatives 1 and 7 would not restrict recreation to address impacts to the Sierra Nevada red fox. These alternatives provide the lowest benefits in terms of addressing recreational impacts to this species.

Roads Alternative 5 gives high priority to reducing road densities across landscapes. This alternative would prohibit new road construction in unroaded areas larger than 5,000 acres. Alternative 5 would also manage ecologically significant unroaded areas between 1,000 and 5,000 acres as reserves. Alternative 5 would provide the greatest benefits to the Sierra Nevada red fox in terms of reducing impacts from roads and related human disturbances. Alternative 3 would prohibit new permanent road construction in unroaded areas larger than 5,000 acres. In other land allocations, new road construction would be offset by reducing existing roads, with the

FEIS Volume 3, Chapter 3, part 4.4, page 40 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 goal of achieving a net reduction in road density at the watershed scale. Alternative 4 has the greatest potential for new road construction. Alternative 4 does not provide for concurrent reduction in existing roads in other areas and therefore poses the most risk from roads to Sierra Nevada red fox persistence. Modified 8 provides for seasonal and multi-year closures, and decommissioning of existing roads and would consider disturbance to wildlife in planning new construction of roads.

Survey Requirements Alternative 5 would provide surveys for Sierra Nevada red fox, but only where Forest Service projects occur. Alternatives 1, 2, 3, 4, 6, 7, 8 and Modified 8 would have the least benefits and pose the greatest risk in regard to survey requirements. None of these alternatives impose survey requirements to determine the current distribution of the Sierra Nevada red fox.

Protected Areas All of the action alternatives except for Modified 8 would establish a protected area for each located maternal or natal Sierra Nevada red fox den site. Alternatives 2, 3, 4, 6, 7, and 8 would establish 250-acre buffers at den site locations while Alternative 5 would establish a 700-acre protected activity center (PAC) at each den site. Alternative 5 would have slightly greater benefits to Sierra Nevada red fox reproductive sites, as it would designate a slightly larger protected area. In addition to providing den site protection, Alternatives 2, 3, 4, 6, 8, and Modified 8 require the evaluation of activities within 5 miles of a verified sighting or detection of a Sierra Nevada red fox and limitation of any activities determined to potentially have an adverse impact.

Summary of Consequences for the Sierra Nevada Red Fox Although the current distribution of the Sierra Nevada red fox in California is uncertain, the species’ range appears to have contracted from the continuous distribution described by Grinnell in the 1930s. Recent sightings of foxes purported to be the Sierra Nevada subspecies have been concentrated in the area around Lassen Peak. Although the habitat requirements of this species are not well understood, its apparent need for montane meadows surrounded by open forests in areas removed from human contact drive this assessment of the proposed alternatives. Alternatives that protect meadow systems, provide mature open forest conditions, and reduce impacts associated with human presence present the greatest benefit to the Sierra Nevada red fox.

The following categories were used to summarize each alternative’s overall risk to the Sierra Nevada red fox: (1) vegetation structure and composition, (2) recreation and roads, and (3) survey requirements and site protection. Vegetation composition and structure reflects how well an alternative maintains or enhances the vegetation elements important to the Sierra Nevada red fox. The “recreation and roads” category considers an alternative’s direction for managing human presence, recreational development, and roads. The “survey requirements and site protection” category considers an alternative’s survey requirements and standards for protecting den sites and detection locations. Habitat for prey species is an additional factor that may affect Sierra Nevada red fox, but it is unclear to what extent the alternatives would produce differential effects on foxes given the diversity of prey items utilized. Uncertainty in predictions of prey habitat within the range of the Sierra Nevada red fox further confuses the issue.

Of all the alternatives, Alternative 5 would have the greatest benefit to Sierra Nevada red fox persistence. Although Alternative 5 does not provide protection for future verified sightings of Sierra

FEIS Volume 3, Chapter 3, part 4.4, page 41 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Nevada red fox, this alternative does provide the highest level of meadow protection, emphasizes reducing road densities across landscapes, and encourages new Sierra Nevada red fox surveys. Alternatives 5 and 3 propose restrictions on recreational activities in unroaded areas. Alternatives 6, 8 and Modified 8 would not require surveys and would have fewer restrictions on recreation and roads. Alternatives 4 and 7 would provide more of the open forest habitat preferred by the Sierra Nevada red fox than Alternative 5; however, these alternatives place fewer restrictions on recreation and would provide moderate reductions in roads. Although Alternative 2 would prohibit OHV and over snow vehicle (OSV) use in den site buffers, Alternative 2 would not require new surveys for the Sierra Nevada red fox.

Environment and Population Outcomes for the Sierra Nevada red fox Environmental and population outcomes were derived by professional opinion to estimate environmental and population conditions that would exist in 50 years for the Sierra Nevada red fox under each alternative. Assigning these outcomes, while inherently subjective, is based on a reasoned thought process and the best available information. The environmental outcome addresses the capability of the environment on Forest Service lands to support population abundance and distribution. The population outcome addresses environmental conditions on all lands within the bioregion and other risk factors that may affect population abundance and distribution.

Table 4.4.1.2f. Outcome ratings for the Sierra Nevada red fox. Environment outcomes evaluate the estimated environmental conditions on Forest Service lands after 50 years under each alternative. Population outcomes evaluate the estimated population conditions based on the environment outcome and other risk factors.

Outcome Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Environment C+ C C+ C+ C B- C+ C C+ C+ Population C C C+ C+ C B- C+ C C+ C+

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Rationale for Outcome Ratings The Sierra Nevada red fox is a habitat generalist that uses a diversity of forest types, principally red fir and lodgepole pine in subalpine and alpine areas, as well as habitat mosaics, including alpine meadows and talus slopes. This species generally occurs at elevations ranging between 7,000 and 12,000 feet, in regions that are remote from human activities and disturbances. The current population status of the Sierra Nevada red fox is unknown, as there are few recent confirmed sightings and this species is indistinguishable in the field from the introduced non- native red fox.

FEIS Volume 3, Chapter 3, part 4.4, page 42 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 High elevation suitable environments for the Sierra Nevada red fox most likely exist across the species’ range in low abundance in a patchy distribution that allows interaction among most subpopulations (Outcome C+). It is uncertain whether this species was habitat-limited, as previous trapping and actions to control this species as a predator of domestic livestock have likely reduced the population. Because the principle habitats for this species occur at higher elevations, forest management activities would have little effect on habitats used by this species. The extent that human disturbance (such as recreation and roads) in high altitude regions adversely affects the Sierra Nevada red fox was unknown, particularly given the adaptability and tolerance of human activities generally displayed by the life history attributes of other fox species. The current distribution of Sierra Nevada red fox was judged to be patchy with some potential for interactions across the range (Outcome C).

For the purposes of this assessment, the assumptions and criteria described in the analysis of environmental consequences above generally apply. Given the uncertainties inherent in the model projections and the lack of spatial location of stand conditions in the future, the effects of implementation of the Standards and Guidelines on distribution and quality of habitat were also considered. Because of the uncertainty regarding the location and implementation of vegetation treatments, alternatives that provided protection at several spatial scales ranked higher in these outcomes.

Environmental Outcome There was very little significant variability among the alternatives in likely outcomes in 50 years for the abundance and distribution of suitable environments for the Sierra Nevada red fox. Whether populations of this species are capable of responding to the apparently suitable environments is unknown. Alternative 5 was judged to provide a minor improvement for the species by maintaining the roadless character of existing unroaded areas larger than 5,000 acres, generally precluding new developments within 5 miles of Sierra Nevada red fox detections, and retaining the greatest stubble in meadows to maintain microtine prey populations. Alternatives 1 and 4 would likely result in a minor decline in habitat conditions with an increased likelihood that subpopulations could become isolated since these alternatives do not minimize human disturbance and provide less consideration for improving meadow habitats.

Population Outcome Three additional assumptions were applied in consideration of the population outcomes: 1) the amount of treatments (e.g. harvest) on private lands in the Sierra Nevada is stable or increasing; 2) human populations are increasing and will lead to increased urbanization; and 3) the popularity of recreational activities is increasing. Since the vast majority of Sierra Nevada red fox habitat occurs on Forest Service lands, the majority of predicted effects on Sierra Nevada red fox population distribution and abundance would also be expected to occur there. These criteria do have some impact, however, in the central and northern Sierra Nevada in areas of mixed ownership where there could be significant reduction in landscape quality and connectivity for Sierra Nevada red fox. Of principal concern in judging population outcomes were: 1) environmental outcome for the alternative, 2) provisions for maintaining connectivity of Sierra Nevada red fox populations.

Alternatives 2, 3 and 5 would manage areas protected from motorized recreation and would provide some buffer from increased recreation pressures. Alternatives 2, 8 and Modified 8

FEIS Volume 3, Chapter 3, part 4.4, page 43 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 would manage substantial acreage for old forest conditions that might provide connectivity for animal movement, particularly when paired with the high elevation wilderness areas. Alternatives 2, 3, 4, 6, 8, and Modified 8 would give consideration to verified sightings and detections of red fox. This standard and guideline focuses on reduction of human disturbance and does not specify any habitat protections. In the absence of a survey requirement in these alternatives, it is unlikely that this standard will be implemented. Only Alternative 5 requires surveys for Sierra Nevada red fox. Consequently, Alternatives 5, 2, 3, 6, 8 and Modified 8 have potential to result in a slight improvement in the population distribution and abundance of red foxes.

FEIS Volume 3, Chapter 3, part 4.4, page 44 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 4.4.1.4. WOLVERINE I. Affected Environment A. Background Habitat Wolverines use a variety of habitats across their range in North America (Banci 1994). This appears to be due to their large home ranges, which include a great diversity of forest and non- forest types. In California, the wolverine once occurred throughout the Sierra Nevada, Cascades, Klamath, and northern Coast ranges in forests in alpine, boreal forest and mixed forest vegetation types (Grinnell et al. 1937, Schempf and White 1977). Wolverines predominately use coniferous forest types (Copeland 1996, Hornocker and Hash 1981), but their significant use of non-forest alpine habitats distinguishes them from the fisher and marten (Banci 1994, Copeland 1996).

Although wolverines use diverse vegetation types, they appear to be particularly selective about two habitat elements. The first element is their natal dens. In Idaho, natal dens occur in high-elevation rocky substrates and are often associated with wood or boulders in cirque basins on north and east slopes where snow persists into the spring (Copeland 1996, Magoun and Copeland 1998). In British Columbia, natal dens also occur in large woody debris piles, often associated with the base of avalanche chutes. Both natal dens described in California were under rock ‘shelves’ at elevations above 10,000 feet (Grinnell et al. 1937).

The second habitat element pertains to human disturbance: wolverines appear to select areas that are free from significant human disturbance, especially during the denning period from late winter through early spring. An empirical wolverine habitat model developed for the Rocky Mountains found that wolverine occurrence was strongly associated with low human population density and low road density (Carroll et al. in press). It is believed that wolverines moved to higher elevations in the summer in Montana to avoid human recreational activity (Hornocker and Hash 1981). Most researchers agree that adult females - particularly during the denning period – are highly sensitive to disturbance (Banci 1994, Copeland 1996, Hornocker and Hash 1981, Magoun and Copeland 1998). If the availability of natal denning areas is as limited as some suggest, then disturbance of wolverines in these areas can affect the population (Copeland and Kucera 1997).

Current and Historic Condition Wolverines are never common anywhere they occur (Banci 1994). Wolverines were part of the early fur harvest in California and were distributed at low densities throughout most of the Sierra Nevada.(Grinnell et al. 1937). In the early 1900s, their populations declined largely due to trapping (Dixon 1925, Seton 1929). By 1933, no more than 30 animals were thought to occur in California (Grinnell et al. 1937). There have been no regular surveys for wolverines since trapping was prohibited in the mid-1900s, and wolverine surveys have not yielded positive results (Kucera and Barrett 1993). Each year, however, there are several sightings in California that appear to be reliable descriptions of wolverines in likely locations. The current wolverine range in California is unknown, largely because it has been over 50 years since verifiable evidence (in other words track, photographs, or carcasses) has been collected in California. The wolverine has been placed in the smallest population size class of Sierra Nevada species, with the most significantly declining trend and the most significantly

FEIS Volume 3, Chapter 3, part 4.4, page 45 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 contracted range. The wolverine is in the highest species vulnerability class; it is viewed as having undergone a major decline since the period of European settlement (Barrett et al. 1994).

Wolverine Diet Wolverines prey largely on ungulate carrion, but also favor marmots and other intermediate- sized vertebrate prey (Grinnell et al. 1937, Banci 1994, Copeland and Kucera 1997). Wolverine can make their own kills but often occur in communities with large carnivores whose kills they scavenge (Banci 1994, Copeland and Kucera 1997).

B. Habitat Risks Factors Recent evidence suggests that wolverines are more common in Oregon and Washington than in California (Banci 1994). Thus, providing adequate quantities of connected forest habitat within the Sierra Nevada and between the Sierra Nevada and the Klamath Mountains and Cascades is critical to the recovery and viability of wolverines in California. Although wolverines readily use non-forest habitat above timberline, a significant portion of their life history needs are met in forest (Banci 1994), and large forest openings are believed to negatively affect their habitat (Banci 1994, Hornocker and Hash 1981). Observations suggest, “wolverines appeared reluctant to cross openings of any size.” (Hornocker and Hash 1981). However, there are indications that wolverines do not avoid all forest openings (Copeland 1996).

Because forest habitat provides important cover for wolverines, the connectivity and distribution of dense forest conditions is assumed to be important for wolverines, especially in the northern Sierra Nevada. This region could be an important linkage between habitat in California and habitat to the north; however, it lacks the alpine regions that can buffer wolverines from contact with human activity. Instead, wolverines must use forested habitat to move north and south. These forests are managed for timber and other amenities, and there are numerous small communities in this region; both factors minimize the separation from humans that wolverines apparently prefer (Hornocker and Hash 1981).

Wolverines that occur in forested areas use dense forest cover for travel and resting, especially in the winter (Copeland 1996, Hornocker and Hash 1981). Individual management projects and natural disturbances (fire) cumulatively affect the distribution of forest patches and the connectivity of forest cover. An empirical spatial habitat model, which addresses effects on wolverine habitat distribution, has been developed from sightings data in the Rocky Mountains and applied to California (C. Carroll unpubl. data). This model cannot be tested in California, however, due to the lack of contemporary wolverine locations; however, the areas with the highest predicted habitat values for wolverine agree generally with the range of wolverine described in the early 1900s by Grinnell et al. (1937). Without a model to address spatial relations of wolverine habitat, it is difficult to assess how the spatial arrangement of management activities could affect wolverine habitat, especially in the northern Sierra Nevada.

Wolverines depend less on large woody habitat structures than fishers or martens (Ruggiero et al. 1994). Woody microhabitat features are rarely mentioned as limiting wolverine populations. The possible exception in the Sierra Nevada may be the northern region where the highest elevations are forested and accessible to people throughout the year. This would mean that wolverines moving through this area may require large woody structures for refuges. With this exception, few of the Forest Service vegetation management activities will have serious effects on microhabitat elements of wolverine habitat. In addition, fuels and mechanical treatments are usually conducted at elevations

FEIS Volume 3, Chapter 3, part 4.4, page 46 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 below optimal wolverine habitat. For these two reasons, management actions typically have minimal impacts on the amount or quality of microhabitat features used by wolverines.

Non-Habitat Risk Factors Wolverine biologists increasingly believe that human disturbances, especially snowmobile operation during late winter or early spring, can affect wolverine use of high country areas (Copeland and Kucera 1997). New, more powerful snowmobiles have had an increasing effect on wolverine habitat elsewhere in the United States, but the potential effect of alpine recreation on wolverines or their habitat has not been evaluated in California. Wolverine biologists suggest that hiker use of backcountry areas may also affect the behavior and distribution of wolverines (Hornocker and Hash 1981). The increasing popularity of high country recreation (vehicular and non-vehicular) and the burgeoning population of California (Duane 1996, Sierra Science Review 1998) are well documented.

Non Forest Service Risk Factors Recreational activities in the high country of national parks are outside scope of this Forest Plan Amendment; however, these activities may affect wolverines. The southern Sierra Nevada national parks, in particular, receive more recreational use than any other region in the world (Duane 1996). The intensity of backcountry use during the summer may be detrimental to the recovery of wolverines. However, the parks do not receive the level of winter snowmobile use that occurs on National Forests so the threat to denning females in the parks would be minor. The region from Yosemite National Park south to Sequoia-Kings Canyon National Parks has been identified as a area likely to harbor wolverines if, in fact, they persist in California (Knudson 1994).

Because wolverines appear to avoid humans, rural communities or small urban areas located in the highest elevations in the Sierra Nevada can be viewed as a threat to wolverine recovery and persistence. In addition, areas of concentrated recreational activities, such as ski resorts and snow parks, can lead to increased road density, increased vehicle traffic, and increased human presence in surrounding areas. Wolverine recovery and persistence could also be adversely affected to the extent that human developments displace ungulate populations, which provide important food for wolverines (Banci 1994, Copeland 1996). Deer carrion provides an important component of the wolverine diet (Copeland and Kucera 1997). Excessive hunter harvest of deer may adversely affect wolverine recovery and persistence (Copeland and Kucera 1997). Thus, the management of deer populations and harvest by California Fish and Game can be viewed as a non-FS threat to wolverine recovery. Unrecovered hunter-killed carcasses, however, can be viewed as a subsidy to wolverines. The state regulation of mountain lions may also affect wolverines, negatively because lions can prey on wolverines and positively because wolverines scavenge the ungulates killed by lions.

Wolverine habitat has critical structural and non-structural elements. Of the latter, biologists familiar with wolverines suggest that freedom from significant human disturbance is one of the most important. The increase in human population in California and the increase in wilderness use, recreational activities (vehicular and non-vehicular), and non-recreational activities (wilderness stock permits) appear to be the greatest threat to the recovery and persistence of wolverines in the Sierra Nevada. Wolverine populations probably declined in the Sierra Nevada due to the combination of trapping, poisoning, and human activities. The sensitivity of wolverines to human disturbance, particularly during the late winter and early spring denning

FEIS Volume 3, Chapter 3, part 4.4, page 47 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 period, has only recently come to light and additional evidence is accumulating (Copeland 1996, Copeland and Kucera 1997).

II. Environmental Consequences Currently, there is no verifiable evidence that wolverines still reside in the Sierra Nevada. However sightings continue to be reported.

A. Measures Used to Assess Environmental Consequences As described in the preceding “affected environment” section, a variety of factors influence the wolverine and its habitat. These factors are listed here, along with measures that were used to assess each alternative’s effects on the wolverine.

1. Retention of dense forest canopy Measure: canopy closure

2. Management of human presence and high country activities Measure: recreation Measure: roads

3. Occurrence of wolverines Measure: survey requirements

4. Management of reproductive sites and protected areas. Measure: protected areas in the vicinity of wolverine detections

5. Quality, quantity, and distribution of habitat of prey species Measure: forest density, size, and canopy

B. Assumptions and Limitations Dense forest cover supplies the overhead canopy and large woody structure that may provide necessary rest sites for wolverines as they forage and travel. Most wolverine habitat exists at high elevations where either no forest occurs or the forest types that do occur are not proposed for significant fuels treatments. The exception may be the northern Sierra Nevada where forests occur throughout most of the high elevation regions that wolverines would be expected to use. It is assumed that management that does not include a plan for retaining and restoring blocks of continuous forest cover throughout the northern Sierra Nevada and southern Cascades will affect this area’s value as a conduit for wolverine movements between the Sierra Nevada and habitat to the north and west. It is also assumed that, in regions where dense forest canopy is most important, the effects analysis of dense forest canopy for the marten (see Part 4.4.1.1.2.) applies to this component of wolverine habitat as well.

Woody microhabitat features are rarely mentioned as limiting wolverine populations. The possible exception in the Sierra may be in the northern region, where the highest elevations are forested and accessible to people throughout the year. This would mean that wolverines moving through this area might require large woody structures for refuges. With this exception, few management activities are expected to have serious effects on microhabitat elements of wolverine habitat. Given the reduced importance of large woody structures to wolverines compared to fishers and martens, and the fact that

FEIS Volume 3, Chapter 3, part 4.4, page 48 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 all of the alternatives seek to increase the density of large trees, the environmental consequences for this habitat feature are not presented.

Recreational activities can impact wildlife species, but this relationship is poorly understood outside of a few species (Knight and Gutzwiller 1995). Non-motorized recreation is often assumed to have lower impacts than motorized recreation, but this may not be the case. It is unclear how results from studies on other species might apply to forest carnivores; however, biologists who have studied wolverines are concerned about the effects of backcountry recreation on the species.

Most wolverine habitat in the Sierra Nevada occurs in the highest elevation forests and the alpine regions where existing road densities are generally lower than in low elevation areas. However, roads are an important issue for wolverine habitat in the northern Sierra Nevada, particularly since this region lacks the area of treeless and roadless alpine areas of the central and southern portions of the range. In the northern Sierra Nevada, forests and forest management occur in the highest elevation zones. Roads provide access for vehicular and non-vehicular recreational purposes.

Mountain lions may be a positive influence on the potential for wolverine recovery by making available the uneaten remains of deer kills. Livestock grazing in the high country may benefit wolverines through the occasional availability of cow or sheep carcasses.

It is assumed that wolverines are either at their historically lowest population levels or they are extirpated in the Sierra Nevada. With sufficient effort to survey for wolverines, a reasonable level of confidence in the conclusion that wolverines do not occur in the Sierra Nevada can be reached.

C. Effects of Alternatives on the Wolverine Dense Forest Canopy Alternatives 5 and 8 would retain the highest amounts of dense forest conditions. Alternatives 6 and 8 would pose a reduced risk to westside forests and an increased risk to eastside forests. Of these two alternatives, Alternative 8 poses a lower level of risk because it would retain and recruit large trees through its stand structural standards for retaining specific levels of canopy cover, basal, area, and large trees in suitable California spotted owl habitat. These standards would be particularly important in maintaining wolverine habitat in higher elevation forested areas in the northern Sierra Nevada. Alternative 4 poses a greater risk to dense canopy forest through the establishment and maintenance of defensible fuels profile zones (DFPZs) and strategically placed area treatments (SPLATs). However, high elevation true fir forests (which would be of greatest concern for the wolverine) would presumably be lower priority for these treatments.

Recreation Alternatives 3 and 5 would restrict human disturbance in unroaded areas. Alternative 5 would prohibit off highway vehicle (OHV) use in unroaded areas. Alternative 3 directs managers to evaluate existing OHV routes in unroaded areas to determine their compatibility with other resource objectives. These alternatives would therefore pose a lower risk to wolverines and their habitat. Alternatives 2, 3, 4, 6, 8, and Modified 8 would evaluate activities within 5 miles of a verified sighting or detection of a wolverine. Alternative 5 would prohibit new recreation development within 5 miles of a wolverine detection.

FEIS Volume 3, Chapter 3, part 4.4, page 49 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Roads Alternative 5 gives high priority to reducing road densities across landscapes. This alternative would also prohibit new road construction in unroaded areas larger than 5,000 acres and manage ecologically significant unroaded areas between 1,000 and 5,000 acres as reserves. Alternative 5 would provide the greatest benefits to the wolverine in terms of reducing impacts from roads and related human disturbances. Alternative 3 would prohibit permanent road construction in unroaded areas larger than 5,000 acres. In other areas, new road construction would be offset by reducing existing roads with the goal of achieving a net reduction in road density at the watershed scale. Modified 8 provides for seasonal and multi-year closures, and decommissioning of existing roads and would consider disturbance to wildlife in planning new construction of roads. Alternative 4 has the greatest potential for new road construction; it does not provide for concurrent reduction in existing roads in other areas. Alternative 4 therefore poses the greatest road-related risk to wolverine recovery and persistence.

Survey Requirements Alternative 5 provides for wolverine surveys but only where projects are planned. Alternatives 1, 2, 3, 4, 6, 7, 8 and Modified 8 would not provide for wolverine surveys. Alternatives 1, 2, 3, 4, 6, 7, 8 and Modified 8 would have the least benefits and pose the greatest risk in regard to survey requirements. None of these alternatives impose survey requirements to determine the current distribution of the wolverine.

Protected Areas None of the alternatives would establish special emphasis areas for wolverines. Alternatives 2, 3, 4, 6, 8 and Modified 8, require the evaluation of activities within 5 miles of a verified sighting of a wolverine and the limitation of any activities determined to potentially have an adverse impact.

Wolverine Prey Species The CWHR habitat utility values for each alternative integrate the CWHR categories of forest type, density, and tree size class. Insofar as the habitat utility values reflect the habitat quality for wolverine prey (mule deer and yellow-bellied marmots in particular), they can forecast the changes in prey habitat quality over the next 50 years. For mule deer, projected changes in habitat utility values under all alternatives indicate a decline in deer habitat. Projected changes in habitat utility values range from a low of – 10.6 percent (Alternative 2) to a high of – 5.3 percent (Modified 8), with a mean of –8.3 percent change across all alternatives. For marmots, the habitat utility values also forecast declining habitat: the average 50-year projected change across alternatives is –9.4 percent, and changes in utility values range from a low of –15.6 percent (Alternative 3) to a high of 0.2 percent (Modified 8). Although deer habitat would decrease the least under Modified 8, the differences between alternatives are so modest as to have little value for comparing the effects of alternatives on wolverine foraging success. Marmot habitat would decrease the most under Alternatives 3 and 6 least under Modified 8. Modified 8 would appear to result in least decline in habitats for wolverine prey species.

Summary of Environmental Consequences to the Wolverine Management and protection for wolverines varies among the proposed alternatives, as do the projected consequences for this species. Wolverines occur at extremely low densities in the Sierra

FEIS Volume 3, Chapter 3, part 4.4, page 50 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Nevada, if they occur at all. Given this status, the alternatives were evaluated with two equal criteria in mind: (1) the retention and persistence of individual wolverines that still occur in the Sierra Nevada bioregion and (2) the recovery of the wolverine population throughout their historic range in the bioregion. Alternatives that favored the measures considered most important to the persistence and recovery of wolverines were identified as having the highest likelihood of increasing the wolverine population in the Sierra Nevada bioregion.

To summarize the overall risk to wolverines for each alternative, the measures evaluated in the preceding sections were grouped into three categories: (1) vegetation structure and composition, (2) recreation and roads, and (3) survey requirements and site protection. Vegetation composition and structure reflects how well the alternative maintains or enhances the vegetation elements important to wolverines. The recreation and roads category considers how human presence, wilderness use, recreational development, and roads under each alternative could impact wolverines. Under the survey requirements and site protection category, the alternatives are evaluated based on requirements for surveys and level of protection provided to den sites and detection locations. Although prey habitat value is important, the diversity of prey items and the uncertainty of associated with habitat value projections for these prey precluded including prey habitat quality in the summary.

Dense forest canopy was the single vegetation measure used to evaluate potential impacts on wolverines. Because wolverines occur in forest and alpine regions, and because the vegetation treatments expected under this Forest Plan Amendment are predominately at elevations below the range of wolverine habitat, changes in vegetation structure were weighed less heavily than the other two categories. Alternatives 1, 4, and 7 have the greatest potential to affect continuous forest cover, especially in the region of the northern Sierra where alpine habitat is lacking. However, this variation did not significantly influence conclusions because none of the alternatives substantially affected the vegetation element of wolverine habitat. Based on CWHR projections, wolverine habitat suitability shows modest increases under all alternatives ranging from 5.4% under Alternative 2 to 9.1 percent under Modified 8.

Consequences to wolverines were influenced most by the two other categories: (1) recreation and roads and (2) survey requirements and site protection. Based on the combined categories, Alternative 5 appears to represent the greatest benefit to wolverine persistence and recovery. Alternative 5 and 3 would restrict recreational activities in unroaded areas. Alternative 5 would emphasize reducing road densities, would encourage new surveys, and prohibit new recreation development within 5 miles of a detection. Although roads are not an issue in the alpine regions in the southern Sierra, they are a concern in the higher elevations of the northern Sierra and southern Cascades. Alternative 3 would not have the same value as Alternative 5 due to the absence of survey requirements. Alternatives 2, 3, 4, 6, 8, and Modified 8 do not require surveys but would limit activites around verified wolverine sightings. Alternative 2 would have more risks related to the effects of roads and surveys than Alternative 5, but would generally provide greater benefits to wolverines than Alternatives 1, 4, and 7. Alternatives 1, 4, and 7 would not encourage surveys, and they would have greater potential for new road development than the other alternatives.

Alternative 5 would have the greatest chance of increasing the abundance and distribution of wolverines.

FEIS Volume 3, Chapter 3, part 4.4, page 51 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Environment and Population Outcomes for the Wolverine Environmental and population outcomes were derived by professional opinion to estimate environmental and population conditions that would exist in 50 years for the wolverine under each alternative. Assigning these outcomes, while inherently subjective, is based on a reasoned thought process and the best available information. The environmental outcome addresses the capability of the environment on Forest Service lands to support population abundance and distribution. The population outcome addresses environmental conditions on all lands within the bioregion and other risk factors that may affect population abundance and distribution.

Table 4.4.1.1d. Outcome ratings for the wolverine. Environment outcomes evaluate the estimated environmental conditions on Forest Service lands after 50 years under each alternative. Population outcomes evaluate the estimated population conditions based on the environment outcome and other risk factors.

Outcome Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Environment C- D D+ C- D C- D+ D D+ D+ Population E E E E E E E E E E

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Rationale for Outcome Ratings There has been no documented evidence of wolverine presence in the Sierra Nevada for the last 50 years, although sightings are made each year and many of these sightings are considered reliable. The decline of the wolverine population from historical levels has been attributed to trapping, shooting wolverines as predators, “predicide,” and the presence of an increasing human population, which may have driven the remaining wolverine population into higher elevation, less- disturbed sites. The capability of the remaining wolverine population to recover is unknown.

Approximately 80% of wolverine habitat occurs on Forest Service lands. The current environmental conditions for wolverines on national forests are primarily patchy with isolation among some subpopulations (Outcome C-). Human disturbance, particularly around den sites in cirque habitat, appears to have a strong negative effect on wolverines. Conditions could be improved by reducing human disturbance to low levels and providing high levels of canopy cover.

For the purposes of this assessment, the assumptions and criteria described in the analysis of environmental consequences above generally apply. Given the uncertainties inherent in the model projections and the lack of spatial location of stand conditions in the future, the effects of

FEIS Volume 3, Chapter 3, part 4.4, page 52 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 implementation of the Standards and Guidelines on distribution and quality of habitat were also considered.

Alternatives 3 and 5 were judged to best provide for wolverine habitat, resulting in a small potential for improvement. Alternatives 3 and 5 would maintain the roadless character of currently unroaded areas larger than 5,000 acres and Alternative 5 would also protect selected ecologically significant unroaded areas between 1,000 and 5,000 acres. Alternative 5 would prohibit significant new development within 5 miles of wolverine detections, and would require surveys for wolverines and other threatened, endangered, and sensitive species prior to implementation of projects. Alternative 3 would limit detrimental activities within 5 miles of a verified wolverine sighting but does not require surveys. Alternatives 2, 6, and 8 were judged to result in slight declines in conditions for wolverines. Alternative 3 maintains fewer acres of unroaded areas than Alternative 5; provisions for integrated biodiversity reserves in Alternative 2 would provide some buffering of human disturbance of wolverines. Conditions were projected to worsen under Alternatives 1 and 4, which could degrade forested conditions in the northern Sierra Nevada for wolverine.

Population Outcome Three additional assumptions were applied in consideration of the population outcomes: 1) the amount of treatments (e.g. harvest) on private lands in the Sierra Nevada is stable or increasing; 2) human populations are increasing and will lead to increased urbanization; and 3) the popularity of recreational activities is increasing. Since the vast majority of wolverine habitat occurs on Forest Service lands, the majority of predicted effects on wolverine distribution and abundance would also be expected to occur there. These criteria do have some impact, however, in the central and northern Sierra Nevada, particularly in areas of mixed ownership where there could be significant reduction in landscape quality and connectivity. Of principal concern in judging population outcomes were: 1) environmental outcome for the alternative, 2) provisions for maintaining connectivity of habitat for potential wolverine movement.

None of the alternatives was judged to significantly improve conditions for wolverine distribution and abundance. Reserves under Alternative 2 may provide some buffering from the effects of human presence and recreation activities as may the roadless areas under Alternatives 3 and 5. It is unlikely that the standards and guidelines under any of the action alternatives are sufficient to result in any improvement in the distribution or abundance of this rare carnivore.

FEIS Volume 3, Chapter 3, part 4.4, page 53 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 4.4.1.5. PALLID BAT (Antrozous pallidus) Affected Environment Life History Maternity colonies form in early April and may contain 12 to 100 individuals (CWHR). Males may roost elsewhere or with the maternity colony (CWHR). An average of 2 young are born between April and July, predominantly in May and June (CWHR); twins most common between May and July (ECMB). Young are weaned in 7 weeks and females and juveniles forage together after weaning (CWHR).

Pallid bats take a variety of insects including beetles, othopterans, homopterans, moths, spiders, scorpions, solpugids, and Jerusalem crickets (large, hard-shelled prey) (CWHR) cicatids and flightless arthropods (ECBM). It has also been known to take lizards and perognathus (ECMB) and to feed predominantly on ground-dwelling arthropods, such as scorpions and sphinx moths in some studies (Brown and Berry 1991). It forages over open ground up to 8 feet, often takes prey on the ground, and is very maneuverable in flight and on the ground (CWHR, ECMB, Brown and Berry 1991). Water is needed, even though this bat has good urine concentration ability (Geluso 1978 - CWHR). Pallid bats prefer rocky outcrops, cliffs, and crevices with access to open habitats for foraging (CWHR). Foraging has been recorded up to 3 miles from the day roost site (CWHR) and pallid bats show a high degree of forage area fidelity (ECMB). Pallid bats typically do not feed over water though will feed in adjacent oak woodlands and have been recorded foraging over agricultural fields (ECMB).

Pallid bats are not known to truly migrate (ECBM), though if temperatures are cold enough, they may migrate locally and elevationally to hibernation sites (CWHR). Activity is infrequent below 2 degrees Centigrade (CWHR). They are believed to hibernate singly or in small numbers, and are known to be active in winter in the southern portions of its range (ECMB). Some maternity colonies remain active all year (ECMB).

Habitat Relationships Pallid bats occur in a variety of habitats ranging from rocky arid deserts to grasslands to higher elevation coniferous forests (ECMB). It is most common in open dry habitats with rocky areas for roosting (CWHR). Pallid bats are most abundant below 6000 feet in elevation in the arid Sonoran life zones but have been recorded up to 10,000 feet in the Sierra Nevada (ECMB).

Roosts consist of caves, crevices, mines, and occasionally hollow trees and buildings (CWHR). Tree roosting has been documented in large conifer snags, redwood and sequoia hollows, and oak cavities (ECBM). It has also been found in stone piles and burlap sacks, in bridges, mud cliffs, rocks and rubble, such as riprap around culverts and talus slopes (ECMB). Roost sites must protect bats from high temperatures since bats move into deeper cover as temperatures rise (CWHR). Night roosts occur in more open areas, such as porches and buildings (CWHR). Night roosts may be in more open sites, such as porches and open buildings (CWHR). Few hibernation sites are known (CWHR).

It appears that group size effects metabolic economy and growth of young because individuals roosting alone had higher recorded weight loss (Trune and Slobodchikoff 1976, 1978 - CWHR). They appear to remain in stable groups and will roost with other species, typically Tadarida and Myotis spp. (ECMB). They show a high degree of roost fidelity (ECMB).

FEIS Volume 3, Chapter 3, part 4.4, page 54 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Status The pallid bat is a California State species of special concern and Forest Service Sensitive species.

Historical and Current Distribution The pallid bat ranges from southern British Columbia, California east to central Texas and Kansas, Mexico, and Baja California (ECMB). It occurs in a wide variety of habitats including grasslands, shrublands, woodlands, and forests from sea level up through mixed conifer forests (CWHR).

This bat occurs on all Sierra Nevada national forests.

Risk Factors The pallid bat tends to be a roosting habitat generalist that utilizes many different natural and man- made structures. Foraging habitat requirements appear to be more restrictive. Pallid bats appear to be more prevalent within edges, open stands, particularly hardwoods, and open areas without trees. The reduction of hardwoods, both from manual removal and competition from conifers, reduces foraging habitat for pallid bats. Hardwood and hardwood conifer stands that contain thick understory vegetation between ground level and eight feet prevents flight. In many areas, chaparral stands are maturing and forming dense contiguous stands that are not easily penetrated by foraging bats. Heavy grazing may also affect prey species hiding and foraging habitat through the reduction of grasses and herbaceous vegetation.

Although considered a roosting habitat generalist, these bats can also be affected by renewed exploration, or closure of mines, recreational caving, and reduction of tree roosts due to their high sensitivity to human disturbance. If a variety of roosting sites are not available in a given area, loss of each individual roost site can have negative effects in the local population. If bats are restricted to fewer roost sites, colonies may become larger, which would make them more susceptible to a single human disturbance. This could cause detrimental affects to maternity colonies or hibernating groups.

Urban expansion and private harvest of hardwoods have removed large amounts of foraging habitat available to the pallid bat. Renewed mining on private lands have also caused abandonment of roost sites.

Insecticide use could potentially reduce the prey base for pallid bats. The effects of insecticides directly on bats are unknown.

Conservation Measures • provide for hardwood stands into the future, including within conifer-hardwood forests, could improve foraging habitat • manage older stands to produce healthy hardwood crowns, or for regeneration through stump sprouting, • reduce overstocked conditions to promote health and vigor, and eliminate conifer encroachment in predominantly hardwood areas • implement vegetation treatments to create open understories that allow for unencumbered flight but that also provide a diverse prey species substrate (hardwood and shrub species) for foraging • adopt mine and cave management plans (currently being formulated at the Pacific Southwest Regional level) that outline specific measures for bat protection • develop mosaic stands of chaparral

FEIS Volume 3, Chapter 3, part 4.4, page 55 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Environmental Consequences Table 4.4.1.5a illustrates the affects of implementing Alternatives 2 through Modified 8.

Table 4.4.1.5a. Effects of implementing Alternatives 2 through Modified 8. Criteria Alts. Fire/Fuels Acres Riparian/ Hardwoods Old Forest Recreation* treated Ave. yearly Meadow ac. Alt. 1 86,225 ac. Existing LRMP Existing LRMP Existing LRMP Project LRMP levels Alt. 2 34,929 ac. Green/gray zone, limits Mtn. All Ret. Large Res. All No change treatments, 1, 20% trees shrubs, limits grazing, >30”dbhWS,21”dbhES, 20% streambank limit 10-15ton/ac. CWD Alt. 3 141,479 ac. Green/gray zone,2, Ret. BO >12” Mtn. LSOG 4 & 5, All No change allow mgnt.limits Ret. MO >15” trees OHV,restricts grazing Fuelwood per. >30”dbhWS,21”dbhES, 100 from rip. Zone 10-15ton/ac. CWD, HDW>15” Alt. 4 159,545 ac. Green/gray zone,2, Allows mgnt 15-20% WS, Ret. All No change early season,20% Fuelwood per. 30”dbh, 20sq ft. snags streambank limit Ret.BO>12”dbh BLO>24”dbhMO>15”dbh Alt. 5 60,527 ac. Green/gray zone, 2, Ret. BO>12” Limited mgnt. 20sq ft. No change Ret. MO>15” snags, All trees Fuelwood per. >30”dbhWS,21”dbhES, HDW>15” Alt. 6 150,592 ac. 1,2, 20% streambank Ret. BO>12” 50% OF, limits mech. No change limit, 20% shrub ut., Ret MO>15” treatment, All trees Fuelwood per. >30”dbhWS,21”dbhES 10-20tons/ac CWD Alt. 7 145,260 ac. 1, 20% streambank Allows mgnt. Fuelwood Ret. Large trees, No change limit,20% shrub ut. per emph. Mgnt. 10- Ret.BO>12”dbh 20tons/ac CWD, BLO>24”dbhMO>15”dbh HDW>15” Alt. 8 98,614 ac. Min. grazing in all Ret. BO>12” Ret. Mtns all existing No change areas,1,2,20% MO>15” CASPO suit., All trees streambank limit, LOP Fuelwood per. >30”dbh, HDW>15 for Yosemite toad, limit Rx fire to 5% Mod 8. 112,860 ac. 20% streambank limit, Ret. BO>8" Ret. Large trees, No change 20% shrub ut, Ret. MO>12" >12"dbh, canopy cover Elim. livestock from Fuelwood per. retention. occ. WIFL sites

* Effects related to recreational caving. 1 – Prohibits livestock within 3 miles of WIFL nesting habitat 2 – 100ft buffer around WIFL nest sites.

Environmental Outcomes Historic Condition. It is thought that the pallid bat was widespread through the planning area is a variety of habitats. The reduction of hardwood habitat may have reduced to limited distribution within the planning area.

Current Condition. The pallid bat occurs on all Sierra Nevada National Forests. Protection, maintenance and enhancement of both westside foothill oaks and montane oaks should provide for pallid bats. All alternatives will lead to an increasing tend in oak species.

FEIS Volume 3, Chapter 3, part 4.4, page 56 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Table 4.4.1.5b Average assessment ratings for the pallid bat. Ratings represent average degree of confidence in each outcome being realized 50 years in the future.

Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Outcome B B B B B B B B B B

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Cumulative Effects Historic Conditions. The pallid bat ranges from southern British Columbia, California east to central Texas and Kansas, Mexico, and Baja California (ECMB). It occurs in a wide variety of habitats including grasslands, shrublands, woodlands, and forests from sea level up through mixed conifer forests (CWHR).

Current Conditions. There is no indication that there has been a change in the range or distribution of the pallid bat. Hardwood and hardwood-conifer stands continue to be lost to urban expansion and timber harvest on private land in the west. Recreational caving and mining will continue to affect the species.

Table 4.4.1.5c. Average cumulative effect assessment ratings for the pallid bat. Ratings represent average degree of confidence in each outcome being realized 50 years in the future.

Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Outcome B B B B B B B B B B

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

FEIS Volume 3, Chapter 3, part 4.4, page 57 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 4.4.1.6. TOWNSEND’S BIG-EARED BAT (Plecotus townsendii) Affected Environment Life History Townsend’s big-eared bats are usually only observed at night. The animals are sensitive to light and movement so if they are disturbed during the day, they awake and their ears begin to move as they try to identify the intruder. If the disturbance occurs for more than a few seconds, the entire group takes flight (Barbour, 1969). In the summer, the females form nesting roosts. Males are solitary during the maternity periods. The maternity colonies consist of one or more small clusters, which rarely exceed 100 bats. Females are alert and active in the maternity roost and prefer dark places for their roost (Kuntz, 1982).

Townsend’s big-eared bats begin to form their hibernation roosts in late October and by January, group size is at its peak. Males usually choose a location warmer than females and are more active than the females in the winter as well (Barbour 1969). In hibernation, the posture of the bat appears to buffer it from environmental extremes. These postures are sensitive to climate change and disturbance. The ears are either erect or coiled. Bats lose weight in the winter because their fat reserves are being depleted. The more frequently they are aroused, the more weight they lose. Townsend’s big-eared bats prefer to hibernate in cold places, usually near the entrance of a cave (Kuntz 1982).

Breeding usually begins within the first three weeks of October (Barbour 1969). Ovulation and fertilization happen in spring, after hibernation, as the females move from their hibernation areas to a nesting roost (Tuttle 1988). Females bats usually only have one young a year. Young bats grow quickly, being able to fly within three weeks. After two months, many of the young bats have left the nursery roosts, with male bats leaving before females. In their first year, male bats are almost certainly incapable of breeding while female bats are able to reproduce at the age of four months (Barbour 1969).

Habitat Relationships Townsend’s big-eared bat seems to use primarily caves or cave-surrogates. These bats require specific microclimatic conditions to roost successfully. Many caves and mines do not have these conditions. Townsend’s big-eared bat is a colonial species that forms maternity colonies of up to several hundred females. Females congregate in March and June with a single pup born between May and July. This bat shows a high degree of roost site fidelity, and, if undisturbed, colonies may occupy the same roost indefinitely. Populations of this bat appear to be quite sedentary with movement to alternate roosts confined to within 12 miles of the primary roost. Townsend’s big-eared bat is very sensitive to disturbance at roost sites and may abandon a roost once disturbed. Disturbance at maternity colonies may also impact adult survival and reproductive success.

Foraging habitat for Townsend’s big-eared bat is varied, but the species forages preferentially over native vegetation. This bat is most abundant in mesic habitats but may be found in a wide variety of habitats throughout the state including grasslands, riparian areas, deserts, and old forests. Habitat associations include desert, native prairies, coniferous forests, mid-elevation mixed conifer, mixed hardwood-conifer forests, riparian communities, and active agricultural areas. They emerge late in the day to feed (Barbour 1969). This bat feeds principally on moths, but may take other insects as well (Kuntz 1982). These aerial foragers concentrate their activity along forested edges and over vegetation. This species requires access to free water.

FEIS Volume 3, Chapter 3, part 4.4, page 58 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Townsend’s big-eared bat is found throughout California, typically below 6,000 feet. Once considered common, Townsend’s big-eared bat is now considered to be uncommon in California. The distribution of the species is patchy and associated with limestone caves, lava tubes, and man-made structures, such as mines and abandoned buildings. This bat is predicted to occur on 10 of the 11 Sierra Nevada national forests.

Status Five distinct subspecies of big-eared bat are presently recognized, Pacific western big-eared bat (Plecotus townsendii townsendii), pallid western big-eared bat (P. t. pallescens), Ozark big-eared bat (P.t. ingens), Virginia big-eared bat (P. t. virginianus), and P.t. australis. The Pacific and pallid bas subspecies occur within the planning area. The Ozark and Virginia subspecies were Federally listed as Endangered in November of 1979.

Townsend’s big-eared bat has suffered a substantial decline in population over the last 40 to 60 years. Surveys conducted in California at historical maternity roosts (located prior to 1980) revealed that 24 of 46 sites (52 percent) were no longer occupied. Nearly 40 percent of the known sites had been destroyed or rendered unusable. In addition, a 55 percent decline in the number of females present in existing populations has been observed. In the Mother Lode region of the Sierra Nevada, the mean colony size has decreased from more than 200 individuals to less than 50. Graham (1966) found no extant colonies in California’s limestone caves and speculated that all had been abandoned due to human activities. Williams (1986) inferred that P. t. pallescens was common in central California into the 1960s, but by the early 1970s was rarely seen. He captured only one individual during 14 years of netting bats in central California.

The Townsend’s big-eared bat is classified as a California State Species of Concern, Forest Service sensitive species, and Fish and Wildlife Service Species of Concern.

Historic and Current Distribution There are two subspecies endemic to the planning area. P. t. townsendii is mainly found on the west coast of North America, from British Columbia to Oaxaca, Mexico. P. t. townsendii tends to occupy the humid, coastal regions of northern and central California and P. t. pallescens occurs in the remainder of the state (Hall 1981). P. t. pallescens can also be found from the central and southern portions of California east into Texas and north to the Dakotas.

There is no indication that the range of this species has changed.

Risk Factors Given the requirement of a specific environment and this bat’s sedentary behavior, it is likely that Townsend’s big-eared bat is limited by roost site availability. Although natural deterioration of caves and mines is expected, the majority of roost loss is related to human activity in the form of disturbance, demolition, renewed mining, hazard abatement, or vandalism. Increased interest in recreational caving has lead to an increased risk for all cave and mine roosting bats. One of the primary distributions of Townsend’s big-eared bat is in the Foothills of the central Sierra Nevada with its limestone cave formations and old mine workings. This area is predicted to have one of the highest human population growth rates in the Sierra Nevada. There have been significant declines in populations of Townsend’s big-eared bat in this area, and this trend will likely continue if human population growth projections are realized.

FEIS Volume 3, Chapter 3, part 4.4, page 59 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Inactive mines provide an important analog to cave habitat for many bat species. Safety concerns have lead to mine closure programs in many areas, often without adequate consideration of the biological value of old mines. Closure at the wrong time of year can eliminate entire colonies. Many old mine workings have been destroyed to make way for larger pit mine operations that have little value as roosting habitat. Caustic chemicals, such as cyanide, may be used in ore extraction. Waste ponds containing these substances pose a risk to wildlife. Bats accounted for 34 percent of documented wildlife fatalities.

Townsend’s big-eared bat seems to respond readily to protections at roost sites as long as some individuals remain in the local area. When roost sites that have been abandoned are gated properly to prevent human intrusion, bats often re-occupy the site within a relatively short period of time. Gates placed at occupied roosts often result in an increase in population size.

Modification of foraging habitat may also pose a risk to Townsend’s big-eared bat. Conversion of habitat to vineyards and loss of riparian habitat has been implicated in the decline of a number of bat species. The impacts of logging on Townsend’s big-eared bat have not been investigated. Loss of foraging habitat may be locally important in some areas, but Townsend’s big-eared bat seems to be an opportunistic feeder capable of foraging in a variety of open habitats.

Other Risks This species has declined due to direct killing by people and because of abandonment of roosts caused by disturbance due to explorers and vandals. Townsend’s big-eared bat requires adequate habitat remote from human disturbance. Urbanization may have caused population declines along the Colorado River and in other areas. Certainly, where bats roost in proximity to humans there is often a negative consequence for the bat. Humans will generally take pest control measures to eliminate bats that roost in buildings. Although known primarily as a cave species, Townsend’s big-eared bat will occasionally take up residence in buildings. Unlike other bats that roost in crevices, Townsend's big- eared bat roosts in the open on ceilings and walls, making them relatively easy to detect.

Thirteen of the largest known colonies of Townsend’s big-eared bat occur on public land. Of these, six occur in national parks, four occur on national forests, and one on lands managed by the Bureau of Land Management. Currently, the colonies in national parks receive the greatest protection from human disturbance through gating, structural modifications, and education.

Townsend’s big-eared bat has specific structural and non-structural roost site requirements. Few caves or mines satisfy the structural requirements. The single most important non-structural requirement for roost sites for this species is absence of human disturbance. Projected increases in human populations and recreational caving are the greatest risks to the persistence of this bat in the Sierra Nevada. Scientific collection, extermination of bats in human habitations, and vandalism contribute to this risk as well. Failure to protect old mine workings and develop a comprehensive conservation approach for bats has played a role in the present decline.

Conservation Measures • minimization of human disturbance is essential for Plecotus townsendii to remain in existence (Nowak 1994) • protection of maternal roosts and hibernacula.

FEIS Volume 3, Chapter 3, part 4.4, page 60 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Environmental Consequences Measures or Factors Used to Assess Environmental Consequences Because of its close association with caves and cave-analogs, Townsend’s big-eared bat is not a good candidate for analysis using CWHR vegetation projections except for general changes to potential foraging areas. This FEIS does not address recreational caving so the analysis of effects on roost sites is based on each alternative’s treatment of mines and mining. The degree to which alternatives protect roosting areas from mineral-related activities was the factor used to assess consequences of the alternatives on Townsend’s big-eared bat.

Assumptions and Limitations Providing roosting habitat is not adequate if appropriate foraging habitat is unavailable or inaccessible for this relatively sedentary species. Lack of spatial data prevents a thorough analysis of foraging habitat. Finally, human disturbance, primarily through recreational caving, poses a continuing risk to the persistence of Townsend’s big-eared bat in the Sierra Nevada.

Townsend’s big-eared bat is assumed to benefit from withdrawal of areas from mining due to reductions in human disturbance, road building, and the rate of re-entry into old mines. Road building to support mining can disrupt foraging habitat and improve access to caves and mines for recreational cavers. There may be a trade-off between mining-related disturbance and the potential for mining to generate new roost sites. However, the majority of recent mines are pit mines that provide little value as roost sites. Given this and the high sensitivity of this species to disturbance, new mine construction is assumed to be detrimental.

FEIS Volume 3, Chapter 3, part 4.4, page 61 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Effects of the Alternatives on Townsend’s big-eared bat Table 4.4.1.6a. Effects of implementing Alternatives 2 through Modified 8. Criteria Alts. Fire/Fuels Acres Riparian/ Hardwoods Old Forest Recreation* treated Ave. yearly ac. Meadow Alt. 1 86,225 ac. Existing LRMP Existing LRMP Existing LRMP Project LRMP levels Alt. 2 34,929 ac. Green/gray zone, Mtn. All Ret. Large No change limits treatments, 1, Reserves, All trees 20% shrubs, limits >30”dbhWS, grazing, 21”dbhES, CWD 20% streambank 10-15 ton/ac limit Alt. 3 141,479 ac. Green/gray zone,2, Ret. BO >12” Mtn. LSOG Ranks No change allow mgnt.limits Ret. MO >15” 4 & 5, All tree OHV,restricts Fuelwood per. >30”dbhWS grazing 100 from 21’dbhES, rip. Zone HDW>15”,CWD 10-15 ton/ac. . Alt. 4 159,545 ac. Green/gray zone,2, Allows mgnt 15-20% WS, Ret. No change early season, 20% Fuelwood per. All 30”dbh, 20sq.ft. streambank limit Ret.BO>12”dbh snags BKO>24”dbhMO >15”dbh Alt. 5 60,527 ac. Green/gray zone, Ret. BO>12” Limited mgnt,all No change 2, check Ret. MO>15” tree >30”dbhWS Fuelwood per. 21’dbhES, HDW>15”,CWD 10-15 ton/ac,8snag/ac Alt. 6 150,592 ac. 1,2, 20% Ret. BO>12” Ret 50% OF, All tree No change streambank limit, MO>15” >30”dbhWS 20% shrub ut., Fuelwood per. 21’dbhES, HDW>15”,CWD 10-15 ton Alt. 7 145,260 ac. 1, 20% streambank Allows mgnt. Ret. Large trees, No change limit, Fuelwood per emphasizes 20% shrub ut. Ret.BO>12”dbh Mgnt.10- BKO>24”dbh 20ton/ac.CWDHD MO>15”dbh W>15” Alt. 8 98,614 ac. Min. grazing in all Ret. BO>12” Mtns all existing No change areas,1,2, 20% Ret. MO>15” CASPO suit., All streambank limit, Fuelwood per. trees >30”dbh, LOP for Yosemite HDW>15” toad, limit Rx fire to 5% Mod 8 112,860 ac. 20% streambank Ret. BO>8" Ret. Large trees No change limit, 20% shrub ut Ret. MO>12" >12" dbh Elim. livestock from Fuelwood per. Cnpy. cov. ret. occ. WIFL sites

* Effects related to recreational caving. 1 – Prohibits livestock within 3 miles of WIFL nesting habitat 2 – 100ft buffer around WIFL nest sites.

Summary of Consequences to Townsend’s Big-Eared Bat Alternative 2 provides the best protection for roosting habitat in mines. Alternative 2 would affect foraging habitat, as it would likely trend toward more dense conditions. Vegetation projections indicate a 4 percent reduction in habitat suitability over 50 years under this alternative. While this seems a small percent change, for a single habitat component (foraging habitat) for a species already in decline, this may represent a more substantial impact. In addition, the vegetation projections may not be able to adequately portray the changes in the habitat types preferred by this species. Consequently, the actual change in suitability over time may be smaller or greater. Alternatives 4 and

FEIS Volume 3, Chapter 3, part 4.4, page 62 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 7 provide for foraging habitat by ensuring the presence of a variety of stand structures on the landscape.

Environmental Outcomes Historic Condition. P. t. townsendii is mainly found on the west coast of North America, from British Columbia to Oaxaca, Mexico. P. t. townsendii tends to occupy the humid, coastal regions of northern and central California and P. t. pallescens occurs in the remainder of the state (Hall 1981). P. t. pallescens can also be found from the central and southern portions of California east into Texas and north to the Dakotas.

Current Condition. This bat occurs on all Sierra Nevada National Forests.

Table 4.4.1.6b. Average assessment ratings for the Townsend’s big-eared. Ratings represent average degree of confidence in each outcome being realized 50 years in the future.

Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Outcome C C C C C C C C C C

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Cumulative Effects Historic Conditions. P. t. townsendii is mainly found on the west coast of North America, from British Columbia to Oaxaca, Mexico. P. t. townsendii tends to occupy the humid, coastal regions of northern and central California and P. t. pallescens occurs in the remainder of the state (Hall 1981). P. t. pallescens can also be found from the central and southern portions of California east into Texas and north to the Dakotas.

Current Conditions. There is no indication that there have been reductions in the historic range or in the habitat upon which they depend. Hardwood habitat through there range is declining but to what extent this is affecting the species is uncertain. There are fluctuations in the development and use of miles and caves.

FEIS Volume 3, Chapter 3, part 4.4, page 63 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Table 4.4.1.6c. Average cumulative effect assessment ratings for the Townsend’s big-eared bat. Ratings represent average degree of confidence in each outcome being realized 50 years in the future.

Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Outcome D D D D D D D D D D

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

FEIS Volume 3, Chapter 3, part 4.4, page 64 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 4.4.1.7. SIERRA NEVADA SNOWSHOE HARE (Lepus americanus) Affected Environment Life History The breeding season is primarily controlled by day length, but the effect of weather and phase of the population cycle can alter the date of the first litter by three weeks in Alberta (Cary and Keith 1979). Two litters per year are normal in the far north and in the south at high altitudes; in central parts of its distribution three or four litters are produced (Keith 1981). Total production per year per female ranges from 5.7 to 17.8. Litter size increases by about one from the first litter of the season to latter litters, and regionally from about 2.2 to almost six (Keith 1981).

Young snowshoe hares leave the location of birth within a few days and scatter into the surrounding undergrowth. Young snowshoe hares in Minnesota spent the days in separate hiding places and came together once a night to nurse. It was found that captive snowshoe hares began to feed on vegetation at 10 to 12 days of age and wild hares become independent at two weeks of age (Tomossi and others 1995).

In spite of adequate spring herbaceous growth, a shortage of winter browse can affect the reproductive performance of females throughout the summer. The deleterious affect suffered by the females affects in turn the survival of juveniles in the summer (Tomossi and others 1995).

Habitat Relationships Given their enormous geographic range, it is not surprising that snowshoe hares use a wide range of forest types including: conifers, aspen, birch, beech, maple and mixed hardwoods. They show a marked preference for subclimax forest, transitional zones, and swamp edges; hence fires are important modifiers. Peak numbers follow fires when the shrub and regrowth become dense. Over time, as the forest matures and the ground cover becomes over-shaded, numbers decline.

The home range is smaller in thick cover, averaging 1.5 to 3.2 acres in different studies (Bittner and Rongstad 1982). Males have larger ranges than females, and juveniles have smaller ranges than adults (Adams 1959). Population cycle, with peak densities every eight to eleven years, are up to 300 times higher than the troughs, but the normal range is 10 to 30 fold (Keith 1981). Cycles are synchronous over wide geographic regions. At peak numbers (7.7 per acre in Alberta), (Keith and Windberg 1978) snowshoe hares have an enormous effect on the vegetation and predators in their habitat. The food eaten by snowshoe hares changes from grasses, sedges, dandelions and various herbs in summer, to birch, spruce, willow, tamarack, and pine in the winter.

In California, they are primarily found in montane riparian habitats with thickets of alders and willow and in stands of young conifers interspersed with chaparral. Habitats used by snowshoe hares include early seral stages of mixed conifer, subalpine conifer, red fir, Jeffrey pine, lodgepole pine, and aspen.

In the Sierra Nevada they live only in boreal zones, typically inhabiting riparian communities with thickets of deciduous trees and shrubs such as willows and alders. They also frequent dense thickets of young conifers and chaparral composed of Ceanothus and Arctostaphylos (Orr 1940). In the Lake Tahoe region in summer, Orr (1940) noted signs of snowshoe hares only near brush adjacent to both meadows and riparian deciduous vegetation.

FEIS Volume 3, Chapter 3, part 4.4, page 65 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Status Judging from circumstantial evidence, Sierra Nevada snowshoe hares are more abundant than other subspecies of snowshoe hares, but still are relatively uncommon. There is no evidence of a populations decline, but L. americanus is vulnerable to widespread alternations due to some logging activities and use of meadows for agriculture. The State recognized the subspecies as one of special concern, although the Sierra Nevada snowshoe hare remains a harvest species in the California.

Historical and Current Distribution The snowshoe hare is the most widespread of all New World hares, extending from Alaska to Newfoundland, and penetrating far south into America down the Coastal Range, the Rockies, and the Appalachians, to mid-California, northern New Mexico and Tennessee. On open tundra in the far north of Canada it gives way to the arctic hare Lepus arcticus. The snowshoe hare seems reluctant to cross open country but is not found as a relict species in small isolated patches of forest (Lomolino et al. 1989).

Within the planning area, the Sierra Nevada snowshoe hare range along the mid-elevations of the Sierra Nevada from the vicinity of Mt. Lassen southward to Mono and Tulare Counties. They are known from Nevada only in the vicinity of Lake Tahoe (Hall and Kelson 1959). They occupy altitudes above 4000 feet in the north of their range and 5000 feet in the south. Upper elevation limits are unknown, but these hares generally occur below 8000 feet. The only national forest in the project area not having the Sierra Nevada snowshoe hare is the Sequoia National Forest.

Risk Factors This species is vulnerable to widespread habitat alternations due to logging activities and use of meadows for agriculture, grazing, and other activities. Furthermore, snowshoe hares are a resident small game species in California.

Conservation Measures • maintain well distributed habitat of young (early seral) dense conifer forest stands interspersed with chaparral • maintain healthy meadow systems • maintain riparian systems which provide thickets of alders and willows.

Environmental Consequences Assumptions and Limitations This assessment assumes that Sierra Nevada snowshoe hares are associated with early seral stage of conifer stands and that future manager activities will maintain these types. Fire and Fuels treatments will lead to short term reductions in habitat suitability for the Sierra Nevada snowshoe hare. Changes in young stand (plantation, 2D, 2M. 3D, 3M and shrub) densities will reduce the suitability of snowshoe hare habitat.

Effects of the Alternatives Table 4.4.1.7a compares the alternatives in terms of how they would manage the various activities or provide standards and guidelines that would maintain or enhance Sierra Nevada snowshoe hare habitat.

FEIS Volume 3, Chapter 3, part 4.4, page 66 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Table 4.4.1.7a. Comparison of management activities that could affect habitats for the Sierra Nevada snowshoe hare by alternative. Alt. Grazing Fire/Fuels** Riparian/Meadow Mgnt 1 Existing LRMP levels 86,225 acres treated Varies per NF w/SMZ and RMAs 2 45% max. utilization, 4” ST* 34,929 acres treated 150ft green zone, 412-760ft gray zone 3 Meet proper functioning condition 141,479 acres treated 150ft green zone, 412-760ft gray zone 4 <45%utilization, meeting AMS goals 159,545 acres treated 150ft green zone, 412-760ft gray zone 5 5% bank trampling,50%shrub cover, 5-7”ST 60,527 acres treated 150ft green zone, 412-760ft gray zone 6 <45%utilization, meeting AMS goals 150,592 acres treated 300ft on perennial, 100ft On seasonal 7 <45%utilization, meeting AMS goals 145,260 acres treated 300ft on perennial, 100ft On seasonal 8 30%utilization, ST>4” 98,614 acres treated 300ft on perennial, 150 intermittent, 75 ephemerals Mod 8 20% streambank limit, 20% shrub ut 112,860 acres treated 300ft on perennial, 100ft On seasonal. Elim. livestock from occ. WIFL sites

*ST = Stubble height **Modeled treatment acres include brush (fire and mechanical), plantation thin, and biomass thin for the 1st decade.

Environmental Outcomes Historic – Within the planning area, the Sierra Nevada snowshoe hare range along the mid-elevations of the Sierra Nevada from the vicinity of Mt. Lassen southward to Mono and Tulare Counties. They are known from Nevada only in the vicinity of Lake Tahoe (Hall and Kelson, 1959). They occupy altitudes above 4000 feet in the north of their range and 5000 feet in the south. Upper elevation limits are unknown, but they generally occur below 8000 feet.

Current – The Sierra Nevada snowshoe hare remains on all national forest except the Sequoia National Forest. Fuels reduction should maintain younger stands in a more open condition, younger shrub fields, and better protection riparian area across all alternatives.

Table 4.4.1.7b represents the assessment ratings for the Sierra Nevada snowshoe hare.

Table 4.4.1.7b. Average assessment ratings for the Sierra Nevada snowshoe hare. Ratings represent average degree of confidence in each outcome being realized 50 years in the future. (See Chapter 4, Part 4.1.5.) Total score for each alternative is 100.

Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Outcome C C B B C B B C B B

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

FEIS Volume 3, Chapter 3, part 4.4, page 67 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Cumulative Effects Population Outcomes Historic Conditions. The snowshoe hare is the most widespread of all New World hares, extending from Alaska to Newfoundland, and penetrating far south into America down the Coastal Range, the Rockies, and the Appalachians, to mid California, northern New Mexico and Tennessee. On open tundra in the far north of Canada it gives way to the arctic hare Lepus arcticus.

Current Condition. The species Lepus americanus remains wide spread and numerous throughout most of its’ historic range. The subspecies L. a. tahoensis remains an uncommon resident at the upper elevation in the Cascade Mountains in Siskiyou County south through the Sierra Nevada to Mariposa, Mono, and Madera Counties. A small insular population also has been reported in the Warner Mts., Modoc County. There is no evidence of a population decline (Williams 1986). The current and project human population growth, and associated land use and development on private lands, would be expected to continue a trend in potential habitat degradation or loss. It is uncertain how much or at what rate habitat loss would occur.

Table 4.4.1.7c represents the estimated population outcomes through the planning horizon for the Sierra Nevada snowshoe hare.

Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Outcome C C B C C B C C C B

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

FEIS Volume 3, Chapter 3, part 4.4, page 68 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 4.4.2. Birds 4.4.2.1. CALIFORNIA SPOTTED OWL Affected Environment The California Spotted Owl: A Technical Assessment of Its Current Status (Verner et al. 1992), referred to here as the Technical Report, provides the most comprehensive assessment of the status of the California spotted owl. Much of the information contained in this analysis is summarized from the Technical Report. In addition, much of the research on Northern spotted owls is relevant to discussions about California spotted owls, given the similarity in concerns and methods used to study these birds. Papers in Demography of the Northern Spotted Owl (Forsman et al. 1996) provide excellent overviews of studies on that subspecies and are frequently referenced herein below.

Species Background Population Status and Trend Range and Distribution. The range of the California spotted owl includes the southern Cascades south of the Pit River in Shasta County, the entire Sierra Nevada Province of California (and extending into Nevada), all mountainous regions of the Southern California Province, and the central Coast Ranges at least as far north as Monterey County (Grinnell and Miller 1944, Gould 1977, cited in Verner et al. 1992). Within this range, the owl occurs on 15 National Forests/Management Units administered by the Forest Service, four National Parks, several State Parks and Forests, private timberlands, and scattered Bureau of Land Management lands. The elevation of known nest sites ranges from about 1000 feet to 7700 feet, with about 86 percent occurring between 3000 and 7000 feet.

The California spotted owl population has two major geographic groups, one inhabiting the Sierra Nevada Province and the other in the Southern California Province, with Tehachapi Pass as the dividing line between the two. These regions are distinct geographically, and owl populations in the two provinces probably seldom exchange individuals. Connectivity may exist, however, through the Tehachapi Mountains and the Liebre/Sawmill area east of Interstate Highway 5.

The California spotted owl's range adjoins that of the northern spotted owl in Siskiyou, Shasta, and Modoc Counties as described in the 1990 U.S. Department of Interior's final ruling listing the northern spotted owl as threatened. The range of the northern spotted owl includes a small portion of both the Modoc and Lassen National Forests, generally north of the Pit River and west of Highway 139. Although most of the northern spotted owl's range is covered under the Northwest Forest Plan, this EIS covers a small portion of its range. The primary concern for northern spotted owls in this area is the condition of dispersal habitat and the possible effects of activities in the Modoc National Forest adjacent to areas covered by the Northwest Forest Plan. This EIS does not change any requirements for northern spotted owl consultation with the U.S. Fish and Wildlife Service in this area.

Population Size and Distribution. Information on the historic distribution, abundance, and habitat associations of California spotted owls in the Sierra Nevada is unavailable (Verner et al. 1992). Thus, it is not possible to determine how current population numbers and distribution may have changed relative to historic conditions. Based on records from the California Department of Fish and Game recorded through 1999, a total of 1,323 owl sites are known on FS lands within the project area, with another 129 owl sites reported on non-FS lands within

FEIS Volume 3, Chapter 3, part 4.4, page 69 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 the boundaries of the project area (Table 4.4.2.1a). The Lassen, Plumas, Tahoe, Eldorado, Stanislaus, Sierra, and Sequoia National Forests have major populations of spotted owls, with 99 percent of the total known owl sites on FS lands in the project area occurring within these forests. These seven National Forests include the vast majority of suitable habitat for spotted owls in the Sierra Nevada. Numbers of California spotted owls are low on the Modoc, Inyo, and Humboldt-Toiyabe National Forests, and in the Lake Tahoe Basin Management Unit, and reproduction is infrequent. Private land comprises a portion of the home ranges of some owl sites on FS lands, with more than 15 percent of the owl sites on FS lands having greater than 15 percent of their home range within privately owned lands. Four National Parks, scattered Bureau of Land Management lands, industrial timberlands, and private timberlands provide the remainder of the estimated suitable habitat and additional spotted owl pairs.

These numbers represent an incomplete count of the California spotted owl population in the Sierra Nevada since not all areas have been surveyed and survey results, particularly on industrial forest lands, have not always been reported to the Department of Fish and Game (Gould 1999, pers comm.). Forest Service biologists estimate an additional 160 to 218 sites (singles and pairs) on National Forest System lands based on unsurveyed suitable habitat. The number of nonterritorial adults in the population remains unknown.

Table 4.4.2.1a. Number of California spotted owl sites by status known on Forest Service lands and non-Forest Service lands, since 1987, within the boundaries of the National Forests in the project area reported in the California Department of Fish and Game database (Fall 1998). National Forest Lands ( Non-NF Lands) National Pairs Territorial Singles Pairs Territorial Singles Total NF Forest Singles Singles (NF+pvt) Eldorado 160 36 13 12 5 5 209 (231) Inyo 1 1 1 0 0 0 3 (3) Lassen 99 18 20 6 0 2 137 (145) Modoc 1 0 0 0 0 0 1 (1) Plumas 171 53 31 8 2 4 255 (269) Sequoia 72 44 17 6 1 1 133 (141) Sierra 141 31 45 9 3 3 217 (232) Stanislaus 113 52 30 20 6 6 195 (227) Tahoe 107 26 24 17 4 8 157 (186) Lake Tahoe 8 0 4 1 1 0 12 (13) Basin Humboldt- 4 0 0 0 0 0 4 (4) Toiyabe* Total 873 185 261 73 36 26 1323 (1452)

*data supplied by forest in 1999

California spotted owls are currently distributed relatively continuously and uniformly throughout their range in the Sierra Nevada (Verner et al. 1992, Noon and McKelvey 1996), although concern exists for fragmentation effects at finer scales due to habitat alteration (Gutiérrez and Harrison 1996). Estimates of mean crude density (that is, number of owls divided by the total acreage of the study area) reported in the Technical Report from four study areas in the Sierra Nevada ranged from 0.526 owls per square mile on the Sierra National Forest to 0.259 owls per square mile on the Eldorado National Forest (Verner et al. 1992:175). Subsequent research has demonstrated that estimates of mean crude density varied annually during 1990-1998, ranging between 0.313-0.530 owls per square mile on the Sierra National Forest and 0.415-0.615 in Sequoia National Park (Steger et al. 1998). Although California spotted owls appear to be distributed throughout their historic range in the Sierra

FEIS Volume 3, Chapter 3, part 4.4, page 70 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Nevada at the present time, eight geographic areas of concern were identified in the Technical Report as having the potential to contribute to future gaps in owl distribution within the Sierra Nevada (Verner et al. 1992). Concerns associated with each of these areas of concern are described in this assessment under the section on “Risk Factors.”

Population Trends. Estimates of California spotted owl population trends are available from demographic studies conducted at four study areas across the range of the owl in the Sierra Nevada (Lassen NF, Eldorado NF, Sierra NF, Sequoia/Kings Canyon National Park). All four studies reported statistically significant declining trends over the duration of each study based on estimates of lambda (Table 4.4.2.1b.) (Blakesley and Noon 1999, Gutierrez et al. 1999, Steger et al. 1999). These estimates suggest rates of decline during the periods of study that range from 6 to 11 percent per year. Extrapolating the annual decline over each study period results in expected overall declines of 49 to 73 percent. The estimate of lambda from the Lassen NF is very close to the overall average (0.923). Using vital rates from the Lassen study (Blakesley and Noon 1999), the cohort replacement rate (R) is calculated to be 0.53, suggesting that the population is declining by nearly half each generation.

By comparison, 11 demographic studies of northern spotted owls exhibited an average annual decline of 8.3 percent (standard error of 1.3 percent) over the period 1985 through 1998. Using estimates of juvenile emigration to adjust the vital rates, an adjusted estimate of lambda suggests an overall rate of decline of 3.9 percent for the northern spotted owl during the study period (Franklin et al. 1999).

Table 4.4.2.1b. Estimates of the finite rate of annual population change (lambda) from four California spotted owl demographic studies conducted in the Sierra Nevada, 1986-1998. Overall change is computed by extrapolating lambda over the period of study. Study Area Years Lambda 95% C.I. S.E. Overall change* Lassen NF 1990-1998 0.923 0.888-0.958 - -51.4% Eldorado NF 1986-1998 0.930 - 0.027 -61.1% Sierra NF 1987-1998 0.898 - - -72.5% Sequoia NP 1988-1998 0.940 - - -49.4%

There is reason to believe that the lambda estimates unadjusted for emigration may overstate the rate of decline in California spotted owls (Franklin et al. 1999). There is no evidence, substantiated or anecdotal, to suggest that actual spotted owl abundances declined by the amount indicated in the “Overall change” column in Table 4.4.2.1b. There are two study areas where the model-based estimates are complemented by census data, the Sierra NF and the Sequoia NF. Based on data in Steger et al. (1999) the number of nesting pairs counted on the Sierra NF study area decreased from 31 to 20 from 1990 to 1998, an annual rate of decline of 5.3 percent -less than half the model-based estimate. In contrast, the number of nesting pairs in the Sequoia NF study area increased from 22 pairs in 1990 to 24 pairs in 1998. The number of nesting pairs counted in both studies peaked at 34 in 1994, suggesting a period of population increase from 1990 to 1994, followed by decline. Only by calculating annual rates of decline in counts of nesting pairs over the shorter period, 1994 to 1998, can one obtain values that approach lambda estimates. Gutierrez et al. (2000) report that there were noticeably fewer territorial individuals encountered on the density area of their study in 1999 than during the previous seven years. They suggest that, although another year or two of study is required to confirm if this drop in territorial owls was due to mortality or detectability, the apparent decline in territorial holders increases concerns about the health of this population.

FEIS Volume 3, Chapter 3, part 4.4, page 71 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Whether lambda estimates obtained in the manner common to spotted owl research are inherently inaccurate (biased) is a much debated proposition. Raphael et al. (1996) discuss five potential sources of bias, both positive and negative: assumptions of the Lotka-Leslie model, duration of studies, senescence (older birds might have reduced reproductive potential), estimating reproductive rates, and emigration. Emigration is usually identified as the major factor biasing estimates. The mark- recapture models used to estimate vital rates cannot distinguish between owls that die and owls that permanently leave the area and survive elsewhere. This confusion can lead to overestimates of mortality. Recognizing this, researchers often calculate the emigration rates that would be necessary in a stable population to reproduce the estimated mortality rates (e.g., Burnham et al. 1996, Blakesley and Noon 1999). Even allowing for reasonable estimates of emigration, the demographic projections suggest declining populations.

In summary, the demographic studies strongly suggest population declines in California spotted owls. The declines are sufficiently severe to warrant concern, even in light of uncertainties in the magnitude of the declines.

Habitat Status and Trend Habitat Preferences Five vegetation types provide spotted owl habitat in the Sierra Nevada Province: foothill riparian/hardwood, ponderosa pine/hardwood, mixed-conifer forest, red fir forest, and the east side pine forest. The mixed-conifer forest type is the predominant type used by spotted owls in the Sierra Nevada: about 80 percent of known sites are found in mixed-conifer forest, 10 percent in red fir forest, 7 percent in ponderosa pine/hardwood forest, and the remaining 3 percent in foothill riparian/hardwood forest and eastside pine.

Six major studies (Verner et al. 1992, Chapter 5) described habitat relations of the owl in four general areas spanning the length of the Sierra Nevada. These studies examined spotted owl habitat use at three scales: landscape; home range scale; and nest, roost, or foraging stand. By comparing the amount of time owls spend in various habitat types to amount of habitat available, researchers determined that owls preferentially used areas with at least 70 percent canopy cover, used habitats with 40 to 69 percent canopy cover in proportion to its availability, and spent less time in areas with less than 40 percent canopy cover than might be expected.

Foraging. In studies referenced by the Technical Report, owls foraged most commonly in intermediate- to late-successional forests with greater than 40 percent canopy cover and a mixture of tree sizes, some larger than 24 inches in dbh. The owls consistently used stands with significantly greater canopy cover, total live tree basal area, basal area of hardwoods and conifers, snag basal area, and dead and downed wood, when compared with random locations within the forest. Studies on the Tahoe and Eldorado National Forests found that owls foraged in stands with large diameter trees (defined as trees greater than 24 inches in dbh in one study and trees 20 to 35 inches in dbh in the other) significantly more than expected based on availability. Owls foraged in stands in the 4G timber stratum significantly more than expected, based on the proportion of that stratum. Several studies have identified foliage height class diversity, or canopy layering, as a stand structural characteristic associated with preferred foraging sites for the northern spotted owl (North et al. 1999, Carey et al. 1992).

FEIS Volume 3, Chapter 3, part 4.4, page 72 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 In general, stands suitable for owl foraging have (1) at least two canopy layers, (2) dominant and codominant trees in the canopy averaging at least eleven inches in dbh, (3) at least 40 percent canopy cover in overstory trees, (30 percent canopy cover in the red fir type), and higher than average numbers of snags and downed woody material. Although canopy covers down to 40 percent are suitable for foraging, they appear to be only marginally so. Radio- tracking data from the Sierra National Forest showed that owls tended to forage more in sites with greater than 50 percent canopy cover than predicted from their availability; stands with 40 to 50 percent canopy cover were used about in proportion to their availability (Verner pers. comm. 1999). Recent analysis by Hunsaker et al. (in press) found that productivity was positively correlated with the proportion of individual owl home ranges having greater than 50% canopy-cover and negatively correlated with the proportion having less than 50% canopy cover, based on aerial photo interpretation. From these correlations, the authors conclude that the threshold between canopy cover values that contribute to or detract from occurrence and productivity is a value near 50 percent. Table 4.4.2.1c summarizes the range of means for foraging stand attributes reported in the Technical Report.

Nesting and Roosting. In studies referenced by the Technical Report, spotted owls preferred stands with significantly greater canopy cover, total live tree basal area, basal area of hardwoods and conifers, and snag basal area for nesting and roosting. Owls used stands in the 4G and 4N timber strata for nesting significantly more than expected, based on the proportion of those strata. In general, stands suitable for nesting and roosting have (1) two or more canopy layers, (2) dominant and codominant trees in the canopy averaging at least 24 inches in dbh, (3) at least 70 percent total canopy cover (including the hardwood component), (4) higher than average levels of very large, old trees, and (5) higher than average levels of snags and downed woody material. Table 4.4.2.1c summarizes the range of means for these attributes reported in the Technical Report.

Table 4.4.2.1c. Range of mean values of some attributes in suitable habitat for spotted owls in Sierra Nevada mixed-conifer forests (from Verner et al. 1992:96). Attribute Nesting and Roosting Stands Foraging Stands Percent canopy cover1 70-95 50-90 Total live tree basal area2 185-350 180-220 Total snag basal area2 30-55 15-30 Basal area of large snags2,3 20-30 7-17 Downed woody debris4 10-15 10-15

1 Mostly in canopy >30 feet high, including hardwoods. 2 Square feet per acre. 3 Dead trees > 15 inches d.b.h. and > 20 feet tall. 4 Tons per acre.

Research on California spotted owls has continued on owl populations on the Lassen, Eldorado, and Sierra National Forests, and in Sequoia/Kings Canyon National Park, since publication of the Technical Report, resulting in an increased number of documented nest sites. Plot data centered on owl nests from these studies provides further information on vegetation characteristics around nest sites. Each of the plots was classified using the R5 FIA classification procedure as described in Appendix B. Standardized classification resulted in classification of the plots to CWHR class and measures of a consistent set of variables that are directly comparable to data analyzed using the same classification procedure from the FIA

FEIS Volume 3, Chapter 3, part 4.4, page 73 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 program. A total of 299 nest and main roost site plots were available for analysis. Seven nest plots from Giant Sequoia stands were excluded from the analysis because of the significant effects that individual large giant sequoia trees can have on plot summary statistics, rendering it inappropriate to compare them to plot data from other forest types.

Classification of the remaining 292 plots to CWHR class using the R5 FIA analysis procedure resulted in approximately 32% of the plots classified as structural class 6, 18% as structural class 5M, 14% as 4D, 11% as 4M, 9% as 5D, 7% as 5P, and 5% as 4P, with 2% or less of the remaining plots as each of the 5S, 4S, 3D, 3M, and 3P classes (Figure 4.4.2.1a). North et al. (2000) suggested that canopy cover, tree density, and foliage volume represent conditions found to be consistent across different forest types and therefore could indicate the basic nest- site conditions selected by owls. Owl nests were consistently located in sites with 75 percent canopy cover, 300 stems/ha, and 40,000 cubic meters/ha of foliage volume.

Figure 4.4.2.1a. Distribution of owl nest plots by CWHR class on the Sierra, Eldorado, and Lassen National Forests.

40

35

30

25

SIE 20 ELD

Percent LAS

15

10

5

0 6 5D5M5P5S4D4M4P4S3D3M3P3S CWHR Class

Nest Tree Characteristics. Spotted owls regularly use five nest types: (1) cavity nests placed in natural cavities resulting from decay; (2) broken-topped trees and snags; (3) platform nests placed on remnant platforms built by other species, or on debris accumulations; (4) dwarf mistletoe brooms; and (5) "undefined nest types" not fitting any of the previous descriptions. In the Sierra Nevada, 63 percent of nests were in live trees, and 37 percent were in snags (Verner et al. 1992:92).

Habitat Composition at Home Range and Landscape Scale. Single stand attributes preferred by spotted owls are relatively easy to measure and describe. Greater difficulty arises

FEIS Volume 3, Chapter 3, part 4.4, page 74 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 in describing habitat attributes at the landscape scale. The Technical Report examined habitat selection at the stand scale and reported on the habitat composition found within home ranges. It could not determine, however, whether the habitats used by owls were adequate to ensure their reproducing at a rate sufficient to maintain their population. This requires mapping of vegetation attributes important to the spotted owl across large areas and analyzing relations between vegetation and spotted owl productivity and survivorship. Such an effort is presently in progress for the demographic study areas in the Sierra and Lassen National Forests, using aerial photo interpretations of vegetation types from 1996 photos. At present, the Forest Service's vegetation inventory based on satellite imagery provides the only large-scale vegetation mapping of the remaining spotted owl demographic study areas in the Sierra Nevada where these types of analyses can be done. Considerable uncertainty remains as to whether this Forest Service vegetation inventory mapping can describe owl habitat with sufficient precision to provide a meaningful description of landscape vegetation characteristics important to the spotted owl.

Home Range Size. Estimates of California spotted owl home range size are extremely variable. All available data indicate that they are smallest in habitats at relatively low elevations that are dominated by hardwoods, intermediate in size in conifer forests in the central Sierra Nevada, and largest in the true fir forests in the northern Sierra Nevada (Verner et al. 1992). Based on an analysis of data from telemetry studies of California spotted owls, mean breeding season, pair home range sizes have been estimated (using 100 percent minimum convex polygon method): 9,000 acres on the Lassen National Forest, true fir type; 4,700 acres on the Tahoe and Eldorado National Forests, mixed conifer type; and 2,500 acres on the Sierra National Forest, mixed conifer type.

Using the FS vegetation inventory data, the Technical Report analyzed the sizes of stands containing nest trees and the cumulative sizes of each nest stand plus all adjoining stands that were in vegetation strata preferentially used by owls for nesting. The mean size of nest stands was about 100 acres; the mean size of the nest stand plus adjacent suitable stands was about 300 acres. In radio tracking studies, the area including half of the foraging locations of owls was found to vary from an average of 317 acres on the Sierra National Forest to an average of 788 acres on the Lassen National Forest (Verner et al. 1992:87). Bingham and Noon (1997) found the "overused" portion of a spotted owl's breeding home range (core area) to be 20 to 21 percent of the home range.

Habitat in Home Ranges. Spotted owls were found to more consistently select for habitat patches with high canopy cover than for large tree size-class (Verner et al. 1992:155). The average proportion of habitat with greater than 40 percent canopy cover within home ranges of spotted owls on the Sierra and Lassen National Forests was found to be 81 and 67 percent, respectively (Verner et al. 1992:153-155).

In its Science Review, the Forest Service Pacific Southwest Research Station (1998) reviewed an analysis by Bart (1995) examining the relation between the amount of an owl pair's home range that is suitable habitat and the productivity and survivorship of northern spotted owls. This analysis “...suggests that removing any suitable habitat within the vicinity of the nest tends to reduce the productivity and survivorship of the resident owls. …It appears that the lambda statistic [essentially the annual balance between birth rate by age class on one hand, and mortality rate by age class on the other] is probably about 1.0 when suitable habitat covers

FEIS Volume 3, Chapter 3, part 4.4, page 75 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 30 percent to 50 percent of the landscape.” Reproduction would drop below replacement rate at some threshold percentage of suitable habitat between 30 and 50 percent in home ranges and in the larger landscape in general.

Recently completed analysis in the Sierra National Forest demographic study area concludes that canopy cover composition within owl home ranges is significantly correlated with owl occurrence and productivity (Hunsaker et al. in press). Owl home ranges were represented as concentric circles surrounding an owl activity center, equal to 50%, 70% and 90% MCP home- range estimates for owl sites. These analysis areas were analyzed in relation to five categories of canopy closure: 0-19%, 20-39%, 40-49%, 50-69% and 70-100%. Productivity was positively correlated with the proportion of the analysis area having greater than 50% canopy- cover and negatively correlated with the proportion having less than 50% canopy cover. The values ranged from 75% of the smallest analysis area (178 acres) with greater than 50% canopy cover to 60% of the largest analysis area (1,062 acres) having greater than 50% canopy cover.

Information on the desired configuration or patchiness of habitat within a spotted owl's home range is lacking for the California spotted owl. Demographic studies on the northern spotted owl in the Klamath Province have found that birds with access to larger blocks of suitable habitat had slightly lower mortality rates, but those with home ranges that were more patchy had slightly higher fecundity (number of young produced per breeding female). A landscape pattern with some fine-scale fragmentation of old forest (small patches of other habitats with convoluted edges) dispersed within and around a main patch of old forest appeared to provide the optimum balance in promoting both high fecundity and high survival (Franklin et al. 2000). A comparison of demographic data from spotted owls on the Sequoia/Kings Canyon National Parks with those on the Sierra National Forest finds that spotted owls on the National Forest average slightly higher fecundity but owls on the National Park had slightly higher annual survival. Although the differences are not significant statistically, the general results are consistent with those found in the Franklin et al. study, assuming that habitat on National Forest lands is patchier than that found on National Park lands (Verner, pers. comm. 1999).

The Science Report highlighted new information on the importance of large, old trees within spotted owl habitat, reporting that "Region 5 data on known ages and diameters of conifers at least 39 inches in dbh (but not owl nest trees) from the seven westside Sierra Nevada National Forests demonstrate that tree ages in different timber strata were 157 to 438 years old, with an average age of 258 years. "Most strata-level age estimates averaged between 250 and 300 years" (Verner and McKelvey 1994). “These findings suggest that the spotted owl requires large conifers 250 or more years old distributed at the landscape scale.”

Habitat Status Trends. Assessing historic to current changes in the amount and quality of California spotted owl habitat in the Sierra Nevada is problematic due to uncertainty about historic conditions, uncertainty about what constitutes high quality owl habitat, and uncertainty regarding current vegetation conditions due to accuracy, resolution, and scale concerns related to current inventory maps. Due to uncertainty about the historic size, distribution, and habitat relationships of California spotted owl populations in the Sierra Nevada, it is not possible to determine if changes in the distribution and amount of habitat have resulted in changes in owl distribution and abundance. As summarized in the Technical Report, the current distribution and abundance of owls in the Sierra Nevada does not suggest that they have declined in their

FEIS Volume 3, Chapter 3, part 4.4, page 76 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 overall distribution in the Sierra Nevada or that they have markedly declined in abundance within any forest type (Verner et al. 1992). Uncertainty exists regarding the suitability of the more open, parklike forest structure that occurred over a larger proportion of the historic landscape for California spotted owls and it has been hypothesized that owls may be more abundant in certain areas of the Sierra Nevada at the present time compared to 100 years ago (Verner et al. 1992).

Assessing changes in habitat quality for California spotted owls is further complicated by the history of vegetation management strategies predominantly used in the Sierra Nevada. In the Pacific Northwest, clearcutting was the predominant timber harvest method used and forest vegetation is generally either owl habitat (i.e., uncut or mature second-growth stands) or not owl habitat (i.e., recent clearcuts or young second-growth stands). Thus, the differences between owl habitat and non-habitat are relatively more straightforward to discern. In contrast, historic timber harvest in the Sierra Nevada was predominantly individual tree selection or high-grading focused on the most desirable trees (mostly large diameter pines). This harvest system, coupled with aggressive fire suppression, resulted in a finer-scale gradient in forest conditions, and subsequently, habitat quality for California spotted owls varies over a continuous gradient of quality (Noon and McKelvey 1996). Thus, compared to the Pacific Northwest, it is more difficult to discern discrete boundaries between habitat and non-habitat.

Current Estimates. California spotted owl occurrence and productivity appears to be significantly correlated with canopy cover composition within owl home ranges. Using the findings reported in Hunsaker et al., the habitat status surrounding individual spotted owl activity center in the Sierra Nevada has been evaluated. The proportion of moderate and dense canopied stands occurring within individual spotted owl home range areas was estimated using Forest Vegetation Inventory data calculated within a circular analysis area surrounding each spotted owl activity center. The analysis area used was equal in size to the breeding season pair home range sizes calculated from various studies and reported above. Based on this analysis, approximately 50 percent of spotted owl home ranges have less than 60 percent of their home range in moderate and dense canopied habitat (approximated as CWHR classes 6, 5D, 5M, 4D, and 4M). Considering the findings reported in Hunsaker et al. (in press), habitat associated with these owl sites may be insufficient to support a self-sustaining population of owls. Fifty-eight percent of owl sites in the Central Sierra Nevada (represented as the Plumas, Tahoe, Eldorado, and Stanislaus national forests) have less than 60 percent of their home range in moderate and dense canopied forest, whereas 32 percent of owl sites in the southern Sierra Nevada (Sierra and Sequoia national forests) have less than 60 percent of their home range in moderate and dense canopied stands. This analysis assumes private lands do not contribute to the proportion of moderate and dense canopied habitat within home ranges, since the future status of that habitat remains uncertain.

At the stand scale, studies described in this assessment have documented that the majority of nest sites occur in CWHR classes 6, 5D, and 5M. Timber strata 4G (similar to CWHR classes 5D and 6) has been documented as being preferentially selected by owls for foraging. These vegetation classes have the highest probability of providing stand structures associated with preferred nesting, roosting and foraging.

FEIS Volume 3, Chapter 3, part 4.4, page 77 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Diet Large differences in observed home range sizes of spotted owls may result from differences in diet. In addition, although not confirmed for California spotted owls, studies of many other owl species from around the world confirm that whether a given pair of owls attempts to nest in a given year, and whether nest attempts are successful, are directly related to prey availability (Verner et al. 1992:74). Understanding how prey availability differs as habitat structure changes is essential to understanding how to manage spotted owl populations by providing suitable habitat for their prey.

Spotted owls above the mid-elevation conifer forests of the Sierra Nevada (about 4,000 to 5,000 feet) prey mainly on flying squirrels. Owls in the mid- to lower elevations of the mixed- conifer zone and the upper part of the ponderosa pine zone prey heavily on both flying squirrels and woodrats. Owls in the Sierra Nevada foothill riparian/hardwoods consume primarily woodrats. The Technical Report concludes that it is important to manage habitat to maintain thriving populations of flying squirrels in Sierra Nevada conifer forests since over 75 percent of all California spotted owl sites occur in these habitats where flying squirrels are the primary prey species (Verner et al. 1992:69). Managing conifer forests in the Sierra Nevada for flying squirrels should emphasize retention of very large and old trees, large snags, and large downed logs, as well as site preparation techniques that minimize disturbance of the soil organic layer.

Breeding Chronology The spotted owl breeding cycle extends from about mid-February to mid- to late September. Egg-laying through incubation, when the female spotted owl must remain at the nest, extends from early April through mid- to late May. Young owls typically fledge from the nest in mid- to late June. In the weeks after fledging, the young are very weak fliers and remain near the nest tree. Adults continue to bring food to the fledglings until mid- to late September. The sensitivity of the spotted owl to nest site disturbance is not well-known. Wasser et al. (1997) found measurements of physiological stress to be significantly higher in owls centered within versus beyond 0.25 miles from a major logging road, or recent timber harvest activity.

Not all pairs of California spotted owls nest every year. In fact, over the ten years of demographic studies in the Sierra Nevada, 1992 was the only year when nearly all study owls nested. It is not unusual for owls in an established activity center to skip several years between one nesting and the next. Sites may be vacant for several consecutive years when the population is in decline, but then be reoccupied to support breeding pairs during a population upswing.

Migration and Dispersal. Information on the dispersal abilities of spotted owls is scant. An understanding of juvenile owl dispersal is essential to understanding effects of increasing distance or spacing between adjacent pairs of spotted owls. Information on the northern spotted owl included in Chapter 4 of the Technical Report, indicates that two-thirds of the juveniles would be expected to disperse at least eight miles.

Unlike northern spotted owls, many California spotted owls migrate altitudinally, moving downslope for the winter. Spotted owls migrated a mean straight-line distance of twenty miles in the Eldorado National Forest and a mean of 12.3 miles in the Sierra National Forest (Verner, et al. 1992:64). Three studies tracked 32 California spotted owls to determine whether they

FEIS Volume 3, Chapter 3, part 4.4, page 78 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 migrated: 44 percent were altitudinal migrants. The reasons why only some individuals migrate are unclear. Altitudinal migration seems not to be related to habitat quality, at least in these studies. Migration may expose owls to greater risk of mortality. Human use and development in traditional wintering areas may also have adverse impacts on the quality of owl habitat.

Risk Factors 1. Habitat Abundance and Distribution. Habitat Abundance. The current distribution and abundance of owls in the Sierra Nevada does not suggest that habitat has markedly declined in abundance within any forest type. However, if habitat abundance is evaluated based upon new information about the proportion of habitat with specific canopy cover values that contribute to or detract from occurrence and productivity, the current abundance and distribution of habitat appears to be of concern. Fifty percent of owl sites in the Sierra Nevada (58 percent in the central Sierra Nevada) have less than 60 percent of their home range in moderate and dense canopied forest, indicating potentially lower productivity for these sites.

The Technical Report cautioned against future management increasing the mean nearest- neighbor distances among spotted owl sites. The Report stated that subtle factors that uniformly decrease habitat quality or increase fragmentation would act to reduce population density and incrementally increase the uncertainties associated with successful dispersal and mate finding. If suitable habitat is allowed to decline and become fragmented, the uncertainty of successful dispersal will become progressively more relevant to the subspecies long term population dynamics and likelihood for persistence (Verner et al. 1992:184).

Geographic Areas of Concern. As summarized in the Technical Report, the current distribution of spotted owls in the Sierra Nevada is characterized by its continuity and relatively uniform density. The Technical Report, nonetheless, described five conditions which give rise to some concern for the integrity of the California spotted owl’s range in the Sierra Nevada (1) bottlenecks in distribution of habitat or owl populations; (2) gaps in the known distribution of owls; (3) locally isolated populations; (4) highly fragmented habitat; and (5) areas of low crude density of spotted owls. Nine areas in the Sierra Nevada were identified in the Technical Report as areas where one or more of these conditions currently limit the owl population. These areas of concern were thought to indicate potential areas where future problems may be greatest if the owl’s status in the Sierra Nevada were to deteriorate. They represent areas where management decisions may have a disproportionate potential to affect the California spotted owl population. Of particular concern are areas of checkerboard ownership and large inclusions of non-federal lands which occur on the Tahoe, Eldorado, and Stanislaus national forests. Habitat projections in areas of checkerboard ownership are highly uncertain and the existing condition is often highly fragmented. The risk and uncertainty associated with maintaining a well-distributed population is certainly higher within these areas of concern.

2. Habitat Quality. Much of the current concern regarding California spotted owl population trends is focused on the effects of vegetation management on the distribution and abundance of important habitat elements. Forest ecologists estimate that old forest conditions have declined from 50 to 90 percent compared to the range of historical conditions. Analyses conducted at both the plot

FEIS Volume 3, Chapter 3, part 4.4, page 79 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 and landscape scales have documented large reductions in late-seral/old-growth forests throughout the Sierra Nevada and reductions in the numbers and distribution of large trees as a result of selective harvesting of large pines, and increases in the numbers of smaller diameter trees, forest understory density, and surface fuels as a result of fire suppression (Laudenslayer 1990, McKelvey and Johnston 1992, Franklin and Fites-Kaufmann 1996, Beardsley et al. 1999, Bouldin 1999). These trends indicate that there has been a reduction in the amount and distribution of mature and older forests, and specific habitat elements such as large trees, snags, and downed logs, used for nesting and foraging by California spotted owls.

The Technical Report discusses five major factors of concern for California spotted owl habitat that have resulted from historical timber harvest strategies: (1) decline in the abundance of very large, old trees, (2) decline in snag density; (3) decline in large-diameter logs; (4) disturbance or removal of duff and topsoil layers; and (5) change in the composition of tree species. Of these concerns, significant changes in diameter distributions of trees in the Sierra Nevada and rapid reductions in the distribution and abundance of large, old, and decadent trees posed the greatest threat to the California spotted owl (Verner et al. 1992). This factor relates to two other factors--the decline in snag density and loss of large-diameter logs. The diameter of nest trees selected by owls in the Sierra Nevada is significantly greater than the average diameters of conifers in the Sierra Nevada. Large trees suitable for owl nesting contribute to the overall quality of owl habitat. Large trees become future large snags and large downed logs, the latter providing important habitat attributes for some prey species. The length of time required to recover old trees and increase their density over the landscape raises the level of concern associated with their decline.

Research by North et al. (2000), suggests that California spotted owl reproduction is influenced by both regional weather conditions and nest-site canopy structure, which protects fledglings from detrimental weather. Forest management practices which do no provide for retaining groups of large, old, high crown-volume trees may be reducing the number of potential owl nesting sites within the forest. Research on the northern spotted owl (North et al. 1999) found snag volume, foliage volume, and canopy layering to be stand attributes significantly associated with owl foraging intensity. Vegetation treatments, such as timber harvest and fuels reduction, that alter these habitat attributes may influence habitat quality for the California spotted owl.

3. Climate Weather conditions, especially late winter and early spring storms, may have been important in the observed downward population trends. Research by Franklin et al. (2000) and North et al. (2000) indicate the importance of annual weather events in determining nesting success and survival. Temporal variation in reproductive output appears to be greatly influenced by annual variation in weather, primarily severe storms during the incubation and nestling periods, although it appears that higher quality habitat, at multiple spatial scales, can function to moderate weather effects on owls (Franklin et al. 2000, North et al. 2000). In demographic studies in the Sierra Nevada, failure of nesting owls has repeatedly been noted following cold, wet storms in the spring months (Verner, pers. comm. 1999). Reproductive rates of owls studied in the Klamath province were found to be negatively affected by increased winter precipitation and positively affected by increased precipitation during the early spring. Survival declined as climatic conditions progressed from an optimal warm, dry spring to a cold, wet spring (Franklin et al. 2000). Franklin et al's results indicated, however, that high

FEIS Volume 3, Chapter 3, part 4.4, page 80 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 habitat quality increased survival of territory occupants above individuals in poorer habitats. If the frequency of El Nino events increases with global warming, habitat conditions at nest-sites could be increasingly important for sustaining spotted owl reproduction (North et al. 2000). Annual variation in prey availability, which may be partially linked to weather, and other factors may also be important.

4. Wildfire The ingrowth of shade-tolerant tree species and the excessive buildup of surface fuels are conditions that have resulted from past forest management and fire suppression, and which increase the risk of high-severity fire--an infrequent and small-scale event before 1850. The Science Report cited studies indicating that the number of multilayered stands with a large proportion of younger trees, and the amount of small downed material has likely increased since 1850. A change in fire severity has resulted, increasing the risk that wildfires will destroy owl habitat. The Science Review noted that twentieth century fire records from the Sierra Nevada show that fire risk is inversely related to elevation (McKelvey and Busse 1996). At the landscape or watershed scale, fire-severity patterns correspond to logging history, stand fuel treatments, topography, and forest type (Weatherspoon and Skinner 1995). These findings indicate that risks and hazards for owl habitat changes with elevation, topography, and forest condition. Such findings are important considerations in designing or implementing long-term management plans for the spotted owl.

Approximately 39 percent of the known owl sites on national forest lands occur in areas of high fire hazard risk (Table 4.4.2.1d). These high fire hazard risk sites include 38 percent of the known national forest pairs, 44 percent of the territorial singles, and 36 percent of the single birds. The known number of California spotted owl sites burned in recent wildfires is low. From 1993 through 1998, only 15 California spotted owl protected activity centers (PACs) or spotted owl habitat areas (SOHAs) burned in wildfires. (Three of the 15 are known to remain occupied.). This represents an annual rate of loss of about 0.2 percent of the PACs and SOHAs on national forests in the Sierra Nevada over a 6-year period. Only limited inferences may be drawn from this short period.

Table 4.4.2.1d. Distribution of known California spotted owl sites by reproductive status and fire hazard risk rating. Fire Hazard Risk Rating (Hazard Class) Reproductive Status Low (3& 4) Moderate (5 & 6) High (7-9) Total Pairs 137 409 333 879 Territorial Singles 40 107 115 262 Singles 38 84 74 196 Total 215 600 522 1337

5. Breeding Site Disturbance Management as well as recreational activities has the potential to disrupt spotted owl nesting efforts and reproductive success. In recent years this risk has been diminished by applying protections to known nest stands and limiting disruptive activities during the spotted owl breeding season within a distance of known nest sites. Habitat disturbance surrounding the nest site has been diminished through designation and protection of 300 acre PACs.

FEIS Volume 3, Chapter 3, part 4.4, page 81 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Conservation Measures Conservation measures for the California spotted owl must provide the environmental conditions needed to establish a high likelihood of maintaining populations of the California spotted owl, well- distributed across the National Forests within the Sierra Nevada planning area. A primary concern for California spotted owls is the effects of vegetation management on the distribution, abundance, and quality of habitat (Verner et al. 1992, Gutierrez and Harrison 1996, Noon and McKelvey 1996). Conservation measures must consider habitat distribution, abundance, and quality at the landscape, home range, and stand-level scales.

At the landscape scale the issue is to provide for sufficient amounts and distribution of high quality habitat to facilitate natal and breeding dispersal among territories and to maintain California spotted owls well-distributed throughout their historic range in the Sierra Nevada. For this purpose, protecting occupied, as well as suitable but unoccupied habitat, over the long- term is important at this scale. A species with obligate dispersal and experiencing habitat limitation would be expected to show a pattern of less than full occupancy of habitat due to the uncertainty of the search process and the survival costs associated with searching for low-density habitat (Noon, pers comm. 2000). Conservation efforts should therefore consider not only occupied habitat, but also suitable unoccupied habitats, in developing conservation strategies for species for which dispersal may function as a primary limiting factor (Lande 1987, 1988).

At the spatial scale of the individual home range, the issue is to manage for high quality territories that provide sufficient amounts and distribution of nesting and foraging habitat to provide for adequate survival and reproduction rates needed to contribute to stable or increasing populations. At finer scales, conservation measures must also address the amounts and distribution of important habitat elements. Specifically, canopy cover and layering, and large trees and their derivatives, large snags and logs, are important habitat elements that influence the distribution, abundance, and availability of California spotted owl prey species. Individual large trees, and often snags, are used as nesting trees by California spotted owls. Spotted owl foraging success is likely to be improved in stands with high tree height diversity, providing perches at varying levels in the canopy.

Environmental Consequences This analysis considers six primary features of the alternatives which would influence identified risk factors and environmental outcomes for the California spotted owl: (1) Distribution of owl sites among various land allocations, (2) provisions for protection of known or potential nest stands, (3) provisions for habitat abundance at the landscape and home ranges, (4) levels and types of forest management activity, (5) standards and guidelines addressing important elements of habitat quality, and (6) levels of natural disturbance.

1. Distribution of owl sites among land allocations. A. Proportion of owl sites occurring in land allocations where vegetation treatments are limited. Each alternative includes one or more large land allocations where vegetation treatments would be limited (in other words, no vegetation treatments could be conducted or treatments would be limited to prescribed burning or light thinning). These allocations include: wilderness areas, old forest emphasis areas, biodiversity reserves, and large unroaded areas. Where California spotted owl activity centers occur within one of these allocations, vegetation treatments are less likely to degrade habitat suitability within and surrounding a given home range area. Opportunities for maintaining and improving old forest habitat conditions for spotted owls are increased. Providing for a portion of the owl population that will be minimally affected by

FEIS Volume 3, Chapter 3, part 4.4, page 82 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 habitat alteration increases confidence for at least two reasons. First, the response of California spotted owls to vegetation treatments remains largely unstudied (Verner et al. 1992). In particular, uncertainty remains regarding how each of the different vegetation management treatments (e.g., mechanical thinning, prescribed fire, CASPO harvest, etc.) affects the distribution, abundance, and availability of prey to California spotted owls. Secondly, trends for increasing amounts of habitat for the northern spotted owl did not show a threshold value above which little or no increase in productivity or survival occurred. This suggests that removing any suitable habitat (50 percent canopy cover or higher) within the vicinity of the nest tends to reduce the productivity and survivorship of owls (Bart 1995). This also suggests that it should not be assumed that habitat in all home ranges can be reduced to a threshold level without adverse effects on the population. The need to maintain and provide for higher than threshold amounts of habitat within some portion of the spotted owl home ranges is apparent in the Sierra Nevada where data indicates that approximately half of the owl home ranges have less than desired amounts of habitat to begin with.

Table 4.4.2.1e displays the percentage of California spotted owl sites (activity centers) that occur in allocations where limited vegetation treatments are allowed under each alternative. In Alternatives 2, 3 5, and 8 more than half the known spotted owl sites occur in these allocations; slightly less than half do in Alternatives 6 and Modified Alternative 8. In Alternatives 1, 4 and 7, less than 5 percent of owl sites occur outside of allocations where vegetation management would be limited to light thinning.

Table 4.4.2.1e. Percentage of California spotted owl activity centers in land allocations where limited vegetation treatments are allowed. Alt 1 Alt 2 Alt 3 Alt 4 Alt 5 Alt 6 Alt 7 Alt 8 Mod 8 Percentage of spotted owl sites 4 86 54 4 67 46 4 58 49

B. Effects specific to implementation of the Herger-Feinstein Quincy Library Group Project Risk and uncertainty associated with implementing vegetation treatments is reduced by including a substantial proportion of the owl population within an old forest allocation where limited treatments would occur. Implementation of Alternative 2 of the HFQLG Project would include a smaller proportion of owl sites in allocations where vegetation treatments are limited. On the Lassen, Plumas and a portion of the Tahoe national forests, treatments would not occur within Rank 4 and 5 Late Successional Old Growth polygons mapped by the Sierra Nevada Ecosystem Project. These polygons are smaller areas of high quality old forest, distributed within larger old forest emphasis areas. As such, they are unlikely to incorporate all habitat used by a number of spotted owl sites, and do not provide the same “buffer” for risk and uncertainty to the owl population as is provided by the larger old forest emphasis areas under Framework alternatives 2, 3, 5, 6, 8, and modified 8,. In the absence of a strategy which provides for a portion of the owl population to be minimally affected by habitat alteration, the effects of implementating the HFQLG Forest Recovery Act is more uncertain.

2. Provisions for protection of known or potential nest stands. A. Survey Requirements Survey efforts to inventory California spotted owls throughout the Sierra Nevada are incomplete to date and some unknown number of breeding territories have not been documented. An additional 160-220 territories are estimated on FS lands based on unsurveyed

FEIS Volume 3, Chapter 3, part 4.4, page 83 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 suitable habitat. The likelihood of locating (and subsequently protecting) these additional owl territories is lowest under Alternative 1 since surveys are not required. Under Alternative 1 the number of known owl territories would increase slowly over time as new locations were opportunistically discovered. All action alternatives establish standards requiring owl surveys to protocol for all activities that occur in suitable nesting habitat. The proportion of owl breeding territories and nest stands known and potentially protected would be higher under these alternatives.

B. Proportion of California Spotted Owl Breeding Territories Protected All alternatives maintain the existing 1050 PACs established through the 1993 Interim Guidelines for the California Spotted Owl and establish PACs for the approximately 260 spotted owl sites within SOHAs. Unlike alternative 1, all of the action alternatives provide protection for owl sites discovered subsequent to 1992. Alternative 1 maintains the existing number and distribution of PACs (those established in 1993 under the Interim Guidelines), but does not require the creation of additional PACs for newly discovered owl sites. Under this alternative, a biological evaluation would evaluate whether a PAC should be created or a new PAC substituted for an existing PAC, but the absence of criteria for making this determination results in uncertainty about the outcome. Given declining population trends, there is greater risk associated with limiting protection to only a subset of known spotted owl nest sites. The potential for increasing nearest neighbor distances between owl sites is greater under Alternative 1, increasing uncertainties associated with effective dispersal and mate-finding. An additional concern is that the existing number of PACs will decline over time as nest stands are lost to wildfire.

The action alternatives all require the establishment of PACs for newly discovered owl sites (including those discovered between 1992 and present). These alternatives would protect known spotted owl nest sites, and allow the creation of additional PACS, offsetting the potential loss of PACS to wildfire.

All action alternatives maintain PACs in the network unless they are rendered unsuitable by wildfire and protocol surveys indicate that they are no longer occupied. This is important given the high temporal variability of California spotted owl reproductive rates. Owl populations may go through periodic declines followed by major breeding pulses. The loss of available nest sites that become unoccupied during periods of decline may preclude population expansion following breeding pulses. This, in turn, may result in declining populations with lower likelihood of persistence over time. The Sierra Sequoia/Kings Canyon study area documented one instance where an owl site remained unoccupied for eleven consecutive breeding seasons prior to becoming reoccupied (Verner, pers. comm., 1999).

C. Size and Configuration of Protected Activity Centers. The size of PACs and delineation of habitat within PACs is the same across all Alternatives. PACs are 300 acres in size and should consist of the best available habitat, including known and suspected nest stands, in as compact a unit as possible.

A PAC size of 300 acres is based on two criteria provided in the CASPO report that determined that: 1) nest stands plus adjoining suitable nest, and 2) activity centers that encompassed 50% of radio-telemetry locations from owls on the Sierra National Forest, were both approximately 300 acres (Verner et al. 1992:87). Activity centers are areas within which

FEIS Volume 3, Chapter 3, part 4.4, page 84 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 owls find suitable nesting sites and several suitable roost sites, and in which they do a substantial amount of foraging. Uncertainty exists regarding the general adequacy of 300 acre PACs to represent owl activity centers given variation in habitat conditions across the range of the owl in the Sierra Nevada. Radio-telemetry studies conducted on the Lassen National Forest indicated that areas including 50% of foraging location averaged 788 acres (Verner et al. 1992). Subsequent analysis by Bingham and Noon (1997) indicated that core use areas averaged 2,300 acres within individual California spotted owl home ranges on the Lassen National Forest. The degree to which 300 acre PACs will be adequate to maintain owl territories will depend on the distribution and abundance of suitable habitat surrounding PACs. PACs alone are not an adequate conservation strategy for maintaining a viable population of owls. They are important because they do provide protection to nest sites. However, the distribution and abundance of owl habitat around PACs and across the landscape are critical considerations that will determine the ultimate adequacy of a PAC-based conservation strategy for maintaining owl viability in the Sierra Nevada.

Modified alternative 8 does the best job of addressing this issue by applying additional habitat protection measures within a larger home range core area surrounding the PAC within the General Forest allocation (Hunsaker et al., in press). In this alternative, habitat alteration in the general forest is limited within a larger area that is expected to represent the overused portion of the spotted owl’s home range (ranging in size from 2,400 acres in the northernmost portion of the Lassen National Forest, to 600 acres in the southern Sierra Nevada). Alternative 2 largely addresses the issue through incorporation of the vast majority of owl sites in biodiversity reserves.

D. Management within Protected Activity Centers. All Alternatives limit activities within PACs to those designed to improve the suitability or integrity of the PAC. Fuels treatment and vegetation management that can occur within PACs differ between Alternatives. There are two main issues concerning vegetation treatments in PACs. One regards the uncertainty that exists related to the trade-off between treating PACs, with the goal of reducing their susceptibility to stand replacing fires, versus the potential negative effects of treatments on California spotted owl occupancy and habitat quality. It seems reasonable to hypothesize, given historic fire patterns in the Sierra Nevada, that light underburns similar to those likely to have occurred prior to the late 1800s, would not typically result in territory abandonment. However, no studies have been conducted that address the effects of fuels treatments on California spotted owl occupancy, survival, and reproduction in PACs. The second issue regards uncertainty about how different treatments or combinations of treatments affect fire risk and severity within PACs or in areas surrounding PACs. This uncertainty stems from differences in the ability of mechanical thinning and/or prescribed fire to reduce surface fuel loads and subsequent risk of stand-replacing wildfire (Weatherspoon et al. 1992, Van Wagtendonk 1996). Thus, given both types of uncertainty it is difficult to evaluate the consequences of the different proposed treatments in the Alternatives.

Given this scientific uncertainty, all alternatives entail degrees of risk associated with the management of PACs. Alternatives that provide the opportunity to address this uncertainty through the use of formal adaptive management projects to better understand the effects of treating PACs have the potential to advance our scientific knowledge and improve our ability to manage California spotted owl habitat. Uncoordinated efforts to conduct treatments within PACs outside of a formal adaptive management framework would provide only limited

FEIS Volume 3, Chapter 3, part 4.4, page 85 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 opportunity to measure management effects at best and would result in a squandered opportunity to address an important area of scientific uncertainty. Additionally, the rates at which treatments would occur also need to be factored into the assessment. Treating a high proportion of territories without knowledge of how treatments will affect habitat quality nor how they will affect short and long-term fire risk, and without a formal monitoring strategy, entails high risk and uncertainty. Alternatives 2, 4, 5, 6, 7 and 8 require formal adaptive management approaches to assess the effects of treatments on PACs.

Alternative 2 permits no habitat altering activities within PACs unless associated with a formal research project. Only light underburning and hand clearing are permitted in Alternative 5 and Modified Alternative 8, unless associated with a formal research project to assess effects. Thus, these alternatives provide the strongest direction to establish an adaptive management strategy for addressing uncertainty. Alternative 8 permits light underburning and mechanical treatment of trees <12” dbh, with a BE required to justify management activity within a PAC. Alternatives 1, 3, 4, 6, and 7 permit sufficient fuels treatment in up to 30% of the PAC to meet fuels objectives while attempting to minimize reductions in habitat suitability. Alternatives 1 and 3 do not require establishment of an adaptive management strategy. A greater number of PACs could be expected to be affected annually under Alternatives 1 and 3 relative to the other alternatives. Alternatives 4, 6, 7, 8 and Modified 8 allow treatment in no more than 10% of PACs per decade per National Forest; if Forests plan treatments affecting a greater proportion of PACs, they must occur as part of a formal research study designed to monitor treatment effects. Given the lack of scientific knowledge about treatment effects, it is not possible to evaluate outcomes unless a formal adaptive management strategy is implemented.

Since the primary intent of treatments within PACs is to reduce the risk of loss to high-severity fire, this risk should be weighed against the uncertainty of such treatments. The risk of losing PACs to high-severity fire is uncertain, and likely varied considerably among PACS. The annual rates of loss are about 0.2 percent of the PACs/SOHAs on national forests in the Sierra Nevada over the past 6-years. While the observed rate seems low, six years is a very short period from which to draw inferences.

E. Effects specific to forests implementing the Herger-Feinstein Quincy Library Group Project Alternative 2 of the HFQLG project would continue to manage existing spotted owl PACs. Lacking requirements to survey for owls or to establish PACs for newly discovered owl sites, however, implementation of HFQLG raises the concern that not all owl sites would receive PAC protection on the Lassen, Plumas, and Tahoe National Forests, as vegetation treatments are implemented over the next 5 years. Over the 5-year timeframe of this project, there would be greater potential for increasing nearest neighbor distances between owl sites on these forests, increasing uncertainties associated with effective dispersal and mate-finding.

3. Provisions for habitat abundance at the landscape and home range scales. A. Modeled Changes in Habitat Abundance. Modeled projections of the change in vegetation classes over time can provide a coarse mechanism for evaluating changes to the overall amount of spotted owl habitat under each alternative. Habitat projections in this assessment are limited to projected changes in the total abundance of California spotted owl habitat under the different alternatives across individual national forests or other large areas within the overall range of the owl in the planning area. It

FEIS Volume 3, Chapter 3, part 4.4, page 86 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 is not appropriate to use vegetation projections generated in this assessment at small scales, such as individual home ranges.

Trends are projected for larger landscapes where statistical sampling properties of the basic information layers improve the accuracy of the comparisons. While vegetation estimates are reported for large areas, these estimates are based are stand-level plots. These plots are congruent with how owls might perceive their environment. Thus, a reported increase in average tree diameter or canopy closure can be correctly interpreted to mean that more stands across the reporting unit will have larger trees and higher canopy closure in the future than today. Furthermore, it is expected that California spotted owls will perceive these differences. This does not suggest, however, that every stand everywhere will exhibit meaningful increases, or even that the average tree diameter or canopy closure in every home range will be higher. The end result is individual variations from the central tendency; some smaller areas are expected to be below average and some above. The concern is that some of these below (or above) average areas may be concentrated or connected in various parts of the landscape, leading to a more patchy distribution of habitat than apparent in the aggregate measures. Indeed, the effect of wildfire or spatially concentrated fuel treatments may create this result. This potential is discussed in following sections.

Approximately 84% of 292 California spotted owl nest vegetation plots were classified as CWHR classes 6, 5D, 5M, 4D, and 4M (see Affected Environment). These CWHR types are also rated as providing high and moderate suitability foraging habitat for owls based on the expert opinion habitat relationship models contained in the CWHR database. Thus, available evidence indicates that these CWHR classes provide high and moderate suitability nesting and foraging habitat for California spotted owls. The regionwide overall trends in total projected amounts of nesting and foraging habitat (CWHR classes 4M, 4D, 5M, 5D, 6) suggest slightly increasing trends over the next 50 years across all alternatives (Figure 4.4.2.1b). However, greater changes in the relative differences within each of the five individual CWHR strata are projected to occur over the next 50 years (Table 4.4.2.1f, Figures 4.4.2.1c to 4.4.2.1g). In general, classes 4M and 4D are transitioning into classes 5M and 5D through growth. As a consequence of the general increasing projected trends in CWHR strata 5D, 5M, and 6, the overall habitat suitability for California spotted owls in the Sierra Nevada based on the existing CWHR habitat suitability models exhibit increasing trends across all alternatives (Table 4.4.2.1g, Figure 4.4.2.1h). All of the above projected trends are similar across Alternatives, although the magnitude of the differences among Alternatives is difficult to interpret with confidence due to uncertainty associated with the vegetation information and the assumptions that underpin the modeling process. Greater differences among Alternatives are observed over longer time periods. However, confidence in these longer-term future projections is further lowered due to additional uncertainty regarding future conditions.

FEIS Volume 3, Chapter 3, part 4.4, page 87 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Figure 4.4.2.1b. Region-wide projected acres of pooled CWHR classes 4M, 4D, 5M, 5D and 6.

Region-wide, all tree types 4D,4M,5D,5M,6

1 6000000 2 5000000 3 4 4000000 5 3000000 6 7 2000000

Acres of conifer types 8 1000000 MLV Pref 0 0 2 4 6 8 10 12 14 16 Decade

FEIS Volume 3, Chapter 3, part 4.4, page 88 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Figure 4.4.2.1c Region-wide projected change in CWHR class 5D.

1 5D 2 2,000,000 1,800,000 3 1,600,000 1,400,000 4 1,200,000 5 1,000,000

Acres 800,000 6 600,000 400,000 7 200,000 0 8 0 5 10 15 MLV Decade Pref

Figure 4.4.2.1d. Region-wide projected change in CWHR class 5M.

1

5M 2 2,000,000 1,800,000 3 1,600,000 4 1,400,000 1,200,000 5 1,000,000

Acres 800,000 6 600,000 7 400,000 200,000 8 0 MLV 0 5 10 15 Decade Pref

FEIS Volume 3, Chapter 3, part 4.4, page 89 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Figure 4.4.2.1e. Region-wide projected change in CWHR class 4D.

1 2,000,000 2 1,800,000 4D 1,600,000 3 1,400,000 4 1,200,000 5 1,000,000 6

Acres 800,000 7 600,000 8 400,000 MLV

200,000 Pref 0 0123456789101112131415

Decade

Figure 4.4.2.1f. Region-wide projected change in CWHR class 4M.

1 4M 2 2,000,000 1,800,000 3 1,600,000 4 1,400,000 1,200,000 5 1,000,000

Acres 6 800,000 600,000 7 400,000 8 200,000 0 MLV 0123456789101112131415 Pref Decade

FEIS Volume 3, Chapter 3, part 4.4, page 90 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Figure 4.4.2.1g. Region-wide projected change in CWHR class 6.

1 6 2 2,000,000 1,800,000 3 1,600,000 1,400,000 4 1,200,000 5 1,000,000

Acres 800,000 6 600,000 400,000 7 200,000 8 0 0123456789101112131415 MLV Decade Pref

Figure 4.4.2.1h. Region-wide projected change in overall CWHR habitat suitability units for the California spotted owl.

3000000 alt_1 2500000 alt_2 2000000 alt_3 1500000 alt_4 1000000 alt_5 alt_6 Habitat Units 500000 alt_7 0 alt_8 1 3 5 7 9 11 13 15 Mod-8 mlv Decade

FEIS Volume 3, Chapter 3, part 4.4, page 91 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Table 4.4.2.1f. Projected percent changes in the amount of high and moderate suitability spotted owl nesting and foraging habitat from the current to 50 years in the future under the FEIS Alternatives. CWHR Strata Alternative 6 5D 5M 4D 4M Total Current 1,120 166 662 1,145 1,206 4,301 (1,000s acres) MLV* 21.5 341.9 46.7 -38.6 -37.3 5.4 1 1.7 – 386.5 77.5 -40.9 -32.8 7.3 2 23.9 382.3 60.9 -38.7 -35.3 10.2 3 9.7 552.6 84.4 -41.8 -29.0 17.6 4 -1.4 – 388.2 107.2 -47.7 -26.2 11.0 5 16.0 434.6 67.7 -42.3 -32.5 10.8 6 13.5 500.9 88.7 -41.8 -31.2 16.7 7 13.2 400.8 85.7 -40.6 -28.9 13.3 8 19.6 401.8 72.7 -41.0 -33.5 11.4 Mod-8 18.4 454.7 67.3 -39.7 -34.2 12.8 Mean Change 13.6 424.4 75.9 -41.3 -32.1

*MLV = No Treatment, Let-Grow Scenario

Table 4.4.2.1g. Projected percent changes in overall habitat suitability scores based on CWHR habitat models from the current to 50 years in the future across the FEIS Alternatives. Alternative MLV* 1 2 3 4 5 6 7 8 Mod.-8 Mean 9.9 25.8 15.0 32.9 34.3 19.6 31.7 27.0 19.7 27.9 24.4

*MLV = No Treatment, Let-Grow Scenario

B. Amount of Habitat Provided in Owl Home Ranges Studies have documented a relationship between the proportion of a landscape covered by habitat and the ability of spotted owl pairs in that landscape to survive and reproduce at replacement rates (Bart, 1995, Franklin et al. 2000, Hunsaker et al. in press). Given declining owl populations, this relationship is particularly important for evaluating opportunity for a particular alternative to stabilize population declines. Existing information suggests that approximately half of spotted owl home ranges in the Sierra Nevada currently provide the amount of moderate- and dense-canopied stands found to be associated with higher levels of owl occupancy and productivity (CWHR classes 6, 5D, 5M, 4D and 4M). On average, spotted owl home ranges in the northern and southern Sierra Nevada provide higher amounts of habitat than those in the central Sierra Nevada, due in part to more contiguous national forest land. Increasing the number of owl sites with desired amounts of habitat is likely important to stabilizing current population declines.

Modeled habitat projections indicate that all alternatives contribute to increasing amounts of spotted owl habitat over a 50-year timeframe. These broad-scale projections do not, however, ensure that the distribution of this habitat will be sufficient to maintain occupancy or productivity within individual spotted owl sites. Alternatives 1, 3, 4, 6, and 7 lack direction that specifies amounts of habitat to be retained within the specific areas known to be utilized by spotted owls (i.e. home range areas). In the absence of this direction, habitat distribution remains a major area of uncertainty in these alternatives. It is difficult to determine the extent

FEIS Volume 3, Chapter 3, part 4.4, page 92 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 to which these alternatives will provide habitat likely to support higher levels of owl occupancy and productivity.

Alternatives 2, 5, and 8 provide a greater degree of certainty that sufficient habitat will be retained within spotted owl home ranges. Alternative 8 retains all existing suitable habitat while Alternatives 2 and 5 provide direction for maintaining 50 percent of each California spotted owl home range in suitable habitat. Only Modified Alternative 8, however, provides suitable habitat within the most used core area surrounding the PAC. (This concern is minor for alternative 2, however, since treatments affect such a small proportion of the landscape).

A number of studies indicate that habitat concentrated in close proximity to the nest or activity center is of greater value since owls are known to concentrate their foraging activities close to the nest (Solis and Gutierrez 1990, Bart 1995, Bingham and Noon 1997, Hunsaker et al. in press). This observation is supported by analyses for the California spotted owl that find greater concentrations of suitable habitat as the analysis area becomes smaller in size surrounding the owl nest site (Hunsaker et al. in press). Alternatives 2, 5 and 8 provide a greater degree of certainty about retention of habitat within home ranges occurring in the general forest, than do the remaining alternatives. Modified alternative 8 increases the effectiveness of this habitat protection, however, by providing direction that would concentrate high quality habitat within a core area closest to the activity center.

C. Amount of Habitat Provided Within Owl Home Ranges Occurring in Geographic Areas of Concern. As described in the Technical Report, several geographic areas of concern for the California spotted owl occur throughout the Sierra Nevada (Verner et al. 1992:45, 47, 48). The Technical Report cautioned that these are areas where management decisions may have a disproportionate potential to affect the spotted owl population. Given documented population declines, the extent to which alternatives provide sufficient habitat to maintain spotted owl sites within the areas of concern is an important consideration. None of the alternatives provides unique management direction specific to these areas. Alternatives which lack objectives for habitat maintenance in spotted owl home ranges (Alternatives 1, 3, 4, 6, and 7), lack assurances that vegetation treatments will not reduce the occupancy and productivity of owl sites in these areas. This is particularly the case in the Areas of Concern that include checkerboard land ownership patterns or fragmented habitat. Management actions have potential to disproportionately impact owl sites in these areas given the existing status of habitat. With past and continuing habitat alteration on non-federal lands (see cumulative effects discussion), alternatives 1, 3, 4, 6 and 7 provide little assurance that owl sites will not decline within these areas of concern, increasing nearest neighbor distances and reducing the likelihood for successful dispersal and mate finding.

In the areas of concern that are fragmented landscapes or have checkerboard land ownership patterns, Alternatives 2, 5 and Modified 8, provide a higher likelihood of providing for replacement rate reproduction for owl sites within these areas by establishing an objective for the amount of habitat in each owl home range. Under these alternatives, even where a large proportion of the spotted owl home range occurs on non-federal lands, the entire habitat objective must be met on national forest land. This is particularly important in areas where national forest lands are highly fragmented since providing sufficient habitat to maintain spotted owl occupancy and productivity in such areas may require that all of the available national forest land be managed to as suitable spotted owl habitat.

FEIS Volume 3, Chapter 3, part 4.4, page 93 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 An additional concern, however, is that a large proportion of the landscapes supporting owl sites within these areas of concern, particularly those areas with fragmented habitat and low population densities (designated in the Technical Report as areas A, B, 1, 4, 5, 7, and 8 occurring on the Lassen, Eldorado, Stanislaus, Sierra, and Sequoia National Forests) already may provide less than the desired amounts of habitat within owl home ranges to maintain non- declining populations. Bart (1995) cautions that it should not be assumed that habitat in all home ranges can be reduced to a threshold level without adverse effects on the population. Alternatives which establish minimum thresholds for the amount of habitat within owl home ranges, may not provide direction sufficient to stabilize owl populations within these geographic areas of concern. In areas of concern where a greater proportion of owl home ranges have less than desired amounts of habitat to begin with, reducing the amount of habitat within the few home ranges that exceed the habitat threshold, prior to increasing amounts of habitat in other owl home ranges, could increase the risk of worsening conditions and increasing nearest neighbor distances for owl sites within these areas.

The strategies proposed in Alternative 2 and Modified Alternative 8 present the least risk of worsening habitat conditions within owl sites occurring areas of concern. Risk is low in Alternative 2 because biodiversity reserves encompass a high proportion of owl sites in areas of concern. Risk is low in Modified Alternative 8 because vegetation treatments are designed to maintain suitable habitat across the landscape (see below); thus, vegetation treatments would not allow for all home ranges within the general forest area to be reduced to a minimum threshold amount of suitable habitat.

D. Effects Upon Habitat in Owl Home Ranges Associated with Implementation of the HFQLG Forest Recovery Act The Biological Evaluation for the HFQLF Forest Recovery Act, estimates (1) a 7 to 8.5% decrease in suitable owl habitat under HFQLG Alternative 2, and (2) an 8 to 11 percent decrease in the number of owl sites with greater than 50 percent suitable habitat within their home range area. This decrease in suitable habitat results in 36% of the owl sites in the project area having less than the amounts of habitat thought to be associated with higher rates of site occupancy and productivity. The Lassen and Plumas National Forests, and Sierraville Ranger District of the Tahoe National Forest support about 30 percent of the known spotted owl sites on National Forest land in the Sierra Nevada. If management actions reduce owl occupancy and productivity across this area (as expected under alternative 2 of the HFQLG), opportunities to stabilize population declines could be substantially compromised.

Population declines that would occur within the three geographic areas of concern located within the HFQLG project area, exacerbate the overall risk to spotted owl population. In particular, Area of Concern 1, occupying a large portion of the Lassen National Forest, is an area where habitat fragmentation decreases the density of spotted owl pairs, making successful dispersal more difficult. Actions proposed under Alternative 2 of the HFQLG will widen gaps between habitat parcels and probably reduce the densities of owls within this area of concern (Biological Evaluation for the HFQLG Forest Recovery Act, 1999).

E. Effects on Habitat Suitability for select prey species of the California Spotted Owl. Projected changes in overall habitat suitability scores for select California spotted owl prey species were estimated using CWHR habitat suitability ratings and vegetation projections (Appendix B). Overall, 65% of the species (11/17) had average projected increases in habitat suitability across the alternatives, while 8 species had projected decreases (Table 4.4.2.1h). Overall the results are

FEIS Volume 3, Chapter 3, part 4.4, page 94 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 consistent with the general projections of increasing amounts and distribution of late-seral/old-growth forest conditions along with an increase in prescribed burning and wildfire. Some of the species that are also abundant in riparian and meadow environments that were not fully modeled in the vegetation projections should remain abundant throughout the Sierra Nevada despite modest projected decreases in habitat suitability within the forested vegetation types.

Table 4.4.2.1h. Projected percent changes in overall habitat suitability scores for select prey species of California spotted owls based on CWHR habitat models from the current to 50 years in the future across the FEIS Alternatives. SPP Species Name Alt1 Alt 2 Alt 3 Alt 4 Alt 5 Alt 6 Alt 7 Alt 8 Mod 8 Mlv Mean Code M080 No. Flying 12.0 4.2 17.3 18.5 7.5 16.1 12.8 7.5 11.9 0.6 10.8 Squirrel M127 Dusky-footed 0.8 -3.9 -3.3 -1.8 -3.0 -3.7 0.6 -3.6 5.2 -3.2 -1.6 woodrat M128 Bushy-tailed 8.3 2.6 5.6 6.4 4.2 5.2 4.1 4.0 9.0 3.3 5.3 woodrat M129 Western Red- 13.3 -0.3 6.0 10.7 6.6 5.4 2.5 0.6 10.8 -2.5 5.3 backed vole M117 Deer Mouse 1.2 -3.3 -2.0 -2.5 -2.2 -2.3 -2.8 -2.8 1.5 2.2 -1.7 M113 Western 215.1 184.8 165.1 163.8 208.0 161.3 162.0 184.3 215.1 210.2 187.0 Harvest Mouse M085 Mountain 39.4 28.8 5.2 7.3 30.6 2.7 10.0 21.4 35.3 46.3 22.7 Pocket Gopher M081 Botta’s 55.4 44.6 16.5 12.0 39.2 11.6 19.9 32.0 43.6 68.4 34.3 Pocket Gopher M134 California 153.5 165.4 54.1 31.9 146.3 59.2 67.0 129.4 126.6 183.6 111.7 Vole M133 Montane Vole 177.6 171.5 171.3 173.2 173.6 171.4 171.0 171.5 178.2 173.0 173.3 M136 Long-tailed -5.4 -13.1 -16.9 -10.7 -13.0 -16.0 -12.0 -13.2 -5.1 -10.8 -11.6 Vole M130 Heather Vole -5.8 -6.4 -9.6 -6.9 -6.6 -8.5 -5.5 -6.9 -4.2 -6.3 -6.6 M006 Ornate Shrew -4.5 -10.3 -1.7 -2.9 -6.3 -2.0 -1.5 -8.3 -0.1 -11.6 -4.9 M004 Dusky Shrew 3.9 1.2 5.4 6.3 2.1 4.9 6.6 2.1 5.1 0.7 3.8 M012 Trowbridge’s 4.0 -6.3 5.3 7.5 -2.4 4.2 1.9 -3.7 3.2 -8.3 0.5 Shrew M003 Vagrant 4.6 -9.8 -12.8 -1.4 -6.7 -12.1 -6.9 -9.1 1.1 -4.9 -5.8 Shrew M018 Broad-footed 142.3 123.7 106.8 113.5 118.6 101.2 114.0 115.6 137.9 140.3 121.4 Mole

4. Levels and Types of Forest Management Activities. A. Acres of Vegetation Treatments As previously discussed, it is not possible to directly assess impacts on individual California spotted owl home ranges using habitat projections. Understanding the types of activities that might occur in proximity to known owl sites and the potential effects of these activities helps identify potential risks. In addition, retaining existing suitable habitat and improving habitat conditions over the next couple of decades may be particularly important for stabilizing owl populations. Research into population dynamics at larger scales has suggested the possible existence of habitat thresholds, below which populations may go extinct in the presence of suitable habitat due to constraints on successful dispersal. With current population declines, vegetation treatment impacts over a short time period may involve risks to the spotted owl population that are not evident by considering longer-term habitat projections alone.

FEIS Volume 3, Chapter 3, part 4.4, page 95 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 All vegetation treatments, from prescribed fire to group selection, are designed to affect stand structure to reduce fuel loads and the risk of high severity wildfire and will in turn affect habitat suitability for owls. Non-treatment also has an effect in that natural processes of growth and mortality invariably change stand structure through time. The effects of treatment vary by treatment type and vary through time; some effects are negative in the short-term and positive in the long run, and vice versa. The shorter the time period between habitat alteration and recovery, the lower the risk associated with implementing a proposed management strategy. Canopy cover, canopy layers, and the average diameter of overstory trees appear to be important attributes within California spotted owl nesting and foraging habitats. Table 4.4.2.1i provides an estimate of the general likelihood of retaining these habitat elements at or above thresholds described for preferred nesting and foraging habitats, immediately following vegetation treatments. Table 4.4.2.1j displays the acres of treatment estimated to occur in the first and second decades under the various Alternatives.

Vegetation treatments anticipated under Alternatives 4 and 7 pose a greater risk of affecting owl sites than treatments under the remaining alternatives. Acres of more intensive vegetation treatments (those treatments with a high or moderate likelihood of changing suitable habitat to unsuitable habitat) are greatest in these alternatives. Alternative 4 and 7 are estimated to result in heavy thinning, group selections, seed tree, or regeneration harvest over approximately 288,000 and 277,000 acres, respectively, across the Sierra Nevada during the second decade. Alternatives 1 and 6 are intermediate in risk, similarly affecting about 155,000 and 113,000 acres respectively. Alternatives 2, 3, 5, 8 and modified 8 would each affect less than 50,000 acres with intensive treatments during the second decade. Alternative 2 and Modified Alternative 8 pose the least risk of reducing the acreage of suitable habitat as these alternatives are projected to result in no heavy thinning, group selections, seed tree or regeneration harvest.

Table 4.4.2.1i. Likelihood (high, moderate, low) of retaining important structural attributes of spotted owl habitat following vegetation treatment prescriptions projected in the alternatives. Treatment Type >70% canopy >50% canopy Two or more >24” average >11”average (prescription #) cover cover Canopy layers DBH of DBH of overstory trees Overstory trees Prescribed fire (11&15) High High High High High Biomass thin (21) High High High High High Light thin (31&35) Moderate High High High High Heavy thin (45, 51,55) Low Low Moderate Moderate High Group selection Low Low Low Low Low Shelterwood/regen (61, 71&81)

FEIS Volume 3, Chapter 3, part 4.4, page 96 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Table 4.4.2.1j. Number of acres (in thousands) scheduled for vegetation treatments by alternative in the first and second decades (excludes brush, plantation, and retreatment acres). Alternative Treatment Type 1 2 3 4 5 6 7 8 Mod. 8 Prescribed fire (11, 15) 376 146 421 358 378 547 376 459 408 266 257 679 389 449 662 266 512 Biomass thin (21) 70 27 124 115 42 114 119 86 208 70 32 148 139 30 134 128 82 Light thin (31, 35) 252 29 110 467 15 142 363 20 52 110 21 63 284 19 127 114 49 Heavy thin (45, 51, 55) 180 0 20 174 10 33 99 12 36 144 0 17 244 40 113 223 50 Group selection, 54 12 47 104 32 44 118 21 0 shelterwood, 11 0 0 44 0 0 54 0 Regeneration (61,71,81)

B. Fragmentation Effects Resulting from Vegetation Treatments Vegetation treatments that create openings or reduce suitable habitat will widen the gaps between habitat patches. Increases in the amount of discontinuous habitat and isolation of habitat patches are concerns within known owl home ranges as well as across the landscape. A reduction in the continuity of habitat between owl activity centers, including the habitat outside known owl home ranges, could limit successful mate finding and dispersal, increasing nearest neighbor distances and affecting population trends. In fragmented landscapes, the high survival costs associated with searching for low-density habitat can create a situation where populations may go extinct in the presence of suitable habitat due to constraints on successful dispersal. Reducing habitat fragmentation and maintaining patches of suitable but unoccupied habitat particularly in areas already identified as geographic areas of concern, is important from this standpoint.

The likelihood of vegetation treatments creating gaps and increasing habitat fragmentation are influenced by 1) the type of vegetation treatments applied, and 2) the scheduling of treatments. Alternatives 4 and 7 have a high likelihood of increasing the fragmentation of habitat considering these factors. Average annual treatments of about 28,000 acres with heavy thinning or group selection harvest, is projected under each of these alternative. Under these prescriptions, reduction in canopy cover will create substantial contrast between treated patches and remaining patches of habitat. Group selections, if implemented in a manner that creates very small, irregularly distributed, low density openings, may not result in fragmentation effects. Neither alternative provides sufficient direction on the frequency, size, and distribution of openings to assume that this is the case, however. The spatial location of treatments is uncertain, however both alternatives prioritize treatments within areas of high fire risk and hazard. Over the first decade, therefore, treatments would be expected to be more extensive within the lower montane zone mixed conifer and pine zone, increasing fragmentation effects in these more productive owl habitats. Vegetation treatments under Alternatives 1 and 6 would also increase habitat gaps, though to a lesser extent. Heavy thinning and a minor amount of group selection would affect about 10,000 to 15,000 acres annually under these alternatives.

Alternatives 3, 5, 8, would have minor amounts of treatment (less than 5,000 acres annually across the Sierra Nevada) that would increase habitat gaps. Alternatives 2 and Modified 8 avoid vegetation treatments that would create habitat gaps. Modified Alternative 8 provides

FEIS Volume 3, Chapter 3, part 4.4, page 97 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Standards and Guidelines explicitly limiting the extent to which canopy cover and structure can be adjusted downward.

C. Location of Vegetation Treatments in Relation to Geographic Areas of Concern To the extent that treatments are concentrated (either in time or space), particularly within certain geographic areas of concern, the overall impacts of the actions upon spotted owl populations may be heightened. Under each of the alternatives, most planning of the spatial location of treatments is left to the national forest or ranger district. However, all of the action alternatives focus and prioritize treatments in vegetation types designated as high fire risk and hazard (typically mid- and lower elevation mixed conifer and pine forests) and place emphasis on treatments within the urban intermix zones. Modified Alternative 8 is most explicit in this direction, specifying that treatments will first occur, and be of higher intensity, in the urban wildland intermix zone. Alternatives 3, 4, 6, 7, 8 and Modified 8 also emphasize fuels vegetation treatments within strategically placed area treatments (SPLATs) in areas of high fire hazard and risk for fuels reduction (often on south and west aspects) Treatment is designed to occur over 30 percent of a watershed area under these alternatives.

Table 4.4.2.1k displays the proportion of spotted owl sites in each geographic area of concern, and in total, that occur within urban zones. In total, four percent of spotted owl activity centers occur within the “defense zone” of the urban intermix (the area within 0.25 miles of communities or developments). An additional 32 percent of owl sites occur within the “threat zone” of the urban intermix, and the remaining 64 percent of owl sites occur outside the urban zones. Within specific geographic areas of concern, the proportion of owl sites in urban zones ranges from 3 percent in AOC 1 on the Lassen National Forest, to as high as 78 percent within AOC 7 on the Sierra National Forest. Areas of concern 5 and 7 have a high proportion (greater than 70 percent) of owl sites occurring within the urban intermix zone, and are therefore likely to be at risk to impacts from vegetation treatments. Areas of Concern 3, 4, and 8 have more than a quarter of the known owl activity centers within the urban intermix zone.

Alternatives 4 and 7, with higher amounts of intensive vegetation treatments (heavy thinning, group selections, seed tree harvest), present the greatest risk to worsening habitat conditions within these areas. Assuming that treatments in the first decade are emphasized within the urban zones, and that vegetation treatments occur over 30% of the landscape, the lack of standards requiring habitat retention in owl home ranges, results in a likelihood that habitat conditions might decline within areas of concern 5, 7, 3, 4, and 8. This could further reduce low owl densities in and adjacent to these areas, decreasing the potential for successful dispersal and population interaction. Alternatives 1, 3, and 6, which do not require retention of habitat in owl home ranges, are also likely to impact these areas of concern. Risks are lower since projected acres of intensive treatments are less.

Alternatives 2, 5, 8 and Modified 8 provide greater assurances about maintenance of high quality habitat, even within urban zones. Home range protections and the extensiveness of reserve areas increases confidence that vegetation treatments under Alternatives 2 and 5 will not impact owl sites; limitations on treatment prescriptions under Alternative 8 ensure retention of suitable habitat. Modified Alternative 8 varies the intensity of treatments based on proximity to communities. Under this alternative it can be assumed that owl activity centers occurring in the urban core defense zone may not be maintained through time, given potential fuels treatment prescriptions. This is four percent of spotted owl sites Sierra Nevada-wide; it

FEIS Volume 3, Chapter 3, part 4.4, page 98 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 represents 22 percent of owl sites in AOC 5. In modified alternative 8, slight reductions in canopy cover could impact habitat for the 32 percent of owl sites occurring in the urban threat zone, but impacts would be subtle. Treatment prescriptions are limited to understory thinning, with retention of at least 50 percent canopy cover. Subtle changes in habitat condition under this alternative are not expected to result in lower owl densities or lower productivity in owl sites. Opportunities for successful dispersal and population interaction will not be reduced. Vegetation treatments outside the urban intermix are lower priority. Such treatments, where 64 percent of owl sites are located, are further limited by spotted owl home range core area or old forest emphasis area protections, and are unlikely to reduce habitat quality in these zones.

Table 4.4.2.1k. Proportion of the spotted owl activity centers that occur within the urban intermix zone, by geographic area of concern. Area of Reason for Concern Number of owl activity centers by fuels management zone Concern Urban core Urban threat Total inside Total defense zone Urban zones Outside zone urban zone AOC 1 Habitat discontinuous, naturally fragmented 0 1 (3%) 1 (3%) 34 (96%) (LNF) and poor quality due to drier conditions and soils

AOC 2 Gap in known distribution, mainly on private 0 4 (13%) 4 (13%) 26 (87%) (LNF) lands, extends east-west almost fully across the width of the owl’s range AOC 3 An area of checkerboard lands; dominated by 3 (6%) 13 (26%) 16 (32%) 33 (68%) (TNF) granite outcrops and red fir forests; both features guarantee low owl densities AOC 4 Checkerboard lands and large, private 2 (3%) 13 (22%) 15 (25%) 44 (75%) (ENF) inholdings; owl densities unknown on some private lands and very low on others AOC 5 Has large private inholdings; owl densities 2 (14%) 8 (57%) 10 (71%) 4 (29%) (STNF) unknown on most private lands. AOC 6 Burned in recent years; the little remaining 1 (4%) 3 (11%) 4 (15%) 23 (85%) (STNF) habitat is highly fragmented AOC 7 Habitat naturally fragmented due partly to low 2 (22%) 5 (55%) 7 (78%) 2 (22%) (SNF) elevations and dry conditions; accentuated by logging AOC 8 Small, isolated populations at the south end of 1 (4%) 5 (22%) 6 (26%) 17 (74%) (SQNF) the Sierra Nevada that are more vulnerable to extinction by local stochastic events Outside 45 (4%) 341 (32%) 386 (36%) 683 (64%) AOC’s Total 56 (4%) 393 (30%) 449 (34%) 866 (66%)

D. Vegetation Treatment Effects Associated with Implementation of the HFQLG Forest Recovery Act The high rates of vegetation treatments occurring over a short time period would result in substantial risk to the distribution and abundance of California spotted owls and owl habitat in the northern Sierra Nevada. Over a 5-year period, Alternative 2 of the HFQLG would create 43,500 acres of group selection openings, 21, 375 acres of which are anticipated to occur within the Westside. An additional 222,600 acres of treatment would occur to create linear DFPZs, characterized by open overstories over open understories, with very little vertical layering. These vegetation treatments are expected to produce a 7 percent decline in suitable owl nesting habitat and a 8.5 percent decline in suitable foraging habitat (Biological Evaluation for the HFQLG Forest Recovery Act, 1999).

FEIS Volume 3, Chapter 3, part 4.4, page 99 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Implementing group selection openings will create low to moderate density openings within each stand and will create additional edge adjacent to or within suitable habitat. Implementing DFPZ treatments will maintain continuous cover but will increase the amount of contrast between treated and untreated stands and associated edge. The Biological Evaluation for the HFQLG project concluded that this alternative increased edge effects, reduced habitat connectivity, and increased habitat gaps. It was rated “low” in minimizing fragmentation.

Implementation of vegetation treatments described in Alternative 2 of the HFQLG would increase the amount of discontinuous habitat and habitat isolation through creation of further fragmentation within Areas of Concern. These actions will widen the gaps between habitat parcels and probably reduce the densities of owls. The Technical Report warned against exacerbating conditions within these areas where, “future problems may be greatest if the owl’s status were to deteriorate (Verner et al. 1992). Alternative 2 would contribute to further habitat fragmentation within three geographic areas of concern (AOCs 1, 2, and 3) where habitat is already discontinuous or naturally fragmented, or where there is little information about owl densities. Such action would be expected to decrease the density of owl pairs making successful dispersal more difficult and reducing the likelihood of rapid colonization of unoccupied habitat by owls.

5. Standards and Guidelines addressing important elements of habitat quality. The Technical Report warned that subtle factors that uniformly decrease habitat quality would act to reduce population density and increase the uncertainties associated with successful dispersal and mate finding. The quality of available spotted owl habitat under each alternative is influenced by specific Standard and Guideline provisions for retention of important structural elements during vegetation treatments.

A. Canopy Cover and Structure Studies have identified canopy cover and layering as stand structural characteristics associated with preferred nesting and foraging sites for the California spotted owl. Hunsaker et al. (in press) conclude that the threshold between canopy cover values that contribute to or detract from occurrence and productivity of California spotted owls is a value near 50 percent (measured through aerial photo interpretation).

Alternative 4 lacks standards addressing retention of canopy cover in any land allocation. Alternatives 1, 3, 6, and 7 have standards addressing canopy cover or basal area retention, but it is unlikely or unclear that the standards included in these alternatives will provide for maintenance of high quality habitat. Alternative 6 establishes canopy cover requirements that are applied as averages over large landscapes. The extent to which such averages (which are as low as 40 percent on most south aspects) will provide for maintenance of canopy cover and structure for high quality owl habitat is uncertain. Similarly, the effectiveness of the general forest prescription in maintaining adequate canopy conditions in Alternative 7 is highly uncertain since the frequency and distribution of group selection openings and their context in the surrounding landscape is not addressed.

Alternatives 2 and 5, provide for retention of canopy cover associated with high quality spotted owl habitat, within spotted owl home ranges and throughout the large areas encompassed by biodiversity reserves and old forest emphasis areas. Alternative 8 and Modified Alternative 8

FEIS Volume 3, Chapter 3, part 4.4, page 100 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 enure that all vegetation treatments maintain a minimum of 50 percent canopy cover, where it exists, thereby retaining owl habitat both within and outside spotted owl home ranges.

Modified Alternative 8 provides further assurances that high quality canopy structure will be maintained for spotted owl sites. This alternative prevents vegetation treatments from resulting in uniform canopy cover at the minimum threshold established by including several additional standards and guidelines. First, canopy retention guidelines set both lower limits and limits upon the degree of change from the existing canopy cover in the stand (limited to 10% change in old forest emphasis areas and owl home range core areas and 20% change in urban areas and the remainder of the general forest). Even after treatment, all prescriptions, except those in the defense zone, retain at least 50 percent canopy cover where it exists representing suitable, and by some standards preferred owl habitat. Second, standards and guidelines in modified alternative 8 maintain existing patches of high capability owl habitat that are greater than an acre in size (defined as CWHR 6, 5D, and 5M). Alternative 5 retains patches greater than 5 acres. Vegetation treatments are limited to removal of small diameter material in these stands. These standards help to avoid uniformity and provide for a diversity of canopy cover conditions throughout spotted owl home ranges and across the landscape as a whole.

B. Large, Old Trees All of the alternatives retain trees greater than 30 inches dbh in westside forests and all trees greater than 24” in eastside forests. Alternatives differ, however, in the stand-level retention standards that will affect recruitment and density of large trees over time. Alternative 4 lacks specific standards requiring retention of smaller trees to provide for future recruitment of large trees. The remaining alternatives provide a mechanism for ensuring continued recruitment of large trees within treated areas. Alternatives 1 and 5 utilize CASPO basal area retention requirements, Alternative 7 relies upon CWHR strata size class retention requirements, and Alternative 8 and Modified Alternative 8 rely largely upon canopy cover retention to ensure a continuing supply of large diameter trees across the landscape. Modified Alternative 8 also has a 20-inch dbh size limit in most vegetation treatments that are designed as understory thinnings, thus specifically retaining the 20- to 30- inch size class for future recuritment of large trees. Alternatives 1, 5, and Modified Alternative 8 limit vegetation treatments in owl habitat to understory thinning prescriptions which provides recruitment of large diameter trees but may, over the longterm, impact continuing recruitment into smaller size classes.

C. Snags and Down Wood All alternatives have standards that require retention of a number of snags greater than 15 inches in dbh in the general forest allocation. Alternatives 2, 4, 6, 8 and modified 8 require retention of the 4 largest snags per acre in mixed conifer habitat and the 6 largest snags per acre in red fir habitats. Alternatives 1, 3, and 5 require retention of at least 20 square feet of basal area in the largest snags available, up to eight snags per acre. These alternatives appear to adequately address snag retention for spotted owl foraging habitat, since they are within the range of the mean values for snag basal area reported by Verner et al. (1992). Alternative 7 does may not provide for adequate retention because it does not require retention of the largest snags available. Retention levels in other allocations such as PACs, old forest emphasis areas, and riparian areas, exceed general forest retention levels under most alternatives.

FEIS Volume 3, Chapter 3, part 4.4, page 101 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 D. Retention of Duff Layer All alternatives would meet Regional soil quality standards. An assumption is made, however, that the more areas treated with mechanical treatments, the greater the potential for disturbance of the duff layer and associated micro- habitat that may be important to spotted owl prey. Under this assumption, Alternatives 4, 7, 1, 6, 3, 5, Modified 8, 8 and 2 result in risk in descending order of magnitude.

6. Level of Natural Disturbance. A. Change in the amount area affected by stand replacing wildfires. Wildfire effects, particularly those associated with large, stand replacing wildfires, are a major source of risk to spotted owl populations. Loss and degradation of habitat, creation of habitat gaps, and lengthy time periods for habitat reestablishment, are some of the impacts that may result from wildfires. Alternatives that are projected to reduce the acreage and/or intensity of wildfires would be expected to provide long-term benefits to spotted owls. Alternatives 3, 4, 6, and 7 are projected to substantially reduce wildfire acres (and especially acres of lethal mortality) over two or more decades; modified Alternative 8 is expected to maintain approximately the existing situation in wildfire acres; and wildfire acres burned are projected to increase under Alternative 8, 1, 5, and 2 (in order of increasing magnitude). An essential question is whether vegetation treatments result in a net gain or net loss of habitat over time when wildfire is factored in. The effects of vegetation treatments upon owl habitat are immediate, for the most part, and relatively easy to quantify. Reductions in the acreage and intensity of future wildfires due to vegetation treatments become apparent over longer timeframes. In addition, due to the stochastic nature of wildfire events, wildfire projections have greater amounts of uncertainty and are heavily dependent upon an array of assumptions for variables that are difficult to quantify. Tradeoffs between habitat lost through treatments versus projected losses to wildfire events are therefore complex and their implications difficult to assess for the spotted owl. However, the relatively light trreatments prescribed in Modified Alternative 8, with limits of 10- to 20-percent reduction in canopy cover, will likely persist for only a short time period until ingrowth re-establishes canopy cover.

Total wildfire acreage and acreage of stand replacing wildfires decreases most dramatically under alternatives 3, 4, 6 and 7, from about 62,000 acres currently to less than 55,000 acres projected to annually at the fifth decade. Under Modified Alternative 8, the average annual acres burned in wildfire is projected to remain about constant with current levels over 50 years. The average annual acreage of stand replacing wildife increases slightly, from about 15,000 acres to about 17,000 acres projected annually at the fifth decade under Modified 8. In the remaining alternatives, projected annual acres of wildfire increase over the 50-year timeframe, with the highest increase projected for Alternative 2 (from about 62,000 acres to about 76,000 projected annually at the fifth decade).

Given the owl’s declining population status, net gains or losses of habitat must be evaluated over short (one to two decades) as well as longer time frames. Shorter-term projections, where the magnitude of change is less influenced by modeling assumptions, may also have lower levels of uncertainty associated with them. Table 4.4.2.1l displays the total acres affected from both fuels treatments and wildfire over the next two decades by adding the total acres of projected wildfire to the acres of vegetation treatments that are unlikely to maintain important spotted owl habitat elements (shown in Table 4.4.2.1i). Based on this comparison, Modified Alternative 8 represents the lowest risk to declining habitat over the first two decades, followed

FEIS Volume 3, Chapter 3, part 4.4, page 102 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 by alternatives 3, 8, 6 and 5 in increasing order of risk. Alternatives 1, 4, and 7 will do less to maintain available habitat over the first two decades, as a tradeoff for greater projected increases in habitat 50 years in the future.

Table 4.4.2.1l. Total of the projected annual acres of wildfire burned and the estimated annual acres of higher intensity vegetation treatments* (in thousands). Annual Acres of Wildfire and Treatment (in thousands) Alt 1 Alt 2 Alt 3 Alt 4 Alt 5 Alt 6 Alt 7 Alt 8 Mod 8 First Decade 93 70 73 89 73 73 86 70 65 Second Decade 88 69 60 80 71 66 82 67 59 Average over Two Decades 90.5 69.0 66.5 84.5 72.0 69.5 84.0 68.5 62.0

*higher intensity vegetation treatments” defined as modeled treatment prescriptions 45 and higher.

B. Change in the amount of area affected by stand replacing wildfire following implementation of the HFQLG Forest Recovery Act. Reduced wildfire acres would be expected if forests were to implement Alternative 2 of the HFQLG Forest Recovery Act EIS. Benefits are expected, but their magnitude remains uncertain. Tradeoffs associated with habitat reductions over the short-term, five-year period, also appear to be substantial (previously described).

7. Cumulative Effects About 2.4 million acres of private lands occur within the Sierra Nevada; of this, about 1.45 million acres are owned and managed as industrial forests, primarily at mid-elevations in the mixed-conifer forest type. National Forests in the Sierra Nevada include approximately 1.4 million acres of private land within their administrative boundaries. Private land inholdings are much greater in extent in the northern National Forests (especially the Lassen, Plumas, and Tahoe) than in the southern Sierra Nevada forests. Much of the private land within the boundary of the Lassen and Plumas National Forests tends to be in contiguous blocks, leaving National Forest lands also fairly contiguous. Most private land on the Tahoe National Forest is in checkerboard ownership, and the Eldorado National Forest has a combination of checkerboard ownership and large contiguous blocks of inholdings. The Sierra and Sequoia have little private land within their administrative boundaries and the four National Parks have negligible amounts.

Industrial forestlands are managed to produce a long-term sustained yield of forest products, primarily saw logs. Management of industrial forest land in California is governed by the forest practice rules of the Z’berg-Nejedly Forest Practices Act, which includes specific requirements for (and restrictions on) aspects of forest management, including the size of clearcut, the tree stocking levels after harvest, protection of wildlife habitat, retention of old growth, etc. These rules also require management of private forests for long-term sustained yield and require preparation and approval of timber harvest plans before logging operations may commence. The Forest practice rules provide protection measures for active nest sites and for late successional forest stands. The size of nest stand buffers is not specified but is designed to protect the immediate nest site and nesting birds from effects of timber operations. Forest practices rules do not establish requirements for maintaining amounts of foraging habitat for owl sites. Management activities planned under the Forest Practices rules do not, therefore, provide assurance that activities will retain the amount and quality of habitat expected to

FEIS Volume 3, Chapter 3, part 4.4, page 103 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 maintain spotted owl occupancy or productivity (Bart, 1995, Hunsaker et al. in press, Verner et al. 1992).

The Petition to List the California spotted owl as a Threatened or Endangered Species (Center for Biological Diversity, Sierra Nevada Forest Protection Campaign, April 2000) reported a total of 299,421 acres of private land timber harvest planned within two miles of known spotted owl sites, based on a review of timber harvest planning documents. Timber harvest on private lands has been and will continue to be a major source of cumulative impact upon spotted owl habitat in the Sierra Nevada. Under current Forest Practices rules, it is assumed that spotted owl habitat on private lands will continue to decline. Since there are few assurances of habitat protection on private lands, analysis of the Alternatives in this EIS does not assume that private lands will continue to contribute habitat to spotted owl sites occurring on National Forest lands. Owl home range habitat requirements under Alternatives 2, 5, 8 and Modified 8 are implemented under this assumption, thereby addressing cumulative impacts to the extent possible. Alternatives 1, 3, 4, 6, and 7, do not provide a specific mechanism that compensates for declining habitat conditions on private lands.

Human population growth and development in the Sierra Nevada is projected to increase substantially over the next few decades. While human development has not been identified as a major source of cumulative impact to date, impacts related to increased urbanization, infrastructure development, and recreation, are likely to increase over time.

8. Environmental Outcomes This section synthesizes the discussion of environmental consequences to arrive at an estimate of the environmental and population conditions that would exist in 50 years for the California spotted owl, under each alternative. The environmental outcomes address habitat distribution and its anticipated consequence to species dispersal and interaction capabilities. Population outcomes factor in the availability of both federal and non-federal habitat and other influences on the spotted owl population that are not accounted for in the environmental outcomes. Assigning these outcomes is inherently subjective, although based on a reasoned thought process and the best available information. Table 4.4.2.1k presents environmental and population assessment ratings for the California spotted owl. These ratings assume implementation of Framework alternatives over the next 50 years, across all forests of the Sierra Nevada, including those forests addressed in the HFQLF Forest Recovery Act). Part 4.1.5 of this chapter provides more detailed descriptions of the five possible outcomes; these outcomes are briefly summarized below Table 4.4.2.1m for the reader’s convenience.

FEIS Volume 3, Chapter 3, part 4.4, page 104 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Table 4.4.2.1m. Environmental outcome ratings for the California spotted owl. Ratings other than “current” represent the outcome most likely to be realized 50 years in the future. (See Chapter 4, Part 4.1.5.) Alternative Outcomes Environmental Population Current B B 1 C D+ 2 C+ C- 3 B- C 4 C D+ 5 B C+ 6 B- C 7 C D+ 8 B+ B- Mod 8 A- B

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Rationale for Ratings The distribution of California spotted owls is currently nearly continuous throughout their range within the Sierra Nevada. However, the declining population trends reported from demographic studies suggest that current habitat conditions are not fully sustaining this distribution. Consequently, the current status of conditions for owls is judged to be primarily well distributed with gaps, allowing operation of metapopulation processes (Outcome B), but that there is some likelihood that existing conditions would result in permanent isolation of some portion of the population (Outcome C).

The following criteria were developed and applied to each alternative to arrive at environmental outcomes: (1) CWHR habitat projections, (2) Standards and Guidelines providing for sufficient amounts and distribution of high quality habitat at landscape and home range scales; and (3) Standards and Guidelines addressing stand-level structure and important habitat elements.

CWHR habitat projections, provide the most synthesized description of environmental conditions in 50 years because their modeling integrates treatment effects, wildfire and other mortality effects, and tree growth effects, into the projection of vegetative conditions. The variability surrounding these projections is high, however, since the results are dependent on assumptions made for a large number of variables, each with varying degrees of certainty, and for which uncertainty is high in many cases. Lower confidence is provided as projections are made further into the future. Additional uncertainty is added because assessment of habitat through a broad vegetation classification system, such as CWHR, invariably miss or allow for

FEIS Volume 3, Chapter 3, part 4.4, page 105 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 misinterpretation of subtle changes in habitat quality that may be important to maintaining productivity of owl sites. (For example, the use of stand averages to evaluate changes in habitat quality can, in some instances, be misleading, as when understory treatments increase the average size class of the stand, yet do not result in a real increase in habitat suitability). Taking these uncertainties, as well as the absence of spatial considerations in CWHR projections, into account, outcomes were judged by considering the two additional criteria. These criteria are intended to provide greater certainty about the quality and distribution of projected habitat.

To arrive at population outcomes, the following criteria were developed and their influence upon the environmental outcomes for each alternative was considered: (1) documented population trends; (2) timber harvest on private lands; and (3) human population growth and development.

Modeling showed all alternatives providing increasing amounts of habitat over 50 years, with greater amounts occurring under alternatives 3, 6, Modified 8 and 8 and slightly lower amounts under alternatives 2, 7, 4, and 1. The magnitude of differences between the alternatives is difficult to interpret with confidence due to variation inherent in the vegetation information and modeling process. The basis for differing environmental outcomes among the alternatives is primarily based upon criteria 2 and 3. The following discussion summarizes findings from the environmental consequences section that influence the environmental outcomes.

Alternative 1: The abundance and distribution of suitable environments for the spotted owl is expected to decline from current conditions, with increased likelihood of population isolation, for the following reasons:

• Alternative 1 lacks provisions addressing the distribution of habitat within owl home ranges, sufficient to maintain occupancy and productivity of spotted owl sites. • Alternative 1 lacks provisions ensuring adequate retention of important structural elements of owl habitat, particularly canopy cover and layering, during vegetation treatments (except within the relatively few acres occurring in PACs). • Ninety-six percent of owl activity centers occur in allocations where more intensive vegetation treatments are permitted occur.

The factors listed above result in uncertainty about the future quality of habitat that would be provided within owl home ranges under Alternatives 1. Currently, suitable environments are estimated to occur in approximately half of the spotted owl home ranges in the Sierra Nevada (considering results reported in Hunsaker et al. in press); there is a likelihood that this proportion would decrease under Alternative 1. Alternative 1 has the potential to result in subtle but uniform decreases in habitat quality across the owl’s range (changes that may not be readily displayed by CWHR habitat projections). Given current range-wide conditions, disproportionate impacts would be anticipated within geographic areas of concern, where, if suitable environments decline, they may become absent or remain present only in low abundance. Habitat projections under alternative 1 do not benefit from decreasing amounts of wildfire; total wildfire acres and high intensity wildfire acres are anticipated to increase from current levels under this alternative. Given these considerations, suitable environments for productive owl sites are estimated to become patchy or unevenly distributed under Alternative 1 and may be reduced to low abundance, particularly within certain geographic areas of

FEIS Volume 3, Chapter 3, part 4.4, page 106 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 concern. Spotted owl population outcomes in 50 years are rated at outcome D, given current population trend estimates and assuming continuation of current levels of timber harvest on industrial timberlands across the Sierra Nevada.

Alternative 2: The abundance and distribution of suitable environments for the spotted owl is expected to decline from current conditions, with increased likelihood of population isolation, for the following reason:

• Wildfire rates (and particularly acres of high-intensity wildfire) are expected to increase substantially under Alternative 2, approaching levels under a “no vegetation treatment” scenario.

Alternative 2 provides a high degree of certainty that vegetation treatments will not adversely affect the distribution or abundance of owl habitat since less than 15 percent of owl activity centers occur in land allocations where more intensive vegetation treatments are permitted occur. Even where owl sites occur outside of large biodiversity reserves, standards and guidelines address the distribution of high quality habitat within owl home ranges. Alternative 2 does not, however, reduce the risk and uncertainty associated with wildfire. Suitable environments for productive owl sites may be reduced to low abundance in certain areas, due to increasing acreage of high-intensity wildfire projected. Spotted owl population outcomes in 50 years are rated at outcome C-, given current population trend estimates and assuming continuation of current levels of timber harvest on industrial timberlands across the Sierra Nevada.

Alternative 3: The abundance and distribution of suitable environments for the spotted owl is expected to remain about the same or decline slightly from current conditions for the following reasons:

• Alternative 3 lacks provisions addressing the distribution of habitat within owl home ranges, sufficient to maintain occupancy and productivity of spotted owl sites.. • Alternative 3 lacks provisions ensuring retention of important structural elements of owl habitat, particularly canopy cover and layering, during vegetation treatments (except within the relatively few acres occurring in PACs). • Forty-six percent of owl activity centers occur in allocations where more intensive vegetation treatments are permitted occur.

Alternative 3 does not include provisions addressing habitat distribution within spotted owl home range areas and does not provide specific standards for retention of structural elements of owl habitat such as canopy cover and structure. These risks are offset, somewhat, by a substantial reduction in wildfire acres estimated under Alternative 3. Alternative 3 does not ensure retention of canopy cover and structure during vegetation treatments. Lack of specificity regarding vegetation treatments in this alternative, increases the uncertainty of effects relative to other alternatives. Since less than 50 percent of spotted owl activity centers occur within land allocations where more intensive vegetation treatments are permitted to occur, treatments are less likely to result in uniform decreases in habitat quality across the owl’s range. Habitat is expected to remain broadly distributed but gaps, where suitable environments are present in low abundance, are likely to increase as a result of vegetation treatments, particularly within geographic areas of concern. Spotted owl population outcomes

FEIS Volume 3, Chapter 3, part 4.4, page 107 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 in 50 years are rated at outcome C, given current population trend estimates and assuming continuation of current levels of timber harvest on industrial timberlands across the Sierra Nevada.

Alternative 4: The abundance and distribution of suitable environments for the spotted owl is expected to decline from current conditions for the following reasons:

• Alternative 4 lacks provisions addressing the distribution of habitat within owl home ranges, sufficient to maintain occupancy and productivity of spotted owl sites.. • Alternative 4 lacks provisions ensuring retention of important structural elements of owl habitat, particularly canopy cover and layering, during vegetation treatments (except within the relatively few acres occurring in PACs). • Ninety-six percent of owl activity centers occur in allocations where more intensive vegetation treatments are permitted to occur.

The factors listed above result in uncertainty about the future quality of habitat that would be provided within owl home ranges. Habitat projections under alternatives 4 benefit from reductions in the acreage of wildfire and stand-replacing wildfire 50 years into the future. Canopy cover retention requirements in alternative 4 are limited to the retention of 30 inch trees; they do not provide for maintenance of high quality owl habitat outside of PACs. The treatment prescriptions modeled typically retain more than 30 inch trees, but, in the absence of specific standards and guidelines, the certainty that actual treatments will resemble the prescriptions modeled is lowered. This reduces the confidence with which one can interprit CWHR projections for alternative 4. Nonetheless, modeled treatments under Alternative 4 include average annual treatment of about 29,000 acres with heavy thinning, group selections, seed tree, or regeneration harvest across the Sierra Nevada. These treatments have greater potential for increasing fragmentation of suitable environments, and isolating patches of suitable habitat. The synergistic impacts associated with habitat fragmentation and edge effects is likely exceed impacts displayed by habitat projections alone. Given current range- wide conditions, disproportionate impacts would be anticipated within geographic areas of concern, where, if suitable environments decline, they may become absent or remain present only in low abundance. Spotted owl population outcomes in 50 years are rated at outcome D, given current population trend estimates and assuming continuation of current levels of timber harvest on industrial timberlands across the Sierra Nevada.

Alternative 5: The abundance and distribution of suitable environments for the spotted owl is expected to remain about the same as current conditions for the following reasons:

• Alternative 5 includes provisions addressing the distribution of habitat within owl home ranges, providing a higher probability of maintaining occupancy and productivity of spotted owl sites. • Alternative 5 includes provisions ensuring retention of important structural elements of owl habitat, particularly canopy cover and layering, within old forest emphasis areas and within spotted owl home ranges in the general forest. • Only 33 percent of owl activity centers occur in allocations where more intensive vegetation treatments are permitted to occur.

FEIS Volume 3, Chapter 3, part 4.4, page 108 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Considering these factors, Alternative 5 provides a higher degree of certainty that vegetation treatments will not adversely affect the distribution or abundance of owl habitat over the next 50 years. Habitat objectives applied to individual owl sites would increase the likelihood of maintaining owl sites that occur within fragmented landscapes and other geographic areas of concern. Environments supporting productive owl sites are expected to remain broadly distributed and to be maintained in abundance across the range of the species; increases in temporary gaps may result from wildfire, however. Spotted owl population outcomes in 50 years are rated at outcome C, based upon current population trend estimates and assuming continuation of current levels of timber havest on industrial timberlands across the Sierra Nevada.

Alternative 6: The abundance and distribution of suitable environments for the spotted owl is expected to remain about the same or decline slightly from current conditions for the following reasons:

• Alternative 6 lacks provisions addressing the distribution of habitat within owl home ranges, sufficient to maintain occupancy and productivity of spotted owl sites. • Alternative 6 lacks provisions ensuring retention of important structural elements of owl habitat, particularly adequate canopy cover and layering, during vegetation treatments (except within the relatively few acres occurring in PACs). • Sixty-four percent of owl activity centers occur in allocations where more intensive vegetation treatments are permitted occur.

Alternative 6 does not include provisions addressing habitat distribution within spotted owl home range areas. Alternative 6 has requirements for canopy cover retention averaged across large landscape areas, but it is unclear that such requirements will provide for maintenance of high quality habitat. Alternative 6 may result in more uniform decreases in habitat quality across the owl’s range, at least across south and west aspects. Habitat projections under alternatives 6 benefit from projected reductions in the acreage of wildfire and stand-replacing wildfire 50 years into the future. Suitable environments for productive owl sites are expected to remain broadly distributed but gaps, where suitable environments are present in low abundance, are likely to increase as a result of vegetation treatments, particularly within geographic areas of concern. Spotted owl population outcomes in 50 years are rated at outcome C, given current population trend estimates and assuming continuation of current levels of timber harvest on industrial timberlands across the Sierra Nevada.

Alternative 7: The abundance and distribution of suitable environments for the spotted owl is expected to decline from current conditions for the following reasons:

• Alternative 7 lacks provisions addressing the distribution of habitat within owl home ranges, sufficient to maintain occupancy and productivity of spotted owl sites. • Alternative 7 lacks provisions for retention of important structural elements of owl habitat, particularly canopy cover and layering, during vegetation treatments (except within the relatively few acres occurring in PACs). • Ninety-six percent of owl activity centers occur in allocations where more intensive vegetation treatments are permitted to occur.

FEIS Volume 3, Chapter 3, part 4.4, page 109 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 The above factors result in uncertainty about the future quality of habitat that would be provided within owl home ranges under Alternative 7. Habitat projections benefit from reductions in the acreage of wildfire and stand-replacing wildfire 50 years into the future. The large number of owl sites allocated to general forest and the lack of specific standards for retention of structural elements of owl habitat such as canopy cover and structure, result in considerable uncertainty about future habitat conditions. Modeled treatments under Alternative 7 include average annual treatment of about 25,000 acres with heavy thinning, group selections, seed tree, or regeneration harvest across the Sierra Nevada. These treatments have greater potential for increasing fragmentation of suitable environments, and isolating patches of suitable habitat. Given current range-wide conditions, disproportionate impacts would be anticipated within geographic areas of concern, where, if suitable environments decline, they may become absent or remain present only in low abundance. Spotted owl population outcomes in 50 years are rated at outcome D+, given current population trend estimates and assuming continuation of current levels of timber havest on industrial timberlands across the Sierra Nevada.

Alternative 8: The abundance and distribution of suitable environments for the spotted owl is expected to remain about the same or increase slightly from current conditions for the following reasons:

• Alternative 8 includes provisions requiring retention of existing suitable habitat both within and outside of known owl home ranges, providing a higher probability of maintaining occupancy and productivity of spotted owl sites. • Alternative 8 includes provisions ensuring retention of important structural elements of owl habitat, particularly canopy cover and layering, within spotted owl habitat. • Forty-two percent of owl activity centers occur in allocations where more intensive vegetation treatments are permitted to occur.

Considering these factors, Alternative 8 provides a higher degree of certainty that vegetation treatments will not adversely affect the distribution or abundance of owl habitat over the next 50 years. Alternative 8 lacks habitat objectives applied to individual owl sites but, instead, relies upon retention of existing suitable habitat across the species range. Lack of habitat objectives reduces the likelihood that habitat conditions will improve within those owl home ranges that currently provide less than suitable environments for occupancy and productivity. This issue would be of greatest concern within fragmented landscapes and other geographic areas of concern where a high proportion of owl sites currently lack suitable conditions. Environments supporting productive owl sites are expected to remain broadly distributed and to be maintained in abundance across the range of the species; increases in temporary gaps may result from wildfire, however. The retention of currently suitable but unoccuppied habitat in this alternative may be important for maintaining well distributed habitat into the future. Spotted owl population outcomes in 50 years are rated at outcome B, based upon current population trend estimates and assuming continuation of current levels of timber harvest on industrial timberlands across the Sierra Nevada.

Modified Alternative 8: The abundance and distribution of suitable environments for the spotted owl is expected to increase above current conditions for the following reasons:

FEIS Volume 3, Chapter 3, part 4.4, page 110 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 • Modified Alternative 8 includes provisions addressing the distribution of habitat within owl home range core areas, providing a higher probability of maintaining occupancy and productivity of spotted owl sites. • Modified alternative 8 includes provisions ensuring retention of important structural elements of owl habitat, particularly canopy cover and layering, across all portions of the landscape except urban core areas. • Fifty-one percent of owl activity centers occur in allocations where more intensive vegetation treatments are permitted to occur.

Considering these factors, Modified Alternative 8 provides a higher degree of certainty that vegetation treatments will not adversely affect the distribution or abundance of owl habitat over the next 50 years. Modified Alternative 8 addresses habitat both within known owl home ranges, and across the landscape as a whole. Importantly, Modified Alternative 8 includes several provisions that ensure vegetation treatments will not result in subtle but uniform decreases in habitat quality across the owl’s range: (1) protection of existing patches of high quality owl habitat across all land allocations, (2) limitations on the amount of change from existing canopy cover conditions (avoiding the potential for uniform canopy cover reductions to a minimum threshold level), and (3) limited habitat alteration within spotted owl home range core areas in the general forest, and within old forest emphasis areas which support about 50 percent of owl sites. Since Modified alternative 8 includes requirements for minimum canopy cover retention across the landscape, increases in the abundance of suitable environments for spotted owls are likely both within and outside of known owl home ranges. This provision increases the likelihood of maintaining owl sites that occur within fragmented landscapes and other geographic areas of concern where maintenance of suitable but unoccupied habitat will improve the opportunity for successful dispersal and optimum use of available habitat. Environments supporting productive owl sites are expected to remain broadly distributed and increasing abundance of suitable environments for the owl should provide opportunity for continuous or nearly continuous intraspecific interactions. The acreage of wildfire is projected to remain about the same as current levels under Modified Alternative 8. Spotted owl population outcomes in 50 years are rated at outcome B, based upon current population trend estimates and assuming continuation of current levels of timber harvest on industrial timberlands across the Sierra Nevada.

9. Areas of Uncertainty 1. The potential benefits of treating PACs with prescribed fire and/or mechanical thinning to reduce the probability for stand-replacing wildfires versus their potential positive or negative effects on California spotted owl occupancy, reproduction, and survival, remains as a major item of uncertainty in this assessment. Related to this concern is uncertainty about the degree to which specific treatments (e.g., mechanical thinning versus prescribed fire) change fire risk (e.g., reduction in surface fuel loads versus reduction small tree density). A commitment to conducting a paired study monitoring spotted owl occupancy and reproductive success associated with treated and untreated PACs, is needed to address this continuing uncertainty.

2. Uncertainty remains regarding the effects of vegetation treatments on prey biomass and availability, and on owl foraging habitat suitability. Knowledgeable evaluation of how each of the different vegetation management treatments (e.g., mechanical thinning, prescribed fire, CASPO harvest, etc.) affects the distribution, abundance, and availability of prey to California spotted owls, will require additional study. The Technical Report identified the need to

FEIS Volume 3, Chapter 3, part 4.4, page 111 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 evaluate spotted owl response to various vegetation treatments. This remains an essential component of any adaptive management strategy associated with management of California spotted owl habitat. A well designed study, investigating owl habitat use in stands affected by varying types of treatments, is needed to address this continuing uncertainty.

3. Understanding where treatments will occur on the landscape is hampered by the fact that the majority of actual decisions will be determined based on local landscape or watershed analyses. Modeling in this assessment has been unable to predict spatial outcomes with any degree of certainty. It is therefore important that a tracking and monitoring mechanism be put in place which will allow for cumulatively assessing the impacts of vegetation treatments over time.

4. Uncertainty exists regarding how the distribution and abundance of habitat at landscape or regional spatial scales affects the number and distribution of owl territories across the landscape and connectivity and dispersal among territories. Understanding these dynamics is important because research on population dynamics at larger scales has suggested the possible existence of habitat thresholds, below which populations may go extinct in the presence of suitable habitat due to constraints on successful dispersal. These concerns are particularly relevant to species such as California spotted owls because of their low fecundity, indicating that populations may require long time periods to recover from low population sizes, and because of the long time periods required to develop the large old trees and late-seral forest stands that comprise owl habitat. Eventually a comprehensive conservation strategy may need to specify guidelines identifying a target number and distribution of spotted owl sites at the Forest, region, and Sierra Nevada scales.

FEIS Volume 3, Chapter 3, part 4.4, page 112 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 4.4.2.2. NORTHERN GOSHAWK Affected Environment Species Background 1. Population Status Northern goshawks are currently listed as a Sensitive Species by the Forest Service in the Pacific Southwest Region. They are listed as a Species of Special Concern by the State of California. The species has been petitioned for federal listing as a threatened or endangered species in part or all of its range in the United States three times in previous years. All listing petitions have been denied, although further litigation is currently pending. There is concern that northern goshawk populations and reproduction may be declining in North America and California due to changes in the amount and distribution of habitat or reductions in habitat quality (Bloom et al. 1986, Reynolds et al. 1992, Kennedy 1997, Squires and Reynolds 1997, Smallwood 1998, DeStefano 1998).

a. Population Size and Distribution Northern goshawks are distributed throughout forest and woodlands of the Holarctic (Brown and Amadon 1968). Within North America, northern goshawks are found in a variety of forested vegetation types, ranging across the boreal forest and extending south through the western mountains into Mexico and, in the East, south through the mixed conifer-hardwood forest to approximately New York/New Jersey at the present (Palmer 1988, Squires and Reynolds 1997). Within the Sierra Nevada, northern goshawks breed from approximately 750 m in the ponderosa pine/mixed-conifer vegetation types through approximately 3050 m in the red fir and lodgepole pine vegetation types, and throughout eastside pine forests on the east slope (Bloom et al. 1986). Additionally, northern goshawks nest in aspen stands occurring within shrub vegetation types on the eastern slope of the Sierra Nevada (Herron et al. 1985, Bloom et al. 1986). Northern goshawks are year-round residents in the Lake Tahoe Region (Keane 1999) and are suspected to be year-round residents throughout the Sierra Nevada, although some limited seasonal altitudinal movements may occur.

Limited information is available to estimate the pre-1850’s distribution, and especially abundance, of northern goshawks in the Sierra Nevada. Grinnell and Miller (1944) reported that the species was distributed during the breeding period throughout the Sierra Nevada south to approximately Tulare County. No data are available to estimate the historic abundance or density of northern goshawks in the Sierra Nevada. Bloom et al. (1986) provided a gross population estimate of approximately 515 territories in the southern Sierra Nevada-White Mountains region, 425 territories in the northern Sierra Nevada-Cascade Range, and 20 territories in the Great Basin region of California based on estimates of the number of territories per township. Although their regional stratification does not overlap exactly with the region addressed in this FEIS the numbers are useful for general comparative purposes. A total of 588 known territories were reported from the ten Sierra Nevada National Forests and Management Units addressed in this FEIS, of which 550 fell on Nationa Forest Service lands (Table 4.4.2.2a). An additional 125 territories were identified in the Sierra Nevada Bioregion from a database maintained by the California Department of Fish and Game (CDFG), of which 71 fell on National Forest Service lands. Maurer (2000) reported an additional 23 territories from Yosemite National Park. Of the 71 territories from the CDFG database that fell on National Forest Service lands, 4 were reported in the 1970’s, 53 in the 1980’s, and 14 in the 1990’s. It is uncertain what proportion of these 71 territories may still be occupied, particularly for territories reported from the 1970’s and 1980’s that have not

FEIS Volume 3, Chapter 3, part 4.4, page 113 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 been re-surveyed. Future surveys will be required to determine if these territories are currently occupied. However, data from the 500+ territories known occupied in the 1990’s on National Forest Service lands indicate that there does not appear to have been a change in the geographic distribution of northern goshawks in the Sierra Nevada relative to the range reported by Grinnell and Miller (1944). It is unknown if there have been changes in the abundance or density of the northern goshawks within the range of the species in the Sierra Nevada. Low numbers of known territories reported from the Sierra and Sequoia National Forests may be a result of lower goshawk densities near the southern edge of the species range in California, reduced numbers due to management, or may be an artifact of lower survey effort on these forests. Future project-level surveys required before vegetation disturbing activities can occur in suitable habitat under all alternatives in this FEIS will provide further data to address this uncertainty.

Table 4.4.2.2a. Number of northern goshawk breeding territories reported by Sierra Nevada National Forests. National Forest Number of Territories Modoc 130 Lassen 113 Plumas 75 Tahoe 52 Eldorado 69 Lake Tahoe Basin 22 Inyo 34 Stanislaus 53 Sierra 12 Sequoia 17 -Outside USFS 11 boundary Total 588

b. Population Trends Population trends of northern goshawks in the Sierra Nevada are unknown, although numbers are suspected to be declining due to habitat reductions and loss of territories to timber harvest (Bloom et al. 1986). Knowledge of northern goshawk demography is limited. Estimates of reproductive parameters are limited to measures of the proportion of territories active among years and of the number of young fledged from only a few short-term studies (e.g., Bloom et al. 1986, Keane 1999). Age- or stage-based survival rates are unknown. Although limited project-level survey and monitoring is conducted, there are currently no rigorous, scientifically defensible monitoring or research efforts being conducted to assess population trends, demographic rates, or effects of habitat manipulations.

2. Habitat Relationships Knowledge of northern goshawk habitat relationships is incomplete. Nest-site habitat characteristics are the best-known aspect of northern goshawk habitat use patterns. Very little information exists regarding foraging habitat use patterns, particularly during winter. No information is available that addresses habitat quality (as measured by fitness metrics (in other words, survival and fecundity)) at any spatial scale.

a. Nest-Site Stand Structure. Nest Tree Scale: Northern goshawks construct stick nests generally located in live conifer or hardwood trees, although nests are sometimes placed in snags. Nests in live trees are usually

FEIS Volume 3, Chapter 3, part 4.4, page 114 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 placed at or just below the lower portion of the canopy usually in a crotch formed by branches against the bole, occasionally on a branch 3-10 ft from the bole. Nest trees are usually among the largest trees in the nest stand. Conifer nest trees in the Lake Tahoe region averaged 32” dbh (sd = 11.8 , range = 15-61, n = 39)(Keane 1999). Nest trees in eastside pine on the Inyo National Forest averaged 34” cm dbh (sd = 10.7 , n = 20)(Hargis et al. 1994). Nest trees in Yosemite National Park averaged 51” cm dbh (sd = 12.6 , range 21-79, n = 26)(Maurer 2000).

Within-stand scale: Within-stand, nest-site habitat structure and composition are among the best-studied aspects of northern goshawk habitat relationships (Squires and Reynolds 1997). Although absolute differences in structural characteristics may differ between vegetation types and geographical regions, relative habitat use patterns are consistent such that northern goshawks use nest-sites with greater canopy cover, greater basal area, greater numbers of large diameter trees, and lower shrub/sapling/understory cover and numbers of small diameter trees, and gentle to moderate slopes relative to non-used, random sites (Hall 1984, Speiser and Bosakowski 1987, Hargis et al. 1994, Squires and Ruggerio 1996, Keane 1999). High canopy cover is the most consistent structural feature similar across studies of northern goshawk nesting habitat (Siders and Kennedy 1996). This habitat provides large trees for nest sites, a closed canopy for protection from predators and thermal cover, and open understories that provide for maneuverability and detection of prey below the canopy. Three studies have described habitat relationships at this scale in the Sierra Nevada (Hargis et al. 1994, Keane 1999, Maurer 2000).

In the Lake Tahoe region, Keane (1999) reported that northern goshawk nest-sites (n = 35) had significantly greater numbers of live trees >40” dbh (mean = 15.8/acre, se = 2.2), >24- 40” dbh (22.1/acre, se = 3.2) and greater canopy cover (mean = 70.4%, se = 3.1), and significantly lower shrub/sapling cover (mean = 9.9%, se = 2.0) and number of live trees >2- 12” dbh (mean = 121.4/acre, se = 12.3) compared to random plots based on 0.25 acre circular plots centered on nest trees and random points.

In eastside pine forests on the Inyo National Forest, Hargis et al. (1994) documented that northern goshawk nest-sites (n = 20) had significantly greater canopy cover (mean = 31%, sd = 13), basal area (37 m2/ha, sd = 9), and numbers of live trees in the 11-18” dbh (mean = 1.4/acre, sd = 0.7), 18-24” dbh (mean = 0.7/acre, sd = 0.4), and > 24” cm dbh (largest size class recognized; mean = 0.5/acre, sd = 0.3) size classes compared to random plots. Mean canopy closure was approximately 64% at northern goshawk nest-sites in eastside pine on the Lassen and Modoc National Forests (B. Turner, Lassen National Forest, unpublished data, B. Woodbridge, Klamath National Forest, pers. comm). Canopy closure at northern goshawk nest sites in eastside pine vegetation in the eastern Sierra Nevada appears to be more variable in absolute measures than those reported for mixed-conifer and red fir vegetation types in the western Sierra Nevada. However, relative differences were similar in that canopy covers were still significantly greater than measures from random sites.

In Yosemite National Park, Maurer (2000) found that northern goshawk nest sites (n = 31) had significantly greater numbers of live trees >40” dbh (mean = 17.3/acre, se = 2.0), and significantly lower numbers of >2-6” dbh (mean = 160.4/acre, se = 27.9) and >6-12” dbh (mean = 66.1/acre, se = 8.7) compared to control plots based on 0.25 acre circular plots centered on nest trees and control points. Canopy cover averaged 65% (sd = 15, range = 39- 100%) at nest sites.

FEIS Volume 3, Chapter 3, part 4.4, page 115 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Generalized habitat models based on best professional opinion contained in the California Wildlife Habitat Relationships (CWHR) database rate the following affected vegetation types and strata as providing high nesting habitat capability: Sierran Mixed Conifer, White Fir, Montane Hardwood-Conifer, and Montane Riparian (6, 5D, 5M, 4D, 4M); Ponderosa Pine, Jeffrey Pine, Lodgepole Pine, Subalpine Conifer, and Montane Hardwood (5D, 5M, 4D, 4M); Red Fir (5D, 5M). Within CWHR, size class 6 is only recognized for a subset of the forest vegetation types (Sierran Mixed Conifer, White Fir, Montane Hardwood-Conifer, Montane Riparian, Aspen). The following vegetation types and strata are rated as providing moderate nesting habitat capability in CWHR: Aspen (6, 5D, 5M, 4D, 4M); Eastside Pine (5D, 5M, 4D, 4M, 3D, 3M); Red Fir (4D, 4M); Lodgepole Pine and Subalpine Conifer (3D, 3M).

Classification of nest plot data from 35 nest sites from the Lake Tahoe Region (Keane 1999) to CWHR type and size/density class following the analytical methods described in Appendix B resulted in the following distribution by CWHR class: 5D (13), 6 (11), 5M (6), 4D(1), 4M (2), and 4P(2).

b. Foraging Northern goshawks have evolved morphological adaptations for capturing prey in forested environments, but are also capable of ambushing prey in open habitats. Northern goshawks forage primarily by exhibiting short-duration sit-and-wait predatory movements whereby they move through the forest in a series of short flights that are punctuated by brief periods of prey searching from elevated perches. They will also use flush-chase techniques whereby they move through the forest and attempt to surprise and flush prey (Squires and Reynolds 1997).

Foraging habitat preferences of northern goshawks are poorly understood, although limited information from studies in conifer forests indicate that northern goshawks seem to prefer to forage in mature forests (summarized in Squires and Reynolds 1997). Hargis et al. (1994) reported that telemetry points within home ranges of northern goshawks had greater basal area, canopy cover, and trees in larger diameter classes compared to random plots within eastside pine vegetation in eastern California. Austin (1993) found that mature and old- growth habitat (> 52 cm dbh, > 40% canopy cover) were used, and open habitats (seedling and sapling stands, meadows), were avoided in mixed-conifer forests in the southern Cascade Mountains of northern California. Beier and Drennan (1997) documented that foraging sites were selected in forests with greater canopy closure and greater density of large trees (> 41 cm dbh) relative to control plots in Arizona. Similarly, Bright-Smith and Mannan (1994), reported that habitat preferences increased with increasing canopy cover for foraging northern goshawks in Arizona. More work is needed on this aspect of northern goshawk ecology, particularly in regards to winter foraging habitat use patterns.

The key prey species used by northern goshawks in the Sierra Nevada are ground dwellers or spend a large proportion of their time near or on the ground. These characteristics, along with the size of each species, likely renders them particularly vulnerable to goshawk predation. Relatively open shrub and lower canopy layers within forested stands may facilitate prey detection and capture by northern goshawks (Reynolds et al. 1992). This hypothesis requires further research. Douglas squirrels use mature and intermediate conifer stands containing large trees capable of providing cones and fungi, with shrub inclusions and adjacent riparian areas (Zeiner et al. 1990b). Golden-mantled ground squirrels are found in open montane conifer stands with logs, stumps, talus, or other rocks to provide cover, and with abundant fungi, forbs, or shrubs to provide food (Ibid). Belding ground squirrels are

FEIS Volume 3, Chapter 3, part 4.4, page 116 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 found in alpine dwarf-shrub, wet meadow, perennial and annual grassland, and in open, grassy stands of bitterbrush and sagebrush (Ibid). Steller’s jays prefer mature conifer forest with open to moderate canopy cover and large, mature trees (Zeiner et al. 1990a). Northern flickers use open forests and shrub habitats with abundant snags for nesting and ecotones with abundant insects and fruits for foraging (Ibid). American robins use moist, open forests for nesting with herbaceous understories or nearby meadows or clearings for foraging (Ibid).

Generalized habitat models based on best professional opinion contained in the California Wildlife Habitat Relationships (CWHR) database rate the following affected vegetation types and strata as providing high feeding habitat capability: Sierran Mixed Conifer, White Fir, Montane Hardwood-Conifer, and Montane Riparian (6, 5D, 5M, 5P, 5S, 4D, 4M); Ponderosa Pine, Jeffrey Pine, Lodgepole Pine, Subalpine Conifer, and Montane Hardwood (5D, 5M, 5P, 5S, 4D, 4M); Red Fir (5D, 5M); Eastside Pine (5D, 5M, 5P, 5S, 4D). Within CWHR, size class 6 is only recognized for a subset of the forest vegetation types (Sierran Mixed Conifer, White Fir, Montane Hardwood-Conifer, Montane Riparian, Aspen).

c. Habitat Composition at Home Range and Landscape Scales Northern goshawks require mature conifer and deciduous forests with large trees, snags, downed logs, dense canopy cover, and open understories for nesting, and use forests with dense to moderately open overstories, open understories interspersed with meadows, brush patches, riparian areas, or other natural or artificial openings for foraging. Conservation strategies proposed for northern goshawks currently recognize three spatial scales for managing northern goshawk home ranges (Reynolds et al. 1992). The first scale addresses the amount and spatial distribution of nesting habitat, the second addresses the post-fledgling area, and the third addresses the foraging areas (FA) within the remainder of the home range. Reynolds et al. (1992) proposed habitat management guidelines for forests within northern goshawk home ranges in the southwestern United States based on best professional opinion. Most research to date has focused on describing within-stand nest habitat structure and composition. Limited information is available on habitat patterns at larger and multiple scales, and how these patterns affect habitat quality for northern goshawks as measured by survival and reproduction. No empirical tests evaluating the relationship between the southwestern guidelines and northern goshawk survival and reproduction have been published and the degree to which the specific habitat guidelines are applicable to Sierra Nevada forests is uncertain. Further uncertainty exists regarding the desired number and spatial distribution of goshawk territories and habitat patterns at the landscape scale required to provide a high probability of maintaining a viable population.

Home Range Size: Northern goshawks are year-round residents in the Lake Tahoe region (Keane 1999), and likely so throughout the entire Sierra Nevada although some limited seasonal altitudinal movements may occur. Mean breeding period home ranges (95% Adaptive Kernels) are 6664 acres (sd = 2576) for males and 4980 acres (sd = 4174) for females in the Lake Tahoe region and 5928 acres (sd = 2131) for males and 3310 acres (sd = 2001) for females on the Inyo National Forest (Hargis et al. 1994, Keane 1999). Mean nonbreeding period home ranges are 20,317 acres (sd = 12,325) for males and 13,776 acres (sd = 8124) for females (Keane 1999). Nest Stand Size: While the within-stand habitat structure of nest stands has been documented, little empirical information exists regarding the amounts and spatial distribution of nest habitat associated with high quality territories. Forest stands containing nests are often small (25-250 acres) and territories may contain 1-5 alternate nest stands (Woodbridge and Detrich 1994, Squires and Reynolds 1997).

FEIS Volume 3, Chapter 3, part 4.4, page 117 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Woodbridge and Detrich (1994) reported that near 100% territory occupancy rates were observed in territories with nest stand clusters totaling 150-200 acres of nesting habitat in the southern Cascade Mountains of California. Further work is required to determine the relationship between these patterns and northern goshawk survival and reproduction.

Post-Fledging Areas: Post-fledging areas (PFA) surround the nest area and are used by both adults and the young as they learn to hunt from the time of fledging through dispersal (Reynolds et al. 1992, Kennedy et al. 1994). PFAs average about 420 acres (range 300-600) (Kennedy et al. 1994). Reynolds et al. (1992) proposed guidelines regarding the desired amounts of different forest structural classes within PFAs to provide for protective cover and for a diversity of prey species. These guidelines call for 60% of the PFA to be distributed evenly among mid-aged (12-18” dbh trees), mature (18-24” dbh trees), and old (>24” dbh trees) forest stages with canopy covers ranging from >50% to >70% depending on the forest vegetation type. The remainder of the PFA is to be managed to provide approximately even amounts of young forests and seedling-sapling/grass-forb-shrub stages. Currently, no empirical data exist that evaluate how habitat patterns at this scale are associated with northern goshawk survival and reproduction and no data exists to evaluate these guidelines relative to Sierra Nevada forests.

Foraging Areas: Foraging Areas (FA) encompass the remainder of the home range beyond the nest area and PFA. Reynolds et al. (1992) proposed guidelines regarding the desired amounts of different forest structural classes within FAs to provide for a diversity of prey species. They are similar to the desired conditions for the PFA except that canopy cover in the mid-aged, mature, and old forest structural classes ranges from >40% to >60% depending on forest vegetation type. Information on habitat patterns within home ranges is limited for northern goshawks. Hargis et al. (1994) reported that breeding period home ranges had greater vegetative and structural stage diversity compared to random locations for northern goshawks in eastside pine forests in the eastern Sierra Nevada. Understanding how prey availability for northern goshawks varies with stand structure and landscape habitat patterns is essential for understanding how to manage northern goshawk populations by providing suitable habitat for prey.

3. Diet Prey availability is a primary limiting factor for raptor populations (Newton 1979). Northern goshawks prey on a wide variety of species, although several key species/species groups are prevalent in diets from a number of northern goshawk populations: tree and ground squirrels, lagomorphs (cottontails, jackrabbits, hares), and medium and large sized birds (Squires and Reynolds 1997). Specific to the Sierra Nevada, Keane (1999) recorded 12 mammal and 22 bird species in northern goshawk diets during the breeding period in the Lake Tahoe region. Primary species were Douglas squirrel, Spermophilus spp. (golden-mantled, Belding, California ground squirrels), chipmunks (Tamias spp.), Steller’s jay, northern flicker, and American robin. Bloom et al. (1986) also reported that lagomorphs were important prey for northern goshawks in California. The relative importance of specific prey species likely varies between forest types due to variation in prey abundance and diversity within each forest type. Further work is required to document winter diets because winter prey availability differs due to the migration and hibernation patterns of important breeding period prey species. Species active and available year-round such as Douglas squirrels and snowshoe hares may even be

FEIS Volume 3, Chapter 3, part 4.4, page 118 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 relatively more important prey species during winter. Starvation, particularly during winter, is an important cause of death (Kenward et al. 1993).

4. Breeding Chronology The northern goshawk breeding period extends from mid-February through mid-September. The precise timing of breeding varies over a 3-4 week period between years, with egg-laying occurring between mid-April and mid-May. The incubation period is approximately 32-34 days. The nestling period is approximately 42-45 days and, once fledged, juveniles remain in the nest area for a period of 4-8 weeks before dispersing.

Not all pairs of northern goshawks reproduce each year (Squires and Reynolds 1997). Annual variation in reproduction is affected by weather and prey dynamics (Kostrzewa and Kostrzewa 1990, Kenward et al. 1993, Sulkava et al. 1994, Keane 1999). The proportion of territories with active nests has been documented to range from 14-100% and the proportion with successful nests from 36-83% among years in the Sierra Nevada (Bloom et al. 1986, Keane 1999). Temporal variation in reproduction appears to be associated with annual variation in late-winter/early spring weather and prey availability that influence whether a female goshawk can attain the necessary energetic condition to successfully reproduce. Reproduction in the Lake Tahoe region was greatest during a year with a warm and dry late-winter/early-spring and high numbers of Douglas squirrels resulting from high cone crop production the previous autumn (Keane 1999)

5. Migration and Dispersal Northern goshawks appear to be year-round residents in the Sierra Nevada, although limited altitudinal movements by some individuals may occur in winter (see Home Range section above). Information on dispersal for northern goshawks in the Sierra Nevada is not available. Natal dispersal distances from nest sites to breeding sites for 3 females were approximately 10, 15, and 63 miles in the southern Cascade Mountains of northern California (Detrich and Woodbridge 1994, reported in Squires and Reynolds 1997). Reynolds and Joy (1998) reported natal dispersal distances of approximately 13 mi. (sd = 5.7) for female and 10 mi. (sd = 4.1) for male northern goshawks in Arizona. Breeding dispersal also has been recorded, with some adults changing territories among years. Reported breeding dispersal distances are approximately 6 mi. (sd = 1.7) for females and 4 mi. (sd = 1.7) for males in California and 3 mi. (sd = 1.7) for females and 2 mi. (sd = 0.7) for males in Arizona (Detrich and Woodbridge 1994, Reynolds and Joy 1998). Data on dispersal distances should be interpreted with caution as they are likely biased low due to the difficulty of detecting individuals that move long distances and out of the study area.

Risk Factors 1. Habitat Amount and Distribution The major threat to northern goshawks at the present time concerns the effects of vegetation management (e.g., timber harvest, fuels treatments, etc.) and wildfire on the amount, distribution, and quality of habitat. (Bloom et al. 1986, Keane and Morrison 1994, Kennedy 1997, Squires and Reynolds 1997, Smallwood 1998, DeStefano 1998).

Assessing historic to current changes in the amount and quality of northern goshawk habitat in the Sierra Nevada is problematic due to uncertainty regarding: (1) historic vegetation conditions; (2) what constitutes high quality goshawk habitat; and (3) current vegetation

FEIS Volume 3, Chapter 3, part 4.4, page 119 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 conditions due to accuracy, resolution, and scale concerns related to current inventory maps. However, it is possible to qualitatively address these issue based on current general knowledge of northern goshawk habitat relationships and overall changes that have occurred in Sierra Nevada forests in response to predominantly selective timber harvesting and fire suppression policies. Analyses conducted at both the plot and landscape scales have documented large reductions in mature and older forests throughout the Sierra Nevada and reductions in the numbers and distribution of large trees as a result of selective harvesting of large pines, and increases in the numbers of smaller diameter trees and density of forest understories as a result of fire suppression (Laudenslayer 1990, McKelvey and Johnstone 1992, Franklin and Fites- Kaufmann 1996, Beardsley et al. 1999, Bouldin 1999). These trends suggest that there has been a reduction in the amount and distribution of the mature and older forests with large trees and open understories used for nesting by northern goshawks. Greater uncertainty exists regarding changes in foraging habitat, although limited knowledge of northern goshawk foraging habitat use would suggest that these habitat trends would also have led to a reduction in the distribution and amount of foraging habitat. Thus, although uncertainty exists, documented changes in the structure and composition of Sierra Nevada forests are predicted to have lead to a reduction in the types of habitats used for nesting and foraging based on current understanding of northern goshawk habitat relationships. It is not possible to determine if changes in the distribution and amount of habitat have resulted in northern goshawk population changes due to lack of data on historic and current population sizes and distributions.

Concern exists regarding potential loss of breeding territories due to stand-replacing fire. Approximately 17% of known northern goshawk nest sites are distributed in high fire hazard risk areas (Table 4.4.2.2b).

Table 4.4.2.2b. Distribution of northern goshawk territories reported by Sierra Nevada National Forests and in the California Department of Fish and Game database in the Sierra Nevada Bioregion by fire hazard risk rating. Fire Hazard Risk Rating (Hazard Class) Moderate (5 & 6) High (7-9) Outside fire Total Low (3& 4) polygon coverage 227 311 120 55 713

2. Breeding Site Disturbance Three factors have the potential to negatively affect northern goshawks during the breeding period resulting from direct disturbance due to: 1) vegetation treatments; 2) human recreation; and 3) falconry harvest.

Little published information exists regarding the sensitivity of northern goshawks to nest site disturbances from human activities, either recreational or related to vegetation management. No published information exists regarding the direct effects of vegetation management activities that change forest structure and composition on northern goshawk territory occupancy, reproduction, or survival. Potential treatments vary significantly in terms of both their duration and intensity, ranging from short-duration, low-intensity prescribed burns during spring to high-intensity, longer-duration mechanical treatments. Accordingly, it is likely that northern goshawks might respond differentially to treatments of different durations and intensities, although this hypothesis requires validation. Northern goshawks are known to nest in stands that have experienced understory fires that did not reduce canopy cover and numbers

FEIS Volume 3, Chapter 3, part 4.4, page 120 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 of large trees below suitable levels (Maurer 2000). Stand replacing fire events have eliminated nesting territories. Goshawks have continued to use nest stands with 100% insect kill for at least 4-5 years after tree mortality in some instances, although the long-term suitability of these sites has been eliminated.

Potential disturbance from human recreational activities has impacts at the scale of individual territories located in areas that receive recreational use or at subregional scales, such as the Lake Tahoe Basin, that receive extensive human recreational use. The problem appears most acute where active northern goshawks nests are located along trails and in areas that receive heavy foot traffic. Northern goshawks initiate breeding when the ground is still covered with snow and nests are sometimes directly located along roads and trails that provide flight access. Following meltout these sites can be prime candidates for conflict as humans begin using these roads and trails. Northern goshawks are aggressive nest defenders that will attack humans that venture into active nest stands (for example, see Gaines 1988). In some cases humans have responded by returning and shooting the birds or harassing the birds through repeated visit to the nest site. Evidence for shooting (2 sites) and continued harassment (1 site) were documented for 3 territories in the Lake Tahoe Basin, resulting in one site being abandoned (J. Keane, unpub. data). These limited case studies suggest that efforts should be directed towards minimizing the potential for negative northern goshawk-human interactions. Management of recreational activities (e.g., temporary seasonal or permanent trail closures; refusal to establish new trails and roads near northern goshawk territories) and education of recreationists are needed to minimize these potential conflicts. Although these threats are currently relevant at local scales, it is possible they may increase in scope in response to projected human population growth and increased recreational use of the Sierra Nevada in the future.

The third threat with the potential to result in disturbance to northern goshawks is the harvest for falconry. Currently the legal harvest of northern goshawks is low and does not impact the Sierra Nevada population. However, both the legal, and potentially illegal, harvest of northern goshawks has the potential to negatively impact individual territories that are repeatedly visited and harvested annually, and local populations that may sustain relatively high levels of harvest. Concern exists about this potential threat for northern goshawks in the eastern Sierra Nevada on the Inyo National Forest, though the magnitude of the effect is unknown.

3. Chemical Pollutants Chemical pollutants and heavy metals are not known to be having effects on northern goshawk populations in the Sierra Nevada Bioregion at the present time. The potential effect of Forest Service use of rodenticides and pesticides is unknown. Upslope drift of chemicals from the Central Valley should be assessed to determine their potential impact on northern goshawks, as well as, on other potentially vulnerable species.

4. Climate Weather, in conjunction with prey dynamics, appears to be a primary factor affecting northern goshawk reproduction, and potentially survival (Kostrzewa and Kostrzewa 1990, Kenward et al. 1993, Sulkava et al. 1994, Keane 1999). Climatic changes that result in colder, wetter winters and springs, or that increase the frequency and severity of spring storms has the potential to affect northern goshawk demography.

FEIS Volume 3, Chapter 3, part 4.4, page 121 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Conservation Measures 1. Current Management Direction for Northern Goshawks in the Pacific Southwest Region Pacific Southwest Region – Sierra Nevada Bioregion: Current Region 5 standards and guidelines provide for 50-125 acre management areas for northern goshawk nesting territories. No explicit standards and guidelines exist for Post-Fledging and Foraging Areas. The Tahoe National Forest has recently adopted northern goshawk management strategies that provide 200 acre nest cores and an additional 300 acre area to be managed as Post-Fledging Areas. The Lake Tahoe Basin Management Unit (LTBMU) manages nest sites to provide for at least 50 acres of nesting habitat and designates a 200 acre secondary protection zone managed for a canopy closure of not less than 40% in mature stands. Additional Tahoe Regional Planning Agency guidelines allows only management activities designed to enhance goshawk habitat within the surrounding 500 acres of goshawk habitat around nest territories. Current Region 5 standards and guidelines utilize a density standard of managing to provide habitat for one northern goshawk territory per 18 square miles.

2. Factors Used to Assess Environmental Consequences The primary threat to northern goshawks at the present time concerns the effects of vegetation management on the distribution, abundance, and quality of habitat. (Bloom et al. 1986, Keane and Morrison 1994, Kennedy 1997, Smallwood 1998, DeStefano 1998). Therefore, past and current Forest Service management direction, and vegetation and northern goshawk management strategies developed within this EIS, have and will continue to significantly contribute to the status and trends of northern goshawk populations in the Sierra Nevada. Concerns about habitat distribution, abundance, and quality are relevant at multiple spatial scales. These scales include the scale of the nest-site, nest stand, post-fledging area, home range, and landscape. At finer scales, concern also exists regarding the amounts and distribution of important micro-habitat elements. Specifically, large trees and their derivatives, large snags and logs, are important habitat elements that influence the distribution, abundance, and population dynamics of northern goshawk prey species. Individual large trees, and often snags, are used as nesting trees by northern goshawks. At the nest-site, nest stand, PFA, and home ranges scales, the issue is to manage for the necessary habitat elements and stand structural and compositional characteristics to provide high quality habitat for individual breeding territories and pairs of northern goshawks. At the landscape scale the issue is to provide for sufficient amounts and distribution of high quality habitat to maintain an adequate number and distribution of northern goshawk pairs necessary to maintain a viable population and to provide for adequate connectivity to facilitate natal and breeding dispersal among territories. Further, the issue is to provide for sufficient amounts and distribution of habitat to maintain northern goshawks well-distributed throughout their historic range in the Sierra Nevada and to avoid the creation of gaps in their current distribution.

This assessment uses three general categories of measures to evaluate the effectiveness of the alternatives to meet the goal of providing the habitat or ecological conditions to support a viable population of northern goshawks well distributed throughout the project area.

(1) Risk relative to the distribution and abundance of northern goshawk territories in the Sierra Nevada; a. survey requirements b. protection of known and newly discovered breeding territories c. size and configuration of Protected Activity Centers (PACs)

FEIS Volume 3, Chapter 3, part 4.4, page 122 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 d. management of PACs e. management of unoccupied PACs f. management of disturbance in PACs g. designation, configuration, and management of Post-fledgling Areas (PFAs) h. home range scale management i. synthesis

(2) Risk relative to the overall distribution and abundance of northern goshawk habitat at multiple scales throughout the Sierra Nevada; a. habitat elements (large trees, snags, downed woody material) b. change in nesting and foraging habitat c. change in habitat suitability of prey species

(3) Risk relative to additional risk factors; a. human disturbance b. chemical pollutants

Assumptions and Limitations Various analysis constraints and uncertainties limit the certainty with which the consequences of the alternatives on the northern goshawk can be assessed.

Lack of spatially explicit, small-scale habitat projections Habitat projections in this assessment are limited to projected changes in the total abundance of northern goshawk habitat under the different alternatives across individual national forests or other large areas within the overall range of the goshawk in the planning area. It is not appropriate to use vegetation projections generated in this assessment at small scales, such as individual home ranges. In regional planning efforts such as this one, accurate fine-scale information is lacking, and no attempt is made to spatially locate individual treatments. Such planning is left to the appropriate administrative unit (national forest or ranger district) that will have the site-specific information and understanding necessary to achieve local resource objectives. In addition, local topography, climate, and natural processes and disturbances, rather than direct management, influence many of the vegetation patterns and responses at smaller scales. Thus, it is never certain that regional management direction will have the intended outcome at each and every local site. While it would be desirable to project vegetation changes at the scale at which individual goshawks or goshawk pairs use the landscape, such resolution is beyond the capacity of this planning effort.

This lack of spatial resolution makes it infeasible to project habitat conditions at watershed or home range scales. Instead, trends are projected for larger landscapes where statistical sampling properties of the basic information layers improve the accuracy of the comparisons. While vegetation estimates are reported for large areas, these estimates are based are stand-level plots. These plots are congruent with how goshawks might perceive their environment. Thus, a reported increase in average tree diameter or canopy closure can be correctly interpreted to mean that more stands across the reporting unit will have larger trees and higher canopy closure in the future than today. This does not suggest, however, that every stand everywhere will exhibit meaningful increases, or even that the average tree diameter or canopy closure in every home range will be higher.

The end result is individual variations from the central tendency; some smaller areas are expected to be below average and some above. The concern is that some of these below (or above) average areas may be concentrated or connected in various parts of the landscape, leading to a more patchy

FEIS Volume 3, Chapter 3, part 4.4, page 123 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 distribution of habitat than apparent in the aggregate measures. Indeed, the effect of wildfire or spatially concentrated fuel treatments may create this result. Understanding this phenomenon tempers the interpretation of the aggregate results.

Assumptions used in habitat modeling projections The results from the modeling of habitat projections are also sensitive to the assumptions underlying the models. The results fom this modeling effort are based on one set of assumptions regarding high wildfire rates and treatment effectiveness. The projections are also deterministic, which is an insightful approach for holding all factors constant for general comparative assessment of treatments across all alternatives and to the “let-grow/no-treatment” scenario (that is, minimum-level alternative). However, including the magnitude of stochastic variation expected in nature would likely add a high degree of uncertainty to the modeling results that would increase over time as projections are made further into the future. Converting the gross projections of CWHR strata into estimates of habitat suitability for a species also incorporates further uncertainty because suitability scores for each strata and life-history category for each species are based on expert opinion, which while being the best available information, have not been empirically validated. Thus, all of these factors add uncertainty to the modeling projections and should further temper the interpretation of the results.

Environmental Consequences 1. Risk relative to the distribution and abundance of northern goshawk territories in the Sierra Nevada. a. Survey Requirements: Survey efforts to inventory northern goshawks throughout the Sierra Nevada are incomplete to date and some unknown number of breeding territories has not been documented. The likelihood of locating (and subsequently protecting) these additional goshawk territories is lowest under Alternative 1 since surveys are not required. Under Alternative 1 the number of known goshawk territories would increase slowly over time as new locations were opportunistically discovered. Alternatives 2, 3, 4, 5, 6, 7, 8, and modified-8 establish standards requiring goshawk surveys to protocol for all activities that occur in suitable nesting habitat. The proportion of goshawk breeding territories and nest stands known, and subsequently protected, would be higher under these alternatives.

b. Proportion of Northern Goshawk Breeding Territories Protected: Current standards and guidelines under Alternative 1 require management to provide habitat conditions to support one northern goshawk territory per 18 square miles. Although desirable, there is no requirement that these management areas be located to coincide with an actual goshawk breeding location on the ground. Thus, some “goshawk management areas” may not, nor ever have, supported breeding goshawks. Alternatives 2, 3, 4, 5, 6, 7, 8, and modified-8 require establishment of Protected Activity Centers (PACs) for all known and newly discovered goshawk breeding territories. In the absence of information and a scientifically defensible basis regarding the desired density and distribution of northern goshawk pairs and territories to maintain a viable population, there is risk associated with not providing protection to all territories, as occurs under Alternative 1.

c. Size and Configuration of Protected Activity Centers. Current standards and guidelines under Alternative 1 require management to provide 50-125 acres of nesting habitat for northern goshawk management areas. PACs are to be 200 acres in

FEIS Volume 3, Chapter 3, part 4.4, page 124 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 size under Alternatives 2, 3, 4, 5, 6, 7, 8, and modified-8. A PAC size of 200 acres is based on criteria reported by Woodbridge and Detrich (1994). They reported that territory occupancy rates were correlated with the amount of available nesting habitat such that occupancy rates of approximately 100% were associated with clusters of available nest stands totaling 150-200 acres. Thus, this standard and guideline is consistent with the best available science. Further research is required to determine how these amounts of nest habitat are related to northern goshawk fitness (survival and reproduction) over longer time periods. The current standards and guidelines under Alternative 1 are not consistent with current scientific information. PACs are to consist of the best available nesting habitat in the largest contiguous blocks possible, and where habitat is patchy, PACs should consist of multiple >30 acres patches within 0.5 miles. d. Management within Protected Activity Centers. All Alternatives limit activities within PACs to those designed to improve the suitability or integrity of the PAC. Fuels treatment and vegetation management that can occur within PACs differ between Alternatives. The main issue concerning vegetation treatments in PACs is the uncertainty that exists related to the trade-off between treating PACs, with the goal of reducing their susceptibility to stand replacing fires, versus the potential negative or positive effects of treatments on northern goshawk occupancy and habitat quality. It is reasonable to hypothesize that, given historic fire patterns in the Sierra Nevada, light underburns similar to those that occurred prior to the late 1800s would not result in territory abandonment provided that high canopy cover and high densities of large trees in nest stands were not affected. Treatments, such as prescribed burning, that maintain high canopy cover and the high densities of large trees in nest stands would be expected to have lower effects on northern goshawks than mechanical thinning that might removes trees in the 12-20” or 20-30” dbh size classes, depending on the conditions of the nest stand. However, no empirical data are available to address the effects of various fuels treatments on northern goshawk occupancy, survival, and reproduction in PACs. Thus it is difficult to evaluate the differences between Alternatives and their different proposed standards and guidelines without empirical data or quantitative models to assess the trade-offs associated with different strategies.

Given this scientific uncertainty, all alternatives entail degrees of risk associated with the management of PACs. Alternatives that provide the opportunity to address this uncertainty through the use of formal adaptive management projects to better understand the effects of treating PACs have the potential to advance scientific knowledge and improve northern goshawk habitat management. Uncoordinated efforts to conduct treatments within PACs outside of a formal adaptive management framework would provide only limited opportunity to measure management effects at best and would result in a squandered opportunity to address an important area of scientific uncertainty. Additionally, the rates at which treatments would occur also need to be factored into the assessment. Treating a high proportion of territories without knowledge of how treatments will affect habitat quality nor how they will affect short and long-term fire risk, and without a formal monitoring strategy, entails greater risk and uncertainty. Alternatives 2, 4, 5, 6, 7, and 8 require formal adaptive management approaches to assess the effects of treatments on PACs.

Alternative 2 permits no habitat altering activities within PACs unless associated with a formal research project. Only light underburning and hand clearing are permitted in Alternative 5, unless associated with a formal research project to assess effects. Alternative 8 permits light underburning and mechanical treatment of trees <12” dbh, with a BE required to justify

FEIS Volume 3, Chapter 3, part 4.4, page 125 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 management activity within a PAC. Alternatives 1, 3, 4, 6, and 7 permit sufficient fuels treatment in up to 30% of the PAC to meet fuels objectives while attempting to minimize reductions in habitat suitability. Alternatives 1 and 3 do not require establishment of an adaptive management strategy. A greater number of PACs could be expected to be affected annually under Alternatives 1, 3, and 5 relative to the other alternatives. Alternatives 4, 6, and 7 allow treatment in no more than 10% of PACs per decade per National Forest. Alternative 8 permits activities in PACs but limit the number of PACs treated annually to 1% of the total number on the Forest; if Forests plan treatments affecting a greater proportion of PACs, they must occur as part of a formal research study designed to monitor treatment effects. The use of BEs to justify treatments in PACs introduces uncertainty because it is difficult to evaluate the quality of individual BEs and the criteria used to reach conclusions, and given the lack of scientific knowledge about treatment effects, it is not possible to evaluate outcomes unless a formal adaptive management strategy is implemented.

Alternative modified-8 limits treatments in PACs located outside of the inner defense zone of the urban buffer to the use of prescribed fire and hand clearing to reduce surface and ladder fuels. Vegetation treatments can occur in up to 5% of PACs per year or 10% per decade unless a formal monitoring and adaptive management approach is developed with the PSW Research Station. LOPs can be waived in up to 5% of PACs per year to allow for early-season prescribed burning. PACs within the inner urban zone receive a 500-ft buffer around the nest trees, with mechanical treatment allowed in the remainder of the PAC. Modified-8 does not require establishment of a formal adaptive management strategy to assess the effects of treatments on PACs.

In summary, Alternatives 2, 4, 5, 6, 7 and 8 most fully address uncertainty regarding treatment effects on goshawk occupancy and habitat quality by requiring formal adaptive management studies. Alternatives 2, 5, 8, and modified-8 provide the clearest direction to limit treatments to prescribed fire and limited thinning within PACs, thereby likely incurring the lowest risk to goshawk occupancy and habitat quality. Alternative 1 likely incurs the greatest risk because it permits a larger number of territories to be potentially treated without a formal mechanism for assessing treatment effects.

While uncertainty regarding treatments effects on goshawk occupancy and habitat quality versus the potential benefits accrued because of decreased fire risk is an issue throughout the Sierra Nevada, it is of particular concern on the Lassen, Plumas, and portions of the Tahoe National Forests under the Herger-Feinstein Quincy Library Group Forest Recovery Act. Direction under the Preferred Alternative in the Final Environmental Impact Statement indicates that approximately 45% (91/202) of northern goshawk management territories would potentially be entered for treatments over the 5 year project period (FEIS 1999, Appendix AA- 52). Treatments would also result in a potential 12% decline in suitable nesting habitat and 13% decline in suitable foraging habitat based on CWHR model projections over the 5 year study project period (FEIS 1999, Appendix AA-51). Given the uncertainty regarding treatment effects and the lack of an adequately defined monitoring strategy, the high rates of activities proposed in this alternative incur high levels of risk and uncertainty.

e. Management of Unoccupied PACs. PACs may be eliminated from the network under all alternatives if rendered unsuitable due to stand-replacing fire events. Alternatives 2, 3, 5, 6, 7, 8, and modified-8 maintain all PACs in

FEIS Volume 3, Chapter 3, part 4.4, page 126 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 the network unless they are rendered unsuitable by wildfire and protocol surveys indicate that they are no longer occupied. Alternatives 1 and 4 permits PACs to be eliminated from the network if surveys determine they are unoccupied for 2 years. The potential to eliminate PACs based on 2 years of survey effort results in much greater increased risk for negatively affecting northern goshawk breeding territories. Two years of survey effort are inadequate to determine if a territory is unoccupied because of several reasons. First, not all northern goshawk pairs reproduce each year due to annual variation in weather and prey availability, though the nonbreeding adults may still be occupying their territory (Keane 1999). Current survey efforts, primarily broadcast calling (Kennedy and Stahlecker 1993, Joy et al. 1994, Watson et al. 1999) conducted during the nestling and fledgling dependency periods, likely have a low probability of detecting nonbreeding adult northern goshawks on occupied territories and adults occupying territories that have experienced early nest failures. Thus, survey efforts conducted during consecutive years of low reproductive activity or success may incorrectly determine that a territory is inactive when indeed it is still occupied and the survey effort does not have a high probability of detecting the nonbreeding adults. The majority of nest failures occurred during the incubation period of the nesting cycle for a population of northern goshawks in the Lake Tahoe Region of the Sierra Nevada (Keane 1999). Surveys are usually conducted during nestling and fledgling dependency periods of the nest cycle to avoid possible disturbance to the birds during the incubation period. Use of dawn vocalization surveys during the courtship period of the nesting cycle may help address this issues (Penteriani 1999). The third reason 2 years of surveys are inadequate is because northern goshawk populations may exhibit temporal variability in size and territories may become occupied and unoccupied over time in response to population fluctuations. f. Management of Human Disturbance in PACs The Alternatives differ in their management of human disturbance in PACs. Alternatives 2, 3, 5, 8, and modified-8 maintain a limited operating period (LOP) prohibiting activities within 1/4 mile around active nests during the breeding period (15 February – 15 September) unless surveys confirm that goshawks are not nesting. Alternatives 4, 6 and 7 have similar LOP requirements except that LOPs can be waived if the activity is needed to protect human health or safety. Under Alternatives 2-8 a Biological Evaluation (BE) can be used to waive the LOP for individual projects of limited scope and duration if they are determined not to result in disturbance to breeding goshawks considering their intensity, duration, timing, and specific location. The use of BEs to justify waiver of LOPs introduces uncertainty because it is difficult to evaluate the quality of individual BEs and the criteria used to reach conclusions, and given the lack of scientific knowledge regarding the types and levels of disturbance that affect breeding northern goshawks. Uncertainty is further increased under Alternatives 4, 6 and 7 because activities can be justified if needed to protect human health and safety. The conditions indicating when this waiver would apply have not been described, rendering it difficult to assess the potential magnitude of it’s effect. Alternative 1 does not require LOPs. The potential for disturbance from human activities (e.g., recreation, spring fuelwood harvest) is a greater risk under this alternative. g. Designation, configuration, and management of Post-fledgling Areas (PFAs) Explicit management direction for PFAs is proposed only under Alternatives 2 and 8. PFAs under these two alternatives are to be 400 acres in size. In westside forests under both alternatives, PFAs are to be managed for 20% of the area to consist of stands with >24” dbh trees and >70% canopy cover, with the remainder to be managed for >11” dbh trees and >50%

FEIS Volume 3, Chapter 3, part 4.4, page 127 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 canopy cover. In eastside forests under both alternatives, PFAs are to be managed for 20% of the area to consist of stands with >24” dbh trees and >50% canopy cover, with the remainder to be managed for >11” dbh trees and >50% canopy cover. Alternatives 3-7 do not incorporate a PFA and therefore, habitat conditions surrounding PACs are more uncertain. Alternatives 3-7 incorporate greater uncertainty because they lack the spatial specificity addressed by Alternatives 2 and 8. Alternatives 2 and 8 more fully address the objective of the PFA concept to provide adequate foraging habitat and prey for juvenile goshawks during the post-fledging dependency period. Whether the desired habitat conditions are adequate or optimal to meet this objective is uncertain and requires validation.

h. Home range scale management Explicit home range scale management is proposed only under Alternatives 2 and 8. Under Alternative 2 home range scale management would be applied to northern goshawk territories located outside of integrated biodiversity reserves on the westside, and for all territories on the eastside. Westside home ranges would be delineated by a 2-mile radius around each nest-site within which 60% of the area would be managed for a mosaic of mid-mature to late- successional forest structure. Eastside territories would be a minimum of 2500 acres in size and managed for 40% of the area in stands with >24” dbh trees and >50% canopy cover. Under Alternative 8, home range scale management would be applied to all northern goshawk territories on the eastside. Home ranges would be designated as a minimum of 2500 acres managed for 40% of the area with >24” dbh trees and >50% canopy cover. No empirical information is available to determine if these standards and guidelines are or are not adequate or optimal to provide high quality habitat at the home range scale. Given the habitat requirements of important prey species, it is reasonable to hypothesize that these conditions would provide for adequate prey populations. However, this hypothesis requires validation. Management of habitat within northern goshawk home ranges beyond the PAC under the other Alternatives would be dictated by the underlying land allocation and associated treatments that are permitted. The proportion of known territories located in major land allocations (wilderness, integrated biodiversity reserves, old forest emphasis areas) that will receive limited treatments (prescribed fire, limited mechanical underthinning) varies across Alternatives (Table 4.4.2.2c).

FEIS Volume 3, Chapter 3, part 4.4, page 128 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Table 4.4.2.2c. Percent of known northern goshawk breeding territories reported by Sierra Nevada National Forests and in the California Department of Fish and Game database that are located in land allocations (wilderness, integrated biodiversity reserves, old forest emphasis areas) designated to receive limited treatment under each FEIS Alternative. Alternative Database 1 2 3 4 5 6 7 8 Mod 8 Total # USFS <1 84 39 18 58 37 <1 60 45 588 CDFG 8 51 30 17 38 37 8 40 42 125

i. Synthesis Table 4.4.2.2d summarizes the consequences of the alternatives relative to eight factors affecting the distribution and abundance of northern goshawk PACs. Uncertainty exists as to whether mechanical treatments, prescribed burning, or both would maintain or enhance northern goshawk habitat. In addition, there is no empirical date available that addresses the effects of various fuel treatments on northern goshawk occupancy, survival, or reproduction in PACs.

Table 4.4.2.2d. Consequences of the alternatives on eight factors affecting northern goshawks. Disturbance and Abundance Alternative Surveys PAC PAC Size Mechanical Fuels Addresses LOP Mgt. of Required Protected Treatments Treatment uncertainty of req. unoccupied treatments PACs 1 No 1 per 18 50-125 ac *Mechanical * underburn Low No **2 years sq. mi. treatments 2 Yes All 200 ac Formal Formal High Yes -Maintain all research only research only 3 Yes All 200 ac *Mechanical *underburn Moderate Yes -Maintain all Treatment 4 Yes All 200 ac *Mechanical *underburn Moderate No **2 years Treatment 5 Yes All 200 ac Hand clearing * underburn High Yes -Maintain all only 6 Yes All 200 ac *Mechanical * underburn Moderate Yes -Maintain all Treatment 7 Yes All 200 ac *Mechanical * underburn Low Yes -Maintain all Treatment 8 Yes All 200 ac *Mechanical * underburn Moderate Yes -Maintain all Treatment Mod 8 Yes All 200 ac *Mechanical * underburn High Yes -Maintain all Treatment

*requires biological evaluation need to demonstrate the need **unoccupied for 2 years and surveyed to protocol -unless rendered unsuitable by wildfire

2. Risk relative to the distribution and abundance of northern goshawk habitat. a. Change in Habitat Elements over Time. Habitat changes based on projections of CWHR classes suggests general increases in mature and late-seral forests, and increases in the numbers of >30” dbh trees. However, concerns exist at finer scales (i.e., within forest stand or CWHR classes) regarding proposed large tree density standards and guidelines. Available vegetation plot data from northern goshawk nest sites in the Lake Tahoe region indicate that northern goshawk nest sites on average are characterized by high densities of large trees. These numbers are based on plot scale data (0.25 acre) and are

FEIS Volume 3, Chapter 3, part 4.4, page 129 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 measures of within stand patches of high large tree density and may not reflect average per acre large tree densities for multi-acre stands. However, they do indicate that these within stand patches of high large tree densities provide nest habitat and are used by northern goshawks.

Within PACs activities are limited to those designed to improve the suitability or integrity of the nesting habitat within the PAC. Thus, there should be no reduction in large tree densities. However, outside of PACs, vegetation management is guided, depending on the Alternative and land allocation within the Alternative, by desired condition for large tree densities, which further vary by vegetation type and site quality (see Table 4.4.2.2c). Desired large tree densities vary between 4-6 trees in areas managed for late-seral forest emphasis areas. It is uncertain if management to meet these proposed standards and guidelines will produce stand structures that meet requirements for northern goshawk nesting habitat for two reasons. First, average large tree densities expressed as desired number per acre do not address the spatial distribution of large trees. Second, densities of 4-6 large trees per acre may not meet the number found in northern goshawk nest stands. For these two reasons it is uncertain if the proposed large tree density standards and guidelines will provide the numbers and patterns of large trees that characterize northern goshawk nest habitat. All alternatives address this uncertainty to some degree by retaining all >30” trees on the Westside and >24” trees on the eastside.

b. Change in Nesting and Foraging Habitat The distribution and abundance of nesting habitat and prey are considered to be an important limiting factor for raptor populations (Newton 1979). Northern goshawks nest in mature, late- seral, and old-growth forests, and these forest structural stages are also rated to provide high suitability foraging habitat (see Affected Environment). The overall regionwide trends in total amounts of high suitability nesting and foraging habitat (CWHR classes 4M, 4D, 5M, 5D, 6) are projected to slightly increase over the next 50 years (Figure 4.4.2.2a). However, greater changes in the relative differences within each of the five individual CWHR strata are projected to occur over the next 50 years (Table 4.4.2.2e). In general, classes 4M, 4D, and 6 are transitioning into classes 5M and 5D through growth (Figures 4.4.2.2b to f). All of the above projected trends are similar across Alternatives, suggesting a general increasing trend towards more late-seral conditions, although the magnitude of the differences among Alternatives is difficult to interpret with confidence due to variation inherent in the vegetation information and assumptions used in the modeling process. Changes in overall habitat suitability across all vegetation classes estimated using the CWHR habitat model suggest only moderate changes in habitat suitability across the Alternatives (Table 4.4.2.2f, Figure 4.4.2.2g). Greater differences among Alternatives are observed over longer time periods in the projections of individual CWHR strata and overall suitability scores. However, confidence in these longer-term future projections is further lowered due to additional uncertainty regarding conditions further into the future. In summary, the modeling projections results suggest general inceasing trends in the larger tree size CWHR strata over time.

FEIS Volume 3, Chapter 3, part 4.4, page 130 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Table 4.4.2.2e. Projected percent changes in the amount of high and moderate suitability northern goshawk nesting and foraging habitat from the current to 50 years in the future across the FEIS Alternatives. CWHR Strata Alternative 6 5D 5M 4D 4M Total Current 1,120 166 662 1,145 1,206 4,301 (1,000s acres) MLV* 21.5 341.9 46.7 -38.6 -37.3 5.4 1 1.7 386.5 77.5 -40.9 -32.8 7.3 2 23.9 382.3 60.9 -38.7 -35.3 10.2 3 9.7 552.6 84.4 -41.8 -29.0 17.6 4 -1.4 388.2 107.2 -47.7 -26.2 11.0 5 16.0 434.6 67.7 -42.3 -32.5 10.8 6 13.5 500.9 88.7 -41.8 -31.2 16.7 7 13.2 400.8 85.7 -40.6 -28.9 13.3 8 19.6 401.8 72.7 -41.0 -33.5 11.4 Mod 8 18.4 454.7 67.3 -39.7 -34.2 12.8 Mean Change 13.6 424.4 75.9 -41.3 -32.1 11.7

*MLV = No Treatment, Let-Grow Scenario

Table 4.4.2.2f. Projected percent changes in overall habitat suitability scores based on CWHR habitat models from the current to 50 years in the future across the FEIS Alternatives. Alternative MLV* 1 2 3 4 5 6 7 8 Mod 8 Mean -4.7 4.1 -2.2 8.1 9.2 0.1 7.3 5.5 0.5 4.6 3.2

*MLV = No Treatment, Let-Grow Scenario

c. Change in Habitat Suitability for Prey Species Projected changes in overall habitat suitability for select northern goshawk prey species were estimated using CWHR habitat suitability ratings and vegetation projections (Appendix B). Overall, 28% of the species (8/29) had average projected increases in habitat suitability across the alternatives, while 21 species had projected decreases (Table 4.4.2.2g). Of particular note is the projected increase in habitat suitability for Douglas squirrels across the alternatives. Overall, the results are consistent with the general projections of increasing amounts and distribution of late-seral/old-growth forest conditions, with decreases in habitat suitability for early-seral associated prey species and increases in species associated with mature forest vegetation. Some species (e.g., American robin, golden-mantled ground squirrel) are also abundant in riparian and meadow environments that were not fully modeled in the vegetation projections and thus should remain abundant throughout the Sierra Nevada despite modest projected decreases in habitat suitability within the forested vegetation types.

3. Risk Relative to Additional Risk Factors a. Human Disturbance Management of human disturbance to northern goshawk breeding territories is addressed under section 1.f. under Environmental Consequences listed above.

FEIS Volume 3, Chapter 3, part 4.4, page 131 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 b. Chemical Pollutants No information is available to suggest that chemical or heavy metals are affecting northern goshawks in the Sierra Nevada at the present time.

4. Synthesis of Environmental Consequences Population trends of northern goshawks in the Sierra Nevada are unknown. Uncertainty exists regarding what constitutes high quality habitat at the home range and landscape scales, and how changes in population density and distribution affect the likelihood of persistence. Uncertainty also results from the lack of ability to conduct spatial analyses that demonstrate how habitat will be distributed at home range and landscape scales under the EIS Alternatives at the present time. Finally, uncertainty exists regarding the sensitivity of model habitat projections to assumptions regarding the expected rates of wildfire and the effectiveness of treatments on wildfire behavior and effects. Therefore, given such high levels of uncertainty, it is difficult to evaluate how each of the Alternatives will affect northern goshawk viability on FS lands in the Sierra Nevada. However, at this time it is possible to synthesize information related to two general factors that provide important contributions to maintaining a viable population of northern goshawks and a high probability of persistence, and are useful for evaluation and comparison among alternatives. The first factor relates to how standards and guidelines are expected to affect the management of individual northern goshawk territories. The assumption is that management directed at protecting and enhancing all territories will provide a greater contribution to the probability of persistence then would a strategy that does not protect all territories given the uncertainty regarding the density and distribution of territories required to provide for a viable population. The second factor relates to projected changes in the amounts of habitat and overall changes in habitat suitability that occur across Forest Service lands in the Sierra Nevada, albeit that habitat projections are available for only one set of wildfire and treatment effectiveness assumptions. The assumption is that management that provides the greatest increase in the amounts and distribution of habitat will provide a greater contribution to the probability of persistence given that northern goshawk populations are suspected to be declining and that habitat distribution, abundance, and quality are considered to be the greatest threat to viability.

Considering factor one, Alternatives 2, 5, 8, and modified-8 require survey efforts, provide protection for all known and newly discovered northern goshawk territories, maintain PACs over time, provide for low rates of vegetation treatment within PACs, and provide for opportunities to conduct adaptive monitoring of treatment effects if higher rates of treatment are desired. These are all positive measures that contribute to maintaining northern goshawk viability in the Sierra Nevada given uncertainty about the density and distribution of territories that is required to provide for a high probability of persistence and about the effects of vegetation treatments on northern goshawk occupancy and habitat quality. Alternatives 3 and 7 incur greater risk relative to the above alternative because it allows for potential higher rates of vegetation treatments within PACs. Alternative 4 incurs greater risk than the above alternatives because although it requires surveys and designates PACs for all territories, it allows PACs to be eliminated from the network based on inadequate survey effort to determine if a PAC is no longer occupied. Further, Alternative 4 allows for potential high rates of vegetation treatments within PACs without requirement for formal adaptive monitoring to ascertain treatment effects on northern goshawk occupancy and habitat quality. Alternatives 1 engenders much greater risk and uncertainty relative to the other alternatives because it does not provide protection for all known and newly discovered territories and allows for potential high rates of treatments within PACs without requirements for formal adaptive monitoring to ascertain treatment effects on northern goshawk occupancy and habitat quality.

FEIS Volume 3, Chapter 3, part 4.4, page 132 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 A major issue of concern across all Alternatives regards the high rates of vegetation treatments proposed over a short time period in the QLG plan. This plan incurs greater risk because of uncertainty regarding treatment effects and a high proportion of northern goshawk territories will be treated in the project area without an explicit and defensible adaptive monitoring strategy to assess treatment effects.

The lack of an explicit strategy to address the uncertainty related to how various fuels treatment techniques affect northern goshawk PACs and habitat is a shortcoming of all Alternatives. Approximately 17% of known northern goshawk territories occur in high fire hazard risk areas. It is important to address this uncertainty because concerns about fire effects will likely intensify in the future. Further, northern goshawk nest stands typically have relatively open understories. Thus, evaluation of the potential positive and negative effects of various treatments or combinations of treatments is critical to understanding how to reduce fire risk and manage for high canopy cover and high densities of large trees characteristic of northern goshawk nesting habitat. All Alternatives fail in this regard because none establish an explicit prioritization system and adaptive monitoring strategy for treating territories at high risk of fire and assessing treatment effects. The uncertainty associated with these issues will continue to be a concern in the future.

Increases in the distribution and abundance of late-seral/old-growth forest and associated habitat elements such large trees, snags, and downed logs, the reintroduction of fire as an ecological process, and the protection afforded to riparian areas, would be expected to have positive benefits for northern goshawks based on current understanding of nesting and foraging requirements. Within the limitations and assumptions of the vegetation modeling conducted as part of this assessment, and the CWHR expert system habitat models and empirical nesting habitat data used, there appear to be general increases in the amount of nesting habitat used by northern goshawks (5M, 5D CWHR Class). Projected trends are similar across all Alternatives, suggesting little variation between Alternatives. To date the analyses have been conducted considering all modeled vegetation types across the entire project area to provide gross estimates of projected changes in habitat. Interpretations of projected changes in habitat may differ at the scale of specific forest types and individual National Forests as compared to results from the gross projections conducted across the entire project area. An additional area of uncertainty related to habitat projections across all Alternatives at finer scales is concerned with the adequacy of proposed large tree density standards and guidelines to an increase in the distribution and abundance of forest stands with the numbers and spatial distribution of large tree that are characteristic of northern goshawk nesting habitat. Alternatives 2, 3, 5, 6, 8, and modified-8 best address this uncertainty by retaining all large trees across all land allocations. Alternative modified-8 also identifies and limits treatment to prescribed fire and light understory thinning in all CWHR strata 5D, 5M, and 6 stands that are one acre or greater in size. Protection of these existing patches of old forest conditions should result in a positive contribution for maintaining northern goshawk habitat.

There is still some unknown degree of uncertainty associated with concluding that any of the Alternatives will provide for a viable population of northern goshawks in the Sierra Nevada. Lack of information related to how northern goshawk survival and reproduction vary in relation to habitat patterns at the home range and landscape scales, and lack of information to assess projected habitat changes at the scale of individual home ranges, precludes a complete, quantitative assessment of the projected effects of the EIS Alternatives on northern goshawk viability in the Sierra Nevada. Explicit and defensible adaptive monitoring strategies and associated research are needed to address current uncertainties and assess management effects.

FEIS Volume 3, Chapter 3, part 4.4, page 133 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Preferred Alternative Survey efforts to locate and establish PACs for all northern goshawk breeding territories under the preferred alternative (modified-8) are positive measures that should contribute to continued persistence of the species well distributed throughout the Sierra Nevada. Treatments within PACs are limited to prescribed burning and light mechanical understory thinning. While these treatments may have short-term effects (for example, one year) on small numbers of individual pairs due to prescribed burning during Spring, the long-term effects of these treatments should be positive due to the reduction of understory vegetation in nest stands and a reduction in the threat of losing the nest stands to high-intensity, stand-replacing wildfire. One missing component of the preferred alternative is an explicit commitment to establish a formal adaptive management study to test these assumptions regarding the response of northern goshawks and vegetation to fuels treatments in PACs.

Uncertainty exists regarding the effects of the preferred alternative on northern goshawk habitat at the scale of the home range. Individual goshawks and pairs have large home ranges and forage in a large area surrounding the PAC. It is unknown how habitat patterns at the home range and landscape spatial scales affect northern goshawk survival, reproduction, and population dynamics, and the degree to which the various fuels and silvicultural treatments affect habitat quality at all spatial scales. No explicit management is directed at managing foraging habitat for northern goshawk pairs beyond the scale of the PAC under the preferred alternative. Thus, the distribution, abundance, and composition of foraging habitat surrounding PACs will be dictated by the underlying land allocation and associated standards and guidelines (Table 4.4.2.2h). The distribution of reported northern goshawk territories by major land allocation provides some general insight into the number of home ranges that occur within each allocation, although most home ranges likely will encompass varying proportions of the different land allocations.

Sierra Nevada-wide, approximately 3% of the reported territories are located in the inner urban zone where they will receive minimal protection because the predominant objective in this zone is to modify vegetation structure to change fire behavior to lower the risk to human health and safety. Approximately 17% of the reported territories are located in the outer urban zone. PACs will be established for each of these territories, while the remainder of the outer urban zone will be managed to establish Strategically Placed Landscape Area Treatments (SPLATs) over 30-40% of the area. Approximately 75% of the known territories are mapped outside of the urban zones, with about 34% of the total located within the old forest emphasis areas (OFEAs). PACs would be established for all of these territories and vegetation treatments within OFEAs would be limited to prescribed burning and light mechanical understory thinning.

About 41% of the reported territories are mapped in the General Forest allocation where a wider range of vegetation treatments are permitted. The primary objective within the General Forest is to establish through treatment or maintain existing conditions such that 30-40% of the landscape is in SPLATs. The combination of treatments used within the General Forest, as well as in the outer urban zone, to meet these landscape objectives will likely have differential effects on the composition of foraging habitat for northern goshawks. If the emphasis is on prescribed burning and limited mechanical treatments (for example, prescriptions 15 and 21) and the emphasis is to take advantage of existing landscape features or conditions (for example, brush fields, meadows, rock outcrops, plantations, old timber harvets units/clearcuts, etc.), there will likely be less emphasis on implementing mechanical treatments (for example, prescription 31) in CWHR strata 4D and 4M. If the emphasis is on mechanical treatments because of smoke, risk, financial or other constraints that inhibit the magnitude of prescribed burning that occurs, then the emphasis could be on treating

FEIS Volume 3, Chapter 3, part 4.4, page 134 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 CWHR strata 4D and 4M stands across up to 35% of the landscape/management unit. This latter scenario incurs greater uncertainty and potentially greater risk to maintaining adequate amounts of mid- to late-seral forest stands to function as foraging habitat in near proximity to northern goshawk breeding territories and PACs.

This spatial uncertainty regarding the distribution of foraging habitat around northern goshawk breeding territories is a greater concern within General Forest on the eastside of the Sierra Nevada. On the westside, protection of CWHR strata 6, 5D, and 5M, designation of OFEAs and riparian buffers, and the standards and guidelines developed as conservation measures for the California spotted owl and fisher throughout the General Forest provide a greater degree of confidence that mid- to late-seral forests used for foraging will be abundant and well-distributed throughout the Sierra Nevada. A total of 256 territories reported by the National Forests plus an additional 37 listed in the CDFG database are mapped as occurring on the eastside of the Sierra Nevada. Of these, 103 are located in OFEA or wilderness land allocations, of which 11 are located within the uban buffer zones. The remaining 190 of the territories are located in the General Forest land allocation or not on National forest lands, of which 20 are located in the urban buffer zones. Protection of one acre and greater stands of CWHR strata 6, 5D, and 5M and riparian buffers will contribute to the certainty of maintaining mid- and late-seral stands throughout the general forest, although greater spatial uncertainty exists regarding the effects of treatments throughout the remainder of the General Forest and the mix of treatments and CWHR strata that will be applied to meet landscape fuels objectives.

In summary, population trends of northern goshawks are unknown and uncertainty exists regarding the relationship between northern goshawk survival and reproduction and habitat patterns at multiple spatial scales, as well as, how treatments affect prey distribution and availability and goshawk occupancy of PACs and home ranges. These magnitudes of uncertainty are commonplace in attempting to develop and assess management strategies for species conservation and dictate that interm strategies that do not foreclose future options be implemented until better information is available to guide management (Ruggerio and McKelvey 2000). Protection of all known northern goshawk territories through establishment and conservative managemnt of PACs should contribute to maintaining northern goshawks throughout the Sierra Nevada. Greater certainty exists that mid- and late-seral forest stands should be well distributed throughout northern goshawk home ranges throughout the westside of the Sierra Nevada. Uncertainty is greater regarding future vegetation conditions in eastside forests, particularly throughout the General Forest, because of the spatial uncertainty regarding the mix of treatments that might be implemented and how changes in the mix would disproportionately target different CWHR strata. Research directed at key uncertainties is an important component in the furthering the development of the scientific knowledge base required to improve management for species conservation (Aubry et al. 2000). Research, monitoring, and adaptive management approaches are needed, particularly in eastside pine and mixed-conifer forest types, to address key uncertainties in the following aspects of northern goshawk ecology in the Sierra Nevada:

1. effects of fuels treatments and wildfire on vegetation structure in PACs and the response of northern goshawks as measured by occupancy, reproduction, and survival; 2. how habitat patterns at multiple scales are related to northern goshawk survival and reproduction; 3. how fuels and vegetation treatments and wildfire affect the distribution and availability of prey and goshawk foraging behavior; and 4. what are the current population trends of northern goshawks.

FEIS Volume 3, Chapter 3, part 4.4, page 135 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Table 4.4.2.2h. Distribution of known northern goshawk territories reported by Sierra Nevada National Forests by major land allocation under the preferred alternative. Numbers in parentheses are territories located in the California Department of Fish and Game database. Allocation In Old Forest Outside Old Forest Total Emphasis Area Emphasis Area Inner Urban Zone 4 (1) 14 (2) 18 (3) Outer Urban Zone 54 (7) 50 (7) 104 (14) Outside Urban Zones 201 (40) 253 (39) 454 (79) Outside Polygon coverages - - 12 (29) Total 259 (48) 317 (48) 588 (713)

Results of the Assessment for the Northern Goshawk Environmental and population outcomes were derived from professional opinion to estimate environmental and population conditions that would exist in 50 years for the northern goshawk under each alternative. Assigning these outcomes is inherently subjective, although based on a reasoned thought process and the best available information. The environmental outcome addresses the capability of the environment on National Forest Service lands to support northern goshawk populations. The population outcome addresses environmental conditions on all lands within the bioregion and other risk factors that may affect population abundance and distribution,

Table 4.4.2.2i. Estimated outcome ratings for the northern goshawk. Environmental outcomes evaluate the estimated environmental conditions on Forest Service lands after 50 years under each alternative. Population outcomes evaluate the estimated population conditions based on the environment outcome and other risk factors. Alternative Outcome Current 1 2 3 4 5 6 7 8 Mod 8 Environment B C B- B+ C+ B+ B+ B- B+ B+ Population B C- C+ B C B B C+ B B

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Rationale for Ratings The northern goshawk is associated with mature conifer and deciduous forests with complex forest structure. Nesting habitat has more dense closed canopies with open understories. Northern goshawks forage in stands with moderately open to dense overstories as well as in habitat mosaics of forest stands with open understories interspersed with riparian areas, meadows, or other openings. The northern goshawk has a nearly continuous distribution throughout its range in the Sierra Nevada. Because of the broad distribution of northern

FEIS Volume 3, Chapter 3, part 4.4, page 136 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 goshawks throughout their range in the Sierra Nevada and their dispersal capabilities, suitable environments are likely to be sufficiently abundant and broadly distributed to permit dispersal and nearly continuous interaction among subpopulations across the planning area at the current time (Outcome B). Because they use a greater diversity of habitats and are more broadly distributed, likely outcomes for the northern goshawk are generally higher than those for the California spotted owl.

The following criteria were used and assessed to arrive at the environmental outcome for each alternative: (1) CWHR habitat projections; (2) anticipated effects of Standards and Guidelines relative to management of individual northern goshawk territories; and (3) the anticipated effects of the Standards and Guidelines relative to the designation and management of the major land allocations. Details relative to these criteria and the alternatives are discussed throughout the risk factor and environmental consequences section of this species account. Two additional factors were factored into the assessment of the population outcome: (1) assumed continued growth of the human population in the Sierra Nevada leading to increased development, infrastructure, and recreation; and (2) stable or increasing timber harvest levels on private lands.

Model projections of CWHR strata rated as high suitability habitat for northern goshawks (5D, 5M, 6, 4D, 4M) suggest a trend toward increasing amounts of these types across all alternatives. Overall habitat suitability was projected to increase slightly under all alternatives except for Alternative 2. Given the uncertainty associated with the modeling results, few differences between alternatives can be distinguished with a high degree of confidence.

Future development, and possible resultant effects (for example, increased fire ignitions), associated with the projected human population growth in the Sierra Nevada and continued or increased timber harvest on private lands are assumed to lead to a greater potential for creating future permanent or temporary gaps in the distribution of habitat across the Sierra Nevada across all alternatives.

Establishment of PACs for all breeding territories is a significant improvement across all of the alternatives relative to the current management reflected in Alternative 1. Efforts to re- establish fire as an important ecological process in Sierra Nevada forests and to lower the potential risk of habitat loss to wildfire, while at the same time acknowledging uncertainty regarding the effects of treatments on habitat quality are also key considerations for assessing how the alternatives may affect northern goshawk environments over the next 50 years. Alternatives 3, 5, 6, 8, and modified-8 were rated the as having the greatest potential to improve environments for northern goshawks over the next 50 years. Establishment of PACs for all territories, and old forest emphasis areas and riparian buffers would likely ensure the broad distribution of some landscapes with suitable foraging habitat across these alternatives. California spotted owl and fisher management would likely ensure that mid- and late-seral forests would be broadly distributed on westside Sierra Nevada forests and eastside forests where owls occur under modified-8. Greater uncertainty is associated with modified-8 regarding management of foraging habitat outside of PACs in the General Forest on the eastside. Alternatives 6 and 8 would establish 2500 acre home range management areas around territories in eastside forests. This measure would provide greater spatial certainty that adequate amounts of foraging habitat would be distributed in core areas around breeding territories.

FEIS Volume 3, Chapter 3, part 4.4, page 137 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Alternative 2 has many features that would contribute to improved environments for northern goshawks over 50 years including substantial acreage in integrated biodiversity reserves and post-fledgling and home range standards and guidelines around each breeding territory. However, concern was judged to exist with this alternative because of low levels of fuels treatments, reliance on fire suppression, and potentially increased risk of high-intensity, stand- replacing wildfires.

Alternatives 4 and 7 incur greater uncertainty because of the projected high rates of mechanical treatments, lack of uncertainty regarding spatial allocations of treatments, and greater proportions of the landscape allocated to General Forest. Alternative 7 does have the beneficial feature of home range scale management for territories in eastside forests. Alternative 4 also allows PACs to be deleted from the network if unoccupied for only 2 years.

Alternative 1 (current management direction) likely has the lowest likelihood of maintaining suitable goshawk environments across the Sierra Nevada because of the lack of explicit management guidance for northern goshawks and the risk of losing breeding territories to vegetation treatments.

Table 4.4.2.2g. Projected percent changes in CWHR habitat suitability for select northern goshawk prey species from the current time to year 50 across the FEIS alternatives. SPP ID Species Name Alt 1 Alt 2 Alt 3 Alt 4 Alt 5 Alt 6 Alt 7 Alt 8 Mod 8 Mlv Avg. B308 PILEATED WOODPECKER 19.8 8.0 26.0 27.9 12.7 25.0 20.6 12.5 20.5 3.1 17.6 M079 DOUGLAS' SQUIRREL 13.0 5.7 19.3 20.6 8.9 18.4 15.4 9.2 13.2 1.8 12.6 B134 BLUE GROUSE 14.6 4.2 12.5 15.5 10.7 11.1 10.0 7.7 14.6 4.6 10.5 B386 HERMIT THRUSH 3.2 -3.2 6.9 8.8 -0.7 6.1 4.7 -0.3 2.9 -5.8 2.3 B306 BLACK-BACKED WOODPECKER 2.7 0.1 2.5 2.3 1.5 2.3 1.6 0.6 1.4 -0.9 1.4 B350 CLARK’S NUTCRACKER -1.2 -1.9 2.8 4.9 -0.6 2.9 2.8 0.8 0.7 -4.1 0.7 B346 STELLER’S JAY 1.5 -4.1 3.5 4.8 -1.8 2.8 1.8 -1.8 1.9 -5.7 0.3 M072 CALIFORNIA GROUND SQUIRREL 6.6 0.7 -4.4 -3.8 1.1 -4.5 -3.1 -0.7 6.5 5.1 0.3 B307 NORTHERN FLICKER 1.3 -6.0 1.4 4.9 -3.3 1.2 1.5 -2.6 3.8 -7.5 -0.5 B251 BAND-TAILED PIGEON 0.6 -8.8 3.6 6.8 -4.8 2.7 0.8 -5.0 2.0 -12.0 -1.4

B141 MOUNTAIN QUAIL -0.5 -5.8 -1.9 -0.8 -3.8 -2.5 -2.5 -4.1 0.4 -5.5 -2.7 M057 ALLEN’S/SHADOW CHIPMUNK 0.6 -8.9 -3.9 4.2 -6.2 -3.4 -0.5 -4.9 3.5 -10.3 -3.0 B471 WESTERN TANAGER -2.9 -9.9 0.3 2.9 -6.7 -0.2 -1.2 -6.3 -2.0 -12.5 -3.8 B299 RED-BREASTED SAPSUCKER -1.8 -9.0 -3.6 0.7 -6.6 -3.5 -2.3 -5.9 0.3 -9.8 -4.1 B304 HAIRY WOODPECKER -2.8 -9.5 -2.9 1.2 -7.0 -3.0 -2.1 -6.4 -1.1 -11.2 -4.5 GOLDEN-MANTLED GROUND 0.9 -5.5 M075 SQUIRREL -0.5 -6.8 -7.3 -3.7 -5.4 -7.0 -4.8 -5.8 -4.6 M055 YELLOW PINE CHIPMUNK 0.0 -9.1 -7.1 0.7 -6.8 -6.3 -5.3 -5.7 1.1 -8.5 -4.7 B300 WILLIAMSON’S SAPSUCKER -2.8 -11.0 -3.2 2.6 -8.8 -3.2 -3.3 -7.7 -2.4 -13.6 -5.3 B305 WHITE-HEADED WOODPECKER -3.9 -11.3 -3.7 0.7 -8.6 -3.8 -3.0 -8.0 -2.0 -13.2 -5.7 M062 LONG-EARED CHIPMUNK 2.3 -13.0 -9.3 2.6 -8.5 -10.1 -5.1 -10.3 4.7 -11.8 -5.9 M077 WESTERN GRAY SQUIRREL -5.7 -12.1 -2.9 2.0 -9.7 -3.3 -2.1 -8.9 -3.0 -15.1 -6.1 M060 MERRIAM’S CHIPMUNK -6.7 -6.7 -6.3 -9.4 -7.1 -7.0 -5.4 -6.3 -2.8 -7.1 -6.5 B140 CALIFORNIA QUAIL -3.4 -10.2 -5.7 -6.0 -8.4 -6.3 -6.6 -8.1 -1.4 -9.3 -6.6 B255 MOURNING DOVE -4.7 -12.0 -7.6 -3.2 -10.0 -7.0 -4.9 -9.1 -0.9 -13.0 -7.2 B475 BLACK-HEADED GROSBEAK -6.2 -13.7 -5.4 -1.3 -11.4 -5.3 -4.2 -9.3 -3.0 -15.6 -7.5 B389 AMERICAN ROBIN -4.8 -11.5 -9.7 -6.2 -9.9 -9.7 -7.6 -9.7 -2.8 -10.8 -8.3 M063 LODGEPOLE CHIPMUNK -5.5 -15.2 -10.4 -3.3 -13.0 -10.6 -7.3 -13.3 -3.6 -16.0 -9.8 B294 LEWIS’ WOODPECKER -7.8 -16.6 -11.3 -7.2 -14.0 -10.9 -11.6 -11.1 -5.8 -15.9 -11.2 M046 NUTTALL’S COTTONTAIL -26.4 -36.1 -26.8 -42.1 -40.5 -26.4 -30.8 -26.4 -26.6 -26.4 -30.9

FEIS Volume 3, Chapter 3, part 4.4, page 138 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Figure 4.4.2.2a. Region-wide projected acres of pooled CWHR classes 4M, 4D, 5M, 5D and 6.

Region-wide, all tree types 4D,4M,5D,5M,6

6000000

5000000 1 2 4000000 3 4 3000000 5 6 2000000 7 Acres of conifer types 8

1000000 MLV Pref

0 0 2 4 6 8 10 12 14 16 Decade

Figure 4.4.2.2b. Region-wide projected acres of pooled CWHR class 5D.

1

5D 2

2,000,000 3 1,800,000 4 1,600,000 1,400,000 5 1,200,000 6 1,000,000 7 Acres 800,000 600,000 8 400,000 MLV 200,000 Pref 0 0 5 10 15 Decade

FEIS Volume 3, Chapter 3, part 4.4, page 139 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Figure 4.4.2.2c. Region-wide projected acres of pooled CWHR class 5M.

1 5M 2 2,000,000 1,800,000 3 1,600,000 1,400,000 4 1,200,000 5 1,000,000

Acres 800,000 6 600,000 7 400,000 200,000 8 0 0 5 10 15 MLV

Decade Pref

Figure 4.4.2.2d. Region-wide projected acres of pooled CWHR class 4D.

1 4D 2 2,000,000 1,800,000 3 1,600,000 1,400,000 4 1,200,000 5 1,000,000

Acres 800,000 6 600,000 400,000 7 200,000 8 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 MLV Decade Pref

FEIS Volume 3, Chapter 3, part 4.4, page 140 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Figure 4.4.2.2e. Region-wide projected acres of pooled CWHR class 4M.

1 4M 2 2,000,000 1,800,000 3 1,600,000 1,400,000 4 1,200,000 5 1,000,000

Acres 800,000 6 600,000 400,000 7 200,000 8 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 MLV

Decade Pref

Figure 4.4.2.2f. Region-wide projected acres of pooled CWHR class 6.

1

6 2 2,000,000 1,800,000 3 1,600,000 4 1,400,000 1,200,000 5 1,000,000

Acres 800,000 6 600,000 7 400,000 200,000 8 0 MLV 0 1 2 3 4 5 6 7 8 9 101112131415 Decade Pref

FEIS Volume 3, Chapter 3, part 4.4, page 141 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 Figure 4.4.2.2g. Region-wide projected change in CWHR habitat utility units for northern goshawks across the FEIS alternatives.

4500000 alt_1 4000000 3500000 alt_2 3000000 alt_3 2500000 alt_4 2000000 alt_5 1500000 alt_6 Habitat Units 1000000 500000 alt_7 0 alt_8 1 3 5 7 9 11 13 15 Mod-8 Decade mlv

FEIS Volume 3, Chapter 3, part 4.4, page 142 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 4.4.2.3. WILLOW FLYCATCHER (Empidonax traillii) Affected Environment Species Background The willow flycatcher (Empidonax traillii) is identified in the Sierra Nevada Forest Plan Amendment Notice of Intent (NOI), November 16, 1998 as one of seven aquatic, riparian, and meadow (ARM)- dependent vertebrate species at risk in the Sierra Nevada bioregion. The willow flycatcher is recognized by the Forest Service Pacific Southwest Region as the highest-priority landbird species in the Sierra Nevada bioregion because it is considered to have “... the highest probability of being extirpated from the bioregion in the near future” (USDA 1996). Three willow flycatcher subspecies breed in California, E. t. adastus, E. t. brewsteri, and E. t. extimus (Phillips 1948, Unitt 1987), and all are included on the Regional Forester's Sensitive Species list in the Pacific Southwest Region (FSM 2670). The willow flycatcher has formal listing status in the State of California as an endangered species (CDFG 1995) and the southwestern willow flycatcher (E. t. extimus) is listed as a Federally Endangered species (Federal Register 1995).

The willow flycatcher is a nearctic-neotropical migrant species that breeds across North America and winters from Mexico to northern South America. The southwestern willow flycatcher (extimus subspecies) winters from Chiapas, Mexico to Costa Rica, possibly western Panama; E. t. adastus and E. t. brewsteri have a broader winter range from the Mexican state of Colima to northwestern Venezuela (Unitt 1999). Currently, approximately half of the willow flycatcher breeding population in California occurs in the Sierra Nevada bioregion (Zeiner et al. 1990, Kus et al. 2000). Of the three willow flycatcher subspecies that occur in California, E. t. adastus and E. t. brewsteri are the only two whose breeding ranges substantially overlap with the planning area (Phillips 1948, Unitt 1987). The only known population of the southwestern willow flycatcher on national forest lands in the Sierra Nevada bioregion occurs along the South Fork of the Kern River at approximately 2,200 feet elevation in Sequoia National Forest. Based on preliminary genetic analyses, the southwestern willow flycatcher recovery team also recognizes willow flycatchers on the Owens River near Bishop as E. t. extimus (S. Leon pers. comm., Paxton 2000). Furthermore, song spectrograms from willow flycatchers along the Owens River, northwest of Bishop, appear to have the vocal signature of the extimus subspecies (J. Sedgwick pers. comm.). Note however that although these Owens River willow flycatchers occur within the planning area, they do not occur on national forest lands.

The Forest Service has engaged in informal consultation with the US Fish and Wildlife Service since E. t. extimus was Federally listed in 1995 (for USFWS response to Forest Service biological assessment, see 1-1-97-I-1498, 1-1-99-I-1364). Because there are no anticipated effects of the Sierra Nevada Forest Plan Amendment Project FEIS alternatives that would significantly affect the downstream habitat and persistence of these two populations, this subspecies is not included in the willow flycatcher analysis of this FEIS and will instead be addressed under local project assessments and the Land and Resource Management Plans of the local national forests, consistent with the Endangered Species Act Section 7 consultation requirements. If, however, results of subsequent analyses or surveys reveal other E. t. extimus populations on national forest lands in the planning area, or adverse effects from national forest lands on adjacent extimus populations, project level consultation will occur with the U.S. Fish and Wildlife Service..

The two willow flycatcher subspecies that are the focus of this analysis, E. t. adastus and E. t. brewsteri, breed in shrubby vegetation in meadow and riparian communities. Serena (1982), Harris et al. (1987, 1988), and Fowler et al. (1991) observed that willow flycatchers were consistently

FEIS Volume 3, Chapter 3, part 4.4, page 143 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 associated with meadows where high water tables resulted in standing water and riparian shrubs (specifically willow) were abundant. Various willow flycatcher researchers in the Sierra report that the shrub layer is typically 6.5 to 13 feet in height, with the lower 6.5 feet comprised of dense woody vegetation. The live foliage density is moderate to high and uniform from the ground to the shrub canopy (Valentine 1987, Sanders and Flett 1989, Bombay 1999). The mean shrub cover within willow flycatcher territories has been documented at 21,529 square feet, but in some cases they use as little as 1,076 square feet (Harris et al. 1987, 1988, Sanders and Flett 1989, Fowler et al. 1991, Bombay 1999). There is usually at least some surface water or saturated soil within defended territories during the early part of the breeding season (Valentine 1987, Sanders and Flett 1989, Bombay 1999). Recent habitat selection modeling confirmed these observations that willow flycatchers are significantly more likely to be detected at sites where the herbaceous community is consistent with high water tables and late seral conditions, and riparian deciduous shrubs are abundant (Bombay 1999). Sites are also significantly more likely to support multiple willow flycatchers, and result in successful breeding efforts, as riparian shrub cover in meadows and willow flycatcher territories increases (Bombay 1999). Other features of sites occupied by willow flycatchers, such as dominant plant species, sizes and shapes of vegetation patches, as well as amount and source of water (for example streams, oxbows, lake margins, springs, seeps) vary widely among sites.

Willow flycatchers have nested in meadows less than one acre (KRCD 1985) and as large as several hundred acres (Serena 1982, Harris et al. 1987, 1988, Bombay 1999) although Serena (1982) and Harris et al. (1987, 1988) report that most willow flycatchers (more than 80 percent) occur in meadows larger than 19.8 acres in size. The most recent compilation of willow flycatcher data in the Sierra Nevada indicates that known sites range in size from 1 to 716 acres, with a mean size of approximately 80 acres. Sierra Nevada surveys for willow flycatchers in 1982 and 1986 as well as subsequent surveys and historic records in the bioregion indicate that willow flycatchers occur at elevations from 1,200 to 9,500 feet, although most of the known willow flycatcher sites (88 percent) occur between 4,000 and 8,000 feet (Serena 1982, Harris et al. 1988, Stafford and Valentine 1985, Bombay et al. 1998, S. Armentrout pers. comm.).

In the Sierra Nevada bioregion, the willow flycatcher breeding season occurs from late May or early June (territory establishment) to the middle of September (fledgling independence) (Stafford and Valentine 1985, Flett & Sanders 1987, Bombay et al. 1999). A recent compilation of multiple years of Sierra Nevada willow flycatcher nesting data (165 nests) reveals that willow flycatchers fledge young between approximately July 15 and August 31 and fledglings remain in territories for 2 to 3 weeks post-fledging (Stafford and Valentine 1988, Sanders and Flett 1989, M. Morrison and H. Bombay, unpubl. data).

Natal dispersal distances and adult between-site, between-year movements for E. t. adastus and E. t. brewsteri in the Sierra Nevada are largely unreported, however some data have recently been obtained. Eleven between-site, between-year movements of 1/3 to 9 miles have been documented for willow flycatchers in the Sierra Nevada since 1984 (mean = 2.63 miles, SD = 2.42 miles) (Stafford and Valentine 1985, Sanders and Flett 1989, Morrison et al. 2000, J. Steele unpubl. data). These movements represent both adult and natal dispersal, and all birds were relocated in the same drainage system in which they were banded. Morrison et al. (2000) report that seven willow flycatchers banded as nestlings between 1997 and 1999 in the Red Lake and Little Truckee River areas, dispersed and were relocated as adults in meadows other than where they fledged. The mean natal dispersal distance for these seven birds was 1.76 miles (SD = 1.07) (Morrison et al. 2000). Also in the Little Truckee River area, Sanders and Flett (1989) report that in 1987 an adult male re-settled

FEIS Volume 3, Chapter 3, part 4.4, page 144 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 approximately 0.6 miles from his 1986 breeding territory. Among adult and juvenile willow flycatchers color-banded at Perazzo meadow in 1994, two were detected 3.5 miles downstream of Perazzo in 1997 (J. Steele unpubl. data, H. Bombay pers. comm.). In the southern Sierra Nevada, Stafford and Valentine (1985) report that one female willow flycatcher (banded as an adult) moved 9 miles from her 1983 territory to a new territory, where she nested successfully in 1984. Research on southwestern willow flycatchers in Arizona indicate that 13-17 percent of adults move to new breeding sites each year (Busch et al. 2000, Paxton 2000, Paxton and Sogge 2000). Between 1997 and 1998, 19 between-site movements with a median distance of 9 miles were documented for southwestern willow flycatchers. Four of these records represent between-drainage movements (Netter et al. 1998). In the 2000 breeding season, between-site between-year movements of up to 186 miles were documented for the southwestern willow flycatcher (Paxton and Sogge 2000). Paxton (2000) reports that “areas with multiple breeding sites that are geographically close have the highest degree of between-site movement, with longer distance dispersal fairly rare. Thus, the frequency of movement is negatively correlated with distance moved”.

The return rates of migratory bird species to the breeding grounds may be used to indirectly indicate the ability of the species to survive migration and the wintering season elsewhere (such as the tropics). In migratory species that exhibit philopatry on the breeding grounds, return rates may be particularly suitable for estimating survivorship because surviving birds are more likely to return to the same area and increase the probability of being detected in subsequent years. Data on juvenile and adult return rates and philopatry for the willow flycatcher population in the Sierra Nevada bioregion are preliminary. An ongoing willow flycatcher demography study in the central Sierra Nevada reports a mean juvenile return rate of 9.6 percent annually (SE=4.2), based on 165 nestlings banded between 1997 and 1999 (Morrison et al. 2000). In Oregon, the juvenile return rate was 13.2 percent of 214 nestlings banded (Sedgwick and Klus 1997), while in Michigan it was 1.4 percent of 147 nestlings banded (Walkinshaw 1966). Juvenile return rates for southwestern willow flycatchers in the southern Sierra Nevada and Arizona are recorded as 34 percent of 38 banded nestlings and 8 percent of 12 banded nestlings, respectively (Stoleson et al. 2000). These juvenile return rates are higher than the 2 to 5 percent rates reported for most other migratory passerines (Shields 1984, Blancher and Robertson 1985, Payne and Payne 1990, Sherry and Holmes 1992, Roth and Johnson 1993, Lemon et al. 1996, Netter et al. 1998).

Willow flycatcher return rates are generally higher for adults. In the central Sierra Nevada, Morrison et al. (2000) reported that for 1999 and 2000 adult survival was 80 and 73 percent respectively (4 of 5, and 8 of 11 banded adults returning, respectively). At two of the same study sites, Sanders and Flett (1989) found that four out of 14 banded adults returned (29 percent). In the southern Sierra, Stafford and Valentine (1985) had four of 12 adults return (33 percent). Adult survival for southwestern willow flycatchers in the southern Sierra Nevada and Arizona is roughly 52 percent for males and 34 percent for females based on 79 returning males and 255 returning females (Stoleson et al. 2000). In Michigan, 30 percent of 53 banded adults returned the following year (Walkinshaw 1966). Similarly, in Oregon 45 percent of birds banded as adults returned over eight years of study (Sedgwick and Klus 1997).

Although the previous discussion of survival indicates how many willow flycatchers survived and returned to any of a number of breeding locations, it does not indicate how many birds return to the same breeding site in sequential years. Philopatry, on the other hand, looks at how faithful willow flycatchers are to specific meadows or even territory locations. Netter et al. (1998) report that 38 percent of adult southwestern willow flycatchers return to the same sites (males 46 percent, females

FEIS Volume 3, Chapter 3, part 4.4, page 145 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

30 percent) in Arizona. Of the birds that returned to the same sites, 55 percent settled within 164 feet of the previous year's territory. Between 1996 and 1998, 32 percent of adults displayed breeding site philopatry in Arizona (Busch et al. 2000). In the southern Sierra Nevada, 61.6 percent of adult male southwestern willow flycatchers (n = 138) and 51.8 percent of adult females (n = 137) returned to the study area (M. Whitfield pers. comm.). Over half of the breeding adults captured in a 1988-1997 study in Oregon returned to the same area and bred again in subsequent years (females: 186/347 [53.6 percent]; males: 138/264 [52.3 percent]) (Sedgwick and Iko 1999). At this study site the median distance for returning males (n = 362) and females (n = 349) from the previous years’ nest site was 82 and 85 feet, respectively (males: mean = 633 ft + 95 SE, range = 0-15,295 ft; females: mean = 764 ft + 121 SE, range = 0-19,442 ft) (J. Sedgwick unpubl. data). Others also report high territory fidelity in willow flycatchers, particularly for males (Stafford and Valentine 1985, Sanders and Flett 1989, J. Steele pers. comm.).

Historic and Current Conditions Historically, willow flycatchers nested throughout California wherever thickets of riparian deciduous shrubs, primarily willow (Salix spp.), occurred (Grinnell and Miller 1944). References in the literature describe willow flycatchers in the Central Valley, the Great Basin, and at elevations below 6,000 feet in the Sierra Nevada (Adams 1907, Bennett 1934, Bent 1942, Davis 1938, Dawson 1916, Grinnell and Miller 1944, Klebenow and Oakleaf 1984, Linsdale 1932, Miller 1941, Ridgway 1907, Salt 1953, Wheelock 1904, Western Foundation of Vertebrate Zoology unpubl. nest records). In addition, there are numerous breeding season records of willow flycatchers, as well as willow flycatcher nest records, occurring at or above 6,000 feet in the Sierra (Bendire 1895, Ray 1903, Ingersoll 1913, Ray 1913, Grinnell 1934, Miller 1941, Bent 1942, Orr and Moffitt 1971, Gaines 1977, 1992, Western Foundation of Vertebrate Zoology unpubl. nest records). In the last four decades, however, willow flycatcher breeding populations have been extirpated from most lower elevation riparian areas in California and it appears that the species may no longer breed at elevations below 3,000 feet in the Sierra Nevada, in the Central Valley, and in the valleys of the central coast (Gaines 1974, Klebenow and Oakleaf 1984, Zeiner et al. 1990, Lynn et al. 1998). Historic records combined with recent survey efforts indicate a long-term decline of willow flycatchers at elevations above 3,000 feet in the Sierra Nevada as well. Examples of higher elevation locations where willow flycatchers have all but disappeared include the historic willow flycatcher population centers associated with the Truckee Marsh and Upper Truckee River system at South Lake Tahoe (Ray 1913, Orr and Moffitt 1971, Erwin et al. 1995, Bombay 1999), the Dinkey Creek area of the Sierra National Forest (KRCD 1985, Valentine et al. 1988, KRCD 1992), and the June Lake area of the Inyo National Forest (Serena 1982, Harris et al. 1988, Gaines 1992, unpubl. nest and survey records).

A review of the literature and nest collection records indicate that willow flycatchers were common in the Sierra Nevada until 1910, and at least locally abundant through 1940. However, noticeable declines occurred after 1950 (Serena 1982, Klebenow and Oakleaf 1984, Harris et al. 1987, 1988, Gaines 1992, Erwin et al. 1995, Bombay 1999, unpubl. Forest Service survey results). There are various likely causes of the willow flycatcher population decline in the Sierra. Livestock grazing, initially by sheep and later by cattle, occurred with heaviest intensity in the Sierra Nevada from the late 1800’s through 1930’s. Grazing altered the hydrology and vegetation of meadows across much of the Sierra Nevada during this period (Ratliff 1985, McKelvey and Johnston 1992, Menke et al. 1996, Kondolf et al. 1996, Kinney 1996). Although quantitative data are limited regarding long-term changes to riparian

FEIS Volume 3, Chapter 3, part 4.4, page 146 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 communities for this period, a recent analysis of stratigraphic pollen records in southern Sierra montane meadows attributes dramatic decreases in willow abundance primarily to the introduction of sheep and cattle during the latter part of the 19th century (Dull 1999).

In the latter half of the 19th century, mining impacted almost every low and mid-elevation stream system on the west slope, dramatically altering hydrology and vegetation (Clark 1970 in Kattelmann and Embury 1996). In addition, intensive upland logging and road building efforts during the 20th century affected hydrologic systems and caused many stream channels to become incised, changing associated meadow communities as a result (Ratliff 1985). Water developments beginning with redirected flows for gold mining and subsequently for hydropower, irrigation, flood control, and municipal supply resulted in hundreds of dams throughout riparian areas of the Sierra Nevada during the first half of the 20th century (Kattelmann et al. 1996, Moyle and Randall 1996). Despite these alterations in aquatic systems, the dramatic willow flycatcher population decline after 1950 is probably the result of additional impacts such as wintering ground deforestation, increased human developments in the Sierra, increased use of chemicals for insect control, or other as yet undocumented factors (Wagoner 1886, Sudworth 1900, Leiberg 1902, Hughes 1934, Unsicker et al. 1984, Gard and Hooper 1995, Graber 1996, Kattelmann and Embury 1996).

The pesticidal properties of synthetic chlorinated hydrocarbon and organophosphate compounds were discovered in the 1940’s and, up until the 1970’s, these chemicals were used increasingly in pest control and public health programs in North America (Gard and Hooper 1995). Although these types of pesticides have been virtually eliminated in North America, they are believed to still be used extensively in Latin America. Though there is no direct evidence that the use of these or other pesticides have impacted or are currently impacting willow flycatchers, it is possible that their past use on breeding, wintering, or migration grounds may have caused reproductive failure and direct mortality, and led in part to willow flycatcher declines in the Sierra Nevada (Gard and Hooper 1995).

Since the 1920s, grazing has remained the primary management activity in meadows although grazing intensity has continued to decline. The expansion of the brown-headed cowbird (Molothrus ater), a brood parasite of the willow flycatcher, into the Sierra Nevada in the 1930's is at least partially attributed to the presence of livestock, forest fragmentation, and rural community expansion occurring during that time (Rothstein et al. 1980, Gaines 1992, Rothstein 1994). Also since the 1920’s, road construction and recreation have increased, bringing with them impacts to stream hydrology and meadow vegetation. Under decreased direct pressure from grazing, some system function in some meadows have improved, while others have held static or continued to decline in condition, especially when road or recreation impacts occur in concert with grazing (Kattelmann and Embury 1996, Menke et al. 1996). Currently, the amount of available willow flycatcher habitat is likely only a fraction of what was available compared to pre-1850 conditions. Many meadows have some standing water and some riparian woody vegetation, however frequently not in amounts or configurations necessary to support willow flycatchers (Serena 1982, Harris et al. 1987, Bombay 1999). Additionally, many historic willow flycatcher habitats should be considered irrevocably lost due to inundation, development, or alteration to such an extent that they could no longer support willow flycatchers (Western Federal Regional Council Interagency Task Force 1979 in Unsicker et al. 1984, Kattelmann and Embury 1996).

FEIS Volume 3, Chapter 3, part 4.4, page 147 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Large source populations for willow flycatchers in places like Yosemite Valley, Truckee Meadows, and South Lake Tahoe were impacted by urbanization and habitat was lost due to draining, channelization and filling (Logan 1948, Klebenow and Oakleaf 1984, Unsicker et al. 1984, McPhee 1993, Kattelmann et al. 1996). It is estimated that between 1900 and 1979, 35 percent of riparian streamside zones, 50 percent of meadows, and 75 percent of marshes in the Lake Tahoe Basin were lost to development, and that between 1969 and 1979 alone, 25 percent of the marshes were developed (Western Federal Regional Council Interagency Task Force 1979 in Unsicker et al. 1984). The declines in, or outright loss of, large source populations (Pulliam 1988) such as these may have led to the willow flycatcher population decline across the Sierra Nevada.

Current estimates of the willow flycatcher population in the planning area range between 300 to 400 individuals (Serena 1982, Harris et al. 1987, 1988, Ritter and Roche 1999, H. Bombay pers. comm.) and the effective population size (number of breeding adults) is likely to be smaller than this. Records compiled from national forests, researchers, scientific literature, and museum collections and dating from 1910 to 2000 document 135 known locations within the planning area boundaries. In most sites, only 1 willow flycatcher territory is recorded (n = 84 or 62 percent) but other sites have 3, 5, or as many as 32 willow flycatcher territories.

Both willow flycatcher migrants and breeding residents occur in the Sierra Nevada bioregion. Migrants may occur in a wide variety of vegetation communities and no precedent for managing migratory stopover habitat exists, even for the Federally endangered subspecies (S. Leon pers. comm.). Consequently, to differentiate between observations of migrating and summer resident willow flycatchers and facilitate identification of breeding sites in the Sierra Nevada bioregion, the FEIS defines “known willow flycatcher sites” as meadows or riparian areas with willow flycatcher observations that meet one of the following criteria:

1. willow flycatcher observed between June 15 and August 1; or 2. willow flycatcher observed between June 1 - June 14, or August 2 - August 15, unless the willow flycatcher was:

• absent during surveys conducted between June 15 and July 15 in the same year, • absent during June 15 to July 15 surveys in multiple subsequent years, or • detected at a site that is clearly outside of known habitat requirements.

For inclusion as a known willow flycatcher site, willow flycatcher(s) must be identified by the “Fitz-bew” song or in-hand examination. Museum skins that are identified as willow flycatchers may also be used if the collection date falls within the range of dates listed above. Nests and egg sets in museum collections infer site occupancy, regardless of collection date. All sites where willow flycatchers were identified using these criteria are included in the dataset regardless of the year of observation or collection, unless the site is known to have undergone an extreme site conversion rendering it incapable of supporting willow flycatchers currently or in the future (for example, wetland conversions or inundation by reservoir). Willow flycatcher records without a specific date are assumed to be breeding season occupants, until more information is available and indicates otherwise. For example, there are 5 willow flycatcher sites that occur on Forest Service lands where the month and day of detection were not specified in the record and the site was only visited in one year; these records will be pursued for additional information or eliminated from the willow flycatcher database if no further information is found.

FEIS Volume 3, Chapter 3, part 4.4, page 148 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

The FEIS defines willow flycatcher “emphasis habitat” as wet meadows larger than 15 acres with at least some riparian deciduous shrubs (woody vegetation). Recall that willow flycatchers have nested in meadows less than 1 acre (KRCD 1985) but most willow flycatchers occur in meadows larger than 19.8 acres (Serena 1982, Harris et al. 1987, 1988). Defining emphasis habitat as meadows greater than 15 acres represents a management decision and compromise. “Potential” habitat includes emphasis habitat, in addition to small wet meadows (less than or equal to 15 acres) with shrubby vegetation.

Figure 4.4.2.3a. Number and Distribution of known willow flycatcher sites in the Sierra Nevada Forest Plan Amendment Project Planning Area by Ownership

20 1 18 11 16 4 55 14 8 S 12 P 4 N 10 1 F 8 1 C 8 1 B 6 12 12 22 10 1 9 4 6 8 7 4 5 55 2 4 111 0 Inyo SEKI YOSE Sierra Tahoe Modoc Lassen LTBMU Plumas Toiyabe Sequoia Eldorado Stanislaus

Known willow flycatcher sites are distributed throughout the Sierra Nevada bioregion and occur on all national forests in the planning area. They are most concentrated on Lassen, Plumas, Tahoe, and Inyo National Forests. Based upon available willow flycatcher records and survey data, the majority of known willow flycatcher sites occur on national forest lands (F, n = 82 or 61 percent); 29 of the 135 sites occur on private property (P). The other known sites occur on National Park Service (N, n = 10), Bureau of Land Management (B, n = 2), State (S, n = 8), or city or county (C, n = 4) lands. As a result of Pacific Southwest Region requests to compile all known data across the bioregion, it appears that survey efforts for willow flycatcher locations in the planning area have been uneven across ownerships. It is possible that the preponderance of known willow flycatcher sites on Forest Service lands may be an artifact of the sampling efforts to date, although this is difficult to determine without being able to compare positive and negative survey results for different ownerships in the bioregion. A compilation of willow flycatcher records on Forest Service lands from 1933 to 2000 indicates a mean number of territory holders of 137, with a maximum estimate of 159 territorial individuals. Presently, there is an estimated 120 to 150 willow flycatcher individuals on national forest lands in the bioregion (Ritter and Roche 1999).

FEIS Volume 3, Chapter 3, part 4.4, page 149 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Known willow flycatcher sites are based upon records from 1910 to 2000; 69 percent of known sites (n = 93) have been surveyed since 1997 and 95 percent of known sites (n = 128) have been surveyed since 1982. Despite the lack of consistent regional surveys (for example, 47 percent of known willow flycatcher sites have been surveyed 2 years or less), the applicability of known willow flycatcher site data to contemporary willow flycatcher site occupancy is as follows: over 68 percent of known willow flycatcher sites (n = 92) are known to have been occupied since 1990 and over 90 percent of sites (n = 122) since 1982. Of the 82 sites on Forest Service lands, 73 percent of known willow flycatcher sites (n = 60) are known to have been occupied since 1990 and over 95 percent of sites (n = 78) since 1982.

Although it has been established that willow flycatchers have declined in the Sierra Nevada since the 1940’s (Serena 1982, Harris et al. 1987, 1988, Erwin et al. 1995), more recent survey results are generally unpublished. Of the 122 sites where willow flycatchers have been documented since 1982, survey records indicate that willow flycatchers were not detected at 46 (38%) of these sites on the most recent survey (see also Bombay 1999). A total of 35 of these 46 sites were on national forest land. Because many known willow flycatcher sites have received only one or two surveys it is difficult to determine exactly how many sites may have become vacant since 1982. Nonetheless, there are 10 sites which may no longer be occupied by willow flycatchers, when examining those sites: 1) with at least 3 years of available survey data, 2) which are known to have supported multiple territories, and 3) which appear to have been vacant on at least the last 2 surveys. These sites are distributed across a large portion of the Sierra Nevada, including Sequoia National Park, Sierra National Forest, Sequoia National Forest, Stanislaus National Forest, Humboldt-Toiyabe National Forest, and Lassen National Forest. These sites are located on national forest lands, with the exception of a Sequoia National Park site, and one privately owned site within the administrative boundaries of Sierra National Forest. While warranting further attention, this preliminary evidence that anywhere from 10 to 46 known willow flycatcher sites have become unoccupied after 1982 should not hasten assumptions of localized extirpations as: 1) some of this information is based upon only one survey year, 2) willow flycatchers may have been present but not detected in the most recent survey year, or 3) willow flycatchers may have re-located to another site.

Mean Mayfield nest success during 1997 through 2000 for willow flycatchers in the central Sierra Nevada was 40 percent (range 25 - 49 percent, SE = 11, n = 208 nests) (Morrison et al. 2000). This mean value falls within the lower half of the range of values observed in other studies of willow flycatcher reproduction (18 - 81.4 percent)(Table 4.4.2.3a).

FEIS Volume 3, Chapter 3, part 4.4, page 150 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Table 4.4.2.3a. Nest success for willow flycatchers across various studies and years. NESTING SUCCESS SOURCE LOCATION 18 b Sogge et al. 1997 Arizona 23 b Whitfield et al. 1999 (pre-brown-headed cowbird trapping) Kern River, CA 24.4 b Harris 1991 Kern River, CA 39 b Whitfield et al. 1999 (post-brown-headed cowbird Kern River, CA trapping) 40 b Morrison et al. 2000 Sierra Nevada, CA 40.7 b Sedgwick and Knopf 1988 Colorado 42 b Sanders and Flett 1989 Sierra Nevada, CA 42.7 a Sedgwick and Iko 1999 Oregon 44.6b King 1955 Washington 65.6 b Walkinshaw 1966 Michigan 81.4 a Berger 1967 Michigan a Success estimated using the egg-to-fledgling ratio b Success estimated using successful nests-to-total number of nests ratio

The current nest success rate for willow flycatchers studied in the central Sierra Nevada of 40 percent is lower than the mean values for cup-type nesting passerines of 49 percent (range = 38-77 percent) reported by Nice (1957) and 55 percent (SE = 5.2 percent) reported by Martin (1992). However the value is higher than the 33 percent (SE = 4.7 percent) reported by Martin (1992) for shrub nesting birds, alone. In the Sierra Nevada, willow flycatcher nest failures are primarily the result of nest depredation (Stafford and Valentine 1985, Bombay 1999, Morrison et al. 2000). From 1997 through 2000 Morrison et al. (2000) estimate a minimum of 72 percent of failures were the result of depredation. Depredation of willow flycatcher nests is also the largest source of nest failure in other western willow flycatcher populations (Sedgwick and Iko 1999, Whitfield et al. 1999). Sanders and Flett (1989) and Morrison et al. (1999, 2000) also report some nest failures as a result of inclement weather. Although a relatively rare event at most willow flycatcher study sites in the Sierra Nevada, brown-headed cowbird parasitism of willow flycatcher nests has been documented at a number of different Sierra Nevada meadows (Gaines 1977, Sanders and Flett 1989, Bombay et al. 1998, Bombay 1999, Morrison et al. 1999). Mean annual brood parasitism rate for 1997 through 2000 was 6.75 percent of willow flycatcher nests (SE = 4.11) (Morrison et al. 2000).

Evidence from a number of studies indicates that yearling recruitment of willow flycatchers as well as other long-distance migrants is often determined primarily by the previous summer’s fledging success (Ricklefs 1992, Sherry and Holmes 1992, 1995, Whitfield and Strong 1995, Whitfield and Enos 1996, Johnson and Geupel 1996). Additionally, Robinson et al. (1993, 1995a, 1995b) postulated that an annual output of 2.23 young per female was the minimum value needed to keep most populations of small passerines stable. In the central Sierra Nevada, the mean annual fecundity rate for willow flycatchers during the period of 1997 through 2000 was 1.67 fledglings per female (n = 246 fledglings) (Bombay 1999, Morrison et al. 2000). At some of the same study sites, Sanders and Flett (1989) reported the number of fledglings per female to be 1.00 for 1986 and 1987 combined (n = 20 fledglings). Similarly, Stafford and Valentine reported 1.04 fledglings per female at their sites in the southern Sierra Nevada in 1983 and 1984 combined (n = 9 fledglings).

In addition to predation, brood parasitism, nest abandonment, or inclement weather, another factor that may influence fecundity is infertile or non-viable eggs. Sanders and Flett (1989)

FEIS Volume 3, Chapter 3, part 4.4, page 151 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

reported two non-viable eggs in two different nests; studies by the Kings River Conservation District reported 7 of 19 eggs as non-viable (Valentine et al. 1988). Morrison et al. (2000) found that 20 out of 87 nests (23 percent) with complete hatching success data had from 1 to 3 unhatched eggs. A total of 28 of 276 eggs (10 percent) were unhatched. In the southern Sierra Nevada, 15 percent of southwestern willow flycatcher eggs were unhatched between 1989 and 2000, with significant declines in hatchability in the last four years (Whitfield 2000). By comparison, willow flycatchers in Ohio (Holcomb 1972) and Wisconsin (McCabe 1991) had only 3 to 4 percent of eggs unhatched. Martin (1992) summarizes hatchability data for 32 species of cup nesting migratory passerines; the mean rate of unhatched eggs was 4.9 percent (SE = 1.2). This preliminary information suggests that the percent of unhatched willow flycatcher eggs in the Sierra Nevada may be high relative to other willow flycatcher populations and other cup-nesting migratory passerines in general.

Although it is normal for a small percentage of eggs to go unhatched, elevated levels of unhatched eggs, as well as other reproductive anomalies, are sometimes associated with the toxic effects of pesticides, pollution, or naturally occurring heavy metals. Because willow flycatchers are insectivorous, they are susceptible to the magnifying effects of chemical and heavy metal bioaccumulation when they consume contaminated insect prey (Gard and Hooper 1995). Many of their prey species have aquatic life stages; this places willow flycatchers at even higher risk because heavy metal bioaccumulation is greatest within aquatic systems (Goyer 1991 in Gard and Hooper 1995). Recent research on levels of certain chemicals and heavy metals at southwestern willow flycatcher breeding areas in Arizona was spurred by reports of willow flycatcher nestling deformities. Results of this examination of insects and surrogate passerine species indicated that although most contaminant concentrations were below thresholds of concern, unusually high concentrations of the heavy metal strontium were found in bird eggs at these sites (Mora et al. 2000). There is no information available on the presence or effects of toxic compounds on Sierra Nevada willow flycatchers, however it is important to note that for long-distance migrants exposure could occur on breeding, migration, or wintering grounds (Gard and Hooper 1995).

Given the small population size, apparent recent vacancies of meadows occupied in the 1980’s, and evidence for moderate to low nest success rate and low fecundity rate, it appears that under existing conditions willow flycatchers are likely still experiencing a declining population trend in the Sierra Nevada bioregion. This is further supported by preliminary calculations of Lambda for willow flycatchers in the central Sierra that are less than 1.0 (1999 = 0.876, 2000 = 0.777) (Morrison et al. 2000). These Lambda values translate to annual population declines of 12.4 and 22.3 percent in 1999 and 2000, respectively, indicating a non-sustainable replacement rate.

Risk Factors There are multiple factors affecting the current habitat, distribution, and abundance of willow flycatchers in the planning area. A number of these factors are influenced by national forest management activities. Factors are discussed below in order of concern and priority for managing the willow flycatcher population.

The primary management activity currently occurring within meadows and riparian areas on national forest lands used by willow flycatchers is livestock grazing. Livestock grazing is a risk factor that is within the control of forest service management and can affect willow

FEIS Volume 3, Chapter 3, part 4.4, page 152 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

flycatchers and their habitat in several ways. Specific research on livestock grazing practices in known willow flycatcher sites in the Sierra Nevada is lacking, however the following studies on willow flycatchers and grazing conducted elsewhere provide relevant information:

1. In Colorado, Stanley and Knopf (in review) report that in 2 pastures receiving late-season grazing on a rest-rotation basis (2 months grazed followed by 2 years of no grazing), willow flycatcher densities approximated those on the 2 ungrazed “control” pastures by the 10th year of the study. They found no statistical differences in vegetation change, although with grazed and ungrazed pastures replicated only once, the statistical power to detect vegetation differences was low. These results indicate that willow flycatcher habitats may be restored while still allowing late-season rest-rotation grazing, although the rate at which the species is recovered will be slower than if all cattle are removed. The authors suggest more severe restrictions on grazing may be preferred where conservation priorities emphasize maximum recovery of the species. 2. In eastern Oregon, studies on grazing in relation to willow flycatchers and other riparian- associated passerines looked at relative amounts of woody riparian vegetation (willow) and bird densities (Taylor and Littlefield 1986, Taylor 1986). Although there are confounding effects of historical (willow removal) and current management (dredging, campground) at some of the nine study transects, the treatments can essentially divided into five groups:

a. no grazing for 40 years (one transect); b. grazing reduced from “extensive” to winter grazing for 3 to 5 years before study (two transects); c. no grazing for 10 years (two transects); d. grazing reduced from “extensive” to late summer grazing in years immediately before study (one transect); e. season long “extensive” grazing or otherwise heavily disturbed up to or including the beginning of study (three transects).

All sites combined had “extensive grazing” of 126,000 AUMs up until 10 years prior to study, with the exception of the site with 40 years exclusion. Taylor (1986) reported that shrub volume in the area with 40 years of rest was 10 times that in the 4 most heavily grazed areas (group d and one transect from group b above). Shrub volume was significantly positively correlated with time since an area was last grazed (r = 0.87, 0.97; Ps<0.01, 0.001). After ranking each transect for the frequency of cattle grazing on an annual basis, Taylor (1986) and Taylor and Littlefield (1986) report that shrub volume was significantly negatively correlated with frequency of grazing (r = -0.79, -0.76; P<0.02). These manuscripts do not explicitly define “grazing frequency”; however, for the purpose of this review it is assumed to mean the amount of time that livestock had access to each study transect annually. The relative abundance of passerines (including willow flycatchers) increased significantly with increasing shrub volume, and even when sites with confounding disturbance (camping, dredging) were removed from analysis, relative abundance of passerines decreased with increased frequency of grazing (r = -0.89, -0.74; P <0.01, 0.001). Willow flycatchers were detected at 4 of 9 transects at this study area. Willow flycatchers were 1.5 to five times more abundant at the site ungrazed for 40 years when compared to the 3 sites that experienced extensive grazing up until 3 to 5 years prior to study (groups b and d, above)(Taylor 1986). In a separate analysis at the same site, Taylor and Littlefield (1986) reported willow flycatcher and yellow warbler abundance relative to grazing intensity (AUMs). Over a ten-year period, yellow warblers

FEIS Volume 3, Chapter 3, part 4.4, page 153 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

increased from 7 birds to 56, and willow flycatchers increased from zero to 30, as grazing intensity decreased from 126,000 to 31,515 AUMs.

Noticeably lacking from these studies is willow flycatcher productivity (reproduction and survival) response in grazed and ungrazed sites. Although there are many regularly or occasionally grazed sites where willow flycatchers are found (Serena 1982, Harris et al 1987, 1988, Marshall and Stoleson 2000, H. Bombay pers. comm., S. Stoleson pers. comm.), there are no data available on the effects of light, moderate, or heavy grazing levels on willow flycatcher fitness and long-term population persistence. In any correlative or causal study of effects on birds it is important to emphasize that occurrence or abundance of the bird species does not necessarily connote successful breeding (for example, unpaired territorial individuals, unsuccessful mated pairs, or failed nests may occur) (Van Horne 1983, Stoleson et al. 2000).

Direct research on livestock grazing effects on willow flycatchers appears to be presently limited to the work above, however, other grazing studies and field observations regarding livestock interactions with meadows or willow flycatchers exist. Although livestock have not been directly seen dislodging willow flycatcher nests, in the southern Sierra Nevada Stafford and Valentine (1985) and Valentine et al. (1988) observed heavily trampled vegetation and broken willow boughs in the location of four overturned willow flycatchers nests while cattle were grazing in the immediate vicinity. It should be noted that two of the four nest failures attributed to livestock occurred within a holding area for cattle, and therefore experienced high livestock densities (Valentine et al. 1988). Additionally, King (1955) states that “cattle trampled two [willow flycatcher] nests” within his study area in Washington. It is unknown whether King (1955) directly observed livestock trampling nests, or inferred it from the nest surroundings. In the central Sierra Nevada, no incidence of direct nest disturbance by livestock has been reported for the 21 nests in 1986 and 1987 (Sanders and Flett 1989) or the 23 nests between 1997 and 2000 (H. Bombay pers. comm.) that were active while livestock were present at the study sites. Although they did not study livestock impacts on willow flycatchers, Loft et al. (1987) reported that under heavy stocking rates, cattle increasingly moved into willow stands as the summer season progressed, resulting in low branches being trampled or broken as the animals pursued herbaceous forage. Photo documentation of the same impacts on a more localized scale also exist for willow flycatcher breeding sites in the central Sierra Nevada (H. Bombay pers. comm.).

Impacts to meadow vegetation and hydrology from livestock grazing can vary widely in scope and scale depending upon a number of factors including vegetation community, soils, stocking rates, season of use, and type of livestock (Ratliff 1985, Clary and Webster 1989, Kinch 1989). Nevertheless, it is documented that livestock grazing can impact the types of meadows used by willow flycatchers by: browsing and grazing vegetation that provides cover for willow flycatcher nests and habitat for insect prey, altering hydrology through soil compaction and streambank erosion, and ultimately by affecting the composition and structure of plant communities in riparian and meadow areas (Duff 1979, Serena 1982, Kauffman et al. 1983a, 1983b, Stafford and Valentine 1985, Ratliff 1982, 1985, Taylor and Littlefield 1986, Flett and Sanders 1987, Loft et al. 1987, Sanders and Flett 1989, Clary and Medin 1990, Dull 1999).

Because willow flycatchers require meadows with a substantial water component, grazing or other activities that impact water tables and result in desiccation pose a threat (Serena 1982, Harris et al. 1987, Bombay 1999). In Oregon, Kauffman et al. (1983a) examined the impacts

FEIS Volume 3, Chapter 3, part 4.4, page 154 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 of late season grazing on streambanks along a 1.86 mile stretch of stream. They compared grazed areas (3.2 - 4.2 acres/AUM) with 5 exclosures (consistently grazed in previous years). Over the following two year period they found that significantly greater stream bank disturbance (undercut collapse and sloughoffs) in grazed areas compared to ungrazed areas when examining both total streambank loss in cm, and mean cm change from pre-treatment readings (Ps<0.05). Similarly, when comparing two montane streams, Myers (1987) reported that active bank erosion affected twice the total bank distance under annual hot season (7/1 to 9/1) grazing versus hot season grazing in only one out of three years after 10 years of study (significance not reported). When examining timing of grazing and streambank effects, Marlow and Pogacnik (1985 in Kinch 1989) reported that most stream bank damage occurs when soil moisture is in excess of 10 percent, which normally occurs prior to late July or early August (in southwestern Montana). When comparing time controlled grazing (10 cow/calf pairs for 3 to 4 days every 28 to 30 days), and deferred grazing (4 cow/calf pairs for 18 to 30 days, late season), with a livestock exclosure control, Marlow (1988) reported no significant difference in stream channel deformation between treatments, or the control. Total AUMs were the same between the 2 treatments, and utilization was held at 15-30 percent. Similarly, Buckhouse et al. 1981 and Hayes (1978 in Kauffman et al. 1983a) reported no significant difference in stream bank erosion between grazed and ungrazed areas.

Because willow flycatchers require willow and other riparian shrubs for nesting and foraging, the persistence of healthy stands through regeneration and recruitment is important. Therefore, the relationship between livestock grazing and riparian shrub regeneration and recruitment is of interest. Quantitative data exists relating livestock grazing to willow germination and recruitment, unfortunately many studies compare no grazing to extremely heavy grazing, or have no ungrazed controls. For this reason the impact of light to moderate grazing on shrub population dynamics is unclear. Nonetheless the following research does provide insight into potential impacts to the woody shrub component on which willow flycatchers rely. Kovalchik and Elmore (1992) describe that first-year seedlings are sensitive to livestock browsing because their small size makes them easily pulled from the ground or trampled. Therefore, willow stands become dominated by mature willows when grazing on young plants retards recruitment (Kovalchik and Elmore 1992). In Oregon, total seedling density of willows and cottonwoods in heavy use pastures was less than one fourth that in moderate spring use, and moderate fall use pastures (Shaw 1988 in Clary and Webster 1989). The same study reported that seedling densities were somewhat greater in spring grazed pastures. Myers (1987) reports that after 10 years of study in Montana, the proportion of willow plants in young age classes (1-20mm basal stem diameter) was roughly 20 percent in areas with moderate stocking annually during July and August, compared to 40 percent in areas with the same grazing intensity and season, but which were rested two out of three years (significance not reported). In the Sierra Nevada, Knapp and Matthews (1996) reported similar results when they created 3 exclosures (two were 4 years old, and one 11 years old) in 2 meadows. All three sites had significantly more willows in all height classes inside the exclosures, and most notably there were significantly more willows in the small (5-30cm) height classes.

Because willow flycatchers are found in meadow with high riparian shrub cover (Serena 1982, Bombay 1999), and have greater success at producing nests in territories with high shrub cover (Bombay 1999), the effect of different livestock grazing strategies on total cover and density of riparian shrubs is important. Knapp and Matthews (1996) report that in the Sierra Nevada the ratio of willows inside to outside of 3 exclosures to be 124:84, 222:22, and 264:11,

FEIS Volume 3, Chapter 3, part 4.4, page 155 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

respectively. In Colorado, Sedgwick and Knopf (1991) reported that after 3 years of late fall grazing at 0.46 ha/AUM, riparian vegetation was generally resilient to grazing when compared to 5 ungrazed control pastures. The only vegetation component that responded negatively to grazing was standing biomass of willow (P<0.01). In Utah, Schulz and Leininger (1990) compared exclosures with moderately grazed areas (65 percent utilization, June to October) in a riparian area that had previously been overgrazed to the point where it was “denuded” of vegetation. After 30 years, individual woody species did not show significant differences in density between treatments, but all woody species combined had significantly higher density in ungrazed areas (P<0.02). Percent canopy closure from willow was 8 ½ times greater in ungrazed areas (P = 0.004), and willows were significantly larger and older than in ungrazed areas (Ps<0.001). This last finding related to age class composition is in opposition to the findings of Knapp and Matthews (1996) and Schulz and Leininger (1990) described above. Similarly, in Nevada, Clary and Medin (1990) did not find statistically significant differences in willow cover and biomass in sites rested for 11 years, versus grazed sites. They did however notice a significant increase in non-willow shrubs in grazed plots, and suggested that these species were replacing willow within the sites.

In addition to the potential to alter vegetation and hydrology, livestock grazing indirectly affects willow flycatchers through the association of brown-headed cowbirds (Molothrus ater) with livestock. Livestock lower the herbaceous layer, disturb soil and create dung, facilitating brown-headed cowbird foraging on both insects and grain (Goguen and Mathews 1999). Verner and Ritter (1983) report that radio-tagged cowbirds are often found feeding with cattle in meadows during the day. This association can result in an increased local abundance of cowbirds and potential brood parasitism when livestock are present (for example Beedy and Granholm 1985, Flett and Sanders 1987, Sanders and Flett 1989, DeSante 1995, Graber 1996, Goguen and Mathews 1999, Shapiro et al. 1999). In addition, nest predation may be higher in grazed versus ungrazed meadows (Ammon and Stacey 1997). Because little is known about the actual effects of these grazing-related factors on willow flycatcher productivity and long-term population persistence in the Sierra Nevada bioregion, there is a high degree of scientific uncertainty and therefore potential management risk to the species.

Also under the control of forest service management are recreation activities which can have effects similar to livestock grazing, though probably to a lesser extent and intensity in many cases. Few correlative or causative studies on willow flycatchers and recreation are available. Blakesly and Reese (1988) reported that in riparian areas in Utah, the presence of willow flycatchers was negatively correlated with campgrounds. Anecdotally, Haas (pers. comm. in Marshall and Stoleson 2000) reported a pair of southwestern willow flycatchers successfully fledged young from a nest located only several meters away from a frequently used picnic table. Other studies of recreation on riparian birds suggest that dispersed and developed recreation activities in meadows and riparian areas have the potential to affect willow flycatchers through disruption of nest contents, vegetation trampling that removes nest cover and disturbs the insect community, altered hydrology through soil compaction and streambank chiseling, modified plant community composition and structure, and habitat fragmentation from trails and campgrounds that also influence the microclimate (Aitchison 1977, Luckenbach 1979, Johnson and Carothers 1982, Davis 1982, Riffell et al. 1996, Gutzwiller et al. 1997, 1998, Hamann et al. 1999). In addition, the supplemental food provided by developed and dispersed recreation and residential developments in close proximity to riparian areas and meadows, as well as movement corridors provided by trails, may indirectly affect willow

FEIS Volume 3, Chapter 3, part 4.4, page 156 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 flycatchers through an increase in local abundance of brown-headed cowbirds as well as nest predators, both native (such as jays, squirrels, chipmunks) and non-native (cats, dogs) (Johnson and Carothers 1982, Blakesley and Reese 1988, Small and Hunter 1988, Hickman 1990, Burkey 1993, Askins 1994, Paton 1994, Rich et al. 1994, Miller et al. 1998).

Other current factors affecting willow flycatcher habitat on national forest lands include: hydrologic and vegetative changes resulting from silvicultural treatments, fire suppression and forest encroachment, fuel treatments, dams or diversions, mining, road construction and maintenance, and other forest service management activities inside or adjacent to riparian and meadow communities that result in the degradation or loss of this habitat. As with recreation effects, the extent of the impact of these activities on willow flycatchers and their habitat has not been quantified. While these factors have not been shown to have direct effects on the decline of the willow flycatcher population in the Sierra Nevada bioregion, their indirect effects on willow flycatcher population viability, via their influence on riparian areas and meadows (e.g. increased nest predator access through meadow desiccation, habitat type conversion, and loss of breeding or foraging habitat) is strongly suggested (Remsen 1978, Garrett and Dunn 1981, Serena 1982, Stafford and Valentine 1985, Flett and Sanders 1987, Valentine et al. 1988).

Willow flycatchers feed primarily on insects, many of which have aquatic larval stages (Beal 1907, Bent 1942, Bombay 1999). Although no inferential or causative studies have examined willow flycatcher prey populations in the Sierra Nevada, research on invertebrates and the interaction between passerines and invertebrates suggest that stream disturbing activities on and off national forest lands, as well as other changes in surface and subsurface water flows, may affect not only the riparian and meadow breeding habitat of willow flycatchers, but may also affect their invertebrate food supply (Voight 1976, Erman 1984, Gray 1993, Kelly and Wood 1996). Management activities that remove emergent vegetation and alter streambanks can disrupt or remove habitat required for insect egg mass and larval stage development and affect stream temperature buffering required for aquatic larval insect stages (Erman 1984, Wohl and Carline 1996, Strand and Merritt 1999). Additionally, adult life stage hiding cover for some insects may be lost when vegetation along streams is altered (Erman 1984). Insect prey populations may also be affected by pesticide programs administered by the Forest Service (Gard and Hooper 1995).

Across most of North America, willow flycatchers are frequent hosts of the brown-headed cowbird (Molothrus ater), a brood parasite, especially in the lowland parts of their range (Grinnell and Miller 1944, Friedman 1963, Whitfield 1990, Whitfield and Enos 1996, Whitfield and Sogge 1999, Marshall and Stoleson 2000). The cowbird lays its eggs in host nests, and the host raises the cowbird young, often to the detriment or death of the host’s young. Although brown-headed cowbird parasitism rates for southwestern willow flycatchers are often quite high (more than 60 percent at times), brood parasitism has a considerably smaller impact on willow flycatchers in most of the Sierra Nevada. Fourteen instances of cowbird brood parasitism are documented for the willow flycatcher in the Sierra Nevada (not including southwestern willow flycatchers along the Kern River), all at elevations above 6,000 feet (approximately 250 nests observed)(Gaines 1977, Valentine et al. 1988, Sanders and Flett 1989, Bombay et al. 1998, Bombay 1999, Morrison et al. 2000). Morrison et al. (2000) report 8 of the 12 parasitism events that they detected occurred in two adjacent sites with only 2 to 5 willow flycatcher territories annually. These parasitized nests accounted for 44 percent of the

FEIS Volume 3, Chapter 3, part 4.4, page 157 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

18 willow flycatcher nests discovered at the two sites between 1998 and 2000. Therefore, although the regional parasitism rate is low relative to brood parasitism rates in other areas of the West, until demographic research is completed the impact of parasitism should not be assumed insignificant, given the small willow flycatcher population size in the Sierra Nevada bioregion (Rothstein 1994).

Within the Sierra Nevada, brown-headed cowbirds associate with pack stations, corrals, supplemental feed, livestock holding facilities, livestock herds, campgrounds, picnic areas, and rural communities to exploit bird feeders, grain, waste grain and insects associated with manure, and stirred-up insects on the ground (Rothstein et al. 1980, Verner and Ritter 1983, Stafford and Valentine 1985, Airola 1986, Flett and Sanders 1987, Laymon 1987, Sedgwick and Knopf 1988, Verner and Rothstein 1988, Harris 1991, Gaines 1992, Rothstein 1994, Coker and Capen 1995, Goguen and Mathews 1999). Note however that cowbirds appear to be in low numbers at pack stations where animals are fed pelletized food (S. Rothstein pers. comm.). Telemetry studies have shown that feeding sites of individual cowbirds are tightly linked to locations of grazing livestock and pack stations (or other livestock facilities) (Verner and Ritter 1983, Verner and Rothstein 1988, Goguen and Mathews 1999, Halterman et al. 1999, Shapiro et al. 1999). Cowbirds in the Sierra Nevada can travel approximately 4 to 6 miles (Verner and Ritter 1983, Rothstein et al. 1984, Verner and Rothstein 1988, Gaines 1992) or more (S. Rothstein pers. comm.) from these feeding and roosting locations to the breeding grounds of host bird species. Verner and Ritter (1983) report that numbers of brown-headed cowbirds were negatively correlated with distance to pack stations or corrals (P < .01) and elevation (P < .01). To provide context for the latter factor, recall that all reported cowbird brood parasitism events on willow flycatchers in the Sierra Nevada have occurred above 6,000 feet. Whether forest openings created by silvicultural treatments, prescribed fire, and mechanical thinning near riparian and meadow areas will increase the opportunity for cowbird brood parasitism in the Sierra remains an open question. Data from the southern Sierra Nevada do not indicate that these vegetation management activities on the west slope will substantially affect cowbird abundance and brood parasitism pressure; Verner and Ritter (1983) discovered that brown- headed cowbirds were rare in all areas except meadows. Their observations of the mean number of cowbirds per count is as follows: meadows = 0.56, clear-cuts = 0.095, logged forest = 0.07, unlogged forest = 0. On the eastside Mammoth Lakes area, however, brown-headed cowbirds were found in meadows and open Jeffrey pine forests (Verner and Rothstein 1988). Any risks from vegetation management will be compounded within active grazing allotments where cowbirds feed and roost in association with livestock (Rothstein et al. 1980, Beedy and Granholm 1985, Airola 1986, Flett and Sanders 1987, Laymon 1987, Sedgwick and Knopf 1988, Harris 1991, Gaines 1992, Rothstein 1994, Coker and Capen 1995, DeSante 1995). Cowbird parasitism is a risk factor that is only partially under the control of Forest Service management. In areas where livestock, pack stations, campgrounds and residential or urban areas occur on non-forest service lands within proximity of willow flycatchers, the forest service cannot manage or alter the activities attracting cowbirds. Additionally, ability to manage cowbird populations on forest service lands may be limited when populations on nearby non-forest service lands are not being managed. Also of concern are the sociopolitical and ethical implications of cowbird trapping, particularly in areas with high human use or habitation (Hall and Rothstein 1999).

Note that nest predation is the major cause of nest failure among willow flycatchers in the Sierra Nevada (Valentine et al. 1988, Whitfield et al. 1999, Morrison et al. 2000). Control of

FEIS Volume 3, Chapter 3, part 4.4, page 158 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 natural predators requires a site-specific and accurate knowledge of willow flycatcher predator species, their densities, and life histories (Cote & Sutherland 1997). Although the removal of predators has sometimes been recommended for conservation purposes (Wagner and Seal 1992, Garrott et al. 1993), efforts are often ineffective and inadvisable due to unintended biological consequences that often attend such actions (Soule et al. 1988, Goodrich and Buskirk 1995, Cote & Sutherland 1997, Ratnaswamy and Warren 1998). Cote and Sutherland (1997) reviewed 20 long-term studies on primarily game bird species, and reported that although hatching success and post-breeding population size were improved by predator removal or exclusion, there were no significant increases in size of the returning breeding bird populations in the next year. Intensive trapping programs may also fail to reduce total predator populations because of associated increases in other non-targeted nest predators (meso-predator release) (Linhart and Robinson 1972, Sargeant and Arnold 1984, Soule et al. 1988, Cote & Sutherland 1997). Similarly, Dion et al. (2000) reported that when medium-sized predators were removed, smaller more numerous predators shifted their predation patterns into vegetation types not used in areas where the medium-sized predators remained. At the same study area in North Dakota, Dion et al. (1999) reported that after the removal of these medium- sized mammalian duck predators, nest success of grassland songbirds did not significantly improve. In addition to shifts in predator assemblage, the ability of some predator species to increase litter size when their population density declines can reduce trapping effectiveness (Sterling et al. 1983, Goodrich and Buskirk 1995). There are also sociopolitical and ethical dilemmas associated with trapping predator species native to the Sierra Nevada (Goodrich and Buskirk 1995).

Due to the limitations of active predator exclusion or trapping, the impact of nest predators may be more effectively limited indirectly by managing meadows so that they provide willow flycatchers habitat with the highest levels of protection from predators (Goodrich and Buskirk 1995, Cote & Sutherland 1997). Martin (1992) suggests that cup-nesting birds suffer less predation when nest sites and territories have high foliage density and cover. Additionally he reported that these characteristics should occur across the entire territory to allow for multiple nest location choices separated by as great a distance as possible to limit the possibility of repeat depredations by the same predator. Bombay (1999) reported that in the central Sierra Nevada willow flycatcher territories with higher riparian shrub cover and deeper water depths were significantly more likely to be successful at producing fledglings. Therefore management that reverses meadow desiccation, minimizes vegetation removal, and supports riparian shrub regeneration and recruitment may limit predator access to nests (Ammon and Stacey 1997). Additionally, management of activities occurring near meadows to ensure that they are not inadvertently increasing predator density or diversity (through providing food, increasing edge effects) could also limit the potential predator community (Martin 1992, Cote and Sutherland 1993).

In the Sierra Nevada, the long-term response of willow flycatchers and their habitat to fire is not known, although wildfire effects are presently under investigation in Arizona (Paxton et al. 1996). The natural role and extent of fire in Sierra Nevada meadows is unclear, but it is likely that under drought conditions historic late summer and fall fires may have at least occasionally influenced meadow vegetation (DeBenedetti and Parsons 1979, Ratliff 1985, Dull 1999). Although the current risk of wildland fire on willow stand mortality in meadows occupied by willow flycatchers in the Sierra is unknown, Agee (1994) asserts that in riparian areas willow re-sprouting stabilizes soils and re-growth takes place. However, even if complete willow

FEIS Volume 3, Chapter 3, part 4.4, page 159 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

regeneration in occupied willow flycatcher habitat were to occur following a fire, the interval between habitat recovery and willow flycatcher territory re-establishment is currently unknown (Paxton et al. 1996). A fire on the Sequoia National Forest during the summer of 2000 burned across a meadow (including willow shrubs) occupied by a willow flycatcher in 1999; same- season willow re-sprouting is reported and vegetation recovery will be monitored (T. Ritter pers. comm.).

The effect of lodgepole pine encroachment into meadows used by willow flycatchers is currently unknown. There is uncertainty about how and when lodgepole pine entered meadows historically, and whether current encroachment is part of natural succession, inter-decadal weather (rainfall) oscillations, the result of human-induced soil disturbance and meadow desiccation, fire suppression, or a combination of all of these things (Benedict 1984, Ratliff 1985, Dull 1999, W. Woolfendon pers. comm.). In meadows currently occupied by willow flycatchers, lodgepole pines are used by males as singing and foraging perches (Sanders and Flett 1989), however proximity to trees is also implicated in increased nest failure (Bombay 1999). Additional data is needed to determine if and when lodgepole pines should be actively managed.

Please note that all of the factors mentioned above that threaten the willow flycatcher population and their habitat in the Sierra Nevada bioregion may occur both on and off national forest lands although the Forest Service can only control or mitigate activities within its boundaries.

Other potential threats to the willow flycatcher population in the Sierra Nevada bioregion that are beyond the control of the Forest Service include: wintering and migration habitat loss and degradation, urbanization and habitat conversion, insect control programs near rural mountain communities (Graber 1996), random environmental and demographic events, unforeseen fluctuations in population regulating mechanisms (for example competition, predation, parasites, disease), and historical uses.

Conservation Measures Based on the literature review of potential willow flycatcher risk factors provided above, the following conservation measures applicable to Forest Service lands and within agency control will partially or fully mitigate these risk factors:

1. Conduct repeated surveys according to Pacific Southwest Region protocol (Bombay et al 2000) to determine contemporary willow flycatcher distribution and abundance in the planning area. 2. Determine willow flycatcher breeding habitat criteria and demographic trends. Provide linkage of willow flycatcher demographic parameters with habitat characteristics to inform and implement management of known and potential habitat for willow flycatcher population growth and expansion. 3. Eliminate, reduce, or modify human-induced factors that do not sustain the biotic vigor of riparian and meadow ecosystems, including livestock grazing and recreation, to allow recovery of these systems. 4. Eliminate, reduce, or modify human-induced factors that may contribute to meadow desiccation, by altering meadow hydrology, including roads, livestock grazing, recreation, and vegetation management.

FEIS Volume 3, Chapter 3, part 4.4, page 160 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

5. Actively restore vegetation and hydrology in known willow flycatcher sites as well as emphasis habitat within colonization distances of known sites that cannot achieve site potential via passive restoration. 6. Reduce willow flycatcher nest predation potential by indirectly managing nest predator population through various means, including: limit or reduce predator access to willow flycatcher nests by maintaining or restoring meadow hydrology, ensuring high values of willow cover, foliar density, and recruitment to provide multiple nest sites; reducing human-induced increases in nest predators (rodents, musteHds, corvids) associated with campgrounds, picnic areas, and summer homes adjacent to meadows. 7. Avoid pesticide and herbicide use in known and potential willow flycatcher habitat. 8. Reduction of cowbird brood parasitism pressure in the planning area 9. Employing a before-and-after, control-impact (BACI) experimental design, use prescribed burning within and adjacent to known and potential willow flycatcher habitat to determine fire effects on willowdominated communities as well as willow flycatcher occurrence and productivity is the planning area. 10. Examine where and when passive (prevent soil disturbance) or active (tree girdling, fuels treatments) management of lodgepole pine encroachment is warranted in meadows managed for willow flycatchers.

The California Wildlife Habitat Relationships (CWHR) model does not adequately model habitat for ARMassociated species, requiring the development of a separate approach for analyzing the effects of the Sierra Nevada Forest Plan Amendment Project FEIS alternatives on ARM species and their habitats.

The following discussion of environmental consequences is based primarily upon a comparison of the effects of each Sierra Nevada Forest Plan Amendment Project FEIS alternative in addressing key management issues identified in the NOI that pertain to the willow flycatcher (E. t. adastus and E. t. breusteri) population and their habitat in the planning area. Additional factors that may influence the willow flycatcher population and habitats (for example silvicultural treatments, prescribed burning, recreation or other forest management in or adjacent to riparian and meadow habitats) are discussed. The consequences of the management direction of the HergerFeinstein Quincy Library Group Forest Recovery Act Pilot Project on the willow flycatcher are also evaluated. Sources used to evaluate the environmental consequences on the willow flycatcher population and their habitats include peer-reviewed journal articles, books, professional meetings and symposia, unpublished reports and data, field knowledge and preliminary spatially-explicit analyses of GIS data.

Factors used to assess Environmental Consequences There are four major factors that were identified in the NOI which pertain the willow flycatcher population in the Sierra Nevada bioregion and may be influenced by Forest Service management. These include: livestock grazing, willow flycatcher population monitoring, willow flycatcher habitat restoration, and brown-headed cowbird brood parasitism pressure. The key management issues, or factors, used to evaluate the environmental consequences of each of the Sierra Nevada Forest Plan Amendment Project FEIS alternatives for the willow flycatcher population and their habitat in the planning area are listed below in order of management concern and priority:

1. Levels of direct and indirect effects of livestock grazing (season of use, duration, methods, and utilization) on willow flycatcher populations and their habitat;

FEIS Volume 3, Chapter 3, part 4.4, page 161 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

2. Requirement to implement a monitoring program that includes annual surveys of willow flycatcher breeding success and habitat conditions; 3. Direction to restore degraded areas to desired conditions for willow flycatcher nesting habitat in order to increase opportunities for willow flycatcher population expansion in the Sierra Nevada bioregion; 4. Management actions to lessen the influence of brown-headed cowbird brood parasitism on the willow flycatcher population (includes livestock and recreation facilities).

Assumptions and Limitations The effects analyses of the Sierra Nevada Forest Plan Amendment Project FEIS alternatives and Herger-Feinstein Quincy Library Group Forest Recovery Act Pilot Project management direction regarding the willow flycatcher population and their known sites and other available habitat (emphasis habitat and small, wet meadows less than or equal to 15 acres with woody vegetation) are predicated upon the following assumptions and limitations:

Willow Flycatcher Population Without effective short-term action and long-term management, the currently estimated willow flycatcher population of 300-400 individuals in the planning area is insufficient to provide a high level of confidence in population persistence. Principles of conservation biology suggest that to ensure long-term population persistence the willow flycatcher population in the Sierra Nevada would need to be much larger (Lande 1995, 1998). Small, isolated populations are inherently at risk of extirpation due to demographic and environmental stochasticity as well as deleterious genetic effects. Alternatives that manage for growth and habitat expansion capability of the willow flycatcher population, not maintenance at the current low level, may help compensate for stochastic effects and instill higher confidence in species persistence.

Stochastic events such as severe weather (e.g. drought, flood, late or heavy snowfall), disease outbreaks, or other random environmental perturbation, as well as chance demographic events, can negatively affect the willow flycatcher population independent of Forest Service management activities or actions.

Although the willow flycatcher population in the Sierra Nevada is known to have dramatically declined after 1940, the current direction and magnitude of the demographic trend are uncertain (Serena 1982, Stafford and Valentine 1985, Flett and Sanders 1987, Harris et al. 1987, 1988, Valentine et al. 1988, Sanders and Flett 1989, H. Bombay pers. comm.). However, if available and preliminary nesting site re-occupancy data as well as central Sierra Nevada nest success and fecundity rates are used as a metric for population trend, it appears that the willow flycatcher population in the Sierra has continued to seriously decline during the past two decades (Morrison et al. 2000).

Specific habitat requirements necessary to support long-term willow flycatcher viability in the Sierra Nevada bioregion can be inferred from existing literature but are not known with certainty.

Aquatic, riparian and meadow-associated bird species that maintain site fidelity, such as the willow flycatcher, will be vulnerable to short-term, as well as long-term, disturbances within and adjacent to these habitats.

FEIS Volume 3, Chapter 3, part 4.4, page 162 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

The effects of habitat degradation or loss on willow flycatcher migratory stopover sites, wintering grounds, and subsequent willow flycatcher survivorship cannot be predicted.

Standards and Guidelines Implementation Standards and guidelines will be implemented within one year of revising management direction through this FEIS, except where specified otherwise.

Where bioregional standards are ambiguous or default to local management control, they may be widely interpreted and will have higher uncertainty with respect to implementation and therefore higher potential risks for focal species.

Managers and grazing permittees will know the locations and designations of known willow flycatcher sites and emphasis habitat and will follow the standards and guidelines that pertain to each of the willow flycatcher site categories.

The existing GIS meadow coverages were developed independently by the 11 national forests in the planning area and merged into a Sierra Nevada Forest Plan Amendment Project planning area meadow layer (K. Teuber pers. comm.). Thus, the merged GIS meadow coverage for the planning area is the result of multiple approaches with varied accuracy. Mapping problems inherent in developing a bioregional-scale GIS coverage complicate willow flycatcher site identification and analysis, and therefore increase the level of risk and uncertainty in managing the population (H. Bombay pers. comm.). Recently generated willow flycatcher site maps will require manual inspection to ensure appropriate identification, classification, and management of willow flycatcher habitat.

Fencing If properly maintained, fencing offers the most effective tool in keeping livestock out of riparian areas (Skovlin 1984). Frequent riding, water development, salt, supplemental feed and rubbing posts are not considered as effective as fencing. The limitations of fencing, in addition to cost and maintenance, are that fences will not alleviate the indirect effects of cowbird brood parasitism associated with livestock congregating along the fenceline that a successful herding program might address. Also, fences can present a hazard to other wildlife species (for example Fitzner 1975, Avery et al. 1978, Knight et al. 1980).

Herding Herding can be a positive solution in keeping livestock out of areas that are off limits to grazing, if it is done on a daily basis (Kauffman and Krueger 1984). However, Chaney et al. (1990) report that “herding was successful at keeping cattle out of a drainage but not out of the riparian area once they were in the drainage”. Thus, it is unknown whether herding, even if done on a daily basis wherever applicable in the Sierra Nevada, will be effective in keeping livestock out of restricted willow flycatcher habitat.

Utilization Due to the variation in forage availability in planning area meadows in a given year, it is uncertain how specific utilization standards will affect willow flycatcher habitats where grazing is allowed.

FEIS Volume 3, Chapter 3, part 4.4, page 163 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Although livestock grazing and willow flycatchers may presently co-occur in known willow flycatcher sites, it is unknown whether the grazing utilization standards and guidelines outlined in this FEIS (for example stubble height and percent utilization of grass and shrubs) will sustain the long-term persistence of the willow flycatcher population in the Sierra Nevada bioregion, or the willow-dominated communities upon which they rely. In the Sierra Nevada, Loft et al. (1987) found that in willow-dominated areas ungrazed by livestock, deer browsed between 12 and 17 percent of annual willow shoots. Therefore, if total shrub utilization does not exceed 20 percent, it is assumed that impacts to the shrub community will not differ substantially from those experienced with deer alone.

Loft et al. (1987) found that as livestock grazing intensity increased, the amount of willow utilization by both cattle and sheep increased. They also observed that willows in areas that were heavily grazed (roughly 0.5 to 0.7AUMs/acre, season-long) in repeated years took on an inverted umbrella shape (highlining). As the season progressed, cattle increasingly moved into willow stands, where low branches were trampled or broken as the animals penetrated into the shrub canopy in pursuit of ungrazed herbaceous forage as well as browse (Loft et al. 1987). By requiring maximum herbaceous utilization values of between 30 - 40 percent under season- long grazing, it is assumed that livestock will be less likely to begin browsing on willows or otherwise damage outer branches (Ratliff 1985, Loft et al. 1987, Clary and Webster 1989).

Season of use Late season grazing (for example, following the breeding season in willow flycatcher sites) may be detrimental to willow flycatcher habitat because willows, and riparian vegetation in general, are more palatable and highly preferred by livestock compared with upland plant communities at this time (Kauffman et al. 1983b). Early season grazing raises concerns about streambank shearing (Marlow and Pogacnik 1985 in Kinch 1989) and increased cowbird brood parasitism rates as a result of introducing livestock, and any associated cowbirds, at a time when willow flycatcher nests are most susceptible to parasitism (in other words, during the egg laying period) (Verner and Ritter 1983). Grazing during the willow flycatcher breeding season presents potential risks of livestock overturning willow flycatcher nests and decreasing nest concealment through grazing and browsing (Valentine et al. 1988).

Stream and Watershed Management Where livestock grazing is allowed in known willow flycatcher sites or emphasis habitat, streambank disturbance could occur under all alternatives except Alternative 3, regardless of season of use.

Where recreation (hiking, fishing, camping, OHV use) occurs in known willow flycatcher sites or emphasis habitat, streambank disturbance could occur under all alternatives.

Willow flycatcher sites that are located in drainages with high road densities and other ground disturbing activities (for example silvicultural and mechanical fuel treatments) are at greater risk of hydrologic alteration and habitat degradation from sedimentation and erosion. Similarly, poorly designed stream crossings place hydrologic systems at risk of erosion and stream course alteration when culverts or bridges improperly redirect normal flows, or fail during flood events.

FEIS Volume 3, Chapter 3, part 4.4, page 164 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Willow flycatchers may be more vulnerable to nest depredation as wet meadows become dry and nests are more accessible to mammalian predators; meadow desiccation may occur as the summer breeding season progresses either from natural causes or as a result of human-caused hydrologic changes.

Riparian and meadow buffer zones will mitigate some of the erosion or sedimentation effects resulting from vegetation management (such as silvicultural treatments, mechanical thinning, salvage logging).

Brown-headed Cowbirds Willow flycatchers are at greater risk of cowbird brood parasitism where pack stations, corrals, supplemental feed, livestock holding facilities, livestock herds, campgrounds, picnic areas, rural communities or other brown-headed cowbird-associated locations occur within at least 5 miles (cowbird foraging distance) of occupied willow flycatcher sites (Rothstein 1980, Verner and Rothstein 1988).

Where large or numerous mountain communities or private recreation sites, grazing, or timberland holdings occur within 5 miles of occupied willow flycatcher sites, the ability to control brown-headed cowbirds at these sites via cowbird trapping, seasonal facilities closures, or re-location of human-related structures on national forest lands may be limited.

Recreation Recreation activities in willow flycatcher habitat can have effects similar to livestock grazing, although to a lesser extent and intensity in many cases. In addition, the supplemental food provided by developed and dispersed recreation in close proximity to riparian areas and meadows, as well as movement corridors provided by trails, may indirectly affect willow flycatchers through an increase in local abundance of brown-headed cowbirds as well as nest predators, both native (such as jays, squirrels, chipmunks) and non-native (cats, dogs) (Johnson and Carothers 1982, Blakesley and Reese 1988, Small and Hunter 1988, Hickman 1990, Burkey 1993, Askins 1994, Paton 1994, Rich et al. 1994, Miller et al. 1998).

Insect Prey Willow flycatcher feed on insects, many of which have aquatic larval stages (Beal 1907, Bent 1942, Bombay 1999). Watershed management that improves water quality may enhance conditions for aquatic insects (Strand and Merritt 1999). However, disturbance within the riparian buffer zone can adversely affect willow flycatcher insect prey populations (Erman 1984, Wohl and Carline 1996, Strand and Merritt 1999). Insect prey populations may also be affected by pesticide programs (Gard and Hooper 1995).

Fire and Fuels Wildland fire in riparian and meadow vegetation is typically a low frequency, low intensity event (Agee 1994, DeBenedetti and Parsons 1979, Ratliff 1985, Dull 1999) with the exception of flare-ups in areas concentrated with down woody debris (Skinner and Chang 1996, B. Bahro pers. comm.).

The role and effects of fire and fuel treatments in riparian and meadow communities should be examined on a case-by-case basis (Weatherspoon et al. 1992).

FEIS Volume 3, Chapter 3, part 4.4, page 165 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

The effects on meadows of fuel treatments in adjacent forests are unknown. For example, timber thinning and mechanical fuels treatments in the forest adjacent to meadows, in addition to possibly creating canopy openings that could have indirect effects on the willow flycatcher population (via increased brown-headed cowbirds), can affect meadow hydrology (for example erosion, sedimentation) and microclimate (desiccation), with possible accompanying effects on willow flycatcher prey or predator abundance or composition, and may put willow flycatcher habitat suitability at risk. Conversely, such treatments could make more water available and increase meadow wetness with possible beneficial effects for willow flycatchers (such as increased prey, decreased nest predation, increased willow flycatcher productivity).

Effects of Alternatives The four major factors identified in the NOI as affecting the willow flycatcher population in the Sierra Nevada bioregion and over which the Forest Service has influence are evaluated in order of management priority. For each factor, there follows an analysis of the unique and similar effects and environmental consequences of the Sierra Nevada Forest Plan Amendment Project FEIS Alternatives 1-8, in numeric order. To facilitate cross-referencing between alternatives and their specific Standards and Guidelines (s&g), the s&g id code numbers are included in parentheses. An evaluation of additional factors is also included. A separate effects analysis of Herger-Feinstein Quincy Library Group Forest Recovery Act Pilot Project is the final section. A concluding summary of these analyses is provided.

1) Levels of direct and indirect effects of livestock grazing (season of use, duration, methods, and utilization) on willow flycatcher populations and their habitat

a. Livestock Grazing: Season of Use, Duration, Methods To assist with the comparison of the effects of the alternatives in this section, combined acreage estimates of known willow flycatcher sites and potential willow flycatcher habitat (emphasis habitat and small wet shrubby meadows) for national forest lands within the planning area boundaries are also included. Although GIS analyses indicate that emphasis habitat and small meadows possess water and riparian shrub attributes, it is unknown how many of these meadows currently contain these attributes in amounts and configurations necessary to support willow flycatchers. Unless ground-truthing or meadow mapping refinements prove otherwise, emphasis habitat and small wet, shrubby meadows are assumed to have the potential to provide willow flycatcher habitat. For the purpose of these analyses, all known willow flycatcher sites are meadows or riparian areas with documented willow flycatcher presence during the breeding season, specifically, either:

1. willow flycatcher observed between June 15 and August 1; or 2. willow flycatcher observed between June 1 - June 14, or August 2 - August 15, unless willow flycatcher was:

• absent during surveys conducted between June 15 and July 15 in the same year, • absent during June 15 to July 15 surveys in multiple subsequent years, or • detected at a site that is clearly outside of known habitat requirements (see criteria in “Affected Environment” above).

The affected acres related to willow flycatcher habitat are based upon available data and subject to revision. For example, with subsequent investigation some sites may be found to

FEIS Volume 3, Chapter 3, part 4.4, page 166 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

have undergone a habitat type conversion while other sites may be discovered. In addition, it is expected that estimates of emphasis habitat and small meadow acres on national forest lands will change, subject to consistent regional meadow mapping standards required by this FEIS (see s&gs B46 and B46A). Note however that potential willow flycatcher habitat does not include wet meadows without riparian shrub vegetation or dry meadows, and thus acreage estimates for these other meadow types are not part of these analyses.

Table 4.4.2.3b. Acres of shrubby wet meadows containing known or potential willow flycatcher habitat in active, inactive, non-allotments, and outside allotments on Forest Service land within Sierra Nevada Forest Plan Amendment Project Planning Area. Acres in Acres in Acres in Non- Acres Total acres Active Inactive Allotments Outside Allotments Allotments Allotments Willow Flycatcher Habitat Category Known sites 6,704 101 191 3,679 10,675* Emphasis Habitat2 36,783 5,930 61 9,921 52,6951 Small wet, shrubby meadows3 12,603 3,754 136 5,813 22,3061 Total Shrubby Wet Meadow Habitat 56,090 9,785 388 19,413 85,6761

*Additional sites may be discovered. 1= This is the best estimate with currently available meadow maps. After meadows are completely and consistently mapped throughout the planning area, the acreage will likely increase by an unknown factor. It is unknown at this time whether these additional acres occur inside or outside grazing allotments. Note that potential willow flycatcher habitat does not include wet meadows without woody vegetation or any dry meadow types, and thus acreage estimates for these other meadow types are not part of these analyses. 2= Emphasis Habitat are meadows greater than 15 acres in size with water present and a woody riparian shrub component. 3= Small wet, shrubby meadows are less than or equal to 15 acres in size with water present and a woody riparian shrub component.

Under existing conditions (Alternative 1), Forests in the planning area apply standards and guidelines to protect actual willow flycatcher nesting areas, or to protect all riparian and meadow areas from activities that diminish hydrologic or vegetative function. S&Gs are generally qualitative in nature, with the exception of some livestock utilization standards. The Stanislaus and Tahoe National Forests have made specific modifications to Allotment permits to limit livestock grazing (limited operating period, LOP) in some known willow flycatcher sites until after August 15 each year. In addition to an LOP, the Tahoe National Forest traded other forest lands in order to acquire habitat known to support willow flycatchers, with the intention of joint management for livestock forage and willow flycatcher habitat (Ritter and Roche 1999). An estimate of 24,000 acres affected by existing riparian community-related s&gs was provided by Modoc National Forest; Modoc is the only forest in the Sierra Nevada bioregion that has made riparian habitat a land allocation with reduced outputs in their existing forest plan. Since the Modoc allocation estimate is comprised of riparian habitat and not restricted to known and potential willow flycatcher habitat, this estimate precludes comparison with the affected acres of the other alternatives discussed below.

Current estimates of known and potential willow flycatcher habitat acres on all national forest lands within the Sierra Nevada Forest Plan Amendment Project planning area are 85,676, with 56,090 acres (65 percent) inside active grazing allotments. Known willow flycatcher sites occur in 64 range allotments, including 56 active allotments. Available analyses for 56 percent of the 82 known willow flycatcher sites on national forest lands indicate known sites inside active grazing allotments range in size from 1 to 667 acres with a mean size of 87 acres. Similarly, available analyses for 53 percent of approximately 1,457 emphasis habitat meadows on national forest lands indicate emphasis habitat meadows within active grazing allotments

FEIS Volume 3, Chapter 3, part 4.4, page 167 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

range from 15 to 371 acres with an average size of 41 acres. Reporting for 80 percent of approximately 5,240 small wet, shrubby meadows on Forest Service land, sites inside active grazing allotments vary from 0.10 to 15 acres with an average size of 5 acres.

Table 4.4.2.3c. Acres of Wet Shrubby Meadows Containing Known or Potential Willow Flycatcher Habitat Inside and Outside 5-mile Buffers in active, inactive, non-allotments and outside allotments on National Forest lands in Sierra Nevada Forest Plan Amendment Project Planning Area 5-mile Buffers Habitat in Habitat in Habitat in Habitat Total Active Inactive Non- Outside acres Allotments Allotments Allotments Allotments Inside 5-mile Buffers Known sites 6,704 101 191 3,679 10,675 Emphasis Habitat 13,750 1,830 33 4,607 20,220 Small wet, shrubby meadows 5,263 756 114 3,884 10,017 Outside 5-mile Buffers Known sites 0 0 0 0 0 Emphasis Habitat 23,033 4,100 28 5,314 32,475 Small wet, shrubby meadows 7,340 2,998 22 1,929 12,289 Total Wet Shrubby Meadow Habitat 56,090 9,785 388 19,413 85,676

Alternative 2 prohibits livestock grazing year-round in “suitable” willow flycatcher habitat within 5 miles of known willow flycatcher sites. Although complete knowledge of willow flycatcher habitat associations are lacking, for the purpose of this alternative, “suitable habitat” is defined as both known willow flycatcher sites and potential willow flycatcher habitat (meadows with standing or running water on June 1 and a riparian deciduous shrub component). Alternative 2 offers the highest likelihood of eliminating risk to the willow flycatcher population from the direct effects of livestock grazing in known sites as well as other potential willow flycatcher habitat within the most likely willow flycatcher movement distance (5 miles) (B33). Willow flycatcher site occupancy is not static in the Sierra Nevada (Morrison et al. 2000, KRCD 1992) and managers must be able to respond to shifting occupancy. Between-site, between-year movements of up to 186 miles have been recently documented for the southwestern willow flycatcher (Paxton and Sogge 2000), which has received considerably more study than birds in the Sierra. However, the 5-mile radius is within the 1/3 to 9-mile range of natal dispersal distances and adult between-site, between-year movements (n = 11 observations, x = 2.63 miles, SD = 2.42) documented to date for willow flycatchers in the Sierra Nevada (Stafford and Valentine 1985, Sanders and Flett 1989, H. Bombay pers. comm.). It has been observed that areas with multiple willow flycatcher breeding sites that are geographically close have the highest degree of between-site movement (Paxton 2000). Alternative 2 is the only alternative that eliminates livestock grazing in the entire meadow of all potential willow flycatcher habitat within close proximity of known willow flycatcher sites. Results of the latest available research on livestock grazing and willow flycatchers indicate that willow flycatcher habitats may be restored while still allowing late- season grazing on a rest-rotation basis, although the rate at which the species is recovered will be slower than if cattle are removed (Stanley and Knopf in review). For example, willow flycatcher densities approximated those on ungrazed pastures by the 10th year of the study. The authors suggest a more severe restriction on grazing may be preferred where conservation priorities emphasize maximum recovery of the species (Stanley and Knopf in review). Furthermore, no specific grazing system has been documented to sustain the biotic vigor of riparian ecosystems (Behnke 1979 in Stanley and Knopf in review). Within designated sites, Alternative 2 is the only alternative that eliminates livestock grazing from the entire meadow

FEIS Volume 3, Chapter 3, part 4.4, page 168 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

(eliminating possible effects on meadow hydrology, herbaceous layer, predator or prey composition and abundance, willow flycatcher productivity). Year-round exclusion of livestock grazing in willow flycatcher sites is supported by willow flycatcher habitat selection and demographic data from the Sierra indicating that territories with a greater proportion of shrub cover are significantly more likely to produce fledglings (Bombay 1999). Furthermore, livestock grazing exclosures have proven successful in maintaining riparian shrub communities; results from long-term studies (at least 10 years) indicate that successful woody vegetation (for example, willow) germination and recruitment occurs in the absence of disturbance provided by livestock and report significant increases in woody species cover and biomass (Taylor and Littlefield 1986, Taylor 1986 (eastern Oregon), Clary and Medin 1990 (northeastern Nevada), Schulz and Leininger 1990 (Rocky Mountains, Colorado), Knapp and Matthews 1996 (southern Sierra Nevada)). Note, however, that this alternative does not apply to willow flycatcher emphasis habitat or small, wet, shrubby meadows outside the 5-mile radius of known willow flycatcher sites. Because willow flycatchers can re-locate distances greater than 5 miles and the level of livestock grazing that is compatible with long-term willow flycatcher productivity and population persistence in the Sierra Nevada is unknown, there is some degree of risk in allowing livestock grazing in potential willow flycatcher habitat beyond the 5-mile buffers. Although Alternative 2 provides the highest likelihood of eliminating the direct effects of grazing in applicable willow flycatcher sites, indirect effects of livestock grazing (such as livestock and brown-headed cowbird association) are not addressed under this alternative as livestock could graze adjacent clearcuts, plantations, or other nearby, non- applicable meadow types during the breeding season. Under alternative 2, livestock grazing would be eliminated year-round in 100 percent of known willow flycatcher sites including 25,717 acres (46 percent) of potential willow flycatcher habitat that are currently inside active grazing allotments.

Table 4.4.2.3d. Acres of Shrubby Wet Meadows Containing Known or Potential Willow Flycatcher Habitat Inside and Outside 100-foot Stream Buffers in active, inactive, non- allotments, and outside allotments on National Forest lands in Sierra Nevada Forest Plan Amendment Project Planning Area 100-foot Stream Buffers Habitat in Habitat in Habitat in Habitat Total Active Inactive Non- Outside acres Allotments Allotments Allotments Allotments Inside Stream Buffers Known sites 2,049 35 71 1,077 3,232 Emphasis Habitat 9,967 1,587 28 4,013 15,595 Small wet, shrubby meadows 3,949 1,109 36 2,108 7,202 Outside Stream Buffers Known sites 4,630 66 120 2,601 7,417 Emphasis Habitat 28,656 4,541 40 6,113 39,350 Small wet, shrubby meadows 6,843 2,447 92 3,503 12,885 Total Shrubby Wet Meadow Habitat 56,094 9,785 387 19,415 85,680

Alternative 3 prevents livestock grazing in a 100-foot wide buffer on each side of streams in meadows deemed “suitable” for occupancy by willow flycatchers with the use of fencing and other methods (B48). Under this alternative, “suitable” willow flycatcher habitat is defined as all known willow flycatcher sites, and potential willow flycatcher habitat (emphasis habitat and small wet meadows with a riparian deciduous shrub component). The definition of “stream” (for example oxbows, braided and secondary channels) has not yet been regionally established and thus the interpretation and implementation of this alternative could vary by national forest unit. Once this definition is resolved it is unlikely that willow flycatcher territories will occur

FEIS Volume 3, Chapter 3, part 4.4, page 169 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

entirely within this narrow buffer zone (Bombay 1999, H. Bombay pers. comm.) and therefore the utility of this buffer in eliminating risks to willow flycatchers and their habitat from the direct effects of grazing is highly uncertain. For example, based upon willow flycatcher field data and using “stream” to indicate any channel appearing on USGS topographic maps, only 3 percent of 66 willow flycatcher territories were entirely within 100 feet of streams. Another tally using the same definition of stream estimates approximately 25 percent of willow flycatcher nests were within 100 feet of streams (H. Bombay pers. comm.). For willow flycatchers outside the Sierra Nevada, others report nests or territories at varying locations, both inside and outside the proposed stream channel buffer (for example Taylor and Littlefield 1986 in Oregon, Sedgwick and Knopf 1992 in Colorado). Furthermore, depending upon the season of use outside the buffer zone, the buffer would not alleviate the indirect effects of cowbird brood parasitism associated with cattle or possible effects on meadow hydrology, herbaceous layer, and predator or prey composition. Due to the uncertainty of risks to willow flycatcher territories with this buffer zone (see also assumptions and limitations regarding fencing, herding, season of use, and utilization), this alternative has the potential for adverse direct and indirect livestock grazing effects to occur in willow flycatcher habitat. Compared with Alternative 2, livestock grazing under Alternative 3 would be eliminated year-round in a smaller percentage (31 percent) of known willow flycatcher sites as well as potential willow flycatcher habitat (15,965 acres or 28 percent) that are currently inside active grazing allotments.

Most willow flycatcher nests are located in the lower branches of willow or other shrubs that are within reach of livestock (Flett and Sanders 1987, Valentine et al. 1988, Allen-Diaz et al. 1999, Bombay 1999, Western Foundation of Vertebrate Zoology unpubl. nest records). Although not directly witnessed, six instances of nests disturbed by livestock have been reported (King 1955, Valentine et al. 1988). A recent compilation of multiple years of Sierra- wide willow flycatcher nesting data reveals that willow flycatchers fledge young between approximately July 15 and August 31 and fledglings remain in territories for 2 to 3 weeks post- fledging (158 nests; Stafford and Valentine 1985, Sanders and Flett 1989, H. Bombay and M. Morrison unpublished data). Prior to the compilation of these nesting data, and based on an earlier recommendation by Valentine (1987), Valentine et al. (1988) and Harris et al. (1987, 1988), some Sierra Nevada meadows were grazed using a limited operating period (LOP) intended to correspond with the end of the willow flycatcher nesting period and set to end annually on August 15 (in other words, to eliminate risk of direct disturbance to nest sites). The more recent analysis incorporates all available willow flycatcher nesting data for the Sierra Nevada and indicates that the willow flycatcher nesting period throughout the Sierra extends from June 1 to August 31.

FEIS Volume 3, Chapter 3, part 4.4, page 170 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Figure 4.4.2.3a. Cumulative percent of successful and unsuccessful Willow Flycatcher nests fledging by date in the Sierra Nevada: 1983, 1984, 1986, 1987, 1997, 1998, and 1999*

120

110

100

90

80

70

60

50

40

30

20 Unsuccessful nests 10 0 Successful nests Cumulative percent of nests 8/2 8/5 8/8 7/15 7/18 7/21 7/24 7/27 7/30 8/11 8/14 8/17 8/20 8/23 8/26 8/29

Fledging dates

* Fledging dates were extrapolated for unsuccessful nests. Nests with unknown outcomes were not included in this figure. Data generated from Sanders and Flett 1989, Stafford and Valentine 1985, and M. Morrison and H. Bombay unpublished data.

This revised LOP date protects approximately 10 percent of all nests that on average fledge after August 15th and also may allow for extreme years when nesting is delayed. In addition, these analyses reveal that the latest annual willow flycatcher fledging date cannot be known with certainty because the length of the nesting season is influenced by willow flycatcher arrival dates, snowpack, summer weather, nest predation, and brown-headed cowbird brood parasitism. Weather, predation, and brood parasitism can result in multiple re-nesting attempts. As many as three nesting attempts in one breeding season have been documented for willow flycatcher territories in the Sierra Nevada (Morrison et al. 1999).

In addition to protecting against direct livestock impacts on nests, the LOP has implications for indirect impacts. The removal of herbaceous cover by livestock may alter nest concealment, movement patterns of predators, and insect prey abundance or composition (Serena 1982, Erman 1984). Furthermore, high mortality during the period shortly after fledging has been reported in passerine birds (Ricklefs 1983, 1993); it is unknown how disturbance may affect fledgling survival. Therefore, to minimize the risk of livestock upsetting nests, and the potential risks associated with vegetation alteration, for Alternatives 4, 5, 6, 7 and Modified Alternative 8, the LOP corresponds with the period of June 1 through August 31, unless regionally reviewed local nest monitoring data collected over multiple years and under different climate conditions can support different dates. It should be noted that although this LOP addresses many issues associated with grazing, potential late season grazing effects are still a concern (see Assumptions and Limitations regarding season of use).

FEIS Volume 3, Chapter 3, part 4.4, page 171 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Alternatives 4 and 7 keep livestock and packstock from coming into physical contact with willow flycatcher nesting areas within known willow flycatcher sites during the breeding season (June 1 – August 31) (B37). In these alternatives, nesting areas are defined as a buffer extending from the streambank to 100 feet beyond riparian shrubs, or to the upland edge if meadow vegetation does not extend 100 feet beyond shrubs. Alternatives 4 and 7 allow for local management flexibility. The impacts of these alternatives could vary by national forest unit, depending upon the effectiveness of the local unit's management of known willow flycatcher sites (assumptions and limitations regarding Standards and Guidelines implementation, herding, and fencing should be reviewed). These alternatives do not apply to other potential willow flycatcher habitat that might facilitate willow flycatcher population expansion or provide replacement habitat. The 100-feet buffer beyond riparian shrubs is within the range of observed foraging distances of willow flycatchers during the breeding season in the central Sierra Nevada. The average foraging distance from territories was approximately 62 feet (n = 30, SD = 28 feet); however, when feeding nestlings and fledglings, willow flycatchers sometimes forage as far as 330 feet or farther from their territory (Sanders and Flett 1989, H. Bombay pers. comm.). There may be some degree of risk in allowing livestock grazing in the remaining meadow during and after the willow flycatcher breeding season (such as possible short- or longer-term effects on meadow hydrology, herbaceous layer, predator or prey composition and abundance, willow flycatcher productivity). Furthermore, these alternatives do not eliminate potential late season grazing effects or streambank trampling within the buffer after the breeding season in known willow flycatcher sites. Late-season grazing poses increased risk of willows being browsed, potentially retarding regeneration and recruitment of young shrubs as well as negatively affecting the form and structure of mature willows. General utilization standards and guidelines are intended to prevent this from occurring, although the risk of accidental overuse is greater than if these meadows were not grazed. While results of the latest research on livestock grazing and willow flycatchers suggest that willow flycatcher habitats may be restored while still allowing late-season grazing on a rest-rotation basis, the rate at which the species is recovered will be slower than if cattle are removed (Stanley and Knopf in review). For example, willow flycatcher densities approximated those on ungrazed pastures by the 10th year of the study. Alternatives 4 and 7 allow early season grazing within known willow flycatcher sites when monitoring shows that livestock avoid nesting areas. This may increase the potential for brown-headed cowbird brood parasitism as well as streambank shearing at the site, which could result in meadow desiccation; see the assumptions and limitations regarding season of use. In addition, livestock contact with willow flycatcher nesting areas is to be avoided but not prohibited under these alternatives so nest disruption remains a possibility. The s&g states that these alternatives are “most appropriate in the northern Sierra Nevada region” and thus it does not provide an adequate bioregional grazing standard and guideline for willow flycatcher habitat. In Alternatives 4 and 7, the increased level of management flexibility increases risk in known or future occupied sites. Under Alternatives 4 and 7, livestock would be kept from coming into physical contact with less than or equal to 100 percent of known willow flycatcher sites during the willow flycatcher breeding season including less than or equal to 6,704 acres (12 percent) of wet, shrubby meadows that are currently inside active grazing allotments. Note that the affected acre estimate is likely to be an overestimate for implementation purposes because the estimate is derived from grazing exclusions in entire meadows whereas Alternatives 4 and 7 avoid grazing only in habitat buffers of known willow flycatcher sites.

FEIS Volume 3, Chapter 3, part 4.4, page 172 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Alternatives 5 and 6 prevent livestock use in a habitat buffer inside known willow flycatcher sites during the breeding season (B35). In these alternatives, the buffer extends from the streambank to 100 feet beyond riparian shrubs, or to the upland edge if meadow vegetation does not extend 100 feet beyond shrubs. This standard & guideline does not apply to the stream channel outside the buffer in known willow flycatcher sites nor does it apply to any willow flycatcher emphasis habitat or small, wet, shrubby meadows that might provide suitable habitat for willow flycatcher population expansion. Note however that Alternative 5 (ACS23) suspends grazing in areas with perennially saturated soils thus willow flycatcher habitat occurring in these areas would not be grazed; it is unknown whether and to what extent these areas would include known willow flycatcher sites or the habitat required by willow flycatchers. The extent of habitat that will be included in the buffer of known willow flycatcher sites is unknown. The 100-feet buffer beyond riparian shrubs is within the range of observed foraging distances of willow flycatchers during the breeding season in the central Sierra Nevada. The average foraging distance from territories was approximately 62 feet (n=30, SD=28 feet); however, when feeding nestlings and fledglings, willow flycatchers sometimes forage as far as 330 feet or farther from their territory (Sanders and Flett 1989, H. Bombay pers. comm.). Thus, it appears that there may be some degree of risk in allowing livestock grazing in the remaining meadow during or after the willow flycatcher breeding season (such as possible short- or longer-term effects on meadow hydrology, shrub herbaceous layer, predator or prey composition and abundance, willow flycatcher productivity). Alternatives 5 and 6 may reduce the direct effects of grazing in willow flycatcher habitat buffers in known sites during the willow flycatcher breeding season; however, these alternatives would not alleviate the indirect effects of cowbird brood parasitism associated with livestock as livestock could graze adjacent clearcuts, plantations or all other meadow types as well as grassy areas immediately outside of the habitat buffer in known willow flycatcher sites during the breeding season with possible effects on meadow hydrology, herbaceous layer, and predator or prey composition and abundance. Furthermore, these alternatives do not address potential late season grazing effects or streambank trampling within the buffer after the breeding season in known willow flycatcher sites. While results of the latest research on livestock grazing and willow flycatchers suggest that willow flycatcher habitats may be restored while still allowing late-season grazing on a rest-rotation basis, the rate at which the species is recovered will be slower than if cattle are removed (Stanley and Knopf in review). For example, willow flycatcher densities approximated those on ungrazed pastures by the 10th year of the study. Fencing, herding, utilization, and season of use assumptions and limitations all apply to these alternatives. Under Alternatives 5 and 6, livestock grazing would be restricted from less than or equal to 100 percent of known willow flycatcher sites during the willow flycatcher breeding season including less than or equal to 6,704 acres (12 percent) of wet, shrubby meadows that are currently inside active grazing allotments. Note that the affected acre estimate is likely to be an overestimate for implementation purposes because the estimate is derived from grazing exclusions in entire meadows whereas s&g B35 restricts grazing only in habitat buffers of known willow flycatcher sites. In addition, for Alternative 5, grazing restrictions in areas with perennially saturated soils in other potential willow flycatcher habitat could not be estimated, because the location of perennially saturated areas that provide willow flycatcher nesting habitat is unknown, but could potentially increase the number of affected acres inside active grazing allotments.

In Alternative 8, livestock grazing would be eliminated year-round in a habitat buffer in known willow flycatcher sites and willow flycatcher emphasis habitat within willow flycatcher

FEIS Volume 3, Chapter 3, part 4.4, page 173 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

colonization distance (5 miles) of known sites (B35B). In this alternative the habitat buffer extends from the streambank to 100 feet beyond riparian shrubs, or to the upland edge if meadow vegetation does not extend 100 feet beyond shrubs. The 100-feet buffer beyond riparian shrubs is within the range of observed foraging distances of willow flycatchers during the breeding season in the central Sierra Nevada. The average foraging distance from territories was approximately 62 feet (n=30, SD=28 feet); however, when feeding nestlings and fledglings, willow flycatchers sometimes forage as far as 330 feet or farther from their territory (Sanders and Flett 1989, H. Bombay pers. comm.). There may be some degree of risk in allowing livestock grazing in the remaining meadow during and after the willow flycatcher breeding season (such as possible short- or longer-term effects on meadow hydrology, herbaceous layer, predator or prey composition and abundance, willow flycatcher productivity). Regarding willow flycatcher habitat, this alternative does not apply to the stream channel outside the buffer in applicable meadows, willow flycatcher emphasis habitat outside the 5-mile radius of known willow flycatcher sites, nor does it apply to any small, wet, shrubby meadows that might provide willow flycatcher habitat. Because willow flycatchers can re-locate distances greater than 5 miles and the level of livestock grazing that is compatible with long-term willow flycatcher productivity and population persistence in the Sierra Nevada is unknown, there is some degree of risk in allowing livestock grazing in potential willow flycatcher habitat beyond the 5-mile buffers. Although Alternative 8 may eliminate or dramatically reduce the direct effects of grazing in willow flycatcher habitat buffers in designated willow flycatcher sites, this alternative does not address unknown risk of livestock grazing outside the habitat buffer (such as possible effects on meadow hydrology, herbaceous layer, predator or prey composition and abundance, willow flycatcher productivity). Furthermore, indirect effects of livestock grazing (such as livestock and brown-headed cowbird association) may occur as livestock could graze adjacent clearcuts, plantations or other nearby non-applicable meadow types as well as herbaceous areas immediately outside of the habitat buffer of designated willow flycatcher sites during the breeding season with possible effects on meadow hydrology, herbaceous layer, and predator or prey composition and abundance. For these reasons, the long-term suitability of known willow flycatcher sites and applicable emphasis habitat meadows is more uncertain under this alternative, compared to Alternative 2. See assumptions and limitations regarding utilization, herding, and fencing. (The extent of habitat that will be included in the buffer of applicable sites is unknown.) Under Alternative 8, livestock grazing would be eliminated year-round in less than or equal to 100 percent of known willow flycatcher sites including less than or equal to 20,454 acres (36 percent) of wet, shrubby meadow habitat that are currently inside active grazing allotments. Note that the affected acre estimate is likely to be an overestimate for implementation purposes because the estimate is derived from grazing exclusions in entire meadows whereas Alternative 8 eliminates grazing only in habitat buffers of all known willow flycatcher sites and emphasis habitat within 5 miles of known sites.

There are 82 known willow flycatcher sites on Forest Service lands with an estimated population size of 120-150 breeding and non-breeding individuals (Ritter and Roche 1999). Modified Alternative 8 will eliminate direct and most of the indirect effects of livestock grazing in known willow flycatcher sites that are determined to be occupied (see s&g wifl-1). Under existing conditions it appears that willow flycatchers are likely still experiencing a declining population trend in the Sierra Nevada bioregion, given the small population size, apparent recent vacancies of meadows occupied in the 1980’s, and evidence for moderate to low nest success rate and low fecundity rate. This is further supported by preliminary

FEIS Volume 3, Chapter 3, part 4.4, page 174 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 calculations of Lambda for willow flycatchers in the central Sierra that are less than 1.0 (1999 = 0.876, 2000 = 0.777), indicating a non-sustainable replacement rate and decreasing population in 1999 and 2000 (Morrison et al. 2000). Researchers of late-season rest-rotation grazing effects on willow flycatchers suggest “a more severe restriction on grazing may be preferred in areas where conservation priorities emphasize maximum recovery of the species” (Stanley and Knopf in review). To determine site occupancy in known sites and emphasis habitat meadows in active grazing allotments within 5 miles of known willow flycatcher sites, willow flycatcher surveys will be conducted using the Pacific Southwest Region protocol (wifl- 1 and wifl-4). With 82 known sites and an estimated 361 emphasis habitat meadows in active grazing allotments within 5 miles of known willow flycatcher sites, s&gs wifl-1 and wifl-4 in Modified Alternative 8 will require that approximately 200 meadows are surveyed across 11 national forests in the planning area in the first survey year and at least 120 other meadows will be surveyed Sierra Nevada-wide in each of the next 2 survey years. Willow flycatcher surveys consist of 2 site visits between June 1 and July15 (Bombay et al. 2000), however the survey interval is compressed to about 1 month at higher elevations (H. Bombay pers. comm.). Repeated surveys, to detect changes in willow flycatcher site occupancy, would be required up to 2 years every 2 years in known willow flycatcher sites (wifl-1) and 1 year every 3 years in emphasis habitat in active range allotments within 5 miles of known willow flycatcher sites (wifl-4). Note that in emphasis habitat where willow flycatchers are detected, survey requirements would switch to the wifl-1 standard. Surveys must be completed within specified timeframes or management will default to livestock exclusion in known sites (wifl-1) and late- season grazing in emphasis habitat in active range allotments within 5 miles of known willow flycatcher sites (wifl-4). Note that known sites and emphasis habitat meadows in active allotments within 5 miles of known sites that lack standing water on June 1 and a deciduous shrub component will be evaluated as to whether willow flycatcher s&gs should apply (wifl-5).

Under Modified Alternative 8, late-season grazing (following the willow flycatcher breeding season) would be allowed in: 1) known willow flycatcher sites where willow flycatchers are not detected after 2-consecutive-survey years (wifl-1 and wifl-3), and 2) emphasis habitat in active range allotments within 5 miles of known willow flycatcher sites where willow flycatchers are detected (wifl-4 and wifl-3). Stanley and Knopf (In review) suggest that habitats for grazing-sensitive birds such as the willow flycatcher may be restored while still allowing late-season grazing, but the rate at which species recover will be slower than if all cattle are removed. The implication is that late-season grazing is compatible with healthy riparian vegetation (such as late seral condition meadows) which is currently providing good nesting habitats for birds (F. Knopf pers. comm.). “Good nesting habitat” for willow flycatchers is still undetermined. In addition, other studies report that late-season grazing may be detrimental to willows (Kauffman et al. 1983b, Sedgwick and Knopf 1991). S&g FW-G99 in Modified Alternative 8 should ensure that a shift to shrub browsing in excess of 20 percent utilization, which approximates utilization by deer (Loft et al. 1987), does not occur (see discussion on browsing that follows this section) although risk of accidental overuse is more likely than if these meadows were not grazed. In addition to sharing many of the similar concerns with Alternatives 4 through 7, the proportion of applicable meadows in late, mid-, and early seral condition is unknown and habitat condition in applicable meadows would be worthy of consideration before implementing Modified Alternative 8. (For example, if indeed the Stanley and Knopf findings apply to willow flycatchers in the Sierra Nevada and a large proportion of applicable meadows are in less than late-seral condition, then total rest from grazing to accelerate habitat restoration in applicable mid- or early-seral status meadows would

FEIS Volume 3, Chapter 3, part 4.4, page 175 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

incur less risk for willow flycatcher recovery.) The uncertainty and risk of Modified Alternative 8 is evident. To address this, commitments in the Record of Decision include a formal study of the effects of grazing on willow flycatchers (wifl-6) and their habitats (G04C) as well as a 5-year re-evaluation of all standards and guidelines pertaining to the willow flycatcher.

To address some of the uncertainty in the short-term, in willow flycatcher sites receiving late- season grazing under Modified Alternative 8, annual utilization and 3-year willow flycatcher habitat condition monitoring is required to detect changes in habitat condition with assessment data to be included in a GIS meadow coverage. The Rangeland Analysis and Planning Guide (R5-EM-TP-004) describes annual utilization monitoring and see Appendix U for a description of willow flycatcher habitat condition monitoring techniques. If habitat conditions are not supporting the willow flycatcher or are trending downwards, then grazing will be suspended or modified (wifl-2). S&g wifl-2 ensures that habitat conditions in known but “unoccupied” as well as occupied emphasis habitat meadows will be maintained or improved to support willow flycatchers but these assurances do not extend to colonizable willow flycatcher habitat, because these sites lack willow flycatcher habitat condition monitoring.

Under Modified Alternative 8, season-long grazing according to the general grazing s&g G04B would be allowed in emphasis habitat in active grazing allotments within 5 miles of known sites where willow flycatchers are not detected. Beyond the 5-mile survey buffer, only a subset of emphasis habitat meadows in active allotments will be surveyed under the terms and conditions of the ROD. Thus, undetected willow flycatchers in any sites that receive annual season-long grazing will experience direct and indirect effects of livestock grazing and be placed at greater risk. Under Modified Alternative 8, creation of willow flycatcher habitat expansion capability or replacement habitat is uncertain. Available research on late-season rest-rotation grazing suggests that willow flycatchers will recover at a slower rate where grazing is allowed (Stanley and Knopf In review). In addition, none of the grazing standards and guidelines has been proven to support long-term population viability of the willow flycatcher in the Sierra Nevada or elsewhere. Furthermore, there is no known research indicating that season-long grazing will support willow flycatcher population expansion over a viable timeframe, adding further uncertainty and risk to the approach of Modified Alternative 8 (s&g wifl-4) for recovering the willow flycatcher population on national forest lands in the Sierra Nevada. A formal management study that includes year-round exclusion, late-season rest-rotation grazing, annual late-season grazing, and annual season-long grazing at utilization levels corresponding to habitat condition is essential to evaluate the risks and uncertainties of livestock grazing on willow flycatchers and their habitats in the Sierra Nevada bioregion.

b. Livestock Grazing: Forage Utilization and Streambank Disturbance The Sierra Nevada Forest Plan Amendment Project FEIS alternatives vary with regard to allowable forage utilization and streambank disturbance of riparian and meadow habitats by livestock. Although several alternatives eliminate livestock grazing in willow flycatcher nesting areas during the willow flycatcher breeding season, only Alternative 2 eliminates grazing entirely from known willow flycatcher sites year-round. Thus the following livestock grazing s&gs from the action alternatives apply to:

• the portions of known willow flycatcher sites that occur outside habitat or stream buffers during the breeding season (Alternatives 3-8)

FEIS Volume 3, Chapter 3, part 4.4, page 176 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

• entire known willow flycatcher site (for example, the meadow) before or after the willow flycatcher breeding season (Alternatives 4 through 7, Modified Alternative 8) • willow flycatcher emphasis habitat as well as small, wet meadows with a woody riparian shrub component (except where grazing restrictions apply to these sites in Alternatives 2, 3, 8 and Modified Alternative 8).

For all alternatives, the assumptions and limitations regarding utilization should be reviewed. In addition, it is assumed that compliance with, or enforcement of, these s&gs will occur (see Standards and Guidelines Implementation assumptions and limitations).

Currently (Alternative 1), Forests in the planning area manage for the effects of livestock grazing in various ways (see Ritter and Roche 1999). Management is generally in the form of standards and guidelines to protect all riparian and meadow areas from activities that diminish hydrologic or vegetative function. S&Gs are generally qualitative in nature, with the exception of some livestock utilization standards and buffer zones around meadows for recreation, roads, and forestry projects. There are no bioregional forage utilization monitoring standards.

In Alternatives 2 and 8, livestock use of grass and grass-like plants in meadows is limited to 30 percent but may be adjusted upwards to 45 percent, if the range is in good to excellent condition. However, in habitats occupied by TES amphibian or fish species, grazing utilization remains at 30 percent; this will have coincidental benefit for known or potential willow flycatcher habitat that is occupied by these species. A minimum stubble height of greater than four inches must be maintained and highly degraded meadows may receive total rest for an unspecified time period (G04B).

Alternative 3 does not have any specific forage utilization standards. Alternative 3 restricts livestock use in community and energy areas of riparian and meadow habitats (WM13), and allows local managers to determine the level of allowable livestock utilization outside these riparian areas. Mitigation measures are developed in cooperation with livestock grazing permittees. Depending upon the efficacy of the mitigations and extent of willow flycatcher habitat outside the community and energy areas of riparian and meadow areas, this could result in inconsistencies in willow flycatcher habitat management and greater risk to the willow flycatcher population in the Sierra Nevada bioregion. See also concerns regarding shrub browse for Alternatives 2, 4, 6, 7, and 8, below.

In Alternatives 4, 6, and 7, livestock utilization within riparian and meadow areas continues to be determined on individual forests through allotment management planning. Regional standards are applied only where it is determined that current practices are not trending toward or maintaining desired conditions. In such situations, maximum utilization of grass and grass- like plants is 30 percent on early seral as well as highly degraded sites and 45 percent on late seral sites. Site-specific stubble heights may be established. Grazing practices should move conditions towards Aquatic Conservation Strategy (ACS) goals; but if after 3-5 years conditions are not supporting the ACS goals, then the grazing permit will be modified or grazing suspended (G04A, ACS22). The effectiveness of Alternatives 4, 6, and 7 could vary by unit, depending upon whether local managers monitor willow flycatcher habitat appropriately and consistently.

Alternative 5 permits herbaceous perennial plant utilization between 30-40 percent. A 5-7" stubble height is required in healthy areas and rest is required in unhealthy or degraded areas

FEIS Volume 3, Chapter 3, part 4.4, page 177 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

for an unspecified period of time (G07). In addition, this alternative has a number of livestock grazing restrictions relating to ACS goals and other contingencies (see ACS22, ACS23, ACS24, ACS25). For example, ACS22 would not permit livestock grazing in situations where ACS goals are not advanced, ACS23 suspends grazing in areas with perennially saturated soils, and ACS25 ensures that these standards are met. Furthermore, Alternative 5 requires environmental analysis for four grazing allotments per year on each National Forest; Forest Service line officers are held accountable if the required analyses are not completed each year (G08). Livestock grazing use must be monitored over a two year period to determine if ecological goals are being met with management adjustments as needed (G09). Alternative 5 offers the most conservative livestock grazing utilization standards; where livestock grazing is allowed, this alternative is likely to eliminate the most risk to willow flycatcher habitats from the direct effects of livestock grazing.

In Modified Alternative 8 (FW-G04B), under season-long grazing, livestock utilization of grass and grass-like plants is limited to 30 percent (or a minimum 6-inch stubble height) for meadows in early seral condition to a maximum of 40 percent (or a minimum 4-inch stubble height) for meadows in late seral condition. Ecological condition on all key areas monitored for grazing utilization will be determined prior to establishing utilization levels. By requiring maximum herbaceous utilization values of between 30 - 40 percent under season-long grazing, it is assumed that livestock will be less likely to begin browsing on willows or otherwise damage outer branches (Ratliff 1985, Loft et al. 1987, Clary and Webster 1989). Under intensive grazing systems where meadows are receiving a period of rest (such as rest-rotation or late-season), utilization levels may be higher if the meadow is maintained in late seral status and meadow-associated sensitive species are not being impacted. Degraded meadows (early seral, with greater than 10 percent bare soil and active erosion) require total rest from grazing until they have recovered and moved to mid or late seral status. Determination of ecological status is assessed according to the regional range analysis and planning guide. Every 3 to 5 years, meadow ecological status will be analyzed; if determined to be in a downward trend, grazing will be suspended or modified. Available range trend data and annual monitoring data for key areas within allotments will be included in a GIS meadow coverage, and updated annually. Though these standards may go a long way toward improving or maintaining range condition, recall that none have been proven to support the long-term persistence of the willow flycatcher population and their habitats in the Sierra Nevada.

In Alternatives 2, 4, 6, 7, 8, and Modified Alternative 8, livestock use of shrubs is limited to 20 percent of seedlings and 20 percent of annual growth of mature shrubs (AM15, FW-AM15). In willow-dominated areas of the Sierra Nevada ungrazed by livestock, Loft et al. (1987) found that deer browsed between 12 and 17 percent of annual willow shoots. Therefore if total shrub utilization does not exceed 20 percent, it is assumed that impacts to the shrub community will not differ substantially from those experienced with deer alone. Recruitment of riparian deciduous shrubs will provide more willow flycatcher nesting substrate. To ensure willow recruitment into tree or shrub form, Modified Alternative 8 clarifies that browsing on seedlings is limited to 20 percent of individual seedlings (in other words, no more than 20 percent of seedlings may have their growth tip removed) (s&g FW-G99). Furthermore, s&g FW-G99 requires removal of cattle from any area of the allotment when browsing indicates a shift in livestock preference from grazing herbaceous vegetation to browsing woody riparian vegetation; sheep must be herded away from these plants at all times. Note that under Alternative 2 s&g AM15 should not be necessary in potential habitat within 5 miles of known

FEIS Volume 3, Chapter 3, part 4.4, page 178 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 willow flycatcher sites where livestock grazing is restricted year-round from the entire meadow. Alternatives 3 and 7 have no specific utilization standards for woody riparian vegetation (WM13, G04A), so adequate regeneration and recruitment of shrubs on a regional scale is uncertain. Vagaries in guidelines will create uneven implementation and lead to greater uncertainty and potential risk to willow flycatchers and their habitats across the planning area. In addition to undefined standards, Alternative 3 does not specify when grazing utilization monitoring should occur or if practices may be adjusted. Alternative 5 limits utilization of woody species to 5 percent (G07), which is an apparently stricter utilization level than occurs where deer browse alone (Loft et al. 1987). In addition, at least 50 percent of the foliar density in the lower portions of all shrubs must be maintained and recruitment of young riparian deciduous shrubs must occur. The requirement of 50 percent foliar density is intended to prevent highlining and decadence in shrub communities and to maintain foliar cover from predators within the height class where willow flycatchers typically place their nests (Valentine et al. 1988, Sanders and Flett 1989, Bombay 1999). Under Alternative 5, conditions must be assessed throughout the grazing season and allowable utilization levels adjusted, if necessary, to meet this standard (G07A). The 20 percent browse utilization standards described above more than provide for this foliar density requirement (S. Bishop pers. comm.).

Streambank stability affects hydrologic condition of willow flycatcher habitats (see stream and watershed management assumptions and limitations). In Alternatives 2 and 3, streambanks in known willow flycatcher sites and other applicable meadows would be protected year-round from disturbance due to grazing however, under Alternative 2, the streambanks of potential habitat outside the 5 mile buffer zone would be susceptible to grazing related disturbance. Under Alternative 3, the definition of “stream” will determine the extent of streambank protected year-round in known willow flycatcher sites and other potential habitat. Applicable standards are discussed below. Streambank disturbance includes bank sloughing, chiseling, trampling, and other means of exposing bare soil or cutting plant roots. In all riparian and meadow areas, Alternative 3 requires restoration and maintenance of bank stability at or above a site-specific bank stability standard (WM21); in watersheds where the bank stability standard is not met, any activities that adversely affect stability or preclude recovery will be eliminated until monitoring documents a statistically significant improving trend over at least 5 years (WM22). In Alternative 5, streambank trampling and bare soil resulting from livestock grazing are not to exceed 5 percent in riparian areas (G07); furthermore, special use permits for any activity should be modified to the extent allowed and as soon as possible to prevent degradation in riparian zones (ACS08). Under Alternative 5, stricter streambank protections and grazing utilization standards for all meadows increase the likelihood that successful willow flycatcher colonization could occur at any distance from known sites. In general, improved habitat translates into higher likelihood of willow flycatcher population expansion or persistence. Under Alternative 8, streambank disturbance is limited to 10 percent in Critical Aquatic Refuges (CARs) (G04C). Outside CARs, Alternative 8 limits streambank trampling or chiseling from any activity in meadows to 20 percent of the stream reach (AM16B). Alternatives 2, 4, 6, 7, and 8 limit disturbance of colonizing native plants from any activity in riparian areas and meadows to 20 percent of a stream reach but exempt developed recreation sites and non-native undesirable plants and noxious weeds (AM16, AM16A). Although it also includes natural lake and pond shorelines, the 20 percent streambank disturbance standard in Modified Alternative 8 (RCA18) represents a weaker standard than AM16, AM16A and AM16B because it pertains to disturbance caused by resource activities, as opposed to total disturbance. Moreover, developed recreation and designated OHV routes are exempted in

FEIS Volume 3, Chapter 3, part 4.4, page 179 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

RCA18 but not in AM16B. Also, outside of stream reaches that are occupied by aquatic- dependent TES species, there is no general streambank disturbance standard in Modified Alternative 8 akin to AM16 and AM16A. Finally, though the 20 percent streambank alteration standard is employed generally throughout the western United States, there do not appear to be any studies that directly support this standard (S. Bishop pers. comm.). Note that the streambank disturbance standards for Alternatives 2, 4, 6, 7, and 8 allow more disturbance than the 12 percent stream disturbance found in reference sites of stream condition inventory data collected throughout the Sierra Nevada and the 10 percent maximum streambank disturbance typically allowed by USFWS in threatened and endangered fish species habitat. Although Modified Alternative 8 does limit streambank disturbance from livestock to 10 percent of the occupied stream reach for threatened and endangered fish species, this is in addition to natural disturbance and does not include other sources of disturbance caused by resource activities. To address present uncertainty regarding this standard, Modified Alternative 8 indicates that streambank disturbance standards will be developed for aquatic-dependent threatened, endangered, and sensitive species in cooperation with State and other Federal agencies (RCA18). Modified Alternative 8 requires use of the regional streambank assessment protocol in stream reaches with aquatic-dependent TES species and implements corrective actions where streambank disturbance limits have been exceeded (RCA18), otherwise, none of the other alternatives specifies streambank stability monitoring to ensure that standards are met. Alternative 3 implements monitoring if streambank conditions are not met (see s&g WM22). Alternatives 3 and 5 offer more stringent s&gs regarding general streambank disturbance but weaker standards specific to willow flycatcher habitat compared to Alternatives 2 and 8.

2) Requirement to implement a monitoring program that includes annual surveys of willow flycatcher breeding success and habitat conditions

Despite the fact that there are many uncertainties regarding willow flycatcher distribution, abundance, and demographic data in the Sierra Nevada bioregion (see willow flycatcher population assumptions and limitations), current monitoring of willow flycatcher populations and habitat conditions on most Forests in the planning area has been limited to project level surveys and incidental nest monitoring as funding permits (Alternative 1). It is estimated that approximately 60-75 percent of present willow flycatcher locations in the bioregion are known. There is not an annual, bioregional willow flycatcher monitoring program in place although survey efforts are important to locate recently (re-)colonized willow flycatcher sites as well as those willow flycatcher locations that have yet to be detected (T. Ritter pers. comm.). For the past four years, Forest Service Pacific Southwest Region has funded a willow flycatcher demography study in the north-central Sierra Nevada; data from this study have already proven useful in formulating management strategies for this species.

Using a consistent regional protocol, Alternatives 2 through 8 require development of a meadow map for all national forests in the planning area (B46 and B46A), which will aid in identification of meadows to be surveyed for willow flycatchers (see however assumptions and limitations for Standards and Guidelines implementation). The existing GIS meadow coverages were developed independently by the 11 national forests in the planning area and merged into a Sierra Nevada Forest Plan Amendment Project planning area meadow layer. Each of the forests developed their own approach for identifying meadows within national forest lands. Thus, the merged GIS meadow coverage for the planning area is the result of multiple approaches with varied accuracy.

FEIS Volume 3, Chapter 3, part 4.4, page 180 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Alternatives 2 through 8 also require annual surveys of willow flycatcher occupancy and habitat conditions in a subset of approximately 1,500 willow flycatcher emphasis habitat meadows as well as known sites without recent willow flycatcher detections where potential habitat exists on national forest lands. A sampling design will be derived to determine willow flycatcher site occupancy and turnover rates at selected, not all, sites (B46 & B46A). In addition to this requirement, Alternatives 4 and 8 have an accelerated monitoring program in which all 912 willow flycatcher emphasis habitat meadows in active grazing allotments must be surveyed within three years to determine willow flycatcher presence and habitat conditions (B46A). This survey requirement increases the likelihood that currently unknown willow flycatcher sites will be located and appropriately managed. Note that existing meadow mapping methods are coarse for a number of forests in the planning area and these meadow estimates require further refinements. None of the s&gs requires surveys in small meadows, or riparian areas outside meadows, where willow flycatchers could potentially occur. Assessment of habitat condition at the sites surveyed will provide valuable information about the relationship between livestock grazing, habitat condition, and willow flycatcher site occupancy. Any causal effects of livestock grazing on long-term willow flycatcher population persistence will still need to be evaluated, however. Furthermore, surveying known willow flycatcher sites where willow flycatchers have not recently been detected will prevent oversights and allow managers to respond to and facilitate willow flycatcher persistence in re-colonized locations by implementing appropriate management actions (for example livestock grazing, roads, pesticide use, and recreation restrictions).

Alternatives 2 through 8 continue the willow flycatcher demography study and measurement of habitat parameters in the Sierra Nevada to allow the Forest Service to determine willow flycatcher population status and trend as well as willow flycatcher habitat conditions (B47). The population that is being studied is considered a stronghold for the species in the Sierra Nevada bioregion (USDA 1998). Continuation of this study is critical to the long-term conservation of the willow flycatcher as the data collected will enable managers to make informed decisions regarding the future of this species by 1) determining long term population trends, and 2) assisting managers with assessing the effectiveness of standards and guidelines for the willow flycatcher. Furthermore, long-term demographic monitoring will provide measures of willow flycatcher habitat quality at the meadow scale. These data will be obtained from sites producing the most young as well as those with the most breeding pairs over time, two of the primary indicators of habitat quality (Pulliam 1988, Vickery et al. 1992a, 1992b, Hall et al. 1997).

Modified Alternative 8 requires the development of a willow flycatcher conservation assessment within 1 year of the completion of the ROD (RCA-02). In cooperation with other State and Federal agencies, universities, and research scientists, the conservation assessment will synthesize the best available information on life history, habitat associations, and risk factors, and also identify occupied and suitable habitat essential for the conservation of the willow flycatcher. The requirement to develop a conservation assessment for willow flycatchers should focus conservation efforts on meadows where habitat restoration and management may be most effective. In addition, requirements to map potential habitat, identify suitable habitat, and monitor willow flycatcher occurrence, distribution, relative abundance, demographic trend, and habitat criteria have shifted from s&gs (B46A1, B47) to commitments in the Record of Decision.

FEIS Volume 3, Chapter 3, part 4.4, page 181 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

3) Direction to restore degraded areas to desired conditions for willow flycatcher nesting habitat in order to increase opportunities for willow flycatcher population expansion in the Sierra Nevada bioregion.

Currently (Alternative 1), there is no Forest Service Pacific Southwest Region strategy for willow flycatcher habitat restoration in the Sierra Nevada bioregion. As a result, riparian and meadow restoration projects are often implemented without knowledge of whether the project will contribute toward the goal of willow flycatcher population expansion in the bioregion. Because funds for restoration work are limited, it is important to develop a restoration strategy that increases opportunities for willow flycatcher population expansion to occur.

Alternatives 2 through 8 provide a s&g for prioritizing meadow restoration at willow flycatcher emphasis habitat meadows as well as sites previously occupied by willow flycatchers (B45). Modified Alternative 8 not only requires that meadows near or adjacent to willow flycatcher sites receive restoration priority but that the prioritization process will occur over the next 5 years during landscape analysis (B45). It is anticipated that this will aid in developing a cohesive and effective strategy for willow flycatcher site restoration in the Sierra Nevada bioregion, depending on available funding. Note, however, that none of the alternatives prioritizes restoration in known or occupied willow flycatcher sites nor are these s&gs mandates for willow flycatcher habitat restoration. On the other hand, Modified Alternative 8 does provide direction to restore the riparian vegetation community, which could address conifer encroachment where it is identified as a problem (RCA19).

A network of meadows, referred to as Important Bird Areas (IBAs), proposed in Alternatives 2, 6, and 8 (B36, B36A), would be identified and managed for the purpose of enhancing habitat for meadow-associated bird species. IBAs would have the potential to benefit willow flycatcher populations as well as other meadow-associated species. The IBAs would be distributed on national forest lands along the north-south and east-west migration axes of the Sierra Nevada bioregion with their locations focusing on important habitat for multiple species of meadow-associated birds. Meadows in the network would also be selected based upon high concentrations of dispersing juvenile birds, or other important migration, molting or staging areas, in addition to proximity to other IBA meadows. This strategy could benefit willow flycatcher populations in the bioregion, provided that there is elevational overlap between IBAs and willow flycatcher range and also that management of the IBAs is conducive to willow flycatcher habitat conditions. Alternatives 2 and 8 specify that IBAs be regionally selected and established by 2005 and include site-specific management and monitoring to meet or maintain the IBA meadow objective (B36). Furthermore, the regional selection process of B36 helps ensure IBA meadow selection consistency and a bioregional design increases the likelihood of benefiting dispersing and migrating birds. Alternative 6 lacks these provisions (B36A). Alternatives 3, 4, 5, and 7 do not include establishment of IBAs.

Candidate meadow IBAs have been provisionally selected and analyzed with the knowledge that locations could change depending upon final IBA meadow selection. Candidate meadow IBAs were selected using locational records of meadow- and riparian-associated bird species in the Sierra Nevada from the California Natural Diversity Database (CDFG 2000) as well as other spatially- explicit meadow bird survey data (R. Siegel, Institute for Bird Population unpubl. data) or other meadow species data (D. Lipton pers. comm., S. Anderson pers. comm.). Meadows with records of Federal or State TES, NOI, or high vulnerability bird species had the highest priority for selection. In addition, due to the paucity of available spatially-explicit data for riparian- and meadow-associated bird species throughout the Sierra Nevada, candidate meadow IBAs were also selected based upon

FEIS Volume 3, Chapter 3, part 4.4, page 182 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 their size or proximity to other meadows or lands with special management (such as known willow flycatcher sites, other candidate meadow IBAs, or parks). The 160 IBA candidate meadows overlap in only one known case with a willow flycatcher record and this record is unconfirmed. Depending upon the final IBA meadow selection and site management, however, IBA meadows could reduce risk to migratory stopover site habitat for willow flycatchers. S&gs B36 and B36A apply to all meadow types (wet with woody vegetation, wet without woody vegetation, dry with woody vegetation, dry without woody vegetation, and emergent vegetation) and are not restricted to wet meadow types with woody vegetation as is potential willow flycatcher habitat (emphasis habitat and small wet, shrubby meadows). A total of 29,178 acres (18 percent of the known underestimate of 161,830 meadow acres on national forest lands) are included in the IBA candidate meadow analyses (160 IBAs, 10-15 IBAs per forest, with 22,019 acres inside active allotments, 805 acres in inactive allotments, 319 acres in non-allotments, 6,036 acres outside allotments).

Modified Alternative 8 does not specifically address Important Bird Area meadows but the Record of Decision (ROD) will include the provision that important bird area meadows be reviewed among existing uses in Landscape Analysis. Landscape analysis would not provide the full benefits of a regional selection process, however (see discussion above).

4) Management actions to lessen the influence of brown-headed cowbird brood parasitism on the willow flycatcher population (includes livestock and recreation facilities)

Currently (Alternative 1) there is no Forest Service Pacific Southwest Region direction regarding the management of brown-headed cowbird populations. The impact of brown-headed cowbirds varies within the Sierra Nevada bioregion. Brown-headed cowbirds are currently known to impact willow flycatcher populations outside the planning area (Sedgwick and Iko 1999, Whitfield 1990, Whitfield and Enos 1996, Whitfield and Sogge 1999). Although brown-headed cowbirds have impacted less than 7 percent of observed willow flycatcher nests in the Sierra Nevada between 1997-2000, their influence could become greater if willow flycatcher populations decrease, brown-headed cowbird populations increase, or both occur (Whitfield and Sogge 1999, Smith 1999, Morrison et al. 2000, H. Bombay pers. comm.). Given that mountain communities are expanding in many areas, and brown- headed cowbirds are highly associated with human disturbance, brown-headed cowbirds are likely to increase in at least some portions of the bioregion (Verner and Ritter 1983). Because the willow flycatcher population in the Sierra Nevada bioregion is very small, it is possible that even low parasitism rates may affect population trend, especially over the long term (Rothstein 1994). In the Lake Tahoe Basin in 1998 through 2000, high cowbird abundance translated into 8 of 18 nests (44 percent) being parasitized (Morrison et al. 2000). Smith (1999) in a review of recent cowbird studies suggests that management actions to control cowbirds may not be warranted unless the parasitism rate is at least 60 percent, however he lists criteria that might suggest using a lower rate, including: restricted habitat, isolated population, and population in prolonged decline. This indicates that the few remaining breeding locations within the Tahoe Basin may benefit from cowbird management if the current parasitism rate remains consistent or increases (Whitfield and Sogge 1999, Whitfield et al. 1999). Nonetheless, high brown-headed cowbird density and high private land ownership in the area could make control difficult and limit its effectiveness (Citta and Mills 1999, Hall and Rothstein 1999, Smith 1999, Whitfield and Sogge 1999, H. Bombay pers. comm.). Brown-headed cowbird trapping programs and livestock facilities removal or relocations will need to be evaluated based on risk levels and effectiveness, prior to implementation (Verner and Rothstein 1988, Smith 1999, Whitfield and Sogge 1999, Whitfield et al. 1999).

FEIS Volume 3, Chapter 3, part 4.4, page 183 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

In 13 documented cases of brown-headed cowbird brood parasitism of willow flycatcher nests for which dates are known in the central Sierra Nevada, parasitism events occurred from approximately June 17 to August 4 (mean = July 4, SD = 12 days) (Sanders and Flett 1989, H. Bombay pers comm.). These parasitism dates correspond to willow flycatcher incubation initiation dates between June 15 and August 1 (Stafford and Valentine 1985, Sanders and Flett 1989, H. Bombay pers comm.). Regional information on cowbird egg laying dates and willow flycatcher incubation initiation dates will need to be compiled as some regions and elevations of the Sierra Nevada may have different dates. In the Dinkey Creek area of Sierra National Forest, Verner and Ritter (1983) found that cowbirds seldom frequented pack stations prior to the arrival of pack animals. Thus, delaying access to livestock and pack stock facilities in relation to estimated dates of brood parasitism might eliminate or alleviate this threat in some areas of the Sierra Nevada.

For all alternatives, it is important to review brown-headed cowbird assumptions and limitations regarding variable effectiveness of brown-headed cowbird management on Forest Service land.

Alternative 2 provides direction to reduce the likelihood of willow flycatcher brood parasitism by brown-headed cowbirds via removal of pack stations and corrals within 5 miles of known willow flycatcher sites and, where necessary, implement a California Department of Fish and Game- approved brown-headed cowbird control program (B41). Recall that numbers of brown-headed cowbirds were negatively correlated with distance to pack stations or corrals (P < .01) in the southern Sierra Nevada (Verner and Ritter 1983). Relocation would apply to an estimated 89 corrals and pack stations in Sierra Nevada national forests, including several under special use authorization. However, without benefit of bioregional assessment of the effectiveness of these mitigations, Alternative 2 automatically imposes the most punitive measures (facilities removals, cowbird trapping) although their efficacy may be limited or, in other instances, less severe approaches, such as delayed use of facilities, may be equally successful. In other words, relocation or closure of livestock facilities may only need to be considered where these facilities are the primary source of brown- headed cowbirds and other mitigations are unsuccessful. Note that the 5-mile radius is within the range of common foraging distances of brown-headed cowbirds (Rothstein et al. 1984, Verner and Rothstein 1988, Gaines 1992, Coker and Capen 1995, S. Rothstein pers. comm.); this 5-mile distance is intended to help alleviate the cowbird threat to willow flycatchers through elimination of cowbird- stock associations within 5 miles of known willow flycatcher sites. The effectiveness of this alternative will vary from site to site because of large inholdings or adjacent lands with uses that are conducive to cowbirds within the 5-mile zone of known willow flycatcher sites on national forest lands. Though Alternative 2 addresses pack stations and corrals, there is no similar direction for other stock concentrating facilities such as livestock handling and management facilities.

Alternative 3 provides no direction for recreation facilities or cowbird control but it requires that new and existing livestock handling and management facilities be located or moved outside of the community and energy areas of riparian buffers. This would affect an estimated 127 corrals and pack stations. This alternative may not be effective in reducing risk to willow flycatchers from cowbird brood parasitism unless proposed and existing recreation facilities as well as livestock management and handling facilities are located at least 5 miles (foraging distance for brown-headed cowbirds) from known willow flycatcher sites, specifically, or riparian areas in general (WM12) or their timing of use is evaluated in relation to brown-headed cowbird egg laying periods.

Alternatives 4 and 7 require new concentrated stock areas (for example pack stations, equestrian stations, corrals) proposed for locations within 5 miles of known willow flycatcher sites undergo

FEIS Volume 3, Chapter 3, part 4.4, page 184 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 biological evaluations in order to determine if the proposed action will increase the potential for brood parasitism by brown-headed cowbirds (B43A). A regional perspective to evaluate the sum and distribution of local effects of brood parasitism pressure would make these biological evaluations more robust; this s&g lacks a regional overview provision. Also, the efficacy of this s&g could be improved by including all facilities that concentrate livestock, such as livestock management and handling facilities, in the evaluations. Alternatives 4 and 7 have no s&g for existing facilities related to cowbird management or a cowbird control program.

Alternative 5 prohibits the use of supplemental feed between June 1 and August 31 within 5 miles of known willow flycatcher sites. The s&g further stipulates that if this restriction is not feasible, livestock use within 5 miles of known willow flycatcher sites would be prohibited. This alternative may lessen the potential influence of brown-headed cowbird brood parasitism on the willow flycatcher population, but only if supplemental feed is the primary food source of cowbirds within 5 miles of known sites. This alternative does not include a brown-headed cowbird control program or recreation and livestock facilities standard (B42). Since brown-headed cowbirds not only exploit supplemental feed provided for livestock and pack animals, but also waste grain and insects in livestock and pack animal manure (Beedy and Granholm 1985, Goguen and Mathews 1999), this standard might not be effective in reducing cowbirds where supplemental feed is prohibited but livestock and pack animals congregate within 5 miles of known willow flycatcher sites or other conditions that are conducive to cowbirds (for example summer tract homes, livestock and pack animal facilities off forest service lands) exist within this zone. Also note that cowbirds appear to be in low numbers at pack stations where animals are fed pelletized food (S. Rothstein pers. comm.).

Alternative 8 prohibits new pack stations or corrals within 5 miles of known willow flycatcher sites (B43), although this universally-applied measure may not achieve the desired results of reducing cowbird brood parasitism potential for the reasons outlined in Alternative 2 above. As with Alternatives 4 and 7, Alternative 6 requires that new concentrated stock areas (such as pack stations, equestrian stations, corrals) proposed for locations within 5 miles of known willow flycatcher sites undergo biological evaluations to determine if the proposed action will increase the potential for brood parasitism by brown-headed cowbirds (B43A). In addition, Alternatives 6 and 8 require that the location of all existing corrals and pack stations throughout the Sierra Nevada be reviewed to identify areas where removal, movement, or other mitigations would be likely to provide the greatest benefits in reducing cowbird populations and protecting important bird habitats (B44, B44A). Only alternative 8 requires that the assessment be completed at a Sierra Nevada-wide scale to address cumulative effects and ensure consistency, and that implementation of a mitigation plan occurs by the year 2003 (B44). For s&gs B44 and B44A, it is assumed that “important bird habitats” refers to both known willow flycatcher sites and IBAs. The approach of Alternatives 6 and 8 for evaluating existing recreation facilities throughout the bioregion addresses the limitation of facilities mitigations in reducing brown-headed cowbird brood parasitism pressure in areas of multiple land ownership at the same time as it offers a flexible “tool box” for addressing the problem (for example facilities closures or removals, cowbird trapping, delaying facility operations until after willow flycatcher incubation periods, use of pelletized feed). Under s&g B44, the effectiveness of each of the site- specific mitigation measures will be evaluated at the regional level to determine the likelihood of success in reducing the incidence of cowbird parasitism. For existing facilities, s&g B44 imposes a mitigation timeline, addresses cumulative effects, and ensures regional consistency; this s&g has high potential to reduce brown-headed cowbird populations, and therefore potential willow flycatcher brood parasitism, while also considering constraints of multiple land ownership on efficacy. Greater

FEIS Volume 3, Chapter 3, part 4.4, page 185 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 benefit of regional assessments of existing concentrated stock facilities mitigation plans would be derived if periodic reviews, incorporating updated willow flycatcher data, were conducted.

Modified Alternative 8 does not specifically address existing corrals and pack stations but the Record of Decision (ROD) will include the provision that these be reviewed among existing uses in Landscape Analysis to identify areas where removal or relocation of these facilities, or other mitigations such as cowbird control or delayed access, would be likely to provide the greatest benefits in reducing cowbird brood parasitism pressure. Landscape analysis would not provide the full benefits of a regional overview of cumulative effects, implementation consistency and mitigation efficacy (see discussion above), however. A modified s&g B43A is included in Modified Alternative 8 to evaluate proposals for new concentrated stock areas that include livestock management and handling facilities in addition to pack and equestrian stations and corrals. The biological evaluation lacks the full benefit of a regional review (see discussion above) but must include a broad landscape- level review.

Additionally, independent of willow flycatcher locations, Alternatives 6, 7, 8 and Modified Alternative 8 would consider relocating existing livestock and pack stock gathering facilities and locate new facilities outside of meadows and riparian areas (AM23, RCA42, FW-WM12). This could affect an estimated 127 corrals and pack stations. Also, guidelines are provided to minimize livestock use of riparian areas with off-channel watering devices, fences, and other means (AM20, FW-AM20). Though broader in application, these standards may have limited effectiveness in protecting willow flycatchers and other riparian- and meadow-dependent birds that are cowbird hosts from brood parasitism unless timing of use or location of these facilities accounts for brown-headed cowbird egg laying periods or foraging distances, respectively.

Other factors considered in the evaluation of consequences: To address the uncertainties and risks in the willow flycatcher management approach of Modified Alternative 8, the Record of Decision (ROD) will include the following provisions and commitments:

• Review important bird area meadows among existing uses in Landscape Analysis. • Review the location of all existing corrals and pack stations during Landscape Analysis to identify areas where removal or relocation of facilities, or other mitigations such as cowbird control or delayed access, would be likely to provide the greatest benefits in reducing cowbird brood parasitism pressure and protecting important bird habitats (occupied willow flycatcher habitat and important bird areas). Implementation of mitigations will begin within 3 years. A re-assessment of cowbird-related mitigations will be required every 5 years in response to updated willow flycatcher data. • Requirements to map potential willow flycatcher habitat (emphasis habitat; small, wet meadows with shrubby vegetation; riparian shrub vegetation) using a consistent regional protocol; identify suitable habitat using regionally established criteria; conduct surveys to determine willow flycatcher occurrence, distribution, relative abundance; monitor willow flycatcher demographic trend and habitat criteria in the North, Central and Southern Sierra Nevada provinces; conduct formal management study of the response of willow flycatchers (habitat, site occupancy, demography) to various levels of grazing. • Re-evaluate the effectiveness of all standards and guidelines pertaining to willow flycatcher management under Modified Alternative 8 as well as other willow flycatcher-related commitments included in the ROD in 5 years. At that time, additional information will be

FEIS Volume 3, Chapter 3, part 4.4, page 186 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

available regarding willow flycatcher population occurrence, distribution and relative abundance, demographic and population trend data, habitat condition and trend data for occupied sites, as well as interim results of a formal management study of the effects of grazing intensity and frequency on willow flycatcher habitat, site occupancy, and demography.

With the exception of recreation, pesticides, and roads, no other resource use s&gs for Sierra Nevada Forest Plan Amendment Project FEIS alternatives specifically address willow flycatcher management issues. Other proposed management activities, such as prescribed burning, mechanical fuels treatments in the adjacent forest, dams, and mining may affect the willow flycatcher population and their habitat. Analysis of preliminary, spatially-explicit willow flycatcher data for the Sierra Nevada bioregion shows that two known willow flycatcher sites and approximately 20 percent of willow flycatcher emphasis habitat meadows are coincident, to varying extents, with proposed fuel treatment areas (SPLATS). Though fuel treatment approaches vary across alternatives, SPLATS are consistently mapped for the alternatives in which they apply and these willow flycatcher sites could be compromised by adjacent fuel treatments. Timber thinning, prescribed burning, and mechanical fuels treatment in the adjacent forest can affect meadow hydrology (for example erosion, sedimentation) and microclimate (desiccation), with possible concomitant effects on willow flycatcher prey or predator abundance or composition, and may put willow flycatcher habitat suitability at risk. In addition, these activities may create canopy openings in the adjacent forest that could have indirect effects on the willow flycatcher population (via increased brown-headed cowbirds, especially on the east side (Verner and Rothstein 1988)). The same applies for roads within and adjacent to riparian and meadow communities. Furthermore, dispersed and developed recreation (for example campgrounds, picnic areas, OHV roads, biking or hiking trails) in or directly adjacent to meadow and riparian habitats can have negative effects on willow flycatchers and their habitat due to trampling of vegetation and streambanks, hydrologic changes, direct disturbance of nests, and an increase in nest predators and brown-headed cowbirds associated with supplemental food. Pesticide programs administered by Forest Service are another potential concern.

Alternatives 2 through 8 provide direction to minimize resource impacts from recreation (R01D). Assessments are required to determine if developed sites, dispersed campgrounds, and day use sites seriously degrade habitat for willow flycatchers, seriously hinder support of ACS goals, or threaten water quality. In addition, if dispersed visitor use is inconsistent with desired conditions, the use may be modified through site design, site rehabilitation, site improvement, site closure, or re-directing the use to a more suitable location. During landscape analysis, Alternative 3 requires assessment of existing developed campgrounds and recreation facilities located in riparian zones for possible redesign or relocation, and prohibits campground expansion in riparian zones to protect the integrity of the zone (WM16), and reduces OHV impacts in riparian areas (R09C). Though the linkage between pack station and corral locations and brown-headed cowbird brood parasitism are considered in relation to known willow flycatcher sites for most alternatives (see discussion of brown-headed cowbird management above), only Alternative 3 and Modified Alternative 8 addresses developed recreation where conflicts with willow flycatcher management, in addition to cowbirds, might arise. For example, efforts to reduce roads and trails (edge effects) through meadows or visitor trampling along meadow streambanks as well as education to reduce human “food subsidies” to cowbirds and nest predators (such as chipmunks, squirrels, and jays) or even seasonal

FEIS Volume 3, Chapter 3, part 4.4, page 187 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

(willow flycatcher breeding season) closures of developed campgrounds and picnic areas may be warranted where such actions are likely to reduce brown-headed cowbird brood parasitism or nest predation pressure on breeding willow flycatchers. Note that developed recreation sites are exempted from streambank stability standards along stream reaches under Alternatives 2, 4, 6, 7, 8 and Modified Alternative 8 (AM16, AM16A, RCA18) although Alternative 8 does not include this exemption for meadow streambanks (AM16B). Also, only Modified Alternative 8 specifically addresses new recreation uses, including: new OHV areas should be located outside of Riparian Conservation Areas (RCA38), and other new uses should minimize the risk of sediment delivery into aquatic systems and impacts to ARM-dependent species such as the willow flycatcher (RCA10). Under Modified Alternative 8, s&gs RCA37 and RCA10 require evaluation of existing developed and dispersed recreation sites as well as a variety of other uses during landscape analysis and provide a flexible "tool box" including redesign, re-direction, rehabilitation, relocation, or closure to address and mitigate existing activities that degrade water quality or habitat for ARM-dependent species. See stream and watershed management, brown-headed cowbirds, recreation, and insect prey assumptions and limitations.

Alternatives 2 through 8 avoid pesticide and herbicide use within 500 feet of sites occupied by sensitive amphibian species unless application is consistent with ACS goals (AR24). Alternative 5 urges general avoidance of pesticide and herbicide application in riparian areas, Aquatic Diversity Areas, and Critical Refuges except where project level analyses demonstrate risks to sensitive species and their habitats as well as high value or private resources on adjacent lands (ACS20, ACS31). These three s&gs have been combined in Modified Alternative 8 to include Riparian Conservation Areas and willow flycatchers (RCA-12, RCA12, RCA12A). Willow flycatcher nestling deformities in Arizona have been reported; unusually high concentrations of the heavy metal strontium were found in bird eggs at these sites (Mora et al. 2000). Reports of infertile or non-viable willow flycatcher eggs range from 10 to 15 percent in the Sierra Nevada (Morrison et al. 2000, Whitfield 2000), although only 3 to 4 percent unhatched eggs have been reported for willow flycatchers in Ohio (Holcomb 1972) and Wisconsin (McCabe 1991) and the mean rate of unhatched eggs for 32 cup-nesting migratory passerine species is 4.9 percent (Martin 1992). Willow flycatchers feed primarily on insects, many of which have aquatic larval stages (Beal 1907, Bent 1942, Bombay 1999) and may bioaccumulate pesticides. The role of pesticides and herbicides on the decline of the willow flycatcher population in the Sierra Nevada has not been studied but it seems plausible that their application in known willow flycatcher sites could have a deleterious effect on willow flycatchers, their habitat, eggs, or prey species (Gard and Hooper 1995) and thus precautions under Modified Alternative 8 are warranted. Alternative 5 also limits the application of herbicides in the green zone of riparian areas and meadows to ground-based, vegetation-specific treatments (ACS21). Ground-based treatments in ACS21 presumably will provide site-specific applications of herbicides and pesticides and reduce drift beyond the targeted application area. This requirement has not been carried forward in Modified Alternative 8. Alternative 8 prohibits application of livestock pesticides in riparian areas (AM21), although these practices are not currently in use on national forest lands (S. Bishop pers. comm.). RIP-AM21 in Modified Alternative 8 also prohibits application of livestock pesticides within Riparian Conservation Areas. The effect of pesticide use (for example, noxious weed, tussock moth, or mosquito control) on willow flycatchers, their habitat, eggs, and prey base represents an unknown risk. See insect prey assumptions and limitations.

FEIS Volume 3, Chapter 3, part 4.4, page 188 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Except for Modified Alternative 8, there are no standards that specifically address roads in known or other potential willow flycatcher sites. Alternatives 2 through 8 allow construction of new roads in meadows and meadow reserves where it is consistent with riparian conservation objectives (RD14). It is anticipated that new road construction in general will be minimal, ranging from 0.02 to 0.5 percent of the existing system in the first decade (T. Durston pers. comm.). In addition, Alternatives 2 through 8 require that existing roads be assessed and that roads affecting natural flows in wetlands be considered for reconstruction or relocation (RD02). RD02 also requires the consideration of seasonal or long-term road closures where such closures can mitigate environmental damage. Some re-alignment of existing roads would be expected in all alternatives. Approximately 5 percent of the existing road system would be decommissioned in the first decade under Alternatives 1, 4 and 7, but would be higher for other alternatives, especially Alternatives 2, 3, and 5 (T. Durston pers. comm.). Alternative 3 would manage riparian zones to reduce existing road densities, eliminate unnecessary roads according to a priority schedule identified in landscape analyses, and prohibit building of new roads in riparian zones, among other provisions (WM11). Under Alternative 5, roads posing the greatest risk to aquatic and terrestrial ecosystems and placed in ecologically significant areas would have priority for decommissioning and restoration, potentially reducing the risk of meadow desiccation from erosion and gullying (RD07A). Alternative 5 also requires that road crossings not currently capable of sustaining a 100 year flood event be identified and prioritized for improvement (RD05A). Within CARs, RD06 establishes road density reduction targets. Improvement of road crossings (RD04) upstream and downstream of known or potential willow flycatcher habitat could reduce the long-term erosion risks of catastrophic floods on willow flycatcher habitat. Under Alternatives 6 and 8, roads along streams are a priority for closure or decommissioning (RD07B), potentially reducing the risk of meadow desiccation as a result of erosion and gullying. Modified Alternative 8 provides specific direction that to the extent possible construct no new roads in potential willow flycatcher habitat including occupied and known willow flycatcher sites, emphasis habitat and small, wet shrubby meadows (RD14A). FW-RD03F also includes a standard to avoid road construction in all meadows, in general. RCA14 provides direction to maintain and restore the hydrologic connectivity of meadows by identifying roads and trails that disrupt water flow and implementing corrective action where necessary. In addition, to reduce environmental impacts such as sediment delivery to aquatic systems or disturbance to wildlife, Modified Alternative 8 also includes standards for road maintenance (RCA-9, FW-RD03G), reconstruction (RCA-9, RCA37), conversion (RCA-9, RCA37), closure (FW-RD03A, RD03J, RCA-9, RCA37), and decommissioning (FW-RD03B, RCA-9, RCA37). See stream and watershed management assumptions and limitations.

There are many s&gs in Modified Alternative 8 that address the hydrologic function and condition of aquatic, riparian and meadow systems. Modified Alternative 8 sets the default conservation area buffer width around meadows in which ground-disturbing activities are to be avoided at 300 feet; the buffer width may be adjusted if Landscape Analysis has been completed and a site-specific analysis demonstrates the need for a different width. Peer review processes are necessary for vegetation treatments or other activities that are likely to significantly affect aquatic resources or disturb ground in more than 25 percent of the buffer area (RCA-000). It is unknown whether this buffer will sufficiently mitigate erosion and sedimentation effects from ground disturbing activities near known and potential willow flycatcher habitat but the peer review process and site-specific buffer width analysis should help ensure protection of these meadows. Note that mechanical ground disturbing fuels

FEIS Volume 3, Chapter 3, part 4.4, page 189 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

treatments, salvage harvest, commercial fuelwood cutting, or hazard tree removal is permitted within riparian conservation area buffers when the activity is consistent with riparian conservation objectives (RCA31, RCA33). RCA30 provides direction to minimize the spread of prescribed fire into riparian vegetation. Within CARs, occupied or suitable habitat for T&E species, or occupied habitat for sensitive species, Modified Alternative 8 allows prescribed fire to back into the area but direct lighting is to be avoided and ground-disturbing impacts are to be mitigated (RCA27, FW-RCA-27). Fire suppression and post-wildfire management activities must minimize adverse effects to ARM-associated species (RCA32, RCA35). In addition, multiple standards exist to secure appropriate in-stream flows for hydroelectric projects (during FERC re-licensing) in order to maintain, restore, or recover ecological conditions (RCA23, RCA24, RCA25). Mining is also addressed (FW-M02, FW-M06A, FW- M10, FW-M11, FW-M12, RIP-M04, RIP-M08, RCA45). There are standards to limit soil compaction to no more than 5 percent of the Riparian Conservation Area and reduce sediment delivery to aquatic systems from management activities (RCA4, FW-S01) as well as maintain and restore the hydrologic function of aquatic systems, including meadows (RCA15, RCA39). In addition, where potential and existing management activities such as road building, recreation, grazing and timber harvest contribute to degradation of aquatic systems or ARM- associated species habitat, preventative and restoration measures are to be implemented (RCA40, RCA6, RCA17). The cumulative effects of these s&gs should result in improved conditions for known and potential willow flycatcher habitat.

All action alternatives are intended to achieve desired conditions (ACS goals) of ARM- dependent species viability but rely on different approaches. Alternatives 2, 4, 6, 7, and 8 have standards that are intended to protect habitats for all ARM-associated species (AM16, AM16A, AM16B) but do not prohibit activities such as livestock grazing, pesticide use, fish stocking and road or trail construction as does the amphibian reserve system proposed in Alternatives 2, 3, and 5 (AM12, AM13). Modified Alternative 8 excludes grazing in Yosemite toad habitat during the breeding and rearing season (RCA-41). The spatially-explicit locations of sensitive amphibian species are unknown at this time and thus an estimate of affected acres regarding amphibian protections is unavailable, and precludes a spatial evaluation of effects on the willow flycatcher population. There are specific riparian and meadow habitat s&gs intended for other sensitive species (Great Gray Owl (for example, FW-B15, FW-B16, FW-B17 in Modified Alternative 8)) or areas (Bogs and Fens (FW-P06 in Modified Alternative 8), Critical Refuges and Critical Aquatic Refuges (s&gs unique to CARs in Modified Alternative 8 include RCA4, CAR-M15, RCA-000, RCA27); but due to GIS data layer attribute inaccuracies (Great Gray Owl), lack of spatially-explicit data (Bogs and Fens) or major revisions (CARs), their spatial overlap with known willow flycatcher sites and potential habitat could not be evaluated. Thus, depending upon the extent of elevational and habitat condition overlap, these other species or area-specific s&gs could increase habitat expansion capability for the willow flycatcher by some unknown degree.

Herger-Feinstein Quincy Library Group Forest Recovery Act Pilot Project Among the Herger-Feinstein Quincy Library Group Forest Recovery Act Pilot Project alternatives drafted, the BA/BE indicates that the selected Alternative 2 has the highest risk of adverse impact to the willow flycatcher. The Herger-Feinstein Quincy Library Group Forest Recovery Act Pilot Project BA/BE states that Alternative 2 creates the greatest risk of willow flycatcher brood parasitism by the brown-headed cowbird. Brown-headed cowbirds exploit edge habitat along openings and the perimeter of the forest canopy; although the amount of

FEIS Volume 3, Chapter 3, part 4.4, page 190 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 forest edge generated by Defensible Fuel Profile Zones (DFPZs) depends upon their number and shape (area-to-perimeter ratio), Alternative 2 treats the highest number of acres and potentially produces the greatest amount of edge. In addition, roads built to allow access to DFPZs for firefighters will also increase access for cowbirds and nest predators to the forest interior, and potentially willow flycatcher nest sites. Note that data from the southern Sierra Nevada do not indicate that DFPZs west of the Sierra Nevada ridgeline would have a significant effect on cowbird abundance and brood parasitism; Verner and Ritter (1983) discovered that brown-headed cowbirds were rare in all areas except meadows. The mean number of cowbirds per count was: meadows = 0.56, clearcuts = 0.095, logged areas = 0.07, and unlogged areas = 0. Whereas, in the east side Mammoth Lakes area, Verner and Rothstein (1988) found cowbirds in meadows and open Jeffrey pine forests. If sheep grazing is employed to maintain surface fuels in treated DFPZs, this may also increase local cowbird abundance through their association with livestock (Sanders and Flett 1989, Lowther 1993).

A spatially-explicit overlay of the proposed Herger-Feinstein Quincy Library Group Forest Recovery Act Pilot Project DFPZs and preliminary willow flycatcher distribution data in the Herger-Feinstein Quincy Library Group Forest Recovery Act Pilot Project planning area displays at least three, and possibly six, known willow flycatcher sites edged by DFPZs. In addition, there are numerous emphasis habitat meadows that show the proposed DFPZs immediately adjacent. The effects on individual meadows from timber thinning and mechanical fuels treatments in the adjacent forest are unknown (see fire and fuels assumptions and limitations). In addition to their creation, what effects will maintenance of DFPZs near willow flycatcher habitat have on willow flycatcher persistence and productivity in known sites as well as colonization in potential habitat? One of the known willow flycatcher sites that has proposed DFPZs adjacent to it is along the Little Truckee River in the Sierraville Ranger District of Tahoe National Forest. This meadow complex is identified as one of the highest priority willow flycatcher sites in the Sierra Nevada bioregion because it is “... most likely to serve as colonizable habitat in the near future” (USDA 1998). Furthermore, the Science Review recommends that priority is accorded all sites that have had willow flycatcher sightings in recent years and specifically considers “what actions might be taken to assure that projects, such as timber harvest and road construction near montane meadows, do not alter markedly their hydrologic functions or sediment loads?”

The proposed DFPZs adjacent to willow flycatcher habitat create uncertainty and risk for both the abundance and distribution of the willow flycatcher population in the Herger-Feinstein Quincy Library Group Forest Recovery Act Pilot Project planning area as well as the Sierra Nevada bioregion and should be prohibited or highly restricted nearby willow flycatcher habitat, until the effects of DFPZs on willow flycatcher surrogate species and their breeding habitat are better understood. Recommendations include: DFPZs should be located away from known willow flycatcher sites or prohibited until more information is available about the effect of DFPZs on brown-headed cowbird brood parasitism rates, willow flycatcher prey or predator abundance or composition, hydrology, and microclimate in the adjacent meadows as well as effects on willow flycatcher surrogate species productivity. To avoid jeopardizing potentially suitable habitat for willow flycatcher colonization, DFPZs near emphasis habitat meadows should only be implemented under an adaptive management basis to study the effects of DFPZs on the adjacent meadows (for example cowbird parasitism rates, willow flycatcher prey or predator abundance and composition, hydrology, microclimate, willow flycatcher surrogate species productivity). Furthermore, these willow flycatcher emphasis habitat meadows should

FEIS Volume 3, Chapter 3, part 4.4, page 191 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

be monitored prior and subsequent to DFPZ implementation, not only to establish baseline data but also to obtain information about the influence of DFPZs on willow flycatcher colonization.

Environmental Outcomes The following environmental outcomes described for each alternative represent a scoring of the estimated abundance and distribution of suitable environments for willow flycatchers on national forest lands in the planning area after 50 years of full implementation. These environmental outcomes do not represent an actual prediction of future population occurrence, size, density, or other demographic characteristics but rather represent a subjective assessment of the capability of the environment on national forest lands to support population abundance and distribution. Environmental outcomes may not account for the effects of interspecific competition, disease, predation, illegal taking, pesticides, air pollution, or current population status. These additional factors, in combination with factors attributed to nonfederal lands, are considered in the population outcomes described below. For more background on these outcomes, see the detailed analyses and research reviews provided in the previous affected environment and environmental consequences sections of this document.

Table 4.4.2.3e. Environmental outcomes for willow flycatchers by alternative. ALTERNATIVE ENVIRONMENTAL OUTCOME Historic Condition B Current Condition E No Action E Alternative 2 C+ Alternative 3 D- Alternative 4 E+ Alternative 5 D+ Alternative 6 D Alternative 7 E Alternative 8 C Modified Alternative 8 D+

Cumulative Effects Table 4.4.2.3f. Estimate of Wet Shrubby Meadows Containing Known or Potential Willow Flycatcher Habitat (acres) by Ownership within Sierra Nevada Forest Plan Amendment Project Planning Area. Ownership Known Emphasis Small wet, Total Sites Habitat shrubby meadow Forest Service 10,675 52,695 22,306 85,676 Non-Forest Service 5,354 21,304 3,179 29,837 Outside Forest Service boundary (inclusions) 943 8,229 270 9,442 Unassigned 3,045 366 8 3,419 Total Wet Shrubby Meadow Habitat 20,017 82,594 25,763 128,374

Based upon limited available information for identifying wet meadows with a riparian shrubby vegetation component in the Sierra Nevada Forest Plan Amendment Project planning area, there are currently over 128,000 acres of known and potential willow flycatcher habitat. For these analyses, ownership could only be reported in broad categories. Although the Forest Service administrative ownership coverage is more accurate than the Statewide ownership coverage, it lacks specific attributes for State, National Parks, City or County, or other ownership. The information in the table

FEIS Volume 3, Chapter 3, part 4.4, page 192 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 above is based upon existing GIS meadow coverages which were developed primarily for national forest lands. Meadow polygon identification was focused on known willow flycatcher sites on national forest lands, primarily, and off national forest lands secondarily. The third and fourth levels of meadow identification priority were emphasis habitat as well as small wet, shrubby meadows on national forest lands and off national forest lands, respectively. Thus, meadow acreage estimates, especially for emphasis habitat and small wet meadows, off national forest lands should be treated as preliminary and will invariably represent underestimates because meadows were identified primarily on national forest lands. In addition, as a result of Pacific Southwest Region requests to compile all known data across the bioregion, it appears that survey efforts for willow flycatcher locations in the planning area have not been even across ownerships. It is possible that the preponderance of known willow flycatcher sites on national forest lands may be an artifact of the sampling efforts to date, although this is difficult to determine without being able to compare positive and negative survey results for different ownerships in the bioregion.

Table 4.4.2.3g. Estimate of Known Willow Flycatcher Sites (acres) by Ownership in active, inactive, non-allotments and outside allotments within Sierra Nevada Forest Plan Amendment Project Planning Area. Known Willow Flycatcher Sites by Ownership In Active In Inactive In Non- Outside Total Allotments Allotments Allotments Allotments Forest Service 6,704 101 191 3,679 10,675 Non-Forest Service 1,504 218 0 3,632 5,354 Outside Forest Service boundary (inclusions) 5 0 0 938 943 Unassigned 4 0 0 3,041 3,045 Total Known Willow Flycatcher Sites 8,217 319 191 11,290 20,017

The 135 known willow flycatcher sites in the Sierra Nevada Forest Plan Amendment Project planning area comprise an estimated total of 20,000 acres. Recall that 82 known sites occur on national forest lands, with the remaining 53 sites on non-national forest lands distributed among private (n=29), National Park service (n=10), State (n=8), City or County (n=4) and Bureau of Land Management (n=2) lands. A majority of the acreage of known willow flycatcher sites (53 percent) occurs on national forest lands; 27 percent occurs on non-national forest lands, a small percentage (5 percent) occurs on inclusions that are outside forest service administrative boundaries but within the Sierra Nevada Forest Plan Amendment Project planning area, and 15 percent of the acres are not assigned ownership in the existing GIS coverage (ownership currently unknown).

Irrespective of ownership, over 40 percent of known willow flycatcher site acres in the Sierra Nevada Forest Plan Amendment Project planning area occur within active grazing allotments. The majority of these acres in active allotments will be managed according to Forest Service standards and guidelines; representing 33% of all known willow flycatcher site acres in the planning area, this underscores the importance of the FEIS decision in determining the cumulative effects of livestock grazing in known willow flycatcher sites in the Sierra Nevada. In addition, over 1500 acres (8 percent) of known willow flycatcher sites occur off Forest Service land and may or may not adhere to the same livestock grazing s&gs, depending upon management arrangements. Note that off Forest Service land, however, the majority of known willow flycatcher site acres occur outside grazing allotments in the planning area.

The combined effects of gold and gravel mining, livestock grazing, timber harvest, and roads on hydrologic systems over the last 150 years have left an indelible impression on riparian and meadow ecosystems in the Sierra Nevada bioregion (McKelvey and Johnston 1992, Menke et al. 1996,

FEIS Volume 3, Chapter 3, part 4.4, page 193 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Kattelmann 1996, Kinney 1996, Kondolf et al. 1996). Research on invertebrates and the interaction between passerines and invertebrates suggest that stream disturbing activities on and off national forest lands, as well as other changes in surface and subsurface water flows, may affect not only the riparian and meadow breeding habitat of willow flycatchers, but may also affect their invertebrate food supply (Voight 1976, Erman 1984, Gray 1993, Kelly and Wood 1996). Furthermore, there are indications that changes in meadow hydrology may influence nest predation by either limiting or allowing ground predators access to nests (M. Morrison pers. comm.). Thus, it seems plausible that water depth may influence the invertebrate community, nest predator access, or both in willow flycatcher habitat.

Adding to the list of cumulative effects in Sierra Nevada meadows are fire suppression, conifer encroachment, reservoirs, land clearance for farming and residences, diversions of water for irrigation, land drainage, railroads, mosquito abatement programs, groundwater abstraction, recreation, and the introduction of beavers (Castor canadensis) (Ingles 1965, Gibbens and Heady 1964, Kondolf et al. 1996, Kattelmann 1996, Graber 1996). Present efforts undertaken by any Sierra Nevada land owner to exclude all activities (such as roads, trails, dispersed and developed recreation, livestock grazing, pesticide use) that may affect the meadow community (for example, hydrology, vegetation, cowbirds, prey and predator species) may still not increase willow flycatcher productivity or facilitate population expansion without active site restoration necessary to overcome the legacy of these various affectors.

Population Outcomes The following population outcomes described for each alternative represent a scoring of the estimated abundance and distribution of willow flycatchers within the planning area after 50 years of full implementation. These population outcomes represent a subjective assessment of the effects on the geologic, hydrologic, and vegetative attributes of habitat (as they pertain to population demographics) as well as the effects of interspecific competition, disease, predation, illegal taking, pesticides, air pollution, or current population status. For more background on these outcomes, see the detailed analyses and research reviews provided in the previous affected environment, environmental consequences, and cumulative effects sections of this document.

It should be noted that the population outcome scores below are generally slightly lower than the scores for environmental outcomes provided previously in this document. This is in part due to the assumption that some unknown impacts to the species are occurring on wintering and migration grounds, and that these impacts are not likely to be reduced in the future. In addition, impacts in the Sierra Nevada on and off national forest lands are also more likely to increase than decrease as the human population and human use of the region grows. These factors may limit the ability of the willow flycatcher population to increase at the same rate as breeding habitat improves in quality or increases in amount. These factors off national forest lands are assumed to be constant between alternatives for the purpose of this analysis.

FEIS Volume 3, Chapter 3, part 4.4, page 194 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Table 4.4.2.3h. Population outcomes for willow flycatchers by alternative ALTERNATIVE POPULATION OUTCOME Historic Condition B Current Condition E No Action E- Alternative 2 C Alternative 3 E+ Alternative 4 E Alternative 5 D Alternative 6 D- Alternative 7 E Alternative 8 C- Modified Alternative 8 D

FEIS Volume 3, Chapter 3, part 4.4, page 195 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.2.4. GREATER SANDHILL CRANE Life History Zeiner and others (1990) describe nesting sites as the following; remote portions of extensive wetlands, or sometimes in shortgrass prairies. On dry sites, nests are scooped out depressions lined with grasses. More commonly, nests are large mounds of wetland plants, in shallow water screened by tall tules, cattails, or shrubs. Natural hummocks or muskrat houses are often used. Ideal sites are on small islands screened by tall tules, cattails, or shrubs.

Cranes do not breed until their fourth year, but then usually mate for life (Johnsgard 1975). Nesting activities begin with courtship in April. Peak breeding occurs in May through July, with nesting usually completed by late August. Average clutch size is two, ranging from one to three. Incubation takes approximately 30 days. Shortly after the second egg hatches, adults lead the young from the nest site and begin feeding them. Each adult generally feeds one chick. Chicks are aggressive toward each other, and shortly after hatching one becomes dominant. Often this dominance leads one chick to be pushed away from the adults. This may cause the chick to starve or be consumed by a predator (Zeiner and others 1990, CDFG 1994). Young fledge at about 70 days, but remain with their parents for up to one year (Harrison 1978). They forage in open shortgrass fields and pastures, as well as open wetlands. Foods consist of grasses, forbs, cereal crops, roots, tubers, earthworms, insects, mice, small birds, snakes, frogs and crayfish.

Habitat Relationships Sandhill cranes are primarily birds of open freshwater wetland and shallow marshes, but the different subspecies utilize a broad range of habitat types, from bogs, sedge meadows, and fens to open grasslands, pine savannahs, and cultivated lands. During the breeding season, the three migratory subspecies may be found in a wide variety of northern wetland communities. Habitats along migration routes tend to be large, open palustrine and riparian wetlands near agricultural areas, while wintering habitats include riparian wetlands, wet meadows, seasonal playa lakes, and pastures. The non-migratory subspecies use seasonally variable wetlands, grasslands, and palm and pine savannahs. Sandhill cranes are omnivorous, feeding on a wide variety of plant materials (including waste grains) and small vertebrates and invertebrates, both on land and in shallow wetlands.

Status Six subspecies of sandhill cranes are currently recognized: lesser sandhill crane (G.c.canadensis), Canadian sandhill crane (G.c. rowani), greater sandhill crane (G.c. tibida), Florida sandhill crane (G.c. pratensis), Mississippi sandhill crane (G.c. pulla), and Cuban sandhill crane (G.c. nesiotes). The taxonomic status of, and the relationships among, the sandhill crane subspecies have been discussed frequently in the literature (North American Crane Working Group 2000). The G.c. canadensis-rowani-tibida group is probably clinal, with grandual changes in morphological characters and no positive means of distinguishing among them. Random pairing among the three subspecies has been documented, and intergrading occurs along the limits of their ranges. G.c pulla was described as a subspecies in 1972, based mainly on color differences between it and G.c pratensis. The existing population of G.c. pulla in Mississippi was probably more widespread in the past, and may have intergraded with G.c. pratensis and G.c. nesiotes to the east (North American Crane Working Group 2000).

G.c tibida is subdivided into five populations in. However, the morphological differences among the populations have not yet been analyzed in terms of taxonomic significance.

FEIS Volume 3, Chapter 3, part 4.4, page 196 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

The greater sandhill crane is a California State Threatened species. This subspecies was listed as sensitive by the Pacific Southwest Region of the Forest Service on June 8, 1998. Historically, this species was a common breeder in the northeast plateau of California (Zeiner and others 1990). Monitoring of this species has been inconsistent over the years. The last State-wide survey was in 1988. Survey results showed there were 276 pairs recorded at 60 sites, with 163 (59 precent) in Modoc County, 75 (27 precent) in Lassen County, 29 (11 precent) in Siskiyou County, 7 (3 precent) in Plumas County, and single pairs (0.4 precent) in Shasta and Sierra Counties (Herziger 2000). Ownership of the sites was as follows: private 186 (67 precent), Fish and Wildlife Service (41 (15 precent), California Department of Fish and Game 36 (13 precent), Forest Service 13 (5 precent).

The data from the four national forests with greater sandhill crane shows that there were only 5 successful nesting attempts in 1997 with 6 colts observed. In 1998 the number of successful nests was 6 and 11 colts were observed.

Herziger and Ivey (2000) recently completely surveying California. The results of this survey show substantial increases (465 pairs verses 276 pairs) in the number of pairs over the 1988 survey efforts. It is unclear as to whether this increase was a result of an actual increase in the State-wide breeding population or survey effort in 1988 (Herziger pers. comm.). There were 190 sites surveyed and 465 pairs were recorded at 127 sites. Eighty-eight pairs were found at 72 “new” sites. The numbers of pairs by county is: Modoc 252 pairs (54 precent), Lassen 122 (26 precent), Siskiyou 51 (11 precent), Plumas 20 (4 precent), Shasta 10 (2 precent), and Sierra 10 (2 precent). Land ownership of sites was as follows: private 296 (63 precent), Forest Service 69 (15 precent), California Department Fish and Game 55 (12 precent), Fish and Wildlife Service 41 (9 precent), Bureau of Land Management 5 (1 precent) (Herziger, in press).

Historical and Current Distribution: With a total estimated population of more than 500,000, the sandhill crane is the most abundant of the world’s cranes. It is widely (though intermittently) distributed throughout North America, extending into Cuba and far northeastern Siberia. Six subspecies have been described. The three migratory subspecies, the lesser, greater, and Canadian sandhill cranes, are relatively abundant. They are distributed across a broad breeding range in northern North America and eastern Siberia, with wintering grounds in the southern United States and northern Mexico. The other three subspecies, the Mississippi, Florida, and Cuban sandhill crane exist as small, non-migratory populations with restricted ranges in the southern United States (Mississippi, Florida, and southern Georgia) and Cuba. The total population is increasing in numbers, although some local populations may be declining (North American Crane Working Group).

The breeding range of the greater sandhill crane spans mid-continental North American from the Great Lakes to the Pacific Ocean. Scientists generally divide the greater sandhill crane into four distinct regional populations (North American Crane Working Group). The Central Valley population breed mainly in south-central and southeastern Oregon and northeastern California, with additional areas up to southern British Columbia and Vancouver Island. In the winter, these cranes migrate to the Central and Imperial Valleys of California (North American Crane Working Group).

At the time of European settlement the species was probably more widely distributed than at present. The remote arctic and subarctic breeding grounds of the lesser and Canadian sandhill cranes have been relatively free of human impacts. However, the wintering grounds of these subspecies have

FEIS Volume 3, Chapter 3, part 4.4, page 197 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 been extensively altered. Hunting, agricultural expansion, drainage of wetlands, and other habitat changes in the 18th and 19th cunturies led to the extirpation of the greater sandhill crane from many parts of its breeding range in the United States and Canada. The population and range of the non- migratory sandhill cranes in the southern United States have also diminished due to hunting, loss of wetlands, and other changes in its habitat. The Cuban sandhill crane was probably more widely distributed in the Cuban archigelago than at present (North American Crane Working Group).

Within the project area, the greater sandhill crane occurs on the Modoc, Lassen, Plumas and Tahoe National Forests during the summer breeding season and during migration. It is found in medium to large wetlands and short grass valley bottoms. The Eagle Lake Ranger District has the most habitat and the most nesting attempts of the four national forests (39 sighting locations, 17 nesting locations consisting of 47 nesting attempts producing 15 colts). The most important wetlands being Pine Creek, Bullard Lake, Poison Lake, and Papoose Meadows. The Hat Creek Ranger District has numerous wetlands on the east side and has had some nesting. The eastside of the Plumas National Forest has numerous meadows with suitable habitat and several sightings, but no documented nesting success. Within the Tahoe National Forest, a breeding population of approximately 11 pairs occurs within Carman Valley and Kyburz Flats on the Sierraville Ranger District.

In Plumas County, nesting cranes have been documented at several locations on private land in American Valley around Quincy, Indian Valley, and Sierra Valley. Cranes have also been documented in Red Clover Valley (within the national forest boundary), but no nesting attempts on Plumas National Forest lands have been documented. The majority of sightings within Plumas County consist of migrating flocks flying overhead in the Spring and Fall.

Risk Factors Factors that may have led to the decline of the species are disturbance by humans and cattle, loss of hiding and nesting cover due to grazing, predation, and loss of wetlands due to erosion and water diversion. The species is particularly sensitive to human disturbance when nesting, especially within a mile of the nest site (Zeiner and others 1990).

Herziger (pers. comm.) identified a number of threats including the fact that a majority of the pairs nest on private lands which could be converted other uses like agriculture using pivot irrigation or late flood irrigation practices, urbanization and grazing.

Environmental Consequences Effects of the Alternatives

FEIS Volume 3, Chapter 3, part 4.4, page 198 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Table 4.4.2.4a. Comparison of management activities that could affect habitats for the greater sandhill crane by alternative. Alt. Grazing Fire/Fuels**annual Riparian/Meadow Mgnt average acres 1 Existing LRMP levels 86,225 acres treated Varies per NF w/SMZ and RMAs 2 45% max. utilization, 4” ST* 34,929 acres treated 150ft green zone, 412-760ft gray zone 3 Meet proper functioning condition 141,479 acres treated 150ft green zone, 412-760ft gray zone 4 <45%utilization, meeting AMS goals 159,545 acres treated 150ft green zone, 412-760ft gray zone 5 5% bank trampling,50%shrub cover, 5-7”ST 60,527 acres treated 150ft green zone, 412-760ft gray zone 6 <45%utilization, meeting AMS goals 150,592 acres treated 300ft on perennial, 100ft On seasonal 7 <45%utilization, meeting AMS goals 145,260 acres treated 300ft on perennial, 100ft On seasonal 8 30%utilization,ST>4” 98,614 acres treated 300ft on perennial, 150 intermittent, 75 ephemerals Mod 8 20% streambank limit, 20% shrub ut, 112,860 acres treated 300ft on perennial, 100ft On seasonal. Elim. livestock from occ. WIFL sites

*ST = Stubble height **Modeled treatment acres include brush (fire and mechanical), plantation thin, and biomass thin for the 1st decade.

Environmental Outcomes Historic. Within the project area, the greater sandhill crane occurs on the Modoc, Lassen, Plumas and Tahoe National Forests during the summer breeding season and during migration. It is found in medium to large wetlands and short grass valley bottoms.

In Plumas County, nesting cranes have been documented at several locations on private land in American Valley around Quincy, Indian Valley, and Sierra Valley. Cranes have also been documented in Red Clover Valley (within the national forest boundary), but no nesting attempts on Plumas National Forest lands have been documented. The majority of sightings within Plumas County consist of migrating flocks flying overhead in the Spring and Fall.

Current. The Eagle Lake Ranger District has the most habitat and the most nesting attempts of the three forest area (39 sighting locations, 17 nesting locations consisting of 47 nesting attempts producing 15 colts). The most important wetlands being Pine Creek, Bullard Lake, Poison Lake, and Papoose Meadows. The Hat Creek Ranger District has numerous wetlands on the east side and has had some nesting. The eastside of the Plumas National Forest has numerous meadows with suitable habitat and several sightings, but no documented nesting success. Within the Tahoe National Forest, a breeding population of approximately 11 pairs occur within Carman Valley and Kyburz Flats on the Sierraville Ranger District.

FEIS Volume 3, Chapter 3, part 4.4, page 199 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Table 4.4.2.4b. Average assessment ratings for the greater sandhill crane. Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Outcome B C B B C B C C B B

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Cumulative Effects Population Outcomes Historic Conditions. At the time of European settlement the species was probably more widely distributed than at present. The remote arctic and subarctic breeding grounds of the lesser and Canadian sandhill cranes have been relatively free of human impacts. However, the wintering grounds of these subspecies have been extensively altered. Hunting, agricultural expansion, drainage of wetlands, and other habitat changes in the 18th and 19th cunturies led to the extirpation of the greater sandhill crane from many parts of its breeding range in the United States and Canada. The population and range of the non-migratory sandhill cranes in the southern United States have also diminished due to hunting, loss of wetlands, and other changes in tis habitat. The Cuban sandhill crane was probably more widely distributed in Cuban archigelago than at present (North American Crane Working Group).

Current Condition. Herziger and Ivey (2000) recently completed a survey of California that found substantial increases (465 pairs verses 276 pairs) in the number of pairs over the 1988 survey efforts. It is unclear as to whether this increase was a result of an actual increase in the State-wide breeding population or differences in survey efforts (Herziger pers. comm.). There were 190 sites surveyed in which 465 pairs were recorded at 127 sites. Eighty-eight pairs were found at 72 “new” sites. The numbers of pairs by county found in the 2000 survey (Herziger pers. comm.) were as follows: Modoc 252 (54 percent), Lassen 122 (26 percent), Siskiyou 51 (11 percent), Plumas 20 (4 percent), Shasta 10 (2 percent), and Sierra 10 (2 percent). The sites surveyed were distributed on the following lands: private 296 (63 percent), Forest Service 69 (15 percent), California Department Fish and Game 55 (12 percent), U.S. Fish and Wildlife Service 41 (9 percent), Bureau of Land Management 5 (1 percent) (Herziger pers com. 2000).

FEIS Volume 3, Chapter 3, part 4.4, page 200 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Table 4.4.2.4c represents the estimated population outcomes through the planning horizon for the greater sandhill crane. Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Outcome C C B C C B C C C B

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

FEIS Volume 3, Chapter 3, part 4.4, page 201 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.2.5. CALIFORNIA YELLOW-BILLED CUCKOO (Coccyzus americanus occidentalis) The yellow-billed cuckoo has declined throughout its range in southern Canada, the United States, and northern Mexico. It is nearly extinct west of the Continental Divide having disappeared from British Columbia in the 1920s, from Washington in the 1930s, from Oregon in the 1940s, and from northern-most California in the 1950s. It is extremely rare in the interior West. Its only remaining Western breeding populations are three small populations in California, scattered populations in Arizona, and New Mexico, and an unknown number of birds in northern Mexico (Hughes 1999, USFWS letter 04/14/2000, Laymon and Williams 1999). Current Breeding Bird Survey (BBS) data also demonstrates that the eastern yellow-billed cuckoo has suffered recent (past 30 years) significant declines in 17 states where it occurs, with the average decline for these states being 72 percent and with some declines (Connecticut) reaching 99 percent (BBS 2000, USFWS 04/14/2000).

Life History The yellow-billed cuckoo is a neotropical migrant, usually arriving from South American wintering areas in June and departs by late August or early September. In California, most eggs are laid in mid- June to mid-July. The cuckoo is monogamous and has a clutch averaging 2 to 4 eggs (ranges 1 to 5). Both sexes incubate with incubation lasting 11 days and hatching is asynchronous. Both sexes care for the altricial young. Young may leave the nest at 6 to 9 days (Bent 1940, Hamilton and Hamilton 1940, Potter 1980, Laymon pers.comm.).

Habitat Relationships Historic nesting locations in California range from near sea level in southern California to 4620 feet in the Owens Valley near Big Pine. In California, breeding populations of the yellow-billed cuckoo are now restricted to isolated sites in Sacramento, Amaragosa, Kern, Santa Ana, and Colorado River valleys (Roberson 1980, Hughes 1999, Laymon and Halterman 1987). The two significant breeding sites’ elevations range from 49.5 to 264 feet in the Sacramento Valley and from 2590 to 2904 feet at the South Fork Kern River (Laymon and Williams 1999).

The yellow-billed cuckoo is known to inhabit deciduous riparian thickets or forests with dense, low- level or understory foliage. Willow is almost always a dominant component of the vegetation in breeding habitat. In California, the yellow-billed cuckoo has a very strong preference for nesting in willows (96 of 97 nests), even when cottonwoods and other tree species are present (Laymon and others 1997). On the South Fork of the Kern, vegetation surveys of nest sites have determined that the average nest tree height was 31 feet with the shortest nest tree at 8.25 feet and the tallest at 58.7 feet (Laymon 1999). The DBH of the average nest tree was 9.9 inches and ranged from 1.2 to 35.1 inches (Laymon and others 1997). Willow species utilized as nest trees were primarily Goodings black willow (Salix gooddingi) and red willow (Salix lavigata).

It is probable that microclimate is an important part of habitat selection for this species. The two known breeding areas in California both showed a decrease in temperature and an increase in humidity closer to the nest (Launer and others 1990). Patch size also appears to be an important component of suitable habitat with larger patches of suitable habitat being occupied at a higher rate than smaller patches (Laymon and Halterman 1989). On the South Fork of the Kern, the land adjacent to occupied yellow-billed cuckoo breeding habitat is primarily flood irrigated pasture and dry rangeland (Laymon and Williams 1999). In this area breeding cuckoos are found more often at upland sites early in the season during wet years, but not in dry years (Laymon and Williams 1999).

FEIS Volume 3, Chapter 3, part 4.4, page 202 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

It is thought by local researchers that flooding in wet years reduces the survival of the larvae of the preferred prey (katydids and sphinx moths) that winter underground (Laymon and Williams 1999).

Species Status The California yellow-billed cuckoo is the western subspecies of the yellow-billed cuckoo (Grinnell and Miller 1944, Laymon and Williams 1999). This species is listed as a California Endangered Species and a Forest Service Sensitive Species. Formal petitions have been filed by environmentalists in 1987 and again in 1998 calling for the yellow-billed cuckoo to be listed west of the Continental Divide as either a subspecies or as a population which is geographically, morphologically, behaviorally, and ecologically distinct from the eastern yellow-billed cuckoo. On February 17, 2000 the U.S. Fish and Wildlife Service (USFWS) published an initial finding that Endangered Species Act protection may be needed for the western subspecies of the yellow-billed cuckoo, either as a subspecies or as a unique population. Environmentalists filed a lawsuit on July 31, 2000 against the USFWS for missing its deadline for deciding whether to propose or deny federal listing status for the California yellow-billed cuckoo.

Historical and Current Distribution Historically, the yellow-billed cuckoo was a common breeding species in riparian habitat throughout much of lowland California (Grinnell and Miller 1944). Grinnell and Miller (1944) described the cuckoo’s range as the coastal valleys from Bakersfield and Weldon, Kern County, north to Redding, Shasta County. Small populations were also found in Northern California along the Shasta River, Siskiyou County, and in Surprise Valley, Modoc County. Populations were also found in suitable habitat east of the Sierra Nevada in the Owens Valley and along the Colorado and Mojave Rivers.

Currently in California recent surveys have shown that the only two primary breeding populations that have persisted each year are along the Sacramento River and along the South Fork of the Kern River (Laymon and Williams 1999). The population on the Sacramento River ranged between 23 and 35 pairs from 1987 to 1990 (Halterman 1991), while the South Fork Kern River population has varied between 2 and 24 pairs from 1985 to 1999.

Small (1 to 4 pairs) populations of cuckoos that breed or possibly breed also occur at the following locations in California: Feather River (from Oroville to Verone; Butte, Yuba and Sutter Counties), Prado Flood Control Basin, Amargosa River near Tecopa, Inyo County, Owens Valley near Lone Pine and Big Pine, Inyo County, Mojave River near Victorville, San Bernardino County; and Colorado River from Needles, San Bernardino County, to Yuma, Imperial County (Laymon and Halterman 1987, Laymon and Williams 1999, S. Laymon pers. comm.). None of these areas are occupied every year (S. Laymon, pers.comm.).

The current distribution of the species range within National Forests System (NFS) lands in the planning area is limited to the Sequoia National Forest on the Greenhorn Ranger District (S. Laymon pers.comm.). The breeding habitat on NFS lands is located within the South Fork Wildlife Area, a 1,200-acre riparian forest that is located at the confluence of the South Fork of the Kern River and Isabella Reservoir. National Forest lands along the South Fork of the Kern has supported between 1 and 11 pairs of yellow-billed cuckoos each year between 1985 and 1996, with the average of 5 pairs each year (Laymon and others 1997). With the average number of pairs for the entire survey area in the South Fork Kern River being 10.4 pairs (Laymon and Williams 1999), the South Fork Wildlife Area supports, on average, approximately 50 percent of the South Fork population.

FEIS Volume 3, Chapter 3, part 4.4, page 203 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Risk Factors The principle factor that places this species at risk is habitat loss and fragmentation. From 1977 to 1987, yellow-billed cuckoo populations declined by 65 to 96 percent in California due to massive loss of riparian gallery forest (less than 1 percent of original forests remaining). Eggshell thinning also threatened the species during the time when DDT was used (Gaines and Laymon 1984, Laymon and Halterman 1987).

Habitat risk factors within the control of the Forest Service include protection of the remaining breeding habitat within the South Fork Wildlife Area of the Greenhorn Ranger District, Sequoia National Forest from the effects that could be caused by livestock grazing. Currently the South Fork Wildlife Area is fenced and is not utilized for grazing. Trespass livestock from adjacent private property occasionally get into the South Fork Wildlife Area. Improved fence maintenance on the east and west sides of the area would help reduce this problem. There are no other risks to the habitat on NFS lands at this time that are within the control of the Forest Service.

Habitat risk factors that are not within the control of the Forest Service include the operation of Isabella Reservoir. On high water years the Reservoir can flood habitat for the yellow-billed cuckoo, making it unavailable for that year and with prolonged flooding rendering the habitat unavailable permanently. It is also likely that flooding in wet years reduces the survival of the larvae of the preferred yellow-billed cuckoo prey (katydids and sphinx moth) that winter underground (Laymon pers. obs.). This forces the cuckoos to forage in upland areas that were not flooded until the prey base in the lower floodplain begins to recover later in the breeding season (Laymon and Williams 1999). The fact that most of the key riparian breeding habitat for the yellow-billed cuckoo is in the primary floodplain of the South Fork Kern River could cause a large reduction in the prey base and be a risk factor for the populations breeding there (Laymon and Williams 1999).

Non-habitat risk factors that are within the control of the Forest Service include recreational use within breeding area on the South Fork Wildlife Area of the Sequoia National Forest. Recreation use in the area is currently low impact, with personal watercraft only allowed to utilize the area at speeds less than 5 miles per hour. Recently access roads that led into the South Fork Wildlife Area were closed to vehicles, allowing for terrestrial travel walking access only.

Non-habitat risk factors that are not within the control of the Forest Service include the hunting of Mourning doves in the South Fork Kern Valley. Mourning doves and the yellow-billed cuckoos have similar profiles and could be confused, especially in early morning and late evening light. The fall migration period for the yellow-billed cuckoo coincides with the early part (September 1 to 15) of mourning dove hunting season. During this migration period the cuckoo is more likely to be active in areas on the edge of its preferred habitat, which would be similar areas utilized by mourning doves. There is a definite risk of a yellow-billed cuckoo being confused for a mourning dove and shot.

Airborne drift of pesticides and herbicides is not considered to be high in the South Fork Valley area; however, it is possible that airborne drift of toxic chemicals from pesticides and herbicides could occur near yellow-billed cuckoo nesting areas further upstream. What the effect of this would be on the cuckoo is not known.

Another potential non-habitat related risk factor is casualties related to television towers during the yellow-billed cuckoo’s migration period (S.Laymon pers.comm.). The yellow-billed cuckoo migrates to South America, making its travel direction south of the Sierra Nevada Bioregion. While there may

FEIS Volume 3, Chapter 3, part 4.4, page 204 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 be television towers on NFS lands that contribute to this risk factor, it would not be applicable to the planning area being analyzed for the Sierra Nevada Forest Plan Amendment Project. It is not known what the effects of television tower casualties has had and will continue to have on this species.

Conservation Measures The only breeding population of this species that occurs on NFS lands within the Sierra Nevada National Forests occurs in the Sequoia National Forest’s South Fork Wildlife Area. This area is a large riparian forest at approximately 2,000 feet in elevation, and is not directly affected by the actions being analyzed under the Sierra Nevada Forest Plan Amendment Project. The South Fork Wildlife Area was transferred to the Sequoia National Forest in 1990 as part of a land transfer between the Forest Service and the Army Corps of Engineers. The area is designated as a State Wildlife Area and is currently being managed for the conservation of wildlife.

Monitoring of the two remaining California yellow-billed cuckoo populations is important to continue to determine effective population sizes necessary for future conservation programs (Hughes 1999). There is no Federal protection for the western population in spite of support by scientists throughout the species range for listed status.

The Sequoia National Forest could develop management guidelines for the South Fork Wildlife Area, where this species occurs on this Forest including guidelines that focus on the long-term conservation of this species.

Environmental Outcomes Historic Condition. The yellow-billed cuckoo is known to inhabit deciduous riparian thickets or forests with dense, low-level or understory foliage. Willow is almost always a dominant component of the vegetation in breeding habitat. In California, the yellow-billed cuckoo has a very strong preference for nesting in willows (96 of 97 nests), even when cottonwoods and other tree species are present (Laymon and others 1997).

Current Condition. The only breeding population of this species that occurs on NFS lands within the Sierra Nevada National Forests occurs in the Sequoia National Forest’s South Fork Wildlife Area. This area is a large riparian forest at approximately 2,000 feet in elevation, and is not directly affected by the actions being analyzed under the Sierra Nevada Forest Plan Amendment Project.

FEIS Volume 3, Chapter 3, part 4.4, page 205 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Table 4.4.2.5a. Average assessment ratings for the California yellow-billed cuckoo. Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Outcome D D D D D D D D D D

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Cumulative Effects Population Outcomes Historic Conditions. Historically, the yellow-billed cuckoo was a common breeding species in riparian habitat throughout much of lowland California (Grinnell and Miller 1944). Grinnell and Miller (1944) described the cuckoo’s range as the coastal valleys from Bakersfield and Weldon, Kern County, north to Redding, Shasta County. Small populations were also found in Northern California along the Shasta River, Siskiyou County, and in Surprise Valley, Modoc County. Populations were also found in suitable habitat east of the Sierra Nevada in the Owens Valley and along the Colorado and Mojave Rivers

Current Condition. The only breeding population of this species that occurs on NFS lands within the Sierra Nevada National Forests occurs in the Sequoia National Forest’s South Fork Wildlife Area. This area is a large riparian forest at approximately 2,000 feet in elevation, and is not directly affected by the actions being analyzed under the Sierra Nevada Forest Plan Amendment Project.

Table 4.4.2.5b represents the estimated population outcomes through the planning horizon for the California yellow-billed cuckoo. Alternative Current 1 2 3 4 5 6 7 8 Mod 8 Outcome D D D D D D D D D D

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

FEIS Volume 3, Chapter 3, part 4.4, page 206 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.3. Amphibians AQUATIC, RIPARIAN, AND MEADOW ASSOCIATES

Widely distributed and “Notice of Intent” Forest Service Sensitive species received a full individual analysis in the FEIS that included evaluating their ecological needs, especially focusing on habitat associations, and identifying threats to survival. The FEIS alternatives and Standards and Guidelines were then reviewed to determine how they would accommodate the species’ needs and reduce threats to survival. For each of the key management activities identified for each species, there follows a short narrative summary of the differences among FEIS alternatives 2 through 8 and modified 8. An overall qualitative summary of the risks of all alternatives to species viability is also provided. For details of this analysis and a qualitative evaluation of the Standards and Guidelines, see Tables 4.3.3.1a and 4.4.3.6a. Forest Service sensitive species that occur on only one or two national forests received limited analyses in which they are grouped by habiatat type.

WIDELY DISTRIBUTED AND N.O.I. SPECIES

4.4.3.1. FOOTHILL YELLOW-LEGGED FROG Affected Environment Species Background The foothill yellow-legged frog (Rana boylii) historically occurred in foothill and mountain streams from northern Baja California to southern Oregon west of the Sierra-Cascade crest to 6000 ft. (1830m) (Table 4.2.4b). This species has experienced significant populations declines especially in the southern part of its range (southern Sierra Nevada and south coastal California) and is currently listed as a California State Species of Special and Forest Service California Region Sensitive Species. In the Sierra Nevada, it apparently has disappeared from 66% of its historic range (Jennings 1996).

Risk Factors In the biphasic life cycle (aquatic eggs and tadpoles; semi-aquatic adults) of this species, tadpoles are primarily herbivores, grazing on algae in the aquatic environment, and adults are invertivores, actively capturing both aquatic and terrestrial insects and arachnids. Cover requirements also differ by life history stage. Eggs and tadpoles require course rocky substrates in streams and rivers with water depths, water velocities, and water temperatures falling into a relatively narrow range (Hayes and Jennings 1988, Fuller and Lind 1992, Kupferberg 1996, Lind and Welsh unpublished data). Adults use both in stream and riparian environments, though use of riparian areas (and adjacent uplands) is poorly understood. The primary impacts to habitat the foothill yellow-legged frog are activities that affect water flow regimes (volume, depth, velocity, temperature) and sediment regimes (channel morphology, fine sediment loads) and thus could alter oviposition and rearing habitats. These activities include dams/diversions, vegetation management and mechanical fuel treatment, roads (especially in riparian areas), livestock grazing, and mining (Lind et. al. 1996). For a detailed discussion of the historic and current conditions of aquatic, riparian, and meadow environments in the Sierra Nevada see Part 3.6.

Calling, which is used to initiate contact between males and females for breeding, and the condition of breeding habitat, are the primary factors influencing mating success and reproduction. The primary impacts to foothill yellow-legged frog reproduction are any management activities or human disturbance that through noise production could cause disruption of male/female communication

FEIS Volume 3, Chapter 3, part 4.4, page 207 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 during the breeding season (March through June, depending on locale). Because this species calls primarily underwater, activities occurring within stream channels, such as suction-dredge mining, would be of primary concern.

The above activities primarily influence the habitat or reproductive behavior of foothill yellow-legged frogs. There are also activities that may directly affect this species. One major management activity that has been documented by recent research is off-seasonal high flow releases from dams. The flows can dislodge eggs masses and tadpoles and result in no or lower survival for these life history stages (Lind et. al. 1996). Other potential direct impacts that were considered in this risk analysis: (1) increases in fine sediments which can affect oxygen uptake by eggs, (2) recreation which may cause direct disturbance or trampling (i.e. pack stock) of frogs, eggs, and tadpoles, (3) toxins (herbicides, pesticides, fertilizers) which can have both lethal and sub-lethal effects (Berrill et. al.1997), and (4) livestock grazing which may result in direct trampling of frogs, eggs, and tadpoles. While research on environmental toxin effects on this species has not yet been conducted, closely related species in other regions have show sensitivity to numerous pesticides, herbicides, and fertilizers (Berrill et. al. 1993, Berrill et. al. 1997, LeBlanc and Bain 1997). Because these chemicals are thought to disrupt endocrine systems in amphibians at low concentrations, application of pesticides and herbicides are considered to be a risk factor for this species.

For foothill yellow-legged frogs, the key management activities which the Forest Service can influence are: dams and diversions (through the FERC re-licensing process), mining, livestock grazing, recreation, vegetation management and mechanical fuel treatment, roads, and locally applied chemical toxins (e.g. pesticides and herbicides). In addition, this FEIS proposes substantial prescribed burning at lower elevations. Fire can directly affect amphibians (mortality) as well as alter their habitat. The costs and benefits of burning are not well understood for the foothill yellow-legged frog, so an evaluation of the potential effects of these activities is provided.

Conservation Measures This is a Forest Service Sensitive Species, so currently it must be evaluated in Forest level Biological Evaluations. Through this process, appropriate mitigations are identified, and may be implemented at the discretion of the district ranger. The “Long-term Strategy for Anadromous Fish-producing Watersheds” should provide for protection of habitat and populations on the Lassen National Forest.

In this FEIS, there is direction (Alternative 8 modified) to conduct a Conservation Assessment for foothill yellow-legged frog. Critical Aquatic Refuges (CAR’s) are established for the most at-risk populations of foothill yellow-legged frogs on Forest Service lands, and the decision allows for the addition of new CAR’s as more information is compiled (FEIS Appendix I).

Environmental Consequences Effects of Alternatives For the foothill yellow-legged frog, the following evaluation of standards and guidelines among FEIS alternatives for each risk factor considers both current known sites and historic range that is currently unoccupied.

Chemical Toxins (Locally Applied Pesticides and Herbicides). All Alternatives contain direction to avoid pesticide/herbicide application within 500 feet of known foothill yellow-legged frog sites as well as at other TES species sites. While this approach addresses local applications, it does not address possible drift or downstream movements of these chemicals. It thus provides some

FEIS Volume 3, Chapter 3, part 4.4, page 208 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 improvement over existing direction. Alternatives 8 and modified 8 also include prohibition of livestock pesticides in riparian areas which would likely lower risk to amphibians. Alternative 5 provides additional direction to avoid use in any riparian areas, Aquatic Diversity Areas, Critical Aquatic Refuges, and if used in “green zones” make them ground-based and vegetation-specific. This alternative is thus lowest risk for all aquatic/riparian amphibian species.

Dams and Diversions. All alternatives provide the same baseline standard and guideline that states that the FERC re-licensing process should be used to provide water flows to protect and restore aquatic and riparian resources and move toward Aquatic Management Strategy (AMS) goals. Alternative modified 8 also includes direction to evaluate the natural hydrograph during re-licensing and to ensure that exempt hydroelectric projects maintain instream flows for aquatic species. Alternative 5 also adds a caveat that comparisons to undammed stream systems be used in evaluating re-licensing projects and that National policies on protecting free-flowing rivers be implemented making it somewhat stronger than the rest of the alternatives. Since this particular management issue is of particular concern to the foothill yellow-legged frog, strong direction is warranted here regarding operation of dams and diversion projects so that flow levels result restoration of aquatic habitats and potentially damaging off-seasonal releases are eliminated.

Livestock Grazing. Many of the standards and guidelines incorporate grazing utilization limits for grasses or shrubs. While these may help maintain certain structural features required by amphibians, there are direct impacts to frogs, eggs, and tadpoles (e.g. trampling, degradation of aquatic microhabitats) that are at least of equal concern. Research is also needed to determine how variation in percent utilization benefits or impacts amphibians. In all Alternatives it is possible that varying degrees of livestock exclusions from willow flycatcher habitats may benefit foothill yellow-legged frogs, where these species ranges overlap. Alternatives 4 and 5 also contain limitations on livestock grazing in riparian areas and wet meadows that would benefit foothill yellow-legged frogs. Alternatives 3, 6, 7, 8, and modified 8 include recommendations for moving livestock handling/gathering facilities outside of riparian/meadow areas and providing off-channel watering devices that would reduce, though not eliminate affects on local amphibians. Alternative modified 8 also contains standards for limiting streambank and lake shore disturbance by all activities (not just livestock grazing) to 20% Sierra Nevada wide and 10% in areas with listed fish species.

Mining. Mining standards and guidelines in all alternatives provide some improvement over existing conditions and have special emphasis on reducing impacts to riparian areas. Alternatives 2,5, 6, 8, and modified 8 propose consideration of mineral withdrawals from Critical Aquatic Refuges that will provide added protection for aquatic habitats that harbor amphibians. Alternative 5 provides additional protection for amphibians and other riparian/aquatic species by allowing approval of mining “plans of operations” in Aquatic Diversity Areas and Critical Aquatic Refuges only if AMS goals are met. Alternative modified 8 also prioritizes reclamation of mined areas in Riparian Conservation Areas and Critical Aquatic Refuges. However, perhaps due to legal limitations, none of these standards provide the option of reducing current mining activities (especially suction-dredge mining) and thus the risk to stream-dwelling amphibians, like the foothill yellow-legged frog remains high across all alternatives.

Prescribed Fire. All Alternatives allow fire to “back in” to riparian areas though the extent of habitat that can be burned varies among different standards and guidelines. Alternative 2 prohibits fire within 500 feet of foothill yellow-legged frog sites. Alternatives 3, 4, 6, and 7 allow varying amounts of suitable habitat for this species to be burned each year, Alternative 7 allows the lowest (5%)

FEIS Volume 3, Chapter 3, part 4.4, page 209 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

amount, Alternative 6 allows a moderate level (15%) and Alternatives 3 and 4 allow the highest (25%) suitable habitat that can be burned prescriptively each year. Alternatives 4, 8 and modified 8 provide direction for conducting site-specific analysis in CAR’s and TES habitats that considers both the timing and extent of burns of prior to management. Given that the costs and benefits of burning are not clearly understood for amphibians, the lowest risk alternatives for foothill yellow-legged frogs are Alternative 8 and modified 8 which propose site-specific evaluation of timing and extent prior to burning.

Recreation. All Alternatives contain direction to assess the use and impacts of developed and dispersed recreation sites, trails, OHV trails, and access roads and consider rehabilitation, relocation, or other measures if AMS goals are not advanced or water quality and aquatic habitat objectives (i.e. Riparian Conservation Objectives) cannot be met. This approach will be helpful in protecting important aquatic habitats for all amphibian species. Alternatives 3 and modified 8 also contain direction that prohibits development new off-road vehicle sites in riparian areas unless no other option exists. This would benefit all amphibian species by protecting important aquatic habitats from direct impacts (noise, trampling, etc.).

Roads. All Alternatives contain improvements regarding evaluation of existing roads and improvements of stream crossings that will benefit aquatic amphibians. Modified Alternative 8 provides broad direction to improve conditions of existing roads, close roads at times of the year when use is low, and decommission them if they are causing unacceptable environmental effects. Alternative 8 also emphasizes road de-commissioning in Critical Aquatic Refuges, which would benefit aquatic species occurring in those areas. Alternatives 3, 6, 8, and Modified 8 emphasize reduction of roads in riparian areas. Alternative 5 includes standards for removal of roads in areas where they are high risk to aquatic ecosystems. Implementation of these strategies would help reduce fine sediments in streams and maintain watershed functions, thus improving conditions for foothill yellow-legged frogs.

Vegetation Management and Mechanical Fuel Treatment. All alternatives contain some limitations on vegetation management and mechanical fuel treatment in riparian areas. Alternative 4 provides the least direction. The definitions of riparian areas or zones differ among Alternatives with 2, 3, 4, and 5 using variable widths based on SNEP approach and Alternatives 6, 7, 8, and modified 8 using stream-type flexible widths (e.g. 300 feet along perennial streams and large lakes/ponds, 150 ft along each side of seasonally-flowing streams). Variable width (SNEP type) zones typically provide relatively wide overall Riparian Areas, though different alternatives allow different activities within the “green and grey zones” of these areas. Alternatives 2 and 5 prohibit timber harvest or mechanical fuel treatment in “green zones” (typically 150 feet wide), while Alternatives 3 and 4 appear to allow these activities. Alternatives 6, 7, and 8 prohibit timber harvest in riparian areas (stream-type flexible). Salvage logging is prohibited in riparian areas in Alternatives 2, 3, 4, 5, 6, 7, and 8. Additional direction in Alternatives 2, 4, 5, 6 ,7, 8, and modified 8 limits or mitigates for heavy equipment operation near occupied amphibian sites. Alternative modified 8 allows thinning and other vegetation management (including salvage logging) activities only if it is in line with Riparian Conservation Objectives. In addition landscape analyses and associated peer reviews will serve to provide feedback to managers in Alternative modified 8. Overall, Alternatives 2, 5, 6, 7, 8, and Modified 8, would all reduce impacts to riparian/aquatic habitats and amphibian species to about the same degree. Alternatives 3 and 4 appear to be the highest risk.

FEIS Volume 3, Chapter 3, part 4.4, page 210 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Overall Evaluation of Alternatives. For foothill yellow-legged frogs, Alternatives 2, 5, and modified 8 appear to be the lowest risk and the most effective management approaches for this species persistence and recovery. All the other alternatives appear to have less protection for this species. However, information/research gaps exist, especially in the realm of grass/shrub utilization standards for livestock grazing and their effects on amphibians.

Environmental Outcome Table 4.4.3.1a. Average assessment ratings for the foothill yellow-legged frog. Ratings represent average degree of confidence in each outcome being realized 50 years in the future. (See Chapter 4, Part I.) Total score for each alternative is 100.

Alternative Outcome Current 1 2 3 4 5 6 7 8 Mod 8 A 0 0 0 0 0 0 0 0 0 0 B 0 0 0 0 0 0 0 0 0 0 C 45 12 48 52 38 65 48 45 48 48 D 55 58 42 30 52 35 50 50 50 50 E 0 30 10 18 10 0 2 5 2 2

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Rationale. The foothill yellow-legged frog occurs primarily in lower elevation riparian ecosystems. This species has been extirpated from an estimated 66 percent of its historic range due principally to water and hydroelectric development, grazing, and urbanization that adversely affect sediment and stream flow regimes. Suitable environments for this species generally exist at low abundance and are highly isolated or occur in patches.

Based upon the available information, water development has been the most significant limiting factor for the foothill yellow-legged frog. Water developments have very likely restricted population interaction between riparian systems and adversely affected sediment and stream flow regimes within riparian systems. The Forest Service has little control over water licensing activities; the Forest Service is able to influence water development only to the extent that negotiations with FERC are effective in mitigating adverse effects associated with water developments. Thus, little significant overall difference exist between the alternatives.

Beyond the factor of FERC licensing, other secondary limiting factors provided some discrimination among alternatives. Alternative 1 lacks an aquatic management strategy (AMS) or grazing restrictions; this alternative would likely result in highly isolated populations occurring in very low abundance. Alternative 5 would likely provide some improvement in the distribution and abundance of suitable environments in 50 years through grazing restrictions, particularly in all sites occupied by

FEIS Volume 3, Chapter 3, part 4.4, page 211 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 at-risk amphibians during the past 25 years. Conditions under Alternatives 2 and 8 (and modified 8) would likely remain stable relative to the current condition through application of the AMS, and grazing restrictions in willow flycatcher habitats. Of particular note are the potentially substantial protections provided under Alternative 2.

Cumulative Effects In addition to the Forest Service risk factors discussed above, the following non-Forest Service risk factors have contributed and may still be contributing to the decline of the foothill yellow-legged frog: climate change, flood, and drought (Table 4.3.3.1a). Data limitations preclude a quantitative cumulative effects analysis for this species. However, given the number of risk factors under Forest Service influence, this agency can play a significant role in reduction of many of threats to the species.

FEIS Volume 3, Chapter 3, part 4.4, page 212 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.3.2. MOUNTAIN YELLOW-LEGGED FROG Affected Environment Species Background The mountain yellow-legged frog (Rana muscosa) historically inhabited ponds, tarns, lakes, and streams from 4,500 to over 12,000 ft. (1370 to over 3650 m) (Stebbins 1985) and was once the most common amphibian in high elevation aquatic ecosystems of the Sierra Nevada (Bradford et. al. 1993, Grinnell and Storer 1924). This species is endemic to California and a small area of western Nevada and occurs in two distinct regions – the Sierra Nevada and several mountain ranges of coastal southern California. Recent genetic analysis indicates that southern Sierra Nevada populations are closely related to coastal southern California populations and that divergence levels are high throughout the species range (Macey et. al. In Press). These results have important implications for management of the species (i.e. portions of the Sierra Nevada may need to be treated as distinct management units). Mountain yellow-legged frogs range widely throughout the Sierra Nevada from northern Plumas to southern Tulare counties and it was abundant at many sites into the early 1960s (Table 4.2.4b). Large groups of populations in the northern Sierra Nevada and local populations elsewhere have since become extinct and have disappeared from 70-90% of its historic range in the bioregion (Jennings 1996, U.S. Fish and Wildlife Service 2000c). It is currently a California Species of Special Concern, Forest Service Sensitive Species, and a recent petition to list the species has been accepted by the US Fish and Wildlife Service.

Risk Factors In the biphasic life cycle (aquatic eggs and tadpoles; semi-aquatic adults) of this species, tadpoles are primarily herbivores, grazing on algae, diatoms, and detritus in the aquatic environment, and adults feed primarily on invertebrates but also take tadpoles of other frogs (especially treefrogs, western toads, and Yosemite toads) (Jennings and Hayes 1994, Pope 1999a). Cover requirements also differ by life history stage and seasonally. Not much is known about egg deposition site requirements, but one study found that eggs were deposited in lakes and were attached to bedrock or emergent vegetation in relatively shallow water (0.2 m) (Pope 1999b). With a short summer activity period, and relative cool summer air temperatures, egg placement and larval behavior appears to reflect an affinity for the warmest regions of lakes (Pope 1999b). Adults are highly aquatic and are typically associated with nearshore areas for reproduction, cover, foraging, and over-wintering. They are most abundant along lake shores and low gradient streams with irregular shores and rocks (Mullaly and Cunningham 1956). Overland movements up to 66m occur when frogs move from one body of water to another for breeding, overwintering, and feeding (Pope 1999b). Over-wintering habitat condition is important for both tadpoles and adults; tadpoles do not metamorphose in their first year and in fact may spend 2-3 winters in aquatic habitats. During winter dormancy, tadpoles can withstand anoxic conditions, while some sub-adult and adult frogs have a lower tolerance and seek winter cover in deep granite crevices along the lake shorelines (Matthews and Pope 1999). Other studies have suggested that adult frogs use the deepest sections of lakes for overwintering (Bradford 1983). While direct habitat degradation has not been cited as a cause of declines of this species, the above characteristics are important for the species survival. Recent and projected future increases in recreational use of the national forests may also affect this species through direct disturbance, trampling by humans and packstock, and habitat degradation. Very little is known about the mating system of this species, though calling is thought to occur primarily underwater and breeding seems to be highly synchronized (Ziesmer 1997, Pope 1999b). Activities occurring within breeding areas, such as recreation, would be of primary concern. For a detailed discussion of the historic and current conditions of aquatic, riparian, and meadow environments in the Sierra Nevada see Part 3.6.

FEIS Volume 3, Chapter 3, part 4.4, page 213 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Introduction of exotic predatory fish in aquatic habitats used by this species is likely the primary reason for the species' decline (Bradford, et. al. 1998, Knapp and Matthews 2000, Matthews and Knapp 1999). Exotic fish have been introduced into previously fish-less lakes and streams throughout the range of the mountain yellow-legged frog and evidence suggests that frogs have declined substantially in areas with fish introductions (Knapp 1996). Fish prey on tadpoles, sub- adults and adults of this species and may also be vectors of disease. Measures that are taken to eliminate fish stocking from some basins, if done with knowledge of the distribution of mountain yellow-legged frog populations, would benefit this species. A recently discovered Chytrid fungus may be affecting a number of mountain yellow-legged frog populations but research on this subject is just beginning. While research on environmental toxin affects on this species has not yet been conducted, closely related species in other regions have show sensitivity to numerous pesticides, herbicides, and fertilizers (Berrill et. al. 1993, Berrill et. al. 1997, LeBlanc and Bain 1997). Because these chemicals are thought to disrupt endocrine systems in amphibians at low concentrations, application of pesticides and herbicides are considered to be a risk factors for this species.

For the mountain yellow-legged frog, the key management activities that the Forest Service can influence are: exotic fish stocking, pack stock use and access, recreation, and locally applied chemical toxins (e.g. pesticides and herbicides).

Conservation Measures There is currently an interagency team of biologists developing a Conservation Strategy for the mountain yellow-legged frog. A draft of this strategy is now available, with a final version expected in early 2001 (A. Carlson, pers. comm.). This is also a Forest Service Sensitive Species, so currently it must be evaluated in Forest level Biological Evaluations. Through this process, appropriate mitigations are identified, and may be implemented at the discretion of the district ranger.

In this FEIS, there is direction (Alternative 8 modified) to implement recovery plans for listed species as funds allow, and to conduct a Conservation Assessment for the mountain yellow-legged frog. Critical Aquatic Refuges (CAR’s) are established for several known populations of mountain yellow- legged frogs on Forest Service lands and the decision allows for the addition of new CAR’s as more information is gained (FEIS Appendix I).

Environmental Consequences Effects of Alternatives For the mountain yellow-legged frog, the following evaluation of standards and guidelines among FEIS alternatives for each risk factor considers both current known sites and historic range that is currently unoccupied.

Chemical Toxins (Locally Applied Pesticides and Herbicides). All Alternatives contain direction to avoid pesticide/herbicide application within 500 feet of known mountain yellow-legged frog sites as well as at other TES species sites. While this approach addresses local applications, it does not address possible drift or downstream movements of these chemicals. It thus provides some improvement over existing direction. Alternatives 8 and modified 8 also include prohibition of livestock pesticides in riparian areas which would likely lower risk to amphibians. Alternative 5 provides additional direction to avoid use in any riparian areas, Aquatic Diversity Areas, Critical Aquatic Refuges, and if used in “green zones” make them ground-based and vegetation-specific. This alternative is thus lowest risk for all aquatic/riparian amphibian species.

FEIS Volume 3, Chapter 3, part 4.4, page 214 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Exotic fish stocking. All alternatives address this issue in some manner, from simply supporting AMS goals to focusing on the development of a Conservation Assessment/Strategy for mountain yellow-legged frogs (Alternatives 2, 4, 6, 7, 8, and modified 8), to direction that emphasizes removal (with CDFG cooperation) of exotic fish from some mountain yellow-legged frog areas (Alternatives 3, 5, 8, and modified 8). Alternatives 8 and modified 8 appear to provide the most effective approach for conserving mountain yellow-legged frogs, including both the development of a conservation assessment/strategy and emphasis on removal of exotic fish from some areas.

Pack stock. Alternatives 3, 6, 7, 8, and modified 8 include recommendations for moving livestock handling/gathering facilities outside of riparian/meadow areas and providing off-channel watering devices that would reduce, though not eliminate affects on local amphibians. Alternatives 8 and modified 8 contain specific direction that limits packstock use of Yosemite toad breeding sites from during the breeding and rearing season at least until September 1 unless the meadow is dry and breeding is not possible. These standards may also reduce risk to mountain yellow-legged frogs. It is also possible that evaluations of dispersed recreation sites would identify impacts of pack stock.

Recreation. All Alternatives contain direction to assess the use and impacts of developed and dispersed recreation sites, trails, OHV trails, and access roads and consider rehabilitation, relocation, or other measures if AMS goals are not advanced or water quality and aquatic habitat objectives (i.e. Riparian Conservation Objectives) cannot be met. This approach will be helpful in protecting important aquatic habitats for all amphibian species. Alternatives 3 and modified 8 also contain direction that prohibits development new off-road vehicle sites in riparian areas unless no other option exists. This would benefit all amphibian species by protecting important aquatic habitats from direct impacts (noise, trampling, etc.).

Overall Evaluation of Alternatives. For mountain yellow-legged frogs, Alternatives 3, 5, 8, and modified 8 appear to be the lowest risk and provide the most effective management approaches for this species persistence and recovery. All the other action alternatives appear to provide less improvement for the species. Alternative 1 (no action) would result in the least benefits to the species.

FEIS Volume 3, Chapter 3, part 4.4, page 215 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Environmental Outcome Table 4.4.3.2a. Average assessment ratings for the mountain yellow-legged frog. Ratings represent average degree of confidence in each outcome being realized 50 years in the future. (See Chapter 4, Part 1.) Total score for each alternative is 100.

Alternative Outcome Current 1 2 3 4 5 6 7 8 Mod 8 A 0 0 0 0 0 0 0 0 0 0 B 0 0 0 0 0 0 0 0 0 0 C 10 0 25 32 2 38 20 20 25 25 D 72 52 70 68 70 62 70 70 68 68 E 18 48 5 0 28 0 10 10 7 7

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Rationale. The mountain yellow-legged frog historically inhabited ponds, lakes, and streams from 4,500 to 12,000 feet. The mountain yellow-legged frog was the most common amphibian in high- elevation, aquatic ecosystems in the Sierra Nevada. The primary threat to mountain yellow-legged frogs is predation by exotic trout that are stocked in high-elevation aquatic systems. While aquatic habitats for the mountain yellow-legged frog are widely distributed, environments that can actually support the frog were judged to be highly isolated.

Primary features of the alternatives that would influence outcomes for mountain yellow-legged frogs are proposed amphibian reserve systems, the aquatic management strategy, commitments to work with California Fish and Game to remove trout areas occupied by frogs, and development of a conservation assessment/strategy specifically for the frogs. Under Alternatives 2, 3, 5, 6, 7, and 8 (and modified 8), the AMS calls for fish stocking programs to be brought in line with the goals of the strategy. Alternative 2 would provide for an amphibian reserve system. This system would protect all known sites of the mountain yellow-legged frog. Alternatives 3 and 5 would protect all known occupied mountain yellow-legged frog habitat, based on records over the last 25 years. Alternatives 3, 5, and 8 (and modified 8) specifically call for cooperative efforts between the Forest Service and California Fish and Game to remove fish from some sites occupied by mountain yellow-legged frogs. Alternatives 2, 6, and 7 call for the development of a conservation assessment/strategy for mountain yellow-legged frogs, but have no specific commitment to implement the strategy.

Based on these provisions, Alternatives 1 and 4 were judged to lead to continuing deterioration in environmental conditions for mountain yellow-legged frogs, with an increased likelihood that conditions would lead to population isolation and extirpation. Alternatives 3 and 5 were judged to have most positive consequences, with a decreasing likelihood that conditions would lead to extirpation and an increasing likelihood that conditions would provide for less isolated populations.

FEIS Volume 3, Chapter 3, part 4.4, page 216 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Alternatives 2, 6, 7, and 8 (and modified 8) were judged to be intermediate, with some likelihood of improving conditions and providing for less-isolated mountain yellow-legged frog populations.

Cummulative Effects In addition to the Forest Service risk factors discussed above, one additional non-Forest Service risk factor, disease, may be contributing to the decline of the mountain yellow-legged frog (Table 4.4.3.6a). However, the Forest Service may be contributing to this factor by inadvertently transporting the disease on field equipment used in stream and lake environments. Data limitations preclude a quantitative cumulative effects analysis for this species. However, given the number of risk factors under Forest Service influence this agency can play a significant role in reduction of many of threats to the species.

FEIS Volume 3, Chapter 3, part 4.4, page 217 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.3.3. YOSEMITE TOAD Affected Environment Species Background The Yosemite toad (Bufo canorus) is endemic to the Sierra Nevada mountain range and occurs in wet montane meadows from the Blue Lakes region north of Ebbetts Pass in Alpine County south to Kaiser Pass area in the Evolution Lake/Darwin Canyon region of Fresno Co. (Table 4.2.4b). Its known elevational range extends from ca. 6,435 to ca. 11,385 ft (ca. 1950 to ca. 3450 m). Recent work on genetic variation in populations fromYosemite and Kings Canyon National Parks indicates high genetic variation between these two areas and recommends treating them as different management units (Shaffer et. al. 2000). In addition, evidence of genetic differences were seen between major drainages (Yosemite only) and among breeding ponds, suggesting that management at the pond level is warranted (Shaffer et. al. 2000). As of the mid-1990’s it had declined substantially or disappeared from over 50% of the sites where it was known historically (Jennings 1996, U.S. Fish and Wildlife Servic 2000d) and it is currently a California State Species of Special Concern, Forest Service Sensitive Species, and a recent petition to list the species has been accepted by the US Fish and Wildlife Service.

Risk Factors In the biphasic life cycle (aquatic eggs and tadpoles; semi-aquatic adults) of this species, tadpoles are primarily herbivores, grazing on detritus, suspended plant material also zooplankton in the aquatic environment while adults feed primarily on invertebrates (Jennings and Hayes 1994). Yosemite toads are found in high montane and subalpine associations in relatively open wet meadows (standing water of at least 1/10th acre on June 1) surrounded by forests of lodgepole pine or whitebark pines and are primarily active during the late spring, summer, and early fall. (Zeiner, et. al. 1988). Meadows with willow vegetation are also used, a indicating possible overlap in habitat associations with the willow flycatcher. At appropriate elevations, management activities that benefit one species may also benefit the other, but this needs more investigation. Suitable breeding sites are generally found at the edges of meadows or in slow-flowing runoff streams. Short emergent sedges or rushes often dominate such sites (Jennings and Hayes 1994, D. Martin, pers. comm.). Water depth and temperature appear to be important limiting factors in the survival of eggs and tadpoles (Kagarise Sherman and Morton 1993, D. Martin, pers. comm.). Breeding choruses of males begin soon after emergence from winter hibernacula and mating is polygynous (Kagarise Sherman 1980, Zeiner, et. al. 1988). Adults utilize both aquatic and terrestrial environments for foraging and cover. Springs, upslope from meadows, and rodent burrows are two features that appear to be important for adult dispersal and over wintering habitat (Kagarise Sherman 1980, D. Martin, pers. comm.). For a detailed discussion of the historic and current conditions of aquatic, riparian, and meadow environments in the Sierra Nevada see Part 3.6.

Potential impacts to this species and its habitat include livestock grazing, drought, chemical toxins, and possibly recent increases in UV radiation. Of these, the primary factors that Forest Service management can influence are chemical toxins, livestock grazing, and fish stocking. Livestock use of wet meadow habitats can affect this species both directly, through trampling of individuals, and indirectly through: (1) changes to meadow stream hydrology and morphology (increased down- cutting and head-cutting), (2) increased siltation of springs, and (3) changes in micro-topography of egg deposition and larval rearing areas resulting in lowered survival rates of these life stages (D. Martin, pers. comm). While research is lacking on the affects of exotic predatory fish introductions on Yosemite toads, there is some evidence to indicate that toad populations may have been more

FEIS Volume 3, Chapter 3, part 4.4, page 218 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 abundant in pond and lake environments than they are now. Measures that are taken to eliminate fish stocking from some basins may benefit Yosemite toads as well as mountain yellow-legged frogs. While research on environmental toxin affects on this species has not yet been conducted, closely related species in other regions have show sensitivity to numerous pesticides, herbicides, and fertilizers (Berrill et. al. 1993, Berrill et. al. 1997, LeBlanc and Bain 1997). Because these chemicals are thought to disrupt endocrine systems in amphibians at low concentrations, application of pesticides and herbicides are considered to be risk factors for this species.

For the Yosemite toad, the key management activities that the Forest Service can influence are: livestock and pack stock grazing, exotic fish stocking, and locally applied chemical toxins (e.g. pesticides and herbicides). In addition, because recent genetic work on this species indicates high divergence among sites, management approaches should function at a scale that allows conservation of populations that are know to be divergent, so that genetic diversity is not lost.

Conservation Measures This is a Forest Service Sensitive Species, so currently it must be evaluated in Forest level Biological Evaluations. Through this process, appropriate mitigations are identified, and may be implemented at the discretion of the district ranger.

In this FEIS, there is direction (Alternative 8 modified) to implement recovery plans for listed species as funds allow, and to conduct a Conservation Assessment for the Yosemite toad. Critical Aquatic Refuges (CAR’s) are established for several known populations of Yosmite toads on Forest Service lands and the decision allows for the addition of new CAR’s as more information is gained (FEIS Appendix I).

Environmental Consequences Effects of Alternatives For the Yosemite toad, the following evaluation of standards and guidelines among the alternatives for each risk factor considers both current known sites and historic range that is currently unoccupied.

Chemical Toxins (Locally Applied Pesticides and Herbicides). All Alternatives contain direction to avoid pesticide/herbicide application within 500 feet of known Yosemite toad sites as well as at other TES species sites. While this approach addresses local applications, it does not address possible drift or downstream movements of these chemicals. It thus provides some improvement over existing direction, but probably doesn’t go far enough to eliminate this risk to amphibian populations. Alternatives 8 and modified 8 also include prohibition of livestock pesticides in riparian areas which would likely lower risk to amphibians. Alternative 5 provides additional direction to avoid use in any riparian areas, Aquatic Diversity Areas, Critical Aquatic Refuges, and if used in “green zones” make them ground-based and vegetation-specific. This alternative is thus lowest risk for all aquatic/riparian amphibian species.

Exotic fish stocking. All Alternatives address this issue in some manner, from simply supporting AMS goals to focusing on the development of a Conservation Assessment/Strategy for mountain yellow-legged frogs (Alternatives 2, 4, 6, 7, 8, and modified 8), to direction that emphasizes removal (with CDFG cooperation) of exotic fish from some mountain yellow-legged frog areas (Alternatives 3, 5, 8, and modified 8). While these standards and guidelines are not specifically directed at Yosemite toads, the overlap in these two species elevational ranges and habitat associations will likely result in benefits for Yosemite toads as well. Modified Alternative 8 provides a process for the

FEIS Volume 3, Chapter 3, part 4.4, page 219 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Forest Service to work with appropriate State and Federal agencies to eliminate exotic fish stocking when it negatively impacts aquatic species. This could result in reduced risk for the Yosemite toad. Overall, Alternatives 3, 5, 8, and modified 8 appear to provide the most effective approaches for conserving Yosemite toads by including emphasis on removal of fish from some areas.

Livestock Grazing. Many of the standards and guidelines incorporate grazing utilization limits for grasses or shrubs. While these may help maintain certain structural features required by amphibians, there are direct impacts to toads, eggs, and tadpoles (e.g. trampling, degradation of aquatic microhabitats) that are at least of equal concern. Research is also needed to determine how variation in percent utilization benefits or impacts amphibians. In all Alternatives it is possible that varying degrees of livestock exclusions from willow flycatcher habitats may benefit Yosemite toads, where these species ranges overlap. Alternatives 4 and 5 also contain limitations on livestock grazing in riparian areas and wet meadows that would benefit Yosemite toads. Only alternatives 8 and modified 8 contain specific direction that limits packstock use of Yosemite toad breeding sites from during the breeding and rearing season at least until September 1 unless the meadow is dry and breeding is not possible. Alternatives 3, 6, 7, 8, and modified 8 include recommendations for moving livestock handling/gathering facilities outside of riparian/meadow areas and providing off-channel watering devices that would reduce effects on local amphibians. Alternative modified 8 also contains standards for limiting streambank and lake shore disturbance by all activities (not just livestock grazing) to 20% Sierra Nevada wide and 10% in areas with listed fish species. Alternatives 5, 8 and modified 8 have the lowest risks for Yosemite toads.

Pack stock. Alternatives 3, 6, 7, 8, and modified 8 include recommendations for moving livestock handling/gathering facilities outside of riparian/meadow areas and providing off-channel watering devices that would reduce, though not eliminate affects on local amphibians. Alternatives 8 and modified 8 contain specific direction that limits packstock use of Yosemite toad breeding sites from during the breeding and rearing season at least until September 1 unless the meadow is dry and breeding is not possible. It is also possible that evaluations of dispersed recreation sites would identify impacts of pack stock.

Overall Evaluation of Alternatives . For Yosemite toads, Alternatives 8 and modified 8 appear to be the lowest risk and provide the most effective management approaches for this species persistence and recovery. Alternatives 2, 3, and 5 provide relatively low risks to this species, but lack specific direction limiting livestock grazing at Yosemite toad sites. Alternatives 2, 3, and 5 also include provisions for the establishment of an Amphibian Reserve System. If new information on genetic variation in this species is used to design this system, concerns about loss of genetic diversity could be reduced. All other alternatives appear to be less effective in meeting this species’ ecological requirements.

FEIS Volume 3, Chapter 3, part 4.4, page 220 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Environmental Outcome Table 4.4.3.3a. Average assessment ratings for the Yosemite toad. Ratings represent average degree of confidence in each outcome being realized 50 years in the future. (See Chapter 4, Part 1) Total score for each alternative is 100.

Alternative Outcome Current 1 2 3 4 5 6 7 8 Mod 8 A 0 0 0 0 0 0 0 0 0 0 B 0 0 2 10 0 38 0 0 2 2 C 58 32 75 80 38 60 58 58 80 80 D 30 48 23 10 42 2 32 32 18 18 E 12 20 0 0 20 0 10 10 0 0

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

Rationale. Yosemite toads are endemic to the Sierra Nevada. They occur in wet montane meadows at elevations ranging from approximately 6,500 feet to 11,400 feet. Yosemite toads have declined substantially or disappeared from more than 50 percent of the sites where they were historically found. Because the toads breed at the edges of wet meadows and slow-flowing streams, livestock grazing is a primary threat. Suitable environments for the Yosemite toad most likely occur in a patchy distribution with isolation of some populations; future conditions may trend toward more severe isolation and possibly extirpation.

Primary features of the alternatives that would affect outcomes for Yosemite toads include: (1) the AMS under all action alternatives, (2) the amphibian reserve system under Alternative 2 and protections for known occupied habitat under Alternatives 3 and 5, (3) some provisions for willow flycatchers, (4) prohibition of livestock grazing in Yosemite toad breeding sites during the breeding season under Alternative 8 (and modified 8), and (5) prohibition of livestock grazing from perennially wet meadows under Alternative 5. Alternative 5 was judged to substantially improve conditions for Yosemite toads, leading to significant likelihood that suitable conditions for this species would become broadly distributed in the Sierra Nevada. This conclusion was due, in large part, to direction in Alternative 5 that would prohibit grazing in perennially wet meadows. Alternatives 2, 3, and 8 (and modified 8) were also judged to improve conditions for Yosemite toads; these alternatives would increase the likelihood that suitable environments would be as patches and decrease the likelihood that suitable environments would be isolated. Alternatives 1 and 4, with no specific provisions for Yosemite toads, were judged to result in continued degradation of suitable conditions for Yosemite toads, with increasing likelihood that suitable environments would be highly isolated. Alternatives 6 and 7 were judged to maintain current levels of suitable environments for Yosemite toads.

FEIS Volume 3, Chapter 3, part 4.4, page 221 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Cumulative Effects In addition to the Forest Service risk factors discussed above, the following non-Forest Service risk factors have contributed and may still be contributing to the decline of the Yosemite toads: climate change, drought, and increases in UV radiation (Table 4.4.3.6a). The Forest Service may be contributing to the disease factor by inadvertently transporting it on field equipment used in stream and lake environments. Data limitations preclude a quantitative cumulative effects analysis for this species. However, given the number of risk factors under Forest Service influence this agency can play a significant role in reduction of many of threats to the species.

FEIS Volume 3, Chapter 3, part 4.4, page 222 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.3.4. CASCADES FROG

Affected Environment Species Background The Cascades frog (Rana cascadae) historically occurred in the northern Sierra Nevada (Lassen Peak area and north) primarily in Tehema, Butte, and Plumas counties from 760 to 8250 ft. (230 to 2500 m) (Table 4.2.4b). The majority of its range is in the Cascade and Olympic mountains of Oregon and Washington.

Risk Factors Cascades frogs are associated with both permanent and ephemeral streams and ponds, though permanent water bodies are probably required for larval survival (Jennings and Hayes 1994). While evidence of declines is substantial, potential causes have not been well studied. Possible factors that have been cited include: exotic predatory fish, loss of breeding habitat due to drought, and loss of wet meadows possibly as a result of fire supression (Drost and Fellers 1993). For a detailed discussion of the historic and current conditions of aquatic, riparian, and meadow environments in the Sierra Nevada see Part 3.6.

While research has not yet been conducted on Cascades frogs, closely related species in other regions have show sensitivity to numerous pesticides, herbicides, and fertilizers (Berrill et. al. 1993, Berrill et. al. 1997, LeBlanc and Bain 1997). Because these chemicals are thought to disrupt endocrine systems in amphibians at low concentrations, application of pesticides and herbicides are considered to be a risk factors for this species. For the Cascades frog the key management activities that the Forest Service can influence are: livestock grazing, exotic fish stocking, fire suppression, and locally applied chemical toxins (e.g. pesticides and herbicides).

Conservation Measures This is a Forest Service Sensitive Species, so currently it must be evaluated in Forest level Biological Evaluations. Through this process, appropriate mitigations are identified, and may be implemented at the discretion of the district ranger.

In this FEIS, there is direction (Alternative 8 modified) to conduct a Conservation Assessment for the Cascades frog. Critical Aquatic Refuges (CAR’s) are established for several known populations of Cascades frogs on Forest Service lands and the decision allows for the addition of new CAR’s as more information is gained (FEIS Appendix I).

Environmental Consequences Effects of Alternatives For the Cascades frog, the following evaluation of standards and guidelines among FEIS alternatives for each risk factor considers both current known sites and historic range that is currently unoccupied.

Chemical Toxins (Locally Applied Pesticides and Herbicides). All Alternatives contain direction to avoid pesticide/herbicide application within 500 feet of known Cascades frog sites as well as at other TES species sites. While this approach addresses local applications, it does not address possible drift or downstream movements of these chemicals. It thus provides some improvement over existing direction. Alternatives 8 and modified 8 also include prohibition of livestock pesticides in riparian areas which would likely lower risk to amphibians. Alternative 5 provides additional direction to

FEIS Volume 3, Chapter 3, part 4.4, page 223 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 avoid use in any riparian areas, Aquatic Diversity Areas, Critical Aquatic Refuges, and if used in “green zones” make them ground-based and vegetation-specific. This alternative is thus lowest risk for all aquatic/riparian amphibian species.

Exotic fish stocking. Modified Alternative 8 provides a process for the Forest Service to work with appropriate State and Federal agencies to eliminate exotic fish stocking when it negatively impacts aquatic species. This could result in reduced risk for the Cascade frog. The range of this species does not substantially overlap that of the mountain yellow-legged, so actions taken to assess exotic fish stocking and remove fish from some areas are not likely to have major benefits for Cascades frogs. Additional measures would need to be developed for Cascade frogs, if this threat is ultimately identified as significant.

Fire Exclusion. No Alternatives emphasize fire exclusion, though Alternative 2 appears to be the most responsive to controlling wildfires. In general, the approach to wildfire and prescribed fire in all Alternatives should result in the re-introduction of this element to forest ecosystems in the Sierra Nevada to a greater degree than current direction allows. Thus, all Alternatives provide benefits to species, such as the Cascades frog, that may depend on openings that result from fires.

Livestock Grazing. Many of the standards and guidelines incorporate grazing utilization limits for grasses or shrubs. While these may help maintain certain structural features required by amphibians, there are direct impacts to frogs, eggs, and tadpoles (e.g. trampling, degradation of aquatic microhabitats) that are at least of equal concern. Research is also needed to determine how variation in percent utilization benefits or impacts amphibians. In all alternatives, it is possible that varying degrees of livestock exclusions from willow flycatcher habitats may benefit Cascades frogs, where these species ranges overlap. Alternatives 4 and 5 also contain limitations on livestock grazing in riparian areas and wet meadows that would benefit Cascades frogs. Alternatives 3, 6, 7, 8, and modified 8 include recommendations for moving livestock handling/gathering facilities outside of riparian/meadow areas and providing off-channel watering devices that would reduce effects on local amphibians. Alternative modified 8 also contains standards for limiting streambank and lake shore disturbance by all activities to 20 percent Sierra Nevada-wide and 10 percent in areas with listed fish species. Alternatives 5, 8 and modified 8 are lowest risk for Cascades frogs.

Overall Evaluation of Alternatives. For Cascades frogs, Alternatives 5, 8, and modified 8 appear to be the lowest risk and provide the most effective protection measures for this species persistence and recovery. All the other alternatives appear to be less effective in meeting this species’ ecological requirements in some way.

Environmental Outcome This species has experienced significant declines in the Sierra Nevada portion of its range. All of the FEIS alternatives contain an Aquatic Management Strategy that should result in improved aquatic and riparian conditions in the future. However, because there are few populations in the bioregion and there are both local and regional affectors on habitat quality, the likely result for this species under all alternatives is Outcome C (see Section 4.1.3).

Cumulative Effects Declines of this species are not well understood but indications are that habitat alteration, and exotic predatory fish are major affectors. There is also the possibility that disease and/or chemical toxins

FEIS Volume 3, Chapter 3, part 4.4, page 224 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 may be contributing factors. Data limitations preclude a quantitative cumulative effects analysis for this species.

FEIS Volume 3, Chapter 3, part 4.4, page 225 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.3.5. NORTHERN LEOPARD FROG Affected Environment Species Background The northern leopard frog (Rana pipiens) historically occurred in scattered population in the eastern Sierra Nevada (3990 to 4931 ft. / 1216 to 1503 m) (Table 4.2.4b) and also occurs over a wide range in central and eastern North America. Within these limited distributions, these species have both disappeared from 99% of their historic ranges within the Sierra Nevada (Jennings 1996) and are currently State Species of Special Concern and Forest Service Sensitive Species.

Risk Factors Northern leopard frogs are associated with a variety of aquatic habitats including ponds, lakes, tarns, and streams with submergent or emergent vegetation. While evidence of declines in both species is substantial, causes have not been clearly identified. Possible factors are: livestock grazing, exotic predatory fish and bullfrogs (Jennings and Hayes 1994). For a detailed discussion of the historic and current conditions of aquatic, riparian, and meadow environments in the Sierra Nevada see Part 3.6.

Research on environmental toxin affects on the northern leopard frog has shown both sublethal and lethal effects (Berrill et. al. 1993, Berrill et. al. 1997). Because these chemicals are thought to disrupt endocrine systems in amphibians at low concentrations, application of pesticides and herbicides are considered to be a risk factors for this species. For the northern leopard frog, the key management activities that the Forest Service can influence are: livestock grazing, exotic fish stocking, and locally applied chemical toxins (e.g. pesticides and herbicides).

Conservation Measures This is a Forest Service Sensitive Species, so currently it must be evaluated in Forest level Biological Evaluations. Through this process, appropriate mitigations are identified, and may be implemented at the discretion of the district ranger.

In this FEIS, there is direction (Alternative 8 modified) to conduct a Conservation Assessment for the northern leopard frog. No Critical Aquatic Refuges (CAR’s) are established for this species, though the decision allows for the addition of new CAR’s as more information is gained (FEIS Appendix I).

Environmental Consequences Effects of Alternatives For the northern leopard frog, the following evaluation of standards and guidelines among the alternatives for each risk factor considers both current known sites and historic range that is currently unoccupied.

Chemical Toxins (Locally Applied Pesticides and Herbicides). All Alternatives contain direction to avoid pesticide/herbicide application within 500 feet of known northern leopard frog sites as well as at other TES species sites. While this approach addresses local applications, it does not address possible drift or downstream movements of these chemicals. It thus provides some improvement over existing direction. Alternatives 8 and modified 8 also include prohibition of livestock pesticides in riparian areas which would likely lower risk to amphibians. Alternative 5 provides additional direction to avoid use in any riparian areas, Aquatic Diversity Areas, Critical Aquatic Refuges, and if used in “green zones” make them ground-based and vegetation-specific. This alternative is thus lowest risk for all aquatic/riparian amphibian species.

FEIS Volume 3, Chapter 3, part 4.4, page 226 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Exotic fish stocking. Modified Alternative 8 provides direction for the Forest Service to work with appropriate State and Federal agencies to eliminate exotic fish stocking when it negatively impacts aquatic species. This could result in reduced risk for the northern leopard frog. The range of this species does not substantially overlap that of the mountain yellow-legged, so actions taken to assess exotic fish stocking and remove fish from some areas are not likely to have major benefits for northern leopard frogs. Additional measures would need to be developed for northern leopard frogs, if this threat is ultimately identified as significant.

Livestock Grazing. Many of the standards and guidelines incorporate grazing utilization limits for grasses or shrubs. While these may help maintain certain structural features required by amphibians, there are direct impacts to frogs, eggs, and tadpoles (e.g. trampling, degradation of aquatic microhabitats) that are at least of equal concern. Research is also needed to determine how variation in percent utilization benefits or impacts amphibians. In all Alternatives it is possible that varying degrees of livestock exclusions from willow flycatcher habitats may benefit northern leopard frogs, where these species ranges overlap. Alternatives 4 and 5 also contains limitations on livestock grazing in riparian areas and wet meadows that would benefit northern leopard frogs. Alternatives 3, 6, 7, 8, and modified 8 include recommendations for moving livestock handling/gathering facilities outside of riparian/meadow areas and providing off-channel watering devices that would reduce, though not eliminate affects on local amphibians. Alternative modified 8 also contains standards for limiting streambank and lake shore disturbance by all activities (not just livestock grazing)to 20% Sierra Nevada wide and 10% in areas with listed fish species. Alternatives 5, 8 and modified 8 are lowest risk for northern leopard frogs.

Overall Evaluation of Alternatives. For northern leopard frogs, Alternatives 5, 8, and modified 8 appear to be the lowest risk and provide the most effective management approaches for this species persistence and recovery. All the other alternatives appear to be less effective in meeting this species’ ecological requirements in some way.

Environmental Outcome This species has experienced significant declines in the Sierra Nevada portion of its range. All of the FEIS alternatives contain an Aquatic Management Strategy that should result in improved aquatic and riparian conditions in the future. However, because there are few, widely separated, populations in the bioregion and there are both local and regional affectors on habitat quality, the likely result for this species under all alternatives is Outcome D (see Section 4.1.3).

Cumulative Effects Declines of this species are not well understood but indications are that habitat alteration, and exotic predatory fish are major affectors. Exotic predators, like bullfrogs, disease, and chemical toxins may also be contributing to declines. Data limitations preclude a quantitative cumulative effects analysis for this species. However, given the number of risk factors under Forest Service influence, this agency can play a significant role in reduction of many of the threats to the species.

Species occuring on only One or Two Forests Affected Environment The spotted frog may have occurred on the Modoc National Forest in Modoc and Siskyou Counties, California from about 3280 to 4757 ft. (1000 to 1450 m) in elevation (Jennings and Hayes 1994). The Inyo mountains salamander is narrowly distributed in the vicinity of the southeastern portion of the Inyo National Forest at sites ranging from about 1800 to 8600 ft. (550 to 2620 m) in elevation

FEIS Volume 3, Chapter 3, part 4.4, page 227 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

(Jennings and Hayes 1994). The Breckenridge Mountain slender salamander is known from only one location on the Sequoia National Forest at 6300 ft. (1920 m) in elevation (Jennings and Hayes 1994). The current status and ecology of these three species in the Sierra Nevada bioregion are not well known. Table 4.2.4a provides a summary of habitat associations and potential key management issues. For a detailed discussion of the historic and current conditions of aquatic, riparian, and meadow environments in the Sierra Nevada see Part 3.6.

Environmental Consequences Permanent Spring and Seep Associates Species. Inyo Mountains slender salamander (Batrachoseps campi), Breckenridge Mountain slender salamander (Batrachoseps spp.).

The Inyo Mountains slender salamander occurs only on the Inyo National Forest and the Breckenridge Mountain salamander is possibly extinct at the one known locality on the Sequoia National Forest. The status and viability of these two species should be assessed at the forest level due to their limited distributions. These species inhabit permanent springs and seeps with rocky substrates at mid (1640+ ft. / 500+ m) to high elevations (8530+ ft. / 2600+ m).

All of spring and seep associates will benefit from increased protection of aquatic habitats provided by the riparian area component of the Aquatic Management Strategy. The primary issues for this species group are management actions that could change the hydrology of spring/seep systems, such as livestock grazing and road building. Alternatives 4 and 5 provides direction regarding protection of wet meadows and riparian areas from livestock grazing and would thus benefit the amphibian species in this group. Alternatives 3, 6, 7, 8, and modified 8 provide management direction that may limit livestock effects on these species as long as springs/seeps are found and identified as riparian/aquatic habitat. All alternatives recommend avoiding pesticide application near occupied sites for at risk frog species, but only Alternative modified 8 is inclusive of all sensitive species and would cover these species. All Alternatives contain improvements regarding evaluation of existing roads and improvements of stream crossings that will benefit these amphibians. Language in the Record of Decision (tracking with Alternative modified 8) provides broad direction to improve conditions of existing roads, close roads at times of the year when use is low, and decommission them if they are causing unacceptable environmental effects.

Permanent Pond, Lake, and Wetland Associates Species. Spotted frog (Rana pretisosa).

Spotted frogs are known from only five locations in California and they are believed to be extinct at all but one of these, a locality on the Modoc National Forest (Jennings and Hayes 1994). They have a much wider range outside of California, occurring throughout the Cascade mountains and intermountain west. The habitat requirements of spotted frogs are poorly know, though some evidence exists for this species selecting habitats that do not contain introduced exotic predators. Livestock grazing may also negatively affect this species’ habitat (Jennings and Hayes 1994). The status and viability of this species should be addressed at the forest level, due to its limited distribution in the Sierra Nevada bioregion.

This pond, lake, and wetland associate will benefit from increased protection of aquatic habitats provided by the riparian area component of the Aquatic Management Strategy. However, the primary issues for this species are not thoroughly addressed in this FEIS Alternatives 4 and 5 provides

FEIS Volume 3, Chapter 3, part 4.4, page 228 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 direction regarding protection of wet meadows and riparian areas from livestock grazing and would thus benefit the spotted frog. All alternatives recommend avoiding pesticide application near occupied sites for at risk frog species, but only Alternative modified 8 is inclusive of all sensitive species and would cover the spotted frog. Alternatives 3,6,7, 8, and modified 8 provide management direction that may limit livestock effects on this species as well. Controls of exotic fish stocking is addressed in all alternatives though targeted at the mountain yellow-legged frog and there is currently no known overlap in spotted frog distribution with that of the mountain yellow-legged frog. Language in the Record of Decision (tracking with Alternative modified 8) provides for assessment and elmination of fish stocking at sites where aquatic species are negatively impacted.

FEIS Volume 3, Chapter 3, part 4.4, page 229 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

TERRESTRIAL ASSOCIATES (OLD FOREST AND LOWER WESTSIDE HARDWOOD)

4.4.3.6. BATRACHOSEPS RELICTUS SPECIES COMPLEX Affected Environment Species Background The species previously known as the Pacific slender salamander (Batrachoseps pacificus) has recently been split into 4 species (Hell Hollow slender salamander – B. diabolicus, Sequoia slender salamander – B. kawia, Kings River slender salamander - B. regius, and relictual slender salamander - B. relictus) through analysis of allozymes, mitochondrial DNA, and morphology (Jockush et. al. 1998). Of the four species, the relictual slender salamander is the only one that is primarily associated with coniferous habitat, though it is also found in foothill hardwood habitats. This species is narrowly distributed in the Sierra Nevada, occurring only on the Sequoia National Forest. Historically, relictual slender salamanders occurred from the lower Kern River Canyon in Kern County, California to higher elevation areas of the Tule and Kern Rivers in Tulare County, California (1640 to 8200 ft. / 500 to 2500 m) (Jockusch et. al. 1998). Relictual slender salamanders appear to be in decline. No salamanders have been found at the type locality (Lower Kern River Canyon) since 1971 (Jockusch et. al.1998). In addition, no salamanders have been found at eight sites in Kern Canyon where they were known to be common in the 1960s (Jennings and Hayes 1994). The three other species primarily occur at lower elevations (mostly below 1640 ft. [500 m] with some populations of B. kawia at 7200 ft. [2200 m]) in pine/oak woodland and chapparal environments. Some populations of B. kawia also occur in mixed conifer forests

Risk Factors These species have limited ranges and often, small population sizes. The specific threats to them are unknown. In general it is believed that activities that affect terrestrial habitats and/or alter hydrologic regimes of riparian canyons may be major factors. Activities such as road building, dam construction and timber harvest are possible concerns (Jennings 1996).

4.4.3.7. OTHER FOREST SERVICE SENSITIVE SALAMANDERS Affected Environment Species Background The yellow-blotched salamander (Ensatina eschscholtzii croceater), Kern Canyon slender salamander (Batrachoseps simatus), and Tehachapi slender salamander (Batrachoseps stebbinsi) are primarily associated with valley foothill hardwood/conifer habitats (Zeiner 1988). The yellow-blotched salamander is also associated with conifer forests (R. Hansen, pers. comm.). All three species are very narrowly distributed in the Sierra Nevada. The only National Forest on which these species may occur is the Sequoia. The Kern Canyon slender salamander is known from only a few sites in the lower Kern River Canyon, at elevations of 1000 to almost 4000 ft. (305 to almost 1220 m) (Hansen 1997). The Tehachapi slender salamander occurs in scattered localities in the Caliente Creek drainage, and in the Piute Mountains at the southern end of the Sierra Nevada in Kern County from 2500 to 4900 ft. (760 to 1500 m) in elevation. This genetic relationships among populations of B. stebbinsi are currently being examined and it may be split into two species. This will further accentuate the narrow distribution of this species (R. Hansen, pers. comm.). The range of the yellow- blotched salamander extends from the Kern River Canyon (lower slopes of Breckenridge Mountain) and the Piute Mountains southwest to Alamo Mountain at elevations of approximately 1400 to 7500 ft. (427 to 2285 m) in Kern and Ventura Counties, California (Jennings and Hayes 1994,R. Hansen,

FEIS Volume 3, Chapter 3, part 4.4, page 230 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 pers. comm.). The Kern Canyon and Tehachapi slender salamanders are California State Threatened species due to their very narrow distributions and yellow-blotched salamander is considered a State Species of Special Concern because of its narrow distribution though it can be locally abundant (Jennings and Hayes 1994, R. Hansen, pers. comm.). The black salamander (Aneides flavipunctatus) occurs in both conifer and foothill hardwood habitats. Its status is unknown and in fact it may be extirpated from the areas it was previously known to occur (Table 4.2.4d).

The Kern Plateau slender salamander (Batrachoseps sp.) and the limestone salamander (Hydromantes brunus) are associated with a variety of vegetation communities. The Kern Plateau slender salamander is associated with vegetation types from pinyon-juniper to red-fir, while the limestone salamander is a chaparral associate. Both are narrowly distributed, Kern Plateau slender salamander is known from only two sites in the Scodie Mountains of Kern County and from several sites on the eastern slope of the Sierra Nevada in Inyo County. It likely occurs on two National Forests – the Sequoia and Inyo. It ranges in elevation from 6000 to 9800 ft. (1830 to 3000 m) (Hansen 1997). The limestone salamander occurs in the Merced River drainage in Mariposa County and ranges from 836 to 2624 ft. (255 to 800 m) (Zeiner 1988). It likely occurs on two National Forests – the Sierra and Stanislaus. The limestone salamander is listed as a California State Threatened species mainly due to it’s restricted distribution. The Kern Plateau slender salamander is a currently recognized taxon, but as yet is undescribed. It is Forest Service sensitive species because of its narrow distribution.

Risk Factors All of these species have limited ranges and often, small population sizes. Most of the range of the yellow-blotched salamander occurs on private land, thus the primary threats are suburban developments in oak woodland habitats (R. Hansen, pers. comm.). Wood-cutting, mining, and recreation may also impact the species if areas of prime habitat are disturbed by these activities (Jennings 1996). The specific threats to the Kern Canyon slender salamander and the Tehachapi slender salamander are not well known. Road-building and other ground-disturbing activities within the vicinity of occupied or suitable sites for these species will likely have negative effects (Zeiner 1988). Specifically, the Kern Canyon slender salamander may be susceptible to localized livestock grazing in narrow canyons of the south side of the Kern River, where grazing is concentrated in the bottoms of the canyons. In these situations, trampling of leaf and woody litter affects habitat quality for these salamanders. Much of the habitat of the Tehachapi slender salamander occurs on private land where suburban development, road construction, livestock grazing, and mining are major impacts. In addition, heavy equipment used in fire fighting and construction of firebreaks may be of concern to this and other Batrachoseps spp. (R. Hansen, pers. comm.).

The primary threats to the Kern Plateau slender salamander are not currently understood. For the limestone salamander major threats are likely ground-disturbing activities such as mining and road building (Zeiner 1988).

Conservation Measures These are all Forest Service Sensitive Species, so currently they must be evaluated in Forest level Biological Evaluations. Through this process, appropriate mitigations are identified, and may be implemented at the discretion of the district ranger.

FEIS Volume 3, Chapter 3, part 4.4, page 231 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Environmental Consequences All of the terrestrial salamanders are considered as a group for analyses because they are all members of the same family (Plethodontidae) and they share many ecological characteristics. Because published information is limited for most of these species, this analysis reviews their general habitat requirements and major threats and make tentative, qualitative assessments of the effects of the FEIS alternatives and their associated standards and guidelines. Due to the complex nature of the FEIS and the multitude of potential affects from forest management, it was not possible to review every proposed standard and guideline for timber harvest, mechanical thinning, and prescribed fire in detail. For these activities, the analysis focused in on key habitat features, such as canopy cover and dead and downed wood, and on standards and guidelines that were written to provide protection for other focal terrestrial species (e.g. California spotted owl, northern goshawk, and Pacific fisher). I also reviewed the treatment of riparian areas and their value to these species.

The major threats compiled from published information for all species above are habitat altering activities, including road building, timber harvest and firewood-cutting, recreation, and mining. Habitat is also being lost permanently to suburban developments, especially in the foothills. Based on their ecology and physiology, additional suspected threats include: prescribed fire, livestock grazing, and chemical toxins (herbicides, pesticides, fertilizers). These suspected factors derive from two concerns; the effect of prescribed fire and livestock grazing on ground cover, and soil compaction resulting from livestock grazing. Both have potentially negative effects on these ground-dwelling salamander species, though effects will depend largely on the timing and extent of the activities. Amphibians are believed to be sensitive to chemical toxins because of their highly permeable skin and the fact that they may accumulate toxins (bio-accumulation) through ingestion of contaminated invertebrate prey and subsequent storage in their body fat (Barthalmus 1980, Ohlendorf et. al. 1988).

Effects of Alternatives Chemical Toxins. Most studies of impacts of chemical toxins on amphibians have focused on the aquatic forms in aquatic environments. However, studies of adult frogs and toads have indicated both direct (Barthalmus 1980) and indirect (increased disease susceptibility) (Taylor et. al. 1999) effects on these semi-aquatic/terrestrial life stages. Alternatives that limit pesticide/herbicide applications in suitable habitats for these salamanders will be the lowest risk.

Over the last 10 years, herbicides are widely used on Forest Service lands in the Sierra Nevada for the following purposes: conifer release, nursery weed control, seed orchard protection, site preparation, general weed control, and research (J. Klines and C. Herzer, pers. comm.). In this FEIS, noxious weed control will likely result in increased use on herbicides on National Forest lands. While the noxious weed program emphasizes prevention of spreading of weeds and integrated control methods, the larger issue of on-going uses of herbicides for “site preparation” and “conifer release” are not generally addressed in this FEIS These activities are likely to negatively affect terrestrial amphibians as well as other wildlife species.

All Alternatives contain direction to avoid pesticide/herbicide application within 500 feet of known TES species sites. While this approach addresses local applications, it does not address possible drift or downstream movements of these chemicals. It thus provides some improvement over existing direction, but probably doesn’t go far enough to eliminate this risk to amphibian populations. Alternatives 8 and modified 8 also include prohibition of livestock pesticides in riparian areas which would likely lower risk to amphibians, but info on exactly what these chemicals are is needed. Alternative 5 provides additional direction to avoid use in any riparian areas, Aquatic Diversity

FEIS Volume 3, Chapter 3, part 4.4, page 232 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Areas, Critical Aquatic Refuges, and if used in “green zones” make them ground-based and vegetation-specific. This alternative is thus lowest risk for terrestrial amphibian species.

Firewood Cutting. Firewood cutting, which is especially of concern in foothill hardwood habitats, will primarily affect the species associated with these habitats by reducing the amount of dead and down cover items. Alternatives that limit firewood cutting in foothill hardwood habitats will be the lowest risk. All alternatives allow fuel wood gathering using off-highway vehicles within 300 ft. of roads. While this may limit the how far into the forest fuel wood can be gathered, salamanders occurring within the zone adjacent to roads will likely be negatively effected by the off-highway vehicles and the loss of downed woody material. Alternatives 2, 6, 8, and modified 8 provide some limitations on fuel wood gathering in Biodiversity Reserves, Old Forest Emphasis Areas, and southern Sierra Fisher Conservation Areas. The standards associated with these alternatives are not specifically designed for terrestrial salamanders, but should ultimately benefit them. Alternatives 8 and modified 8 also prohibit gathering of fuel wood in Riparian Areas and Critical Aquatic Refuges that would benefit terrestrial salamanders occurring there. In hardwood forests, all Alternatives except 2 allow fuel wood cutting to reduce tree canopy to 40% cover. Standards for canopy in Alternative 2 are not clear. This may have negative effects on terrestrial salamanders associated with these forests (i.e. yellow-blotched salamander, Kern Canyon slender salamander, Tehachapi slender salamander, and arboreal salamander). Alternative 8 appears to be the lowest risk for terrestrial salamanders, followed by alternatives 2 and 6.

Livestock Grazing. Livestock grazing, which is primarily a concern in foothill hardwood habitats, will likely affect the habitat of these species by reducing herbaceous cover, altering its associated cool, moist microclimate, compacting soil, and changing the quality of subterranean refuges (Fleischner 1994). Alternatives that limit livestock grazing in foothill hardwood habitats will be the lowest risk. Most livestock grazing standards provide direction for riparian and meadow areas, but do not specifically identify hardwood forest areas. However, alternatives 5 and modified 8 provide direction that would limit grazing on oaks. Alternatives 4 and 7 provide for changing grazing levels if current approaches aren’t moving toward desired conditions. However, the allowed utilization levels seem high; ground-level microclimates (temperature and moisture) and conditions will likely still be affected thus resulting reduced habitat quality for terrestrial salamanders. Alternatives 2, 3, and 5 propose an amphibian reserve system based on current and historical locations of focal frog species. Within that system, livestock grazing would be prohibited. If terrestrial salamanders occur in these areas they may benefit from such standards. Alternatives 3, 4, and 5 allow livestock grazing in Management Emphasis Areas and/or Riparian Areas only if desired conditions and goals are advanced. The effects of these standards are unknown because there could be substantial subjectivity during implementation. Alternative 8 limits grazing in Great Grey Owl PAC’s from February 1 through August 15. This would benefit terrestrial salamanders if they occur in these PAC’s. Alternatives 5, 8, and modified 8 appear to be the lowest risk for terrestrial salamanders, followed by Alternatives 4, 2, and 3.

Mining. The primary effects of mining are immediate ground disturbance and chemical toxins. Alternatives that reduce or eliminate these effects in suitable habitat for these T.E.S. salamanders will be the lowest risk. All alternatives provide more ecologically sound approaches to mining and attempt to limit mining in special emphasis areas (e.g. Critical Aquatic Refuges, PAC’s, etc.) and in Riparian Areas. These changes to management from current conditions should benefit terrestrial salamanders. Alternatives 2,3,5,6,8, and modified 8 provide additional standards. Alternative 2, 5, 6, 8, and modified 8 propose consideration of mineral withdrawal from Critical Aquatic Refuges (Alt. 5,

FEIS Volume 3, Chapter 3, part 4.4, page 233 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

6, and 8, and modified 8) and Integrated Biodiversity Reserves (Alt. 2) which would benefit terrestrial salamanders that occur in these areas. In addition, Alternatives 3 and 5 limit mining in special emphasis areas if desired conditions and goals are advanced. The effect of this standard is unknown because there could be substantial subjectivity during implementation. Alternative 5 also proposes mineral withdrawal from large roadless areas and that would benefit terrestrial salamanders that occur in these areas. Alternative 5 appears to be the lowest risk for terrestrial amphibians, followed closely by 2, 3, 6, 8 and modified 8.

Prescribed Fire. As fire was once a natural disturbance factor in Sierra Nevada ecosystems, the re- introduction of it will not likely be detrimental on a large scale. The little research that has been conducted on salamanders (primarily in the eastern U.S.) has found both positive and negative effects (e.g. Kirkland, et. al. 1996, McLeod and Gates 1998). The primary concerns for these narrowly distributed salamanders are maintenance of relatively large downed wood and the timing and extent of the fire. These species are active when surface moisture is sufficient for them to survive, forage, and reproduce. In the foothill hardwood and conifer forests of the Sierra Nevada, this is typically during the rainy season, with likely peaks during the warmer parts of the fall and spring seasons (Zeiner 1988). Most prescribed burning will likely be done in the fall, following the first major rain event, and this coincides with a high surface activity time for terrestrial salamanders. If timing of prescribed burning is not flexible (due to other constraints), then alternatives that limit the extent of burning within a species range will be the lowest risk. Alternatives that provide fro some maintenance of large woody debris as refuges for these species will also be lower risk.

In the FEIS, limited operating periods for prescribed burning have been established for focal bird and carnivore species. These periods primarily cover the late winter through summer. While this will provide protection for salamanders that are active in spring rainy periods, fall activity periods are not currently protected. Thus, the concern becomes the extent of burning in any given area. Several of the standards and guidelines written for focal frog species may provide some protection here. At the watershed scale, in any given year, prescribed fire can be used in 5-25% of the suitable habitat for focal frog species; Alternative 7 allows 5%, Alternative 6 allows 15%, and Alternatives 3 and 4 allow 25%. Alternatives 8 and modified 8 calls for site-specific analysis to determine timing and extent of prescribed fire. Alternatives 2 and 5 do not provide this type of direction. Alternative 8 and modified 8’s approach to determining prescribed burning timing and extent on a site-specific basis, while open to some subjectivity, may be the best approach for terrestrial salamanders. In the best case scenario, such an approach would include gathering information on the distributions of species within the area to be burned which would benefit terrestrial salamanders.

In terms of the effects of fire on downed woody material, all Alternatives emphasize retention of snags and minimal loss of downed wood during prescribed burning in spotted owl and goshawk PAC’s. Alternatives 2, 4, 5, 6, 7, 8, and modified 8 also provide for retention of large, early-decay class logs throughout the bioregion except modified 8 does not provide for this in urban/wildland intermix zones. Terrestrial salamanders generally prefer later decay classes of logs, so retention of early stage decaying logs will have long term, but not immediate, benefits. All alternatives provide some approach to maintaining downed wood. Alternatives 2, 6, 7, and 8 also provide for enhancement of downed wood elements in martin and Sierra Nevada red fox elevational ranges that may benefit terrestrial salamanders in these areas. Alterative 3 also includes a standard for moving the general forest toward a SNEP Rank 3 relative to downed logs that may benefit terrestrial salamanders. Alternatives 2, 3, 6, 7, 8, and modified 8 may be lowest risk as they provide the most specific direction for these habitat elements.

FEIS Volume 3, Chapter 3, part 4.4, page 234 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Recreation. These effects are limited to ground-disturbance (e.g. construction of campgrounds, packstock, or other facilities) in suitable habitat for these T.E.S. salamanders. Alternatives that limit construction of new facilities in areas of suitable habitat for these species will be the lowest risk. Alternatives 2, 6, 7, 8, and modified 8 limit development of recreational sites in older forests (with suitable habitat for forest carnivores or large trees) and emphasize natural openings that should benefit terrestrial salamanders. Alternatives 3, 5, 6, 8, and modified 8 limit new OHV/OSV trails in PAC’s and near forest carnivore den sites that should benefit terrestrial salamanders that occur in these areas. Alternative 3 also proposes prohibition of new OHV routes in large roadless areas, Old Forest Emphasis Areas, and Ecologically Significant Areas which should benefit terrestrial salamanders that occur in these areas. All Alternatives contain direction to assess the use and impacts of developed and dispersed recreation sites, trails, OHV trails, and access roads and consider rehabilitation, relocation, or other measures if AMS goals are not advanced or water quality and aquatic habitat objectives (i.e. Riparian Conservation Objectives) cannot be met. Management resulting from this direction would likely benefit terrestrial salamanders. Alternative 3 appears to be the lowest risk for terrestrial salamanders, followed by Alternatives 6, 8, and modified 8.

Roads. The primary effects of road-building are ground-disturbance (removal of cover and compaction) and changes to the hydrologic regime. Even though these species are not tied to surface water per se, they do require moist terrestrial habitats. Roads can interrupt and re-direct water flows on the landscape. Alternatives that limit new road-building and/or decommission and restore existing roads will be the lowest risk for these salamander species.

All alternatives provide direction for integrating roads analysis into landscape/watershed analysis. Language in the Record of Decision (tracking with Alternative modified 8) provides broad direction to improve conditions of existing roads, close roads at times of the year when use is low, and decommission them if they are causing unacceptable environmental effects. If done consistently across Forests, and if terrestrial salamanders are considered, this should lead to better land management and improved habitat for terrestrial salamanders. All alternatives also call for maintenance of the existing characteristics of large roadless areas (1,000-5,000 acres) until such an analysis can be done. Alternatives 2, 6, 7, and 8 include direction to de-commission roads in Integrated Biodiversity Reserves, Forest Carnivore Areas, and Critical Aquatic Refuges. Though some disturbance will be required initially, long-term habitat conditions should be improved for terrestrial salamanders. Alternatives 2, 3, and 5 propose an amphibian reserve system based on current and historical locations of focal frog species. Within that system, new road construction would be prohibited. If terrestrial salamanders occur in these areas they may benefit from such standards. Alternatives 3 and modified 8 provide standards for reducing roads in Riparian Areas which has unknown effects on terrestrial salamanders. Effects could be positive if few new roads are constructed at all, but negative if existing roads were moved and road-building was focused in upland areas where these species may occur. Alternative 5 contains many other standards relating to road de- commissioning, seasonal use restrictions, and improved construction approaches that should primarily benefit terrestrial salamanders. Alternative 5 appears to be the lowest risk for terrestrial salamanders, followed by Alternative 3, modified 8, and then Alternatives 2, 6, 7, 8.

Timber Harvest and Mechanical Fuels Treatment. Timber harvest includes mechanical thinning for fuels reduction as well as harvest for wood fiber. Timber harvest has both direct (ground disturbance from equipment) and indirect (habitat) effects on terrestrial salamanders, though the degree of effect may differ among species. Research from the Pacific Northwest has shown that removal of overstory canopy affects ground-level microclimate (temperature and moisture) and cover

FEIS Volume 3, Chapter 3, part 4.4, page 235 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 availability and that terrestrial salamanders associated with these features occurred in lower abundances on harvested sites (Welsh and Lind 1991, Welsh and Lind 1995). Alternatives that limit the use of ground disturbance, limit the amount of canopy removed, and leave downed woody material will be the lowest risk for these species.

In terms of canopy cover, all alternatives provide some direction, from minimal (Alternative 4) to extensive and detailed (Alternatives 8 and modified 8). Most of the standards are focused on bird PAC’s, forest carnivore den sites, and areas of old-forest. These actions will only benefit terrestrial salamanders if the salamanders occur in these areas. Alternatives 2, 5, 6, 7, 8, and modified 8 include standards for canopy cover that range from 40% to 70% closure depending on exisiting canopy, location (eastside/westside) and focal species habitat (spotted owl, goshawk, forest carnivores). Evidence from other geographic areas with related species of salamanders (cited above) indicates that higher canopy closure provides better habitat for terrestrial salamanders. Alternatives 8 and modified 8 appear to provide the highest canopy closure requirements over a range of areas (PAC’s and forest carnivore areas). Alternatives 2, 5, 6, and 7 include a variety of standards that, given our incomplete knowledge about salamander distributions in these areas, appear to provide approximately equivalent levels of protection for terrestrial salamanders. Alternatives 2, 4, 6, 7, 8, and modified 8 also provide specific direction for maintaining overstory canopy along intermittent and perennial streams and would provide higher quality habitat for terrestrial salamanders. Alternative 5 prohibits timber harvest, salvage, and mechanical fuel treatment in “Green and Grey Zones” which will effectively maintain existing canopy closure.

Ground disturbance in riparian areas is directly addressed in Alternatives 2, 4, 5, 6, 7, 8, and modified 8 with a standard that encourages minimizing or mitigating ground disturbance, especially near habitats occupied by sensitive amphibians. In Alternatives 2, 5, 6, 7, and 8 timber harvest and mechanical fuels treatment are not allowed in the Riparian Areas or “Green Zones” of intermittent and perennial streams. In modified 8 these types of treatments are only allowed if they meet Riparian Conservation Objectives. Alternative 5 also disallows these activities in the “Grey Zones”. Alternative 4 does allow some of these activities in these areas, but prohibits salvage logging.

Retention of downed woody material is required in all alternatives. Alternatives 2, 4, 5, 6, 7, 8, and modified 8 also provide for retention of large, early-decay class logs throughout the bioregion except modified 8 does not provide for this in urban/wildland intermix zones. Terrestrial salamanders generally prefer later decay classes of logs, so retention of early stage decaying logs will have long term, but not immediate benefits. Alternatives 2, 6, 7, and 8 also provide for enhancement of downed wood elements in martin and Sierra Nevada red fox elevational ranges that may benefit terrestrial salamanders in these areas. Alterative 3 also includes a standard for moving the general forest toward a SNEP Rank 3 relative to downed logs that may benefit terrestrial salamanders. For timber harvesting and mechanical fuels treatment, the overall lowest risk alternative for terrestrial salamanders appears to be Alternatives 8 and modified 8, followed by Alternatives 2, 5, 6, and 7.

Overall Evaluation. Evaluating all the major threats to terrestrial salamanders in the context of proposed new land management standards and guidelines, the lowest risk Alternatives for these species appears to be Alternatives 2, 5, 8, and modified 8, in no particular order. Alternatives 3, 6, and 7 appear to be of moderate risk to terrestrial salamanders, because they do not address all the threats at the same level as do Alternatives 2,5, and 8. Alternative 4 is overall the highest risk alternative for terrestrial salamanders.

FEIS Volume 3, Chapter 3, part 4.4, page 236 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Environmental Outcomes There is limited information on the causes of declines for species within the relictual slender salamander complex. In addition, this complex of species potentially occurs on a total of four forests with each member species occurring on only one or two forests. These factors together make it impossible to determine an Outcome for this species complex at the scale of the Sierra Nevada bioregion.

Cumulative Effects Declines of these species are not well understood but indications are that habitat alteration is the major affector. Data limitations preclude a quantitative cumulative effects analysis for these species. However, given that habitat condition is a factor the Forest Service influences, this agency can play a significant role in reduction of the threats to the species.

FEIS Volume 3, Chapter 3, part 4.4, page 237 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Table 4.4.3.6a. A summary of primary risk factors and key conservation measures/ management actions (standard and guideline # in parantheses) across FEIS alternatives for high elevation species, the mountain yellow-legged frog (Rana muscosa) and the Yosemite toad (Bufo canorus), focusing on the aquatic habitat of these species.

ALTERNATIVES CRITERIA CURRENT 1 2 3 4 5 6 7 8 Mod 8 CWHR-Based Habitat N/A - aquatic habitat is key for these species and model does not provide reliable estimates and Change in Utility Values Habitat Distribution Description Mountain yellow-legged frog: High elevation (4500 to > 11,000 ft. / 1370 to >3650 m) tarns, lakes, and streams. Yosemite toad: High elevation (6400 to approx. 11,300 ft. / 1950 to approx. 3450 m) wet montane meadows; historically possibly some shallow lakes. Quantitative Data not available Qualitative 1 Widely Widely Widely Widely Widely Widely Widely Widely Widely Widely distributed, with distributed, with distributed, with distributed, with distributed, with distributed, with distributed, with distributed, with distributed, with distributed, gaps gaps gaps gaps gaps gaps gaps gaps gaps with gaps Protection of Sites/Popn’s Federal Status USFWS has recently accepted petitions for listing of both these species. Standard and guideline states that “recovery plans” for Federally listed species will be implemented as funds allow (RCA03). Both are currently Forest Service Sensitive Species in California, so they must be evaluated in Forest level Biological Evaluations. Other Protections none none Amphibian Amphibian none Amphibian Other Emphasis none Other Emphasis CAR’s are Reserve System Reserve System Reserve Areas (IBA’s, Areas (IBA’s, established for (AM12). (AM13). System. CAR’s, EW’s, CAR’s, EW’s, several at risk (AM13). etc) may provide etc) may provide populations Other Emphasis Other Emphasis protection if protection if and can be Areas (IBA’s, Areas (IBA’s, Other Emphasis species occurs species occurs revised as CAR’s, EW’s, CAR’s, EW’s, Areas (IBA’s, there. there. needed post- etc) may provide etc) may provide CAR’s, EW’s, EIS (see FEIS protection if protection if etc) may provide Appendix I). species occurs species occurs protection if there. there. species occurs Conservation there. Assessments conducted within 1 year of completion of Record of Decision (ROD) for Notice of Intent (NOI) amphibians should contribute to landscape analysis and adaptive management and eventually reduce risk.

FEIS Volume 3, Chapter 3, part 4.4, page 238 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

ALTERNATIVES CRITERIA CURRENT 1 2 3 4 5 6 7 8 Mod 8 FS Risk Factors Chemical Toxins Existing Standards do Avoiding Avoiding Avoiding Avoiding Avoiding Avoiding Avoiding Avoiding conditions are not improve application of application of application of application of application of application of application of application of known to habitat or protect pesticides/ pesticides/ pesticides/ pesticides/ pesticides/ pesticides/ pesticides/ pesticides/ negatively affect species, so high herbicides within herbicides within herbicides within herbicides within herbicides within herbicides within herbicides within herbicides species. risk. 500 ft. of focal 500 ft. of focal 500 ft. of focal 500 ft. of focal 500 ft. of focal 500 ft. of focal 500 ft. of focal within 500 ft. ampihbian ampihbian ampihbian ampihbian ampihbian ampihbian ampihbian of TES populations will populations will populations will populations will populations will populations will populations will species will reduce risk reduce risk reduce risk reduce risk reduce risk reduce risk reduce risk reduce risk (AR24). (AR24). (AR24). (AR24). (AR24). (AR24). (AR24). (RCA12, RCA12a). Avoiding use of Prohibition of pesticides/ use of livestock Prohibition of herbicides in pesticides in use of Riparian Areas, riparian areas livestock ADA’s, CAR’s, should reduce pesticides in should reduce risk (AM21). riparian areas risk (ACS20, should reduce ACS21, ACS31). risk (RIP- AM21).

FEIS Volume 3, Chapter 3, part 4.4, page 239 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

ALTERNATIVES CRITERIA CURRENT 1 2 3 4 5 6 7 8 Mod 8 Exotic fish stocking Existing Standards do Working with Working with Working with Working with Working with Working with Working with Language in conditions are not improve CDFG to bring CDFG to bring CDFG to bring CDFG to bring CDFG to bring CDFG to bring CDFG to bring the Record of known to habitat or protect fish stocking fish stocking fish stocking fish stocking fish stocking fish stocking fish stocking Decision negatively affect species, so high program in line program in line program in line program in line program in line program in line program in line (ROD) species. risk. with AMS goals with AMS goals with AMS goals with AMS goals with AMS goals with AMS goals with AMS goals commits the should lower risk should lower risk should lower risk should lower risk should lower risk should lower risk should lower risk FS to work (AM10). (AM10). (AM10). (AM10). (AM10) (AM10). (AM10). with CDFG to monitor Producing a Working with Producing a Working with Producing a Producing a Producing a mountain Conservation CDFG to Conservation CDFG to Conservation Conservation Conservation yellow-legged Strategy for the remove fish from Strategy for the remove fish from Strategy for the Strategy for the Strategy for the frogs and mountain yellow- some areas with mountain yellow- some areas with moutain yellow- moutain yellow- moutain yellow- assess legged frog may mountain yellow- legged frog may mountain yellow- legged frog may legged frog may legged frog may removal of reduce risk legged frogs reduce risk legged frogs reduce risk reduce risk reduce risk exotic fish (depending on should reduce (depending on should reduce (depending on (depending on (depending on from some implementa- risk (AM06). implementa- risk (AM06). implementa- implementa- implementation). water bodies. tion) (AM03). tion) (AM03). tion) (AM03). tion) (AM03). Working with ROD CDFG to language also remove fish from commits the some areas with FS to work mountain yellow- with CDFG legged frogs and other should reduce agencies to risk (AM06).- eliminate fish stocking that negatively impacts aquatic species.

Working with CDFG to develop a program for removing exotic fish from mountain yellow-legged frog habitat should reduce risk (RCA05).

FEIS Volume 3, Chapter 3, part 4.4, page 240 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

ALTERNATIVES CRITERIA CURRENT 1 2 3 4 5 6 7 8 Mod 8 Livestock Grazing Existing Standards do Livestock 100 ft. buffer Livestock Livestock Movement of Movement of Movement of Exclusion of conditions are not improve exclusions from around WIFL grazing exclusions from livestock livestock livestock livestock from known to habitat or protect WIFL habitat habitat reduces disallowed within riparian areas, handling handling handling wet meadows, negatively affect species, so high reduces risk if risk if if site is 100 ft. of and wet facilities outside facilities outside facilities outside and species. risk. site is also used also used by suitable WIFL meadows of riparian/ of riparian/ of riparian/ associated by toads. toads (B48). habitat should reduces risk meadow areas meadow areas meadow areas streams and reduce risk (ACS22, ACS23, and providing and providing and providing springs during No livestock (B35). ACS24, ACS25, off-channel off-channel off-channel the Yosemite grazing in WIFL Movement of G07). watering devices watering devices watering devices toad breeding habitat within 5 livestock should reduce should reduce should reduce and rearing miles of historic handling risk (AM23). risk (AM23). risk (AM23). season and current facilities outside (through Sept WIFL sites of riparian/ No livestock No livestock No livestock 1), surveys of would reduce meadow areas grazing within grazing within grazing at unoccupied risk if if site is and providing 100 ft. of 100 ft. of Yosemite toad sites, and also used by off-channel suitable WIFL suitable WIFL breeding sites assumption of toads (B33). watering devices habitat should habitat should (April 15 to Sept. occupancy if should reduce reduce risk if if reduce risk if if 15) should surveys not risk (WM12, site is also used site is also used reduce risk done should WM13). by toads(B35). by toads(B35). (AM14). reduce risk (RCA41). No livestock grazing within Location of 100ft. of new livestock occupied WIFL handling or suitable facilities habitat within 5 outside of miles of riparian/ occupied habitat meadow sites would areas and reduce risk if site evaluation of is also used by existing toads (B335B). facilities should reduce risk (RCA42).

Limitation of streambank and lake shore disturbance from livestock grazing and other activities to 20% everywhere and 10% along reaches with listed fish should reduce risk (RCA18).

FEIS Volume 3, Chapter 3, part 4.4, page 241 – Affected Environment and Environmental Consequences

Elimination of Sierra Nevada Forest Plan Amendment – Part 4.4

ALTERNATIVES CRITERIA CURRENT 1 2 3 4 5 6 7 8 Mod 8 Livestock Grazing, cont. Elimination of livestock grazing in known sites (n=82) and late-season grazing in new WIFL sites may reduce risk if amphibians occur in these areas (WIFL1, WIFL2).

Packstock Existing Standards do No specific No specific No specific No specific No specific No specific No livestock Exclusion of conditions are not improve standards to standards to standards to standards to standards to standards to grazing at pack and known to habitat or protect improve habitat improve habitat improve habitat improve habitat improve habitat improve habitat Yosemite toad saddlestock negatively affect species, so high or protect or protect or protect or protect or protect or protect breeding sites from wet species. risk. species, so high species, so high species, so high species, so high species, so high species, so high (April 15 to Sept. meadows, and risk. risk. risk. risk. risk. risk. 15) should associated (but see (but see (but see (but see (but see (but see reduce risk streams and recreation – recreation – recreation – recreation – recreation – recreation – (AM14). springs during evaluation of evaluation of evaluation of evaluation of evaluation of evaluation of the Yosemite dispersed sites dispersed sites dispersed sites dispersed sites dispersed sites dispersed sites Also see toad breeding may provide may provide may provide may provide may provide may provide recreation – and rearing some benefit – some benefit - some benefit - some benefit - some benefit - some benefit - evaluation of season R01D). R01D). R01D). R01D). R01D). R01D). dispersed sites (through Sept may provide 1), surveys of some benefit unoccupied (R01D). sites, and assumption of occupancy if surveys not done should reduce risk (RCA41).

FEIS Volume 3, Chapter 3, part 4.4, page 242 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

ALTERNATIVES CRITERIA CURRENT 1 2 3 4 5 6 7 8 Mod 8 Recreation Existing Standards do Relocation or Relocation or Relocation or Relocation or Relocation or Relocation or Relocation or Assessment conditions are not improve rehabilita-tion of rehabilita-tion of rehabilita-tion of rehabilita-tion of rehabilita-tion of rehabilita-tion of rehabilita-tion of and redesign known to habitat or protect dispersed dispersed dispersed dispersed dispersed dispersed dispersed of access negatively affect species, so high recreation sites recreation sites recreation sites recreation sites recreation sites recreation sites recreation sites roads, trails, species. risk. if not in line with if not in line with if not in line with if not in line with if not in line with if not in line with if not in line with OHV AMS goals AMS goals AMS goals AMS goals AMS goals AMS goals AMS goals trails/staging should reduce should reduce should reduce should reduce should reduce should reduce should reduce areas, risk (R01D). risk (R01D). risk (R01D). risk (R01D). risk (R01D). risk (R01D). risk (R01D). developed and dispersed Reduction of recreation ORV use in sites, etc. to Riparian Areas meet habitat should reduce and water risk (R09C). quality objectives should reduce risk (RCA37).

Location of new OHV sites outside of RCA’s and CAR’s should reduce risk (RCA38).

Non-FS Risk Factors Mountain yellow-legged frog Disease A recently discovered parasitic fungus is currently under investigation as a possible threat to mountain yellow-legged frogs. The Forest Service may contribute to this factor by inadvertently transporting the fungus on field equipment used in stream and lake environments. Yosemite toad Climate chng Extended droughts have apparently caused declines of some populations of this species in recent history. (drought) UV radiation Increases in UV radiation may be occurring due to atmospheric changes. This can affect egg/larval development and immune system function. Disease A recently discovered parasitic fungus is currently under investigation as a possible threat to Yosemite toads. The Forest Service may contribute to this factor by inadvertently transporting the fungus on field equipment used in stream and lake environments. Non-FS Benefits National Parks Yosemite and Sequoia-Kings Canyon National Parks contain populations of both species. These parks have discontinued stocking of exotic fishes and may provide protected areas for reestablishment in the future.

FEIS Volume 3, Chapter 3, part 4.4, page 243 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.4. Fish The following sections present the affected environment for Forest Service Sensitive fish species in the Sierra Nevada. The final section presents environmental consequences for these species.

I. Affected Environment

GOOSE LAKE LAMPREY (Lampetra tridentata ssp.) Life History The life history of this taxon is largely unknown, but presumably the adults live for a year or two in Goose Lake, preying on Goose Lake tui chubs, suckers, and redband trout. It is likely that they migrate up suitable tributary streams in winter or spring for spawning. They have to move up far enough to find gravel for spawning and to have enough suitable soft-bottomed habitat downstream of the spawning area for survival of the ammocoetes. Thus, spawning areas may be as much as 12.4 to 18.6 miles upstream from the lake. Ammocoetes probably spend 4 to 6 years in the streams before metamorphosing into adults and moving into the lake.

Habitat relationships Adults live in shallow, alkaline Goose Lake where they prey on larger fishes. Goose Lake is described in the Goose Lake tui chub account. Like other lampreys, Goose lake lampreys require gravel riffles in streams for spawning, and the ammocoetes require mudding backwater habitats downstream of the spawning areas. However, the habitat requirements of this lamprey have never been specifically studied.

Status Goose Lake lampreys were fairly common in the Goose Lake until the lake dried up in the summer of 1992. They were readily collected from large tui chubs caught in gillnets (R. White, USFWS, pers. comm.). The Goose Lake lamprey has a high probability of becoming extinct during a period of prolonged drought if the lake and lower tributaries are dry for several years in a row. However, the ammocoetes may persis for 3 to 4 years if there is adequate water flowing over the habitats they occupy. The Cottonwood Reservoir population is of unknown size but the reservoir may serve as a refuge, provided a minimum pool is maintained throughout extended drought periods.

Historical and Current Distribution The Goose Lake lamprey is endemic to Goose Lake and its tributaries in Oregon and California. However, the streams most important for spawning and as habitat for the ammocoetes have not been identified with certainty. They have been collected only from Willow, Lassen and Cold (tributary to Lassen Creek) Creeks, Modoc County (G. Sato, pers. com., CDFG unpubl. data), but a thorough search of the tributary streams for lampreys has not been done. It is likely that dams now restrict the distribution of ammocoetes by blocking the migration of adults and by drying up suitable habitats downstream. In Lake County, Oregon, a population apparently exists in Cottonwood Reservoir on Cottonwood Creek (Oregon Dept. of Fish and Wildlife, unpubl. data)

Risk Factors The principal threat to the Goose Lake lamprey is dessication of its habitat, Goose Lake and its tributaries. The combination of severe, extended drought and reduced inflow into the lake presumably resulted in accelerated desiccation of the lake during the 1986 to 1992 drought.

FEIS Volume 3, Chapter 3, part 4.4, page 244 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Diversions, dams, culverts, and other obstructions may be preventing migrating adults from reaching spawning areas in tributary streams, although the reservoirs may also be serving as refuges for the species. The diversion of water from streams for agriculture and other uses may have reduced or dried up habitats required by ammocoetes. Habitat may also have been lost through stream channelization and erosion caused by livestock grazing in riparian areas. The loss of habitat for ammocoetes is likely to be particularly severe in the lower reaches of the inflowing streams. Although Goose Lake has presumably dried up in the past and the lamprey and other fishes have persisted, recent watershed conditions probably have increased the rate of time span of dessication and reduced access to upstream refuges.

CHINOOK SALMON, Fall run (Oncorhyncus tshawytscha) Life History The great majority of fall chinook salmon appear to spawn in the mainstem of the Sacramento River (R. Painter, pers. comm), which they enter from October through February (Vogel and Marine 1991). In the past, these migrating fish were a mixture of age classes ranging from two to five years old. The fall-run is different from the other runs of fish entering freshwater, time and locations of spawning, incubation times, incubation temperatures, and timing of juvenile migration. While migrating and holding in the river, fall chinook do not feed, relying instead on stored body fat reserves for maintenance. Spawning occurs in January, February, and March, although it may extend into April in some years. Eggs are laid in large depressions (redds) hollowed out in the gravel beds. The embryos hatch following a 3 to 4 month incubation period and the alevins (sac-fry) remain in the gravel for another 2 to 3 weeks. Once their yolk sac is absorbed, the fry emerge and begin feeding on aquatic insects. All fry have emerged by early June. The juvenile hold in the river for nearly a year before moving out to sea the following December through March. Once in the ocean, salmon are largely piscivorous and grow rapidly.

Habitat relationships The specific habitat requirements of fall chinook have not been determined, but they are presumably similar to other chinook salmon runs and fall within the range of physical and chemical characteristics of the Sacramento River above Red Bluff.

For other runs, adult numbers holding in an area seem to depend on the volume and depth of pools, amount of cover (especially “bubble curtains” created by inflowing water), and proximity to patches of gravel suitable for spawning (G. Sato, unpubl. data). Sustained water temperature above 27oC are lethal to adults (Cramer and Hammack 1952). The pools in which adults hold are at least 3.3 to 9.9 feet deep, with bedrock bottoms and moderate velocities (G. Sato, unpubl. data; Marcotte 1984). In Deer Creek preferred mean water velocities measured during 1988 were 60-80 cm sec-1 for adults (Sato and Moyle 1988) The pools usually have a large bubble curtain at the head, underwater rocky ledges, and shade cover throughout the day (Ekman 1987). The salmon will also seek cover in smaller “pocket” water behind large rocks in fast water.

Habitat preference curves determined by the USFWS for adult chinook in the Trinity River indicate that pool use declines when depths become less than 7.9 feet and that optimal water velocity ranges between 15-37 cm sec-1 (Marcotte 1984). Spawning occurs in gravel beds with gravel of a size that fish can excavate. Optimum substrate for embryos has been reported as a mixture of gravel, rubble (mean diameter 0.39 to 1.6 inches) and less than 25 percent fines (less than 0.26 inches diameter) (Platts and others 1979, Reiser and Bjornn 1979). Juvenile in Deer Creek were found to prefer runs

FEIS Volume 3, Chapter 3, part 4.4, page 245 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

or riffles with gravel substrates, depths of 7.8 to 46.8 inches, and mean water-column velocities of 20- 40 cm sec-1 (Sato and Moyle 1989).

During downstream migrations in the Sacramento River and Delta, smolts presumably stay close to the banks during the day (near cover) and them move out into open water at night, to migrate. Historically, they may have moved into flooded marshy areas in the Sacramento Delta to feed but there is little evidence of such activity today.

Unlike most tributary streams of the Sacramento and San Joaquin Rivers which now have major water storage facilities that inundate or block hundreds of miles of historical anadromous spawning habitat, upstream habitat in Mill, Deer, and Antelope Creeks is still available for utilization by anadromous fish. In all three drainages, instream habitat conditions for anadromous fish are in good condition overall but are underutilized because of low escapement levels. Battle Creek and Butte Creek are also considered important for anadromous fish and have a high potential for restoration.

Status The historic abundance of fall chinook is not known because it was recognized as a distinct chinook only after the Red Bluff Diversion Dam was constructed in 1966. In order to get past the dam, salmon migrating up the Sacramento River had to ascend a fish ladder in which they could be counted with some accuracy for the first time. The four chinook salmon runs present in the river (fall, late- fall, winter, spring) were revealed as peaks in the counts, although salmon passed over the dam during every mongth of the year. Next to winter-run chinook (now listed as an endangered species by the state and threatened by the federal government), late-fall run chinook are the least numerous run in the Sacramento River and, like the winter-run and spring-run chinook, their numbers have declined since counting began in 1967. In the first 10 years of counting (1967-1976) the run averaged about 22,000 fish (FWS Red Bluff Field Office). There have been no counts of 20,000 fish or more since 1975, although 16,000 fish were counted in 1987. The run in 1991 was 7,089 fish (USFWS 1992). Counts from 1992 through 1998 were not available because the gates at Red Bluff Diversion Dam were opened to allow free passage for winter-run chinook adults and smolts.

Historical and Current Distribution The Sacramento fall run chinook are found mainly in the Sacremento River, and most spawning and rearing of juveniles take place in the reach between Red Bluff and Redding (Keswick Dam). According to Vogel and Marine (1991), however, up to approximately 15 to 30 percent of the total late-fall run can spawn downstream of Red Bluff when “water qualtiy is good”. R. Painter (pers. comm.) indicated that apparent late-fall run chinook have been observed spawning in Battle Creek, Cottonwood Creek, Mill Creek, Yuba River and Feather River, but these are at best a small fraction of the total populations. The Battle Creek spawners are presumably derived from an artificially maintained run into the Battle Creek Fish Hatchery. The historic distribution of the late-fall run is not known, but it probably spawned in the upper Sacramento River and major tributaries in reaches now blocked by Shasta Dam.

Risk Factors For fall and late-fall run chinook salmon, the causes of population decline are poorly understood, but presumably are similar to those of winter-run chinook (Williams and Williams 1991) and spring-run chinook. The principal causes of decline seem to be (1) passage problems over dams, (2) loss of habitat, (3) introgression with other runs, and (4) other factors such as disease and pollutants.

FEIS Volume 3, Chapter 3, part 4.4, page 246 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Dams and Diversions. When Shasta and Keswick Dams were built in the 1940s, they presumably denied access of fall and late-fall run chinook to upstream spawning areas where run-off and spring water originating from Mt. Shasta and other areas kept water temperatures cool enough for successful spawning, egg incubation and over-summer survival of juvenile salmon. The effects of Red Bluff Diversion Dam (RBDD) were more subtle and not recognized until fairly recently (Williams and Willaims 1991). This dam apparently delayed passage to upstream spawning areas and also concentrated predators, increasing mortality on out-migrating smolts (USBOR reports). Kope and Botsford (1990) documented that the overall decline of Sacramento River salmon was closely tied to the construction of RBDD.

Habitat modification. Large dams on the Sacramento River and its tributaries have not only denied salmon access to historic spawning grounds, but they have reduced or eliminated recruitment of spawning gravels into the river beds below the dams and altered temperature regimes. Loss of spawning gravels in the Sacramento River below Keswick Dam is regarded as a serious problem, and large quantities of gravel are now trucked to the river and dumped in, mainly to provide spawning sites for winter-run chinook. However, it is likely that late-fall run also use these gravel deposits (R. Painter, pers. comm.). Overly warm temperatures can be a problem in this reach, mainly during drought years when flows are reduced to save water in Shasta Reservoir. Also, the reduced reservoir volume during drought years and the inability to tap colder levels of the reservoir have meant that water released below the dam is often warmer than desirable. Efforts being made to provide cooler summer flows for winter-run chinook should also benefit fall and late-fall run chinook.

Overexploitation. The actual harvest rates of late-fall chinook are not known, but it is highly likely that they are harvested at the same rates as fall chinook, the principal remaining run in the Sacramento River. In general, chinook salmon are harvested in both ocean and in-river fisheries. Although the fisheries are capturing mainly hatchery fish, they are presumably also taking wild fish at least in proportionate abundance relative to hatchery fish. Given the small size of the remaining runs of wild fish, the take of even a few wild fish may have a significant effect on their populations. It is likely that as many as one-half of the wild fish are taken in the fisheries.

Commercial fisheris also may be affecting the chinook populations indirectly through the continual removal of larger and older individuals. This results in spawning runs made up mainly of three-year- old fish, which are smaller and therefore produce fewer eggs per female. The removal of older fish also eliminates much of the natural “cushion” the populations have against natural disasters such as severe drought, which may wipe out a run in one year. Under natural conditions, the four- and five- year-old fish still in the ocean help to keep the runs balanced and can make up for the fish lost during an occassional catastrophe. Under present conditions, a loss of a run in one year will result in very low runs threes years later, and the loss of runs two or three years in a row can eliminate a population.

Out-migrant mortality. Smolt mortality is probably a factor affecting fall and late-fall chinook abundance as it is for all runs of salmon in the Sacramento-San Joaquin drainage. Small number of out-migrants are presumable entrained at every irrigation diversion along the Sacramento River that is operating during the migration period. At the same time, extensive bank alteration, especially rip- rapping, reduces the amount of cover available to protect the out-migrants from striped bass and other predators. When SWP and CVP pumping rates are high and outflows relatively low, spring chinook smolts are probably entrained in large numbers, consumed by predators in Clifton Court Forebay and other off-channal areas, or are otherwise diverted from their downstream migration.

FEIS Volume 3, Chapter 3, part 4.4, page 247 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Introgression with other chinook salmon runs. The spawning season of late-fall run chinook overlaps somewhat with that of fall-run chinook in January and with winter-run chinook in April. Behavioral or physiological barrier to interbreeding at these times are unlikely, and the extent to which it occurs is not known. Prior to the construction of Shasta Dam, there probably was spatial as well as seasonal segregation among the three runs. However, since these three runs are now forced to spawn in one reach of the Sacramento River, introgression is likely. Introgression of mainstem populations of spring-run chinook with fall-run chinook apparently has resulted in the loss of the distinctiveness of these runs in the Sacramento River, as indicated by the earlier shift in fall-run arrival in the upper river and a protracted fall spawning period (Vogel and Marine 1991). The blurring of run distinctivenss may also be happening with the late-fall chinook.

Late-fall run chinook are reared in small numbers in Coleman National Fish Hatchery on Battle Creek. Hatchery broodstock selection for late-fall chinook includes both fish naturally returning to Battle Creek and those trapped at Keswick Dam. An arbitrary separation date (December 15) is used to designated fish returning to Coleman National Fish Hatchery as late-fall run versus fall-run; however, there undoubtedly is overlap in the run timing. Interbreeding of hatchery with wild fish has the potential to dilute and eventually eliminate the genetic distinctiveness of the remaining naturally reproducing stock.

Pollution. A potential problem is the likelihood of a major spill of water laden with toxic chemicals from the Iron Mountain mine site, if the Spring Creek retention reservoir spills or bursts. These waters could wipe out either migrating adults or, more likely, juveniles holding in the river.

EAGLE LAKE RAINBOW TROUT (Oncorhynchus mykiss aquilarum) Life History It is thought that Eagle Lake trout historically migrated over 20 miles upstream Pine Creek to reach suitable spawning habitat. The timing of migration was dependent upon water flows and water temperatures, and normally occurred between March and May (Moyle and others 1989). After spawning, adults returned to Eagle Lake, and young spent one to two years in the stream before descending Pine Creek and into Eagle Lake. Eagle Lake trout mature at three years, and mature females produce 2,500 to 3,000 eggs (Moyle and others 1989). The upper stretches of Pine Creek contain good spawning media, with long reaches of gravel substrates. The degraded section of lower Pine Creek serves as a migration corridor at this time. As the lower section recovers, it is likely that juvenile rearing habitat will be expanded to include this stretch of Pine Creek.

Habitat relationships In Eagle Lake, adult trout are found throughout the lake during fall, winter, and spring, when more uniform temperatures throughout the water column exist. During the hot summer months when the lake stratifies, trout are found in the deeper (deepest point approximately 100 feet), southernmost basin where cooler temperatures and higher oxygen levels persist. In the early part of this century, widespread introductions of non-native species occurred and were mostly unsuccessful due to high alkalinity (pH 8.6 to 9.4) of this lake. The lower 20 miles of Pine Creek is in poor condition and Eagle Lake trout do not have access to upstream habitat due to an existing trapping facility at the mouth of Pine Creek. The upper seven miles of perennial habitat is occupied by brook trout. The lack of access to the upper portions of Pine Creek has resulted in a lack of understanding of the habitat use by Eagle Lake trout when in the stream environment.

FEIS Volume 3, Chapter 3, part 4.4, page 248 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Previous stream inventories conducted in 1996 and 1997 found habitat upstream of State Highway 44 meets most of the habitat needs listed above. Pools generally have good cover components of overhanging vegetation, and in the forested reaches, an abundance of woody debris. Eagle Lake trout require access to reach their historic spawning grounds. Once at their spawning grounds, clean gravels are necessary for successful reproduction, pools and riffles are necessary for rearing and holding of Eagle Lake trout while in the stream, and isolation from competitors (brook trout) is likely needed to restore their numbers to genetically sustainable levels. Rearing and spawning areas are presently occupied by brook trout. Generally, stream reaches are in a recovering mode based on recent management changes and restoration projects designed to benefit the return of Eagle Lake trout.

Diet. Diet will most likely differ between the stream and the lake environment, and vary between season and age of the fish. Stream dwellers are likely to ingest a wide variety of macroinvertebrates found in Pine Creek. Newly planted Eagle Lake trout in the lake environment feed on zooplankton and benthic invertebrates. At the end of their first summer in the lake, most trout are feeding on young-of-the-year tui chubs (King 1963).

Status The Eagle Lake rainbow trout was petitioned for listing under the Endangered Species Act in June of 1994. In 1995, the US Fish & Wildlife Service (USFWS) found that the petition did not present substantial information to determine whether this subspecies of rainbow trout should be proposed for listing. According to the finding published in the Federal Register on August 7, 1995, the Eagle Lake trout will remain a species of concern. As additional information becomes available, the USFWS may reassess the listing priority for the trout or the need for listing (Pustejovsky 1995). The Eagle Lake rainbow trout was listed as a Forest Service Sensitive species on June 10, 1998. Due to dramatically low population numbers in the 1940s and 1950s, the fish trap was constructed to prevent their unsuccessful upstream migration attempts. A successful hatchery program has resulted in the widespread rearing and planting of this species, as they are quite tolerant of marginal water quality conditions. Close to 200,000 fish are planted into Eagle Lake each year to provide for a trophy trout fishery, as they achieve good size in a short time in the lake (17 to 18 inches and 2 to 3 pounds after 1 year in the lake).

Historical and Current Distribution Eagle Lake rainbow trout are endemic to Eagle Lake and its tributaries in Lassen County in Northeastern California. The majority of its historic habitat is on Federal lands administered by Lassen National Forest and the Eagle Lake Resource Area of the Bureau of Land Management. Presently, this species occupies Eagle Lake, and is prevented from migrating upstream in any of the tributaries to Eagle Lake. A fish trap is present at the mouth of Pine Creek (the main tributary to Eagle Lake) to allow the California Department of Fish & Game to collect the fish for spawning and rearing at hatcheries. Eagle Lake rainbow trout are not known to occur in Pine Creek at this time, but are heavily planted in Eagle Lake and other waters throughout the western United States.

Risk Factors Factors that caused the decline in the 1940s include: habitat alteration from road construction (timber harvest activities), railroad construction (including water diversions), unregulated and, or, improper grazing practices in the early part of the century and beyond, overharvest of the species in the early part of the century, and drought. Introduced brook trout in the uppermost reaches of Pine Creek may outcompete rainbow trout for space and food.

FEIS Volume 3, Chapter 3, part 4.4, page 249 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

VOLCANO CREEK GOLDEN TROUT (Oncorhynchus mykiss aguabonita) Life History Volcano Creek golden trout are opportunistic feeders. They feed on virtually every invertebrate organism that live in or fall into the water (Moyle 1976). These include larval and adult aquatic insects, as well as a few terrestrial forms, in streams. In lakes, food items include caddisfly larvae, chironomid midge larvae, and planktonic crustaceans (Curtis 1934). The abundance of micorcrustaceans has most likely contributed to the success of mountain lake populations (Moyle 1976).

Reproductive maturity is reached by the third or fourth year in golden trout. Spawning occurs in late June and July, when water temperatures are between 7 and 10 degrees Centigrade. Gravel riffles are used for breeding habitat. Spawning usually does not occur within lakes. Females produce between 300 and 2,300 eggs, depending on size (Curtis 1934). With waters at approximately 14 degrees Centigrade, eggs hatch in about 20 days. The fry remain in gravel riffles up to three weeks after hatching and are about 1 inch in total length. Fry spawned from lake dwelling adults, migrate back from tributary streams when they reach about 1.8 inches total length.

Growth rates of Volcano Creek golden trout are relatively slow, presumably a product of a short cold water growing season and low productivity (Knapp and Dudley 1990). Stream forms typically grow 1.2 to 1.6 inches their first year, 2.7 to 3.1 the second year, and 3.9 to 4.3 inches the third year. Growth continues at 0.39 to 0.78 inches per year, reaching a maximum length of 7.4 to 7.8 inches total standard length (Knapp and Dudley 1990). Lake forms reach 1.6 to 1.9 inches their first summer, 3.9 to 5.9 inches during the second, 5.1 to 9.0 inches the third year, and 8.2 to 10.9 inches by the fourth year (Curtis 1934). When left unfished, lake form golden trout will reach 13.7 to 7.2 inches total standard length by the seventh year of life.

Habitat relationships Volcano Creek golden trout occupy streams of the Kern Plateau above 7,590 feet elevation. The valleys within the basin are broad and flat, and filled with alluvial gravel. Streams in the basin run wide, shallow and exposed. Riparian vegetation is sparse but provides suitable cover for the trout. Stream are typically cold ( 3 to 22 degrees Centigrade in summer) and clear, with bottoms composed of sand, gravel, and cobble (Moyle and others 1995).

Status The main threat to the Volcano Creek golden trout populations is degradation of their streams from the grazing of livestock. The introduced brown trout (Salmo trutta) is both predator and competitor. Efforts have been made to remove brown trout from the native habitats of the golden trout, but reintroduction by anglers continues (Moyle and others 1995).

Historical and Current Distribution The Volcano Creek golden trout is native only to the Upper Kern River Basin, Tulare and Kern Counties, California. Early records show presence in the basin in the main Kern River, the lower Little Kern River, Golden Trout Creek (of which Volcano Creek is tributary to), and the South Fork of the Kern River (Schreck and Behnke 1971). These golden trout have been translocated to other waters around the state. As a result of stocking, Volcano Creek golden trout are found in hundreds of high elevation lakes and streams outside of their natural range. Golden trout have also been established in other mountain habitats around the western states and provinces (Moyle 1976).

FEIS Volume 3, Chapter 3, part 4.4, page 250 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Volcano Creek golden trout occur in densities of 8 to 52 individuals per 330 feet of stream, within their native range. The lowest numbers often occur in the more degraded reaches of habitat (Knapp and Dudley 1990). Although common outside of the native range, golden trout hybridize readily with rainbow trout (Moyle and others 1995).

Much confusion and debate has occurred over the taxonomy of this golden trout (Moyle 1976). The description for three separate golden trout was originally given; Salmo aguabonita from the South Fork of the Kern River (Volcano Creek), S. whitei from the Little Kern River, and S. roosevelti from Golden Trout Creek. The South Fork and Little Kern River forms were eventually recognized as subspecies of S. aguabonita, (S. a. aguabonita, and S. a. whitei). S. roosevelti is believed to be a color variation of S. a. aguabonita. Berg (1987) found a greater similarity between the Kern River rainbow trout (O. m. gilbertti) and the two subspecies of golden trout, than the two had between each other. Thus, the Volcano Creek golden trout is believed to be a subspecies of rainbow trout, and was given the name O. m. aguabonita. Golden trout are readily identified by their brilliant colors. The cheeks and belly are bright red to orange, the lower sides are gold, The lateral band is red-orange, and the back is a dark green to olive. Nine to ten parr marks are visible. The lower fins are orange with white tips, were the dorsal fin is dark with a white to orange tip. It is widely speculated that these colors aid in the camouflage against the volcanic substrates on the stream beds. Few natural predators also promotes the highly visible colors (Moyle 1976).

Risk Factors Risk to the Golden trout include the immediate loss of individual fish and loss of specific habitat features such as undercut banks use for cover, increases in sedimentation leading to changes in spawning bed capacity, and the loss of riparian vegetation necessary to maintain adequate temperature regime. The risk factors identified are primarily a result of historic and current grazing practices.

GOOSE LAKE REDBAND TROUT (Oncorhynchus mykiss ssp.) Life History There are two life history strategies present in the Goose Lake redband trout: a lake strategy and a headwater stream strategy. Lake fish live in Goose Lake where they grow to large size and spawn in the tributary streams. Headwater fish remain small and spend their entire life cycle in the streams. It is likely, although unproven, that the two forms represent one population because the aperiodic dessication of Goose Lake presumably has eliminated the lake forms repeatedly in the past.

In California, the lake-dwelling form spawns in the tributaries and headwaters of Lassen and Willow Creeks. If sufficient flows are available, they spawn primarily in Cold Creek, a small tributary of Lassen Creek, and in Buck Creek, a small tributary of Willow Creek. Upstream of its confluence with Cold Creek, a steep, rocky gorge apparently prevents spawners from ascending further up Lassen Creek. In Oregon, they formerly spawned in Thomas Creek and its tributaries and possibly in Cottonwood and Drew Creeks. In recent years, the largest spawning run occurred in Lassen Creek (J. Williams, unpubl. data). Although large spawners had been observed in lower Willow Creek, diversion structures prevented most or all the fish from reaching suitable spawning and rearing habitat in Buck Creek. Buck Creek has been severely degraded by irrigation diversion structures, but it has considerable potential for improvement as a spawning stream. Spawning migrations occurred in Willow and Lassen Creeks during late March 1988. Adults returned to the lake in April. Young trout may spend one or more years in the stream before moving down into Goose Lake. In the lake, the

FEIS Volume 3, Chapter 3, part 4.4, page 251 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 trout presumably feed on the abundant Goose Lake tui chub. Growth seems to be rapid, as scales from 6 spawning fish (10.5 to 18.7 inches) taken in 1967 indicated that they were all 3 years old (CDFG files).

The life history of the stream-dwelling form has not been studied, but it is presumably similar to that other redband and rainbow trout that live in small, high-elevation streams. Such trout typically spawn in thier third spring and live four to five years.

Habitat relationships These fish can survive an extended duration of warm temperature (15 to 20 degrees Centigrade), and the high alkalinity and turbidity that exist in Goose Lake in summer that would be lethal to most other trouts. Redband trout in tributaries have survived short duration temperatures as high as 29 degrees Centigrade. Spawning areas are located in high-elevation sections of tributary streams as much as 24.8 to 31 miles from the lake. Prior to spawning, adults must have access from the lake to spawning areas. In every stream this means negotiating extensive agricultural areas characterized by water diversions, erosion, and channelization. Logjams and beaver dams also may prohibit or restrict upstream movement of spawners during low-flow conditions. After spawning, adults and eventually the young must have passage back to Goose Lake. The spawning sites themselves must be non- degraded reaches of streams with clean gravels and suitable riparian cover for maintenance of cool water temperatures. Goose Lake redband have been observed to spawn in lower reaches of Willow and Lassen Creeks when access to upstream areas is blocked (P. Chappell, pers. comm.), but siltation and high temperatures probably preclude successful recruitment in the lower reaches of these streams. The habitat requirements of the stream-dwelling form are presumably similar to other populations of redband trout that occupy small, cool, high-elevation streams.

Status The Goose Lake redband trout population historically has undergone major fluctuations, being depleted during a series of dry years but recovering in wet periods. In the 1930s, for example, Goose Lake was almost completely dry, which decimated the lake-dwelling subpopulation; reestablishment of the lake population presumably resulted from colonization by stream-dwelling fish (E. Gerstung, pers. comm.). The lake population was severely depleted during the 1976 to 1977 drought, recovered during the wet years in the early 1980s, and again dropped precipitously during the 1986 to 1992 drought. Goose Lake dried up in 1992 and there is little reason to think any of the lake fish survived. In the past, interviews with local residents indicate that both sport and commercial fisheries existed for Goose Lake redband trout and that large runs occurred in local creeks, especially Thomas Creek in Oregon. Together, tributaries in Oregon, at one time, supported runs of several thousand fish, but storage dams, diversion structures, and habitat degradation eliminated most of these runs. Lassen Creek and Willow Creek are presently the only streams still easily accessible to spawning redbands in years when there is water in the lake. Numbers of spawning fish in these streams fluctuated from ten or so individuals to several hundred, but the creeks appear to have the potential to support perhaps a thousand spawning fish under optimal flow conditions (E. Gerstung, pers. comm.). The most recent runs of spawning fish in Lassen and Willow Creeks occurred in years 1997 to 2000, with 300 to 800 spawners present (M. Yamagiwa, pers. comm.)

The stream form of the Goose Lake redband trout apparently exists in about 15 small headwater streams and the number in each population are not known. It is safe to assume, however, that numbers in these streams became reduced during the drought, as stream flows declined.

FEIS Volume 3, Chapter 3, part 4.4, page 252 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Historical and Current Distribution The Goose Lake redband is endemic to Goose Lake and its tributaries (Lassen and Willow Creeks in California and the extensive Thomas Creek system and Crane Creek in Oregon) as well as to smaller streams such as Cottonwood and Pine Creeks in California and Augur, Bauer, Camp, Cox, Drew, Snyder Meadow, Shingle, Mill and Warner Creeks in Oregon. Berg (1987) reported that Joseph, Parker Creek and East Creek, tributaries to the upper Pit River in California, contained trout gentically similar to Goose Lake redband.

Risk Factors Goose Lake redband trout are threatened by many factors, but habitat modification of the streams and the lake are the biggest threats.

Habitat modification. Populations of the lake-dwelling form were initially reduced because spawning habitat in streams was largely unavailable. Most of the former spawning streams have been channelized in their lower reaches, as well as damned and diverted. These changes effectively blocked the large runs that once existed. In some areas, passage problems have been exacerbated by beaver dams. In addition, potential spawning areas have been degraded from human activities in the drainage (for example, livestock grazing, roadbuilding, and logging) which did not pay sufficient attention to maintaining aquatic and riparian habitats. Such activity reduces riparian and pool cover needed by the spawning adults and can cause the accumulation of silt in spawning riffles, smothering the eggs. Because of water diversions and loss of wetland areas in the drainage that once helped to retain water and streamflow, Goose Lake probably dried up faster than usual. However, there is evidence that Goose Lake dried up in the 1420s, in the 1630s and 1926 (with low lake levels from 1925 to 1939). Thus the key to the survival of the Goose Lake trout (and other fishes) was presumably conditions in the lower reaches of the streams. During the dry periods, the lake dwelling trout persisted by either (1) maintaining populations in the lower reaches of the tributary streams, which assumes the streams had year-round flows, or (2) repeated recolonization from the resident populations in the headwaters, which assumes that fish from headwater populations are capable of adopting the potadromous habit.

The headwater streams containing redband trout have been heavily grazed, resulting in reduced riparian cover and, in places, down-cutting to bedrock. The impact of grazing has been reduced in recent years through a combination of fencing, rotational grazing, installation of erosion control structures, and planting willows (H. Jasper, pers. comm.). Roads are also a problem on some streams, especially where culverts may be barriers to fish movement or where the road-cuts are a source of silt. Some streams have multiple problems. For example, Auger Creek in Oregon has poor water quality as the result of road building, channalization, and waste material from uranium mines. Overall, the stream populations are few, small, and isolated and are therefore vulnerable to extinction due to natural events, especially where stream and riparian habitats have been degraded.

Much of the critical stream habitat for Goose Lake redband trout is on private land and much of the water the depend upon is diverted through riparian water rights. Thus, their long-term survival depends on the close cooperation of private landowners with agencies. Most of the remaining habitat is either on the Fremont or Modoc National Forests where cooperative efforts to improve stream conditions have been undertaken.

Overexploitation. When lake-dwelling fish are moving upstream to spawn, they are extremely vulnerable to angling or poaching, especially when confined below an artificial barrier. This may

FEIS Volume 3, Chapter 3, part 4.4, page 253 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 have been a factor in the decline of the Lassen and Willow Creek populations, although the lower reaches of the Goose Lake streams were closed to angling during much of the drought period.

Introduced species. Brook, brown, and rainbow trout have been introduced into streams of the Goose Lake drainage in the past. Brown trout are now well established in two California tributaries, Pine and Davis Creeks. California has not stocked any rainbow trout in the drainage since 1980, when electrophoretic studies indicated that the native trout were distinct; however, planting of hatchery rainbow trout apparently has continued in Oregon tributaries (R. Elliott, pers. comm). The potential for future introductions to disrupt the native trout populations throught disease, hybridization, predation, or competition remains. Numerous attempts have also been make to introduce warm water fishes, including striped bass, into the lake, but they have been largely unsuccessful, presumably because of the lake’s alkalinity. This does not preclude the possibility that at some time fish or invertebrate species could be introduced that would disrupt the lake ecosystem as it exists today. A number of warm water game fish species have already become established in reservoirs and farm ponds within the Oregon portion of the drainage. The Fathead minnow (Pimiphales promelus), an introduced species, is now established within both the Oregon and California portions of the drainage.

MC CLOUD RIVER REDBAND TROUT (Oncorhynchus mykiss ssp.) Historical and Current Distribution The McCloud River redband trout is not known to occur within the project area.

WARNER VALLEY REDBAND TROUT (Oncorhynchus mykiss ssp.) Habitat relationships They are found in California in the upper Dismal and Twelvemile Creek drainages. In Oregon, they are found in Twelvemile, Twentymile, Deep, Honey, Dismal, Mosquito, Mud and Willow Creeks, and in Hart and Crump Lake

Historical and Current Distribution The Warner Valley redband trout in found in the Warner Valley drainage in south-central Oregon and small portions of northwestern Nevada and northeastern California.

Status Populations in Twelvemile Creek on the Modoc National Forest lands are low. Populations in the headwaters of Deep and Honey Creek, Fremont National Forest, appear to be healthy. In September 1997, the USFWS received a petition to list the redband trout as threatened or endangered throughout its range. In March 2000, a 12-month petition finding determined that the listing of the redband trout was not warrented.

Risk Factors Threats resulting for genetic introgression and competition with hatchery rainbow trout is occurring throughout the species limited range. Suitable habitat on the Modoc National Forest is “functioning at risk.” Other threats include but may not be limited to exotic fish, water diversions, channel manipulation, grazing practices and drought.

FEIS Volume 3, Chapter 3, part 4.4, page 254 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

GOOSE LAKE SUCKER (Catostomus occidentalis lacusanserinus) Life History Little is known about the life history of the Goose Lake sucker, except that they spawn during spring in the streams that are tributary to Goose Lake (Martin 1967). Adults can be found in the streams and lake throughout the year. Young suckers 1.6 to 2.8 inches long are very abundant in shallow water during summer in the lake, “packed” in among the aquatic macrophytes. (R. White, unpubl. data). Fish become sexually mature by the second year when they are 3.2 to 3.6 inches. Martin (1967) found several fish (5.6 to 8.6 inches), both male and female, with mature gonads at the beginning of April and concluded that C. o. lacusanserinus breeds during April or May, depending on water temperature. J. Williams observed 9.8 to 17.2 inch fish on a spawning migration in Willow Creek during May 14 to 16, 1984. Goose Lake suckers feed primarily on algae and diatoms (Martin 1967). Like other suckers, it has a long intestine and ventral mouth adaptive to this diet.

Habitat relationships Little information is available on this subspecies. In streams, C. o. lacusanserinus is typically found in water depths of 5.9 to 58.5 inches of moderate to slow velocity (Martin 1967). Streams they inhabit are up to 14.9 feet wide, with summer water temperatures of 15 to 19 degrees Centegrade. Little vegetation is present in the streams. Substrates consist primarily of rock and gravel in headwater sections and mud, silt and gravel in lower sections. Goose Lake, the principal habitat of the fish, is shallow, muddy and alkaline; it is described in the Goose Lake tui chub account. Gillnetting and trawling indicate that the sucker is found throughout the lake. Polulations of Goose Lake suckers are apparently also present in small reservoirs in the Cottonwood and Thomas Creek drainages, Oregon, but the characteristics of these reservoirs are not known. Juvenile fish have been observed in shallow water among emergent vegetation.

Status The drying of Goose Lake in 1992 led to the complete loss of the lake population. It is assumed that refugial populations of suckers from the above named tributaries have repopulated the lake. In 1994, suckers were common in a small portion of Lassen Creek and throughout most of Willow Creek. The most recent spawning run of 200 to 300 fish was observed in Willow Creek during May 5, 2000 (M. Yamagiwa, pers. Comm.)

Historical and Current Distribution The Goose Lake sucker is restricted to the Goose Lake basin and has been reported from Goose Lake and Willow, Lassen, Branch, and Corral Creeks, Modoc County, California; and from Dog, Drews, Cottonwood, Dry, Thomas, Cox, and Warner Creeks, Lake County, Oregon (Sato 1992a). It is also known from Drews and Cottonwood reservoirs in Oregon, but it is not certain if permanent populations are established in these reservoirs. Apparent spawning runs, however, have been recorded recently.

Risk Factors The principal threat to the Goose Lake sucker is destruction of its habitat in Goose Lake and its tributaries. Diversions for irrigation, combined with the loss of natural water-storage areas (for example, wet meadows lost to bank erosion and downcutting of streams) presumably caused the lake to dry up rapidly during a period of prolonged drought. While the lake has dried up naturally in the past, it may do so more quickly now, or more frequently become too alkaline to support freshwater fishes such as the sucker. Diversions, dams, culverts, and other obstructions also presumably prevent

FEIS Volume 3, Chapter 3, part 4.4, page 255 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

migrating adults from reaching spawning areas in tributary streams and reduce stream habitat required for persistence of the fishes during long-term droughts. In addition, many of the streams have experienced some habitat loss due to the effects of logging, grazing and other factors that can degrade watersheds. The presumed populations in Drews and Cottonwood Reservoirs may help this sucker from becoming extinct, provided that the reservoirs are not drawn down to low as well.

LAHONTON LAKE TUI CHUB (Oncorhynchus clarki pectinifer) Life History In Lake Tahoe, nocturnal spawning occurs during May and June, possibly extending into July (Miller 1951). Tui chub may be serial spawners, reproducing several times during the spawning season (Moyle 1976). Reproductive adults spawn near-shore over beds of aquatic vegetation, to which the eggs adhere (Snyder 1917). Young remain near-shore until winter when body size is 0.39 to 0.78 inches long, when they migrate into deeper water. Linear growth of tui chubs occurs about 4 years, then mass is accumulated rapidly. The largest documented length in Lake Tahoe is 5.3 inches total standard length, but longer chub (8.2 inches) have been found in Walker Lake, Nevada (Miller 1951).

Habitat relationships Based on occurrence within such widely diverse habitats as Lake Tahoe and Pyramid Lake, it is believed the species can tolerate a wide range of physicochemical water conditions. The species is known as a mid-water feeder. In Lake Tahoe, larger fish (greater than 6.2 inches total length) occur in deeper water (greater than 165 feet) during the day, moving into shallower water areas at night (Miller 1951). Young fish generally occur in shallow water. It has also been noted that a seasonal migration occurs within the water column with deeper water utilized during winter months and shallower water utillized during summer months (Snyder 1917, Miller 1951). Algal beds in shallow inshore areas seem necessary for spawning, egg hatching, and larval survival.

Lahontan Lake tui chub are schooling fish reaching 13.6 to 16.0 inches total length, which inhabit large, deep lakes (Moyle 1976). Lahontan Lake tui chub feed primarily on zooplankton, especially cladocerans and copepods, but also eat benthic insects when available (Miller 1951, Marrin and Erman 1982). Tui chub are eaten mostly by large trout, and rarely by birds and snakes (Miller 1951).

Status The Lake Tahoe population is the only confirmed population in the Sierra Nevada, with a probable population in Stampede, Boca and Prosser reservoirs. No water bodies within the pilot project area known to contain this species, although due to habitat connectivity, could occur in Weber Lake, Independence Lake, and Lake of the Woods. There is no evidence that this species is within these water bodies (Carlson, pers. comm 1999).

Little study has occurred on the Lake Tahoe population since Miller (1951). Zooplankton levels have changed over this period. Daphnia, an important prey of adult chubs, have been nearly eliminated (Richards and others 1975) by the introduced Kokanee salmon (Oncorhynchus nerka) and opossum shrimp (Mysis relicta) , both of which feed on zooplankton. Marshland degradation along the lake may be taking away vital spawning and nursery areas. Chubs are abundant in the three reservoirs but are increasingly threatened by operations that take little account of resident fish populations.

The Lahontan Lake tui chub was first described as Leuciscus pectinifer by Snyder (1917) who simultaneously described the sympatric form as Siphateles obesus. Hubbs and Miller (1943), however, considered L. pectinifer to be a subspecies of S. obesus and thus called it S. obesus

FEIS Volume 3, Chapter 3, part 4.4, page 256 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

pectinifer. Shapovalov and Dill (1950) recognized that both forms were part of the Siphateles bicolor complex and renamed them S. b. pectinifer and S. b. odesus, respectively. Baily and Uyeno (1964) designated Siphateles as a subgenus of Gila and designated the fine gill raker tui chub as Gila bicolor pectinifer.

Historical and Current Distribution The Lahontan Lake tui chub are a cyprinid subspecies found in Lake Tahoe and Pyramid Lake (Nevada) which are connected to each other by the Truckee River and in nearby Walker Lake (Nevada). Populations of plankton-feeding chub occurring in Stampede, Boca and Prosser reservoirs may also be Lahontan Lake tui chub due to morphological similarities (Marrin and Erman 1982, Moyle and others 1995). These three reservoirs are located on the Truckee Ranger District.

Risk Factors Threats to the Lake Lahontan tui chub include but are not limited to water quality, specifically alkalinity due to diversions of inflowing water, change in prey base due to introduced species, and reservoir and wetland management.

GOOSE LAKE TUI CHUB (Gila bicolor thalassina) Life History The life history of this subspecies has been little studied. Chubs commonly reach 10 inches total length in the lake and fish as large as 12.6 inches have been collected, indicating that his form may be very long-lived in lake habitats. In streams, however, they rarely exceed 4.8 inches. The size distribution of tui chubs sampled from Goose Lake in 1989 show two modes. The great majority (greater than 90 percent) of fish were less than 4.8 inches total standard length (SL), while the remainder were 8 to 12 inches SL (R. White, unpubl. data). Most tui chubs are opportunistic omnivores and consume a wide variety of aquatic invertebrates (Moyle 1976). Tui chubs are a major prey of Goose Lake lamprey; depending on the length class, 20 to 70 percent of the tui chubs greater than 8 inches SL sampled in 1989 had lamprey scars (R. White, unpubl. data).

Habitat relationships Goose Lake is a massive, natural alkaline lake covering approximately 9,633 acres along the Oregon- California border. The lake is shallow, averaging 8.3 feet deep, and is hypereutropic and very turbid (Johnson and others 1985). Physical limnological measurements take by R. White (unpubl. data) reveal that a thermocline may be present, depending on wind conditions. On a calm September day, water temperature at one sampling locality was 17 degrees Centegrade from the surface to 15.6 inches deep, with a sharp drop at 15.6 to 19.5 inches, and 14 to 15 degrees centigrade at between 19.5 and 78 inches deep. At a second locality, temperature decreased from 23 degrees Centigrade at the surface to 15 degrees Centegrade at 13.7 inches, remaining at about 15 degrees Centegrade between 13.6 and 103 inches. At those two localities, dissolved oxygen concentration held constant in deeper water, depending on locality. The drop in oxygen occurred at about 58.5 inches in depth at one locality, and at between 101 and 105 inches at the second locality. On a windy September day, the water temperature was 15 degrees Centigrade throughout the water column measured at one locality. Dissolved oxygen was constant from the surface to 66.3 inches in depth, but dropped abruptly at about 68 to 70 inches.

The surface elevation of Goose Lake fluctuates seasonally, but averages 4,750 feet. In California, no tui chub have been found in streams above 4,755 feet in elevation, although tui chubs have been

FEIS Volume 3, Chapter 3, part 4.4, page 257 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 found above 5,115 feet in Oregon streams (J. Williams, unpubl. data). Chubs prefer pools and are generally not found in swift water, although they have been collected from runs in Battle Creek near the west shore of Goose Lake (J. Williams, unpubl. data). In July 1992, large numbers of chubs have been collected in the lower reaches of Thomas, Willow, and Lassen Creeks (G. Sato. pers. com.), where they may have been attempting to escape from the increasing alkalinity of the drying lake.

Goose Lake tui chubs also occur in several small reservoirs in the Thomas Creek drainage, Oregon, but the limnological characteristics of those reservoirs are not known.

Status This tui chub has been extremely abundant in the lake. During the 1966 gillnetting surveys of Goose Lake, it constituted 88 percent of fishes collected (King and Hanson 1966). In 1984 it constituted nearly 96 percent of gillnet collections (R. White, unpubl. data), and in 1989 was 96 percent of fishes sampled by trawls, gillnets, and seines (R. White, unpubl. data). Large numbers of chubs could be caught with comparatively little sampling effort (for example more than 100 were captured in a 5 minute haul with a small trawl). In 1992, the chubs were eliminated from the lake as it became progressively more shallow and alkaline and then dried up. As lake levels dropped, fish crowded into the inflowing streams where they were extemely vulnerable to predation from white pelicans and other fish-eating birds. Apparently the tui chubs survived in greatly reduced numbers in stream pools and in some upstream reservoirs. When the lake was dry in late June of 1944, chubs were abundant in a small portion of Lower Willow and Lassen Creeks.

Historical and Current Distribution This species is endemic to the Goose Lake Basin. It is confined to but widely distributed and abundant in Goose Lake. The chub also occurs in low-elevation sections of streams tributary to Goose Lake and Everly Reservoir in California, as well as in Cottonwood, Dog and Drews Reservoirs in Oregon (Sato 1992a).

Risk Factors The principal threat to the Goose Lake tui chub is the loss of water in its principal habitat, Goose Lake, accompanied by loss of refuge habitat in tributary streams and reservoirs in the drainage. Although the lake has dried up in the past, diversions for irrigation and loss of natural water storage areas (like wet meadows) presumably caused it to dry up more rapidly during the recent period of prolonged drought. Even in the absence of complete drying, reduction of inflows would increase the likelihood that the lake will periodically become too alkaline to support freshwater fishes such as the tui chub. High alkalinity may be particularly a problem to early life-history stages. The key to the survival of Goose Lake tui chubs in the past presumably was the presence of refuges in the springs and pools of the lower reaches of tributary streams. A number of these refuges still exist but may be in a degraded condition as the result of road building, agricultural activity, reduced streamflow, and other factors. On the other hand, reservoirs created for storage of irrigation water apparently now serve as major refuges for the tui chubs because they have fairly large minimum pools.

When water levels rise in the lake again and alkalinity levels decline, attempts may be made, both officially and unofficially, to introduce exotic game fishes to the lake, as has been done in the past. If such introductions are successful, they could cause a major shift in the Goose Lake ecosystem, away from native fishes such as the tui chub.

FEIS Volume 3, Chapter 3, part 4.4, page 258 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

HARDHEAD (Mylopharodon conocephalus) Life History Hardhead mature after their second year and most likely spawn in the spring (Reeves 1964), judging by the by the upstream migrations of adults into smaller tributary streams during this time of year (Murphy 1947, Bell and Kimsey 1955, Rowley 1955). Estimates based on juvenile recruitment suggest that hardhead spawn by May to June in Central Valley streams and that the spawning season may extend into August in the foothill streams of the Sacramento-San Joaquin River drainage (Wang 1986). Spawning presumably involves mass spawning in gravel riffles (Moyle 1976). Female fecundity varies from 9,500 to over 20,000 eggs (Burns 1966).

Hardhead reach 2.7 to 3.1 inches by their first year, but growth slows in subsequent years. In the American River, hardhead reach 11.7 inches total standard length in 4 years, whereas in Pitt and Feather Rivers, it takes six years to reach that length (Moyle and others 1983, PG&E 1985).

Habitat relationships Hardhead are typically found in undisturbed areas of larger middle and low elevation streams (Moyle and Nichols 1973, Moyle and Daniels 1982). Elevational range is 30 to 4,750 ft (Reeves 1964). Most streams occupied by hardhead have summer temperatures in excess of 20 degrees Centigrade, selecting an optimal range between 24 and 28 degrees Centigrade (Knight 1985). Hardhead are relatively intolerant of low oxygen levels, especially at higher temperatures, a factor which may limit their distribution to well oxygenated streams and the surface water of reservoirs (Cech and others 1990). They prefer clear, deep (greater than 3.3 feet) pools with sand-gravel-boulder substrates and slow water velocities (Moyle and Nichols 1973, Knight 1985, Moyle and Blatz 1985). In streams, adult hardhead tend to remain in the lower half of the water column, rarely moving into the upper levels (Knight 1985), while juveniles concentrate in shallow water close to the stream edges (Moyle and Baltz 1985). Hardhead are always found in association with Sacramento squawfish and usually with Sacramento suckers. They tend to be absent from streams introduced with exotics, especially centrarchids (Moyle and Nichols 1973, Moyle and Daniels 1982), or streams that have been severely altered by human activity (Baltz and Moyle 1993). Studies indicate hardhead populations are maintained when native species compositions remain at 90 percent or higher and small mouth bass are absent (Moyle and Nichols 1973; Brown and Moyle 1993).

Diet. Primarily bottom feeders, hardheads forage for benthic invertebrates and aquatic plant material in quiet water. They will occasionally feed on plankton and surface insects. Smaller fish feed primarily on mayfly larvae, caddisfly larvae, and small snails, whereas larger fish feed mainly on aquatic plants (especially filamentous algae), as well as crayfish and other large invertebrates (Reeves 1964). The ontogenetic changes in teeth structure seem to fit the dietary switch. Reeves (1964) found no remains of fish in the stomachs of large hardhead.

Status The hardhead is a large, native cyprinid (minnow family) that can exceed 24 inches in length (Moyle and others 1995). Hardhead were first described by Baird and Girard (Girard 1854) as Gila conocephala. Ayres (1854) redescribed it as Mylopharodon robustus. G. conocephala was later classified as M. conocephalus and considered to be closely related to M. robustus. Electrophoretic studies by Avise and Ayala (1976) indicate the hardhead to be closely related to the Sacramento squawfish, but different enough to be considered a separate species.

FEIS Volume 3, Chapter 3, part 4.4, page 259 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Hardhead is a non-game species. Hardhead was added to the Pacific Southwest Region Forest Service Sensitive species list (revised June 10, 1998) for the Tahoe and Plumas National Forests. Although not identified as sensitive for the Lassen National Forest, this species is known to occur in a few areas within the national forest boundary. The Tahoe, Plumas and Lassen National Forest Management Plans do not provide specific management guidelines for this species. However, general guidelines direct the forests to improve habitat capability for riparian associated species, and favor riparian dependent resources in cases of competing resource demands (USDA 1988).

Historic and Current Distribution Historically, hardhead have been regarded as a widespread and locally abundant species (Ayres 1854, Jordan and Evermann 1896, Evermann 1905, Rutter 1908, Murphy 1947, Soule 1951, Reeves 1964). Hardhead are still widespread in the foothill streams, but their specialized habitat requirements, combined with widespread alteration of downstream habitats, has resulted in the isolation and localization of populations. These conditions increase the chance for localized extinctions. Hence, hardhead are less abundant than they once were, especially in the southern half of their range.

Hardhead are widely distributed in low to mid-elevation streams in the main Sacramento-San Joaquin River drainage as well as the Russian River drainage. Their range extends from the Kern River, Kern County, in the south to the Pitt River, Modoc County, in the north. Populations are scattered in the tributary streams of the San Joaquin River drainage, but have not been found in the valley reaches of the San Joaquin River (Moyle and Nichols 1973, Saiki 1984, Brown and Moyle 1987). In the Sacramento River drainage, hardhead are present in most of the larger tributary streams as well as the Sacramento River. They are widely distributed in the Pitt River drainage (Cooper 1983, Moyle and Daniels 1982), including the main Pitt River with its series of hydroelectric reservoirs.

Risk Factors Current threats included but are not limited to widespread alteration of downstream habitats, population isolation which increases possibility of local extinction, habitat loss from hydroelectric power developments, and predation by exotic species.

II. Environmental Consequences Fish species of the Sierra Nevada exhibit a high degree of endemism due in part to a dynamic climatic and geological history. Many Sierra Nevada fishes have very restricted distributions, often limited to single drainage basins or in some cases to single streams or springs. Of the 40 fish species that can be characterized as having declined in population size (Table 4.5.2a), 11 have special status as being Forest Service Sensitive Species.

FEIS Volume 3, Chapter 3, part 4.4, page 260 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

The following species sensitive species (focal species) are in the Sierra Nevada bioregion:

Table 4.4.4a. Sierra Nevada bioregion fish focal species. Species Scientific Name Proximity1 Habitats2 Elevational Vulnerability4 Association3 Goose Lake lamprey Lampetra. tridentata ssp FS S F M Chinook salmon, Fall run Oncorhynchus tshawytscha A, L L, F L Eagle Lake rainbow trout O. m. aquilarum FS K H M Volcano Creek golden trout O. m. aguabonita FS S, L, K F H Goose Lake redband trout O. m. ssp. FS S F L McCloud River redband trout* O. m. ssp. NA S F M Warner Valley redband trout O. m ssp. NA S F, H M Goose Lake sucker Catostomus occidentalis FS S, K F M lacusanserinus Lahontan Lake tui chub Gila bicolor pectinifer K L, F, H L Goose Lake tui chub G. b. thalassina FS S F M Hardhead Mylopharodon conocephalus S, L F L

*The McCloud River redband trout is not known to occur within the project area. 1Proximity: FS = distributed primarily within or adjacent to Forest Service boundaries; NA = found outside areas of direct influence; if no class is identified, species are affected to varying degrees depending on distribution of individual populations. 2Principal habitats: A = anadromous, S = small stream, L = large stream, K = lakes and reservoirs, P = springs. 3Elevational Association: L = lowlands, F = foothills, H = high elevations 4Vulnerability: L = low, M = moderate, H = high.

Effects of the Alternatives Management Areas. Alternatives 2, 5, 6, 8, and Modified 8 identify special management areas, selected in part based on consideration of fish habitats. Critical Aquatic Refuges (CARs) are included in Alternatives 2, 6, 8, and Modified 8, while critical refuges are included in Alternative 5. Alternatives 2, 6, and 8 also delineate emphasis watersheds but Alternative Modified 8 has enlarged CARs on scale with emphasis watersheds. Alternative 5 delineates aquatic diversity areas, which generally overlap with emphasis watersheds, but encompass larger areas. These special areas are potentially significant to the viability of native fishes: 22 of these 40 fishes in the Sierra Nevada group with declining numbers can be found in these special management areas, including all that occur on national forest lands. Management in these areas could directly affect the entire current range of 11 fish species. Three additional declining species occur downstream of special management areas, and thus might benefit from planning and management activities in these areas.

Direction for planning and conducting management activities in special management areas varies between alternatives. Alternatives 2, 5, 6, 8, and Modified 8 state that critical aquatic refuges should be priority area for landscape analysis as well as a high priority for watershed restoration activities. Management activities within CARs should be compatible with the goal of sustaining and enhancing habitat for aquatic and, or riparian dependent species. Alternative 5 further emphasizes landscape analyses by requiring that they be completed before activities requiring documentation in an environmental assessment or environmental impact statement are conducted. Alternative 8 emphasizes landscape analyses by requiring them in emphasis watersheds before projects requiring a Decision Notice or a Record of Decision are conducted. Other alternatives emphasize landscape analyses to lesser extents than Alternatives 5 and 8, and require them in situations not specifically targeted to benefit fish.

In Alternative 5, special management areas (critical refuges and aquatic diversity areas) influence planning and management in the form of landscape analyses. Specifically, timber harvest and fuels treatments would be prohibited in these areas unless such activities contributed to the attainment of

FEIS Volume 3, Chapter 3, part 4.4, page 261 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Aquatic Management Strategy (AMS) goals. Pesticide use would be prohibited in special management areas under Alternative 5. These provisions would benefit fish where watershed disturbance or pollution were significant determinants of abundance.

The incorporation of AMS goals in the planning and implementation of management activities should influence management impacts on the population viability of fishes. The AMS goals affect all of the action alternatives (Alternatives 2 through Modified 8) in that mining and pesticide use in riparian areas “must be compatible with the goals of the AMS.” However, in contrast to the other alternatives, Alternative 5 uses the AMS goals as the basis for several standards and guidelines.

Riparian Areas. Linkages between riparian and instream processes and habitat indicate that differences between alternatives, in terms of their effect on fishes, are influenced by the extent of: (1) designated riparian areas and (2) the activities allowed in these areas. Determining how different approaches to delineating riparian areas might affect fishes is complicated by several factors. The variable width approach (used in Alternatives 2, 4, and 5) includes two classes of riparian areas (green and grey zones), with some standards and guidelines applying only to green zones. Both the variable width approach (Alternatives 2, 4, and 5) and Alternative 3 would use site-specific analyses to delineate riparian zones, but the specific outcomes of these analyses are uncertain. While the stream type, flexible width approach to identifying riparian areas (Alternatives 6, 7, 8, and Modified 8) would probably result in larger riparian areas along perennial streams than the variable width approach, the latter might be more effective in reducing cumulative watershed effects.

The alternatives provide a range of riparian area protection guidance. Alternative 5 is the most restrictive. Alternative 5 prohibits all land disturbing activities in green and grey zones unless they benefit riparian-dependent or aquatic species, or water quality, and the benefits are demonstrated via landscape analysis. Most of the other alternatives prohibit salvage or commercial logging in green zones (Alternative 2) or stream type riparian zones along permanent and intermittent streams (Alternative 8). Timber harvesting may be conducted in riparian areas, following different guidelines, under Alternatives 3, 4, 6, 7, and Modified 8. Alternatives 3 and 5 prohibit road building in riparian zones; Alternative 5 further addresses negative effects of roads on streams by requiring that failed road crossings and culverts be identified and have priority for rehabilitation.

Grazing. Effects common to all alternatives are those that contribute to the immediate loss of individual fish and loss of specific habitat features (for example, undercut banks and spawning beds) or localized reductions in habitat quality (like sedimentation and loss of riparian vegetation). One of the greatest risk factors, within the control of the Forest Service, to Forest Service Sensitive fish species in the western United States has been degradation of the aquatic environment, especially those resulting from long term livestock grazing.

Grazing practices in the western United States have led to severe degradation of some riparian areas and have greatly increased the nutrient and sediment export potential in many areas (Karr and Schlosser 1978, Gregory and others 1991). Behnke and Zarn (1976) identified livestock grazing as the greatest threat to the integrity of stream habitats in the western United States. Numerous publications have documented the detrimental effects of livestock grazing on streams and riparian areas. Effects on fish habitat can include nutrient loading, reduction of shade and cover with resultant increases in water temperature, more intermittent flows, changes in stream channel morphology, and the addition of sediment due to bank degradation and off-site soil erosion. Removal of streambank vegetation through grazing decreases shade and cover, which promotes greater water temperature

FEIS Volume 3, Chapter 3, part 4.4, page 262 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4 fluctuations, decreased water storage capacity, and increased erosion potential. Grazing effects are not limited to riparian areas. Grazing of upland vegetation can expose soils to erosive impacts of raindrops, reduce water infiltration, and accelerate runoff. This can erode topsoil and cut rills and gullies, that concentrate runoff, deepen gullies, lower water tables, and increase sediment production (Chaney and others 1993).

When livestock graze directly on streambank vegetation, mass erosion from trampling, hoof slide, and streambank collapse causes streambank soils to move directly into the stream (Platts 1990). Heavy trampling by livestock can compact soils, reducing the infiltration of overbank flows and precipitation. Reduced infiltration and increased runoff may decrease the recharge of the saturated zone and increase peak flow discharge (Platts 1990). Riparian areas in poor condition are unable to buffer the effects of the accelerated runoff. Doubling the speed of streamflow increases its erosive power by four times and its bedload and sediment carrying power by 64 times (Chaney and others 1993). Accelerated runoff can cause unstable stream channels to downcut or erode laterally, accelerating erosion and sediment production (Chaney and others 1993). Lateral erosion results in progressively wider and shallower stream channels that can adversely effect fish populations.

Streambank damage can eliminate habitat associated with banks (Armour 1977), alter stream morphology such as pool/riffle and width/depth ratios (Gunderson 1968, Platts 1979), and cover spawning areas with sediment that reduces survival of fish embryos (Bjornn 1969, Phillips and others 1975). Additionally, undercut banks that normally provide shelter, are often damaged or collapse in grazed areas, thus decreasing the amount of available fish habitat. Increased sedimentation due to bank collapse may decrease pool volume downstream, eliminating other important habitats.

The effects of grazing on woody vegetation are critical because of the importance of woody debris in providing nutrients, structure, pool formation and streambank stability, shading, and favorable microclimates. Grazing can eliminate woody species over time. While mature vegetation approaches senescence, excessive grazing pressures have prevented the establishment of seedlings (Carothers 1977, Glinski 1977). On streams rested from continuous grazing for ten years, Claire and Storch (unpublished) found alders and willows provided 75 percent shade cover over areas that had been devoid of shrub canopy cover before exclosure. Crouch (1978), Duff (1979), and Kauffman (1982) found similar results.

Other direct effects of livestock grazing on aquatic species include wallowing and wading in the stream. Direct wading in streams by livestock can be assumed to induce mortality on eggs and pre- emergent fry at least equal to that demonstrated for human wading (Roberts and White 1992).

In addition, some indirect effects are expected to occur. Trampling affects the hydrology of the watershed. Accelerated runoff temporarily increases streamflows but decreases the amount of water retained in the watershed to sustain base flows. Greater water yields have been demonstrated in grazed compared to ungrazed areas (Liacos 1962, Hanson and others 1970, Lushby 1970). Alderfer and Robinson (1949), Bryant and others (1972), Orr (1960), and Rauzi and Hanson (1966) all found soil compaction increased linearly with increases in grazing intensity. Rauzi and Hanson (1966) found water intake rates on a moderately grazed watershed to be nearly twice that on the heavily grazed watershed. Water intake rates on a lightly grazed watershed were nearly four times that on the heavily grazed watershed and over twice that on the moderately grazed watershed. Heavy grazing compacted the soil and significantly decreased the pore spaces in the top four inches of the soil when

FEIS Volume 3, Chapter 3, part 4.4, page 263 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

compared to light grazing. Twenty-two years of differential grazing resulted in changes in both plant species composition and soil properties (Rauzi and Hanson 1966).

A general reduction in the plant biomass of riparian areas can have multiple consequences. These can be increased water temperature, increased sedimentation, and decreased water storage. Increased sediment loads reduce primary production in streams. Reduced instream plant growth and woody and herbaceous riparian vegetation limits populations of terrestrial and aquatic insects. Alternative Modified 8 would implement grazing standards that limit grazing intensity and control the timing of grazing both for physiological plant needs and streambank protection. Protection would be given to willow flycatcher nesting habitat during the breeding season (June 1 through August 31) by either (1) using permanent or electrical fencing or (2) avoiding occupied habitat during the breeding season. If late season grazing was employed, this could have indirect negative effects on willow species because they are relatively more palatable at this time of year than associated upland vegetation. Cattle could utilize them differentially over other plants.

Environmental Outcomes The environmental outcomes for fish species are based on the implementation of the ACS and associated Standards and Guidelines for each alternative.

Table 4.4.4b. Assessment ratings over the 50 year planning horizon for Forest Service sensitive fish species. Alternatives Species Current 1 2 3 4 5 6 7 8 8 mod Goose Lake lamprey C C C C C C C C C C Chinook salmon- Fall run C C C C C C C C C C Eagle Lake rainbow trout B B B B B B B B B B Volcano Creek golden trout B B B B B B B B B B Goose Lake redband trout B B B B B B B B B B Warner Valley redband trout C C C C C C C C C C Goose Lake sucker D D D D D D D D D D Lahontan Lake tui chub B B B B B B B B B B Goose Lake tui chub B B B B B B B B B B Hardhead C C C C C C C C C C

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

FEIS Volume 3, Chapter 3, part 4.4, page 264 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Cumulative Effects Population Outcomes Table 4.4.4c represents the assessment ratings over the planning horizon for the Forest Service sensitive fish species. Alternatives Species Current 1 2 3 4 5 6 7 8 8 mod Goose Lake lamprey C C C C C C C C C C Chinook salmon- Fall run D D D D D D D D D D Eagle Lake rainbow trout C C C C C C C C C C Volcano Creek golden trout C C C C C C C C C C Goose Lake redband trout C C C C C C C C C C Warner Valley redband trout D D D D D D D D D D Goose Lake sucker D D D D D D D D D D Lahontan Lake tui chub B B B B B B B B B B Goose Lake tui chub C C C C C C C C C C Hardhead C C C C C C C C C C

Outcome A. Suitable environments are broadly distributed and of high abundance across the range of the species. Outcome B. Suitable environments are either broadly distributed or of high abundance across the range of the species; however, there are temporary gaps where suitable environments are absent or only present in low abundance. Disjunct areas of suitable environments are typically large enough and close enough to permit dispersal and interaction among subpopulations across the species’ range. Outcome C. Suitable environments are frequently distributed as patches or they exist at low abundance, or both. Gaps, where suitable environments are either absent or present in low abundance, are large enough that some subpopulations are isolated, limiting opportunity for species interactions. In most of the species range, subpopulations have the opportunity to interact as a metapopulation; however, some subpopulations are so disjunct or of such low density that they are essentially isolated from other populations. Outcome D. Suitable environments are highly isolated or they exist at very low abundance, or both. While some subpopulations associated with these environments may be self-sustaining, there is limited or no opportunity for population interaction. There has likely been a reduction in overall species range from historical conditions, except for some rare, local endemics that may have persisted in this condition since the historical period. Outcome E. Suitable environments are highly isolated and exist at very low abundance. Populations have little or no interaction, resulting in strong potential for local or regional extirpation, and low likelihood of recolonization.

FEIS Volume 3, Chapter 3, part 4.4, page 265 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.5 ReReptilesptiles 4.4.5.1. NORTHWESTERN POND TURTLE (Clemmys marmorata) Affected Environment Life History The breeding season occurs from March to August and 3 to 11 eggs are laid in nests dug in soil at least four inches deep and with a relatively high internal humidity (Zeiner and others 1988). Egg laying usually occurs in June to July and two clutches may be laid in a season though the frequency of this is unknown (Lardie 1975, Goodman 1997). Hatchlings typically over winter in the nest and emerge the following year (Buskirk 1990, Jennings and Hayes 1994). Along large, slow-moving streams, nests are constructed in sandy banks (Buskirk 1990). Along foothill streams, females may move considerable distances up hillsides to find a suitable nest site (Buskirk 1990). In California, nests are usually located on relatively unshaded slopes with largely southern aspects in substrates containing a high clay or silt fraction (Jennings and Hayes 1994). Hatchlings and juveniles have more specialized habitat requirements than do adults and primarily use shallow water with abundant emergent vegetation (Holland 1991). Devin Reese (pers. comm.) believes that low flow rates are more important than depth of water, and that ephemeral habitats in the form of seasonal ponds or seeps that are removed from the main watercourse may provide refuge from predation for juvenile turtles. Turtles are usually absent from steeper, fast moving sections of the streams or deep canyons. Deep, slow pools adjacent to unvegetated gravel bars are typically utilized (Reese and Welsh 1998).

Mats of submergent vegetation provide basking sites for all life stages, and are important to behavioral thermoregulation (Jennings and Hayes 1994). The life history of the western pond turtle is characterized by low fecundity, low hatchling and juvenile survivorship, high adult survivorship, and a potentially long life span (Holland and others 1992). Age to reproductive maturity (first breeding) in California is seven to eleven years (Jennings and Hayes 1994).

Habitat Relationships This species is associated with permanent or almost permanent water in a wide variety of habitat types below 6,000 feet elevation (Zeiner and others 1988). Basking sites, such as partially submerged logs, rocks, mats of floating vegetation, or open mud banks, are important habitat elements. Adult turtles have been known to move up to 1.24 miles as a result of diminishing habitat, but the recolonization potential of western pond turtles following extirpation of a local population is unknown (Jennings and Hayes 1994).

Diet. The western pond turtle primarily feeds on small aquatic invertebrates, but is omnivorous (Holland 1991). Foods include: aquatic plant material, such as pond lilies; beetles; aquatic invertebrates, such as Daphnia; fish; frogs; and carrion (Holland 1985, Jennings and Hayes 1994, Zeiner and others 1988).

Basking is a critical behavior for western pond turtles. The long-term health of western pond turtle populations depends on fires and seasonal scouring by high stream flows. These processes prevent streamside vegetation from becoming too dense for pond turtles to properly thermoregulate. Much of the suitable habitat for western pond turtles in the Sierra Nevada lies downstream of dams. Water releases from dams are a critical influence on the viability of downstream western pond turtle populations

FEIS Volume 3, Chapter 3, part 4.4, page 266 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Status The western pond turtle has been extirpated from or reduced to low numbers in a number of areas, and presently occurs in about 90 percent of the historic range (USFWS 1993b). Factors contributing to the local extirpations and reductions include loss and degradation of wetlands and terrestrial habitats, predation by introduced species, like bullfrogs and bass (Holland and others 1992). This species underwent a Status Review by the USFWS in 1992 to 1993, and they concluded that it did not meet the definitions of Endangered or Threatened Species at that time (USFWS 1993b). However, USFWS felt that concern and continued monitoring was justified.

Many of the western pond turtle populations examined by Dr. Dan Holland in California demonstrate a skewed age structure. They are heavily biased toward adults, indicating that little or no recruitment has taken place for the last 10 to 20 years (Jennings and Hayes 1994).

Eight affected national forests within the range of this species have conducted periodic surveys, although most have been project-related (Forest Wildlife Biologists, Pers. Comm.). Estimated population levels on the eight affected Forests have been reported between 3,500 and 7,000 individuals (Forest Wildlife Biologists, pers. comm.). Although population trends for this species are virtually unknown on these national forests, habitat availability trends over the past ten years are reported to be stable to declining (Forest Wildlife Biologists, pers. comm.). Primary threats to the western pond turtle in the Sierra Nevada national forests are reported to be grazing and large scale wildfire (Forest Wildlife Biologists, pers. comm.).

Historical and Current Distribution The western pond turtle (Clemmys marmorata) was historically found in a wide variety of wetland habitats west of the crests of the Sierra Nevada and Cascade Range (USFWS 1993b). The northwestern pond turtle (C. m. marmorata) presently occurs from the San Francisco Bay to the Puget Sound of Washington, and the southwestern pond turtle (C. m. pallida) is found from Monterey Bay south along the coastal region to northwestern Baja California, including the Mojave River (USFWS 1993b). The south San Francisco Bay area and the San Joaquin Valley have been described as an area of integration between the two subspecies (Seeliger 1945 in USFWS 1993b).

This species inhabits fresh or brackish water in permanent or intermittent ponds, lakes, and rivers from sea level to about 6,000 feet elevation, but is primarily found below 3,000 feet (Holland 1991). It is often restricted to areas near banks or in quiet backwaters where there is a relatively slow current, basking sites, and refugia (Holland 1991). The southwestern and northwestern pond turtles occur on the following National Forests: Modoc, Lassen, Plumas, Tahoe, Eldorado, Stanislaus, Sierra, and Sequoia (Timossi 1990).

Risk Factor Cattle grazing, road building, and logging in riparian areas can negatively impact western pond turtle populations. Protection of nesting sites and rearing habitat for juvenile turtles are key to future recovery and long term viability of western pond turtle populations. Many studies have noted the apparent low recruitment of young turtles. Roads adjacent to streams also provide access to otherwise remote areas, possibly leading to negative impacts from repeated human disruption.

Introduced species, such as the bullfrog, small mouth bass, and largemouth bass, negatively impact turtle populations. This will continue to occur unless their populations can be restricted and reduced.

FEIS Volume 3, Chapter 3, part 4.4, page 267 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

All three of these introduced species are protected by the California Department of Fish and Game as game species and have daily bag limits.

Environmental Consequences A. Assumptions and Limitations Cited literature continually notes low recruitment of juvenile western pond turtles or complete reproductive failure in specific populations. The major assumption in this assessment is that protecting nesting sites and juvenile western pond turtle rearing habitat is essential to the long- term viability of the species.

B. Effects of Alternatives Table 4.4.5.1a compares the alternatives in terms of how they would manage various activities that could alter habitats for the western pond turtle, including roads, fuelwood collection, grazing, prescribed fire, recreation, and vegetation treatments.

Table 4.4.5.1a. Comparison of management activities that could affect habitats for the western pond turtle by alternative. Alt. Veg. Mgnt. Roads Grazing Rx Fire Recreation Fuelwood 1 CASPO As necessary Continues low levels of Rx No restricts Continues fire 2 Amph. Res. Restricts new const. Cont. in green protects green and grey zones zone but not grey 3 Amph. Res. Closes/reroute rds Restricted in Manages nat, Reloc. OHV Rds. in rip. areas green zone, not and Rx fires Out of rip area grey 4 Equip. Manages nat. & Gathering restricted. In Rx fires allowed in green green zone & grey 5 Harvest. In Restricts new rd. LOP for salmonid Manages nat. & green prohib. construction and frogs in Rx fires In green zone green zone none harvest allowed in gray in grey zone. ADA, CAR* Amph. Res. 6 Harvest. in RA Rds. Decom. In Limited restr. in Manages nat. & Gathering prohibited. CAR. RA, fac. moved Rx fires prohibited in RA from RA 7 Harvest in RA Limited restr. in Manages nat. & Gathering prohibited. RA, fac. moved Rx fires allowed in RA from RA 8 Harvest in RA Rds. Decom. In Limited restr, Manages nat. & Gathering prohibited. CAR Harv. in RA Rx fires prohibited. in RA prohib.except in CAR, EW Mod 8 Harvest in RA Restricts new rd. 20% streambank Manages nat. & Gathering prohibited construction limit, 20% shrub Rx fires prohibited. in RA unless support ut, Elim. livestock AMS goals from occ. WIFL sites

*ADA=Aquatic Diversity Areas, CAR= Critical Aquatic Refuges, EW=Emphasis Watershed, RA=Riparian Area, AMS=Aquatic Management Strategy

Overall Assessment for Western pond turtles Protection measures for the foothill yellow-legged frog (Rana boylii) should benefit populations of western pond turtles since these species utilize similar riparian habitats. Alternatives 2, 5, and

FEIS Volume 3, Chapter 3, part 4.4, page 268 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Modified 8 would provide the greatest protection for this species (see Part 4.4.3.1 "Foothill Yellow-legged Frog").

FEIS Volume 3, Chapter 3, part 4.4, page 269 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.5.2. CALIFORNIA LEGLESS LIZARD (Anniella pulchra pulchra) Affected Environment Background Anniella pulchra nigra does not occur in the Sierra Nevada although the more widespread and common Anniella p. pulchra does inhabit the southern Sierra Nevada foothills.

For the purposes of this analysis, this species only occurs in the foothills of the southern Sierra Nevada Mountains. Anniella pulchra is a small fossorial lizard about the size of a pencil. This lizard requires sandy or loose loamy soils to construct burrows and must be able to reach damp soil; it is usually found in chaparral, pine-oak, or deciduous woodlands along perennial and intermittent streams. Anniella pulchra typically inhabits rocks, fallen timber, and leaf litter at the edge of sun- exposed tree canopies. This species probably occurs further upland beyond the green and grey zones of riparian areas wherever soil conditions are suitable. On warm evenings, they can be found on the surface hunting in the leaf litter for insect prey.

Anniella pulchra is a live-bearing species that lays one to four young in September through November. Young lizards grow rapidly reaching maturity in 2 to 3 years. These lizards are fairly long lived. Despite their small litter sizes, this species can reach high population densities where suitable habitat exists.

Historic and Current Conditions Establishing the viability of Anniella pulchra populations is difficult due to the secretive fossorial existence of this species. Anniella pulchra has been studied more in coastal habitats where it occurs on sandy beaches. Less is known about habitat preference and behavior for inland populations, but it is fairly certain that Anniella pulchra cannot exit near or within urbanized areas where surface soil conditions have been radically altered. Suitable habitat for this species in the southern Sierra Nevada has likely been fragmented due to human encroachment, road building, and logging.

Major risks to the California legless lizard include: (1) noxious weeds, (2) cattle grazing and feral pigs, and (3) off road vehicle or equipment use. Soil compaction and introduced plant species present the greatest risks to the future viability of Anniella pulchra populations.

Environmental Consequences A. Assumptions and Limitations Management practices that either lead to soil compaction or foster non-native plant communities will decrease the viability of legless lizard populations. Therefore, reintroducing natural fire regimes that promote native species enhance conditions for this species, but operations that result in soil compaction are detrimental. A limitation to this assessment is that the range of the legless lizard within the national forests is unknown.

B. Effects of Alternatives on the California Legless Lizard Table 4.4.5.2a compares the alternatives in terms of how they would manage various activities that could alter habitats for the California legless lizard, including roads, hardwood management, grazing, prescribed fire, and vegetation treatments.

FEIS Volume 3, Chapter 3, part 4.4, page 270 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Table 4.4.5.2a. Comparison of management activities that could affect habitats for the California legless lizard by alternative. Alt. Veg. Mgnt Emphasis Rx Fires Grazing Hardwood Roads 1 CASPO not emphasized Present levels Existing LRMP As Needed 2 Harv. and mech. Fuel in grey not emphasized Soil compaction, Mtn. All High decom zones limited restr in RA 3 mech. Fuel treatment, harv. Mgn. nat. & Rx Restr. In green zone Ret. BO >12” High decom fires but not grey Ret. MO >15” Fuelwood per. 4 Mech fuel treatment, harv. in Rx fires Limited green/grey Allows mgnt Mod decom RA zone restrictions. Fuelwood per. Ret.BO>12”dbh BLO>24”dbh MO>15”dbh 5 Harv. prohib. in green zone, not Rx fires Limited green/grey Ret. BO>12” High decom grey zone zone restrictions Ret. MO>15” Fuelwood per. 6 Harv. and mech. Fuel treatment Mgn. nat. & Rx Limited restr. in RA Ret. BO>12” Mod decom restr. in RA fires Ret MO>15” Fuelwood per. 7 Harv. prohib. in RA fuels Mgn. nat. & Rx Limited restr. in RA Allows mgnt. Mod decom treatment limited fires Fuelwood per Ret.BO>12”dbh BLO>24”dbh MO>15”dbh

8 Harv. prohib. in RA Mgn. nat. & Rx Limited restr. in RA Ret. BO>12” Mod decom fires, S&W slopes Ret. MO>15” Fuelwood per. Mod 8 Harv. prohib. in RA except Mgn. nat. & Rx Limited restr. in RA Ret. BO>8" Mod decom where support AMS goals fires Ret. MO>12" Fuelwood per.

*ADA=Aquatic Diversity Areas, CAR= Critical Aquatic Refuges, EW=Emphasis Watershed, AMS=Aquatic Management Strategy

Summary of Consequences to California legless lizards Restoration of natural fire regimes would promote native plant and animal populations, and therefore be beneficial to California legless lizard populations. Alternatives 3, 6, and 8 propose the highest amounts of prescribed fire to meet fire hazard reduction and ecological goals, and lesser amounts of mechanical treatments.

FEIS Volume 3, Chapter 3, part 4.4, page 271 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.5.3. SIERRA NIGHT LIZARD Affected Environment Background The Sierra night lizard occurs only on the western edge of the Greenhorn Mountains (Granite Station) in Kern County. This lizard is a strict rock dweller inhabiting granitic rock outcroppings at elevations between 1,350 and 1,500 feet. It is typically found beneath thick, horizontal cap rocks on top of large boulders. Little is known of its biology, but it is assumed to be similar to the other more common subspecies. As a group, night lizards are insectivorous with an ant-dominated diet. They produce one to two live young per season.

Historic and Current Conditions Little is known about the current status of this subspecies though it is probable that the population has been fragmented by human encroachment and urban development. Reptile collectors who pry off cap rocks have probably also impacted populations of this species. Exotic ants probably present the major risk to this species. Human encroachment brings the risk of introduction of Argentine ants into the habitat of the Sierra night lizard. Argentine ants are known to invade disturbed habitats and displace native ant species; this could drastically affect food availability for the Sierra night lizard. Studies have shown that horned lizards did not substitute Argentine ants for native ants in their diet, and the same probably holds true for the Sierra night lizard.

Environmental Consequences Restoration of natural fire regimes would promote native plant and animal populations, and therefore be beneficial to night lizard populations. Alternatives 3, 6, 7, and 8 propose the highest amounts of prescribed fire to meet fire hazard reduction and ecological goals.

The extent of the range of the Sierra night lizard in the national forests is unknown. Granite station lies outside of the national forest boundaries and the current distribution of this species is unknown.

4.4.5.4. PANAMINT ALLIGATOR LIZARD (Elgaria panamintina) Affected Environment Background Little is known about the life history of the Panamint alligator lizard (Elgaria panamintina), but observations suggest that it is similar to the other better-known Elgaria species. The Panamint alligator lizard is believed to be a relic species that was once more widespread. Like other alligator lizard species, it probably has lower temperature requirements than other diurnal lizards, like the western fence lizard, which are usually observed basking. The Panamint alligator lizard typically inhabits the narrow, thickly vegetated riparian strips that are surrounded by more xeric-adapted vegetation, like sagebrush and creosote bush. The narrow canyons occupied by Panamint alligator lizards are typically composed of upland talus and rock that the lizards probably use for foraging.

Historic and Current Conditions Panamint alligator lizards occur only in the Argus, Inyo, Nelson, Panamint, and White Mountains of Inyo and Mono Counties at elevations between 2,280 and 6,200 feet. The health of the 15 known isolated populations is unknown. All except two populations occur on private lands. In the areas where Panamint alligator lizards occur, there are few trees outside narrow riparian corridors.

FEIS Volume 3, Chapter 3, part 4.4, page 272 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Therefore, timber harvesting does not play a role in the future viability of Panamint alligator lizards. Likewise, fire exclusion is not likely negatively impacting this species considering the natural openness of its habitat. Fire exclusion could, however, alter native plant communities and the relationship of the Panamint alligator lizard with other species, especially prey species.

The riparian areas in the ranges occupied by the Panamint alligator lizard do not resemble those in other parts of the Sierra Nevada. These ranges usually receive much less precipitation so riparian zones only occur below a few permanent springs. Because these riparian zones have such a limited distribution, they are generally considered to be more fragile. The integrity of these riparian zones is critical for the long-term viability of Panamint alligator lizard populations

The primary risks to this Panamint Alligator Lizard include: (1) mining, (2) livestock grazing, and (3) off highway vehicle use. Several intermittent streams inhabited by Panamint alligator lizards have dirt roads paralleling them for long distances. These roads potentially lead to activities, such as off- highway vehicle use, that may impact the species or their habitat. In addition, these roads can directly impact streams by increasing erosion and sedimentation.

Environmental Consequences A. Assumptions and Limitations Protecting the fragile riparian corridors within its range from man made disturbance is critical for protecting the long-term viability of the Panamint alligator lizard. Due to the natural open habitat of this region, it is also assumed that fires do not play a critical role in the viability of this species.

B. Effects of Alternatives Table 4.4.5.4a compares the alternatives in terms of how they would manage various activities that could alter habitats for the Panamint alligator lizard, including roads, grazing, and vegetation treatments.

Table 4.4.5.4a. Comparison of management activities that could affect habitats for the Panamint alligator lizard by alternative. Alt. Veg. Mgnt. Emphasis Roads Grazing 1 CASPO Continued use Present condition 2 No treatment Decom. in CAR Allows within green zone

3 Natural fire regimes Decom. in RA Allows within green zone 4 Rx fires and mech. No reduction Allows within green zone treatment 5 Mgn. nat. & Rx fires Limits const., decom in RA Allows within green zone 6 Mgn. nat. & Rx fires Limited reduction Allows within RA 7 Rx fires and mech. Limited reduction Allows within RA treatment 8 Rx fires Decom. in CAR Allows within RA Mod 8 Mgn. nat. & Rx fires Limited reduction Allows within RA

*ADA=Aquatic Diversity Areas, CAR= Critical Aquatic Refuges, EW=Emphasis Watershed

Environmental Consequences Alternatives 3, 6, 7, and 8 propose the highest amounts of prescribed fire while employing lesser amounts of mechanical treatment. Grazing is allowed within riparian areas, but under strict standards and guideline for most of the action alternatives except Alternative 4.

FEIS Volume 3, Chapter 3, part 4.4, page 273 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

4.4.5.5. COAST HORNED LIZARD (Phrynosoma coronatum) Affected Environment Background The coast horned lizard (Phrynosoma coronatum) seems to be decreasing rapidly throughout most of its range. This species occurs in several habitat types from sun-exposed gravel and sand areas with scattered shrubs (especially dry lake beds) to clearings in riparian woodlands and chaparral. Historically, the species was most abundant in relict lake sand dunes and old alluvial fans bordering the San Joaquin Valley. Coast horned lizards are generally a ground dwelling species that prefers open areas with undisturbed sandy soils. At night they retreat into rodent burrows, under rocks, or simply burrow into the loose soil. They eat a variety of small insects with harvester ants making up the majority of their diet. Following a winter brumation, horned lizards breed in the early spring (April and May) and lay as many as 21 eggs shortly thereafter. Hatchlings emerge in August and September.

Historic and Current Conditions Coast horned lizards once ranged throughout the lower elevations (below 3,600 feet) of the Sierra Nevada from Lake Shasta south. They are believed to have disappeared from greater than 35 percent of their former range in northern and central California. Current populations are becoming increasingly fragmented as urban development expands in the Sierra Nevada foothills. Coast horned lizards are only abundant in isolated habitats along the south Coast Mountains and on the Central Valley floor in preserves.

Population Status and Trend Continued fragmentation of coast horned lizard populations will decrease habitat connectivity and put isolated populations at an increased risk of extinction. As habitat disturbance increases, so does the probability that Argentine ants will invade regions with healthy coast horned lizard populations. Coast horned lizards generally do not exist near urban development. Coast horned lizards seem to tolerate cattle; however, extensive grazing, especially when cattle eat the small shrubs that horned lizards often use for cover, can potentially influence populations.

The primary risks to this species include: (1) urban development, (2) exotic ants, (3) roads, (4) off highway vehicle or equipment use, and (5) noxious weeds.

Environmental Consequences A. Assumptions and Limitations This assessment assumes that reintroducing natural fire regimes will promote native plant communities and especially native ant populations, which are the main food of horned lizards. Additionally, due to their sedentary life style and preference for exposed habitat, coast horned lizards are especially susceptible to heavy equipment use, off highway vehicle use, and road use.

B. Effects of Alternatives on the Coast Horned Lizard Table 4.4.5.5a compares the alternatives in terms of how they would manage various activities that could alter habitats for the coast horned lizard, including roads, prescribed fire, and vegetation treatments.

FEIS Volume 3, Chapter 3, part 4.4, page 274 – Affected Environment and Environmental Consequences Sierra Nevada Forest Plan Amendment – Part 4.4

Table 4.4.5.5a. Comparison of management activities that could affect habitats for the coast horned lizard by alternative Alt. Veg. Mgnt. Emphasis Roads Rx Fire 1 CASPO Continued use Fire suppression 2 No treatment Decom. in CAR Fire suppression 3 mech. Fuel treatment, harv. Decom in RA Natural fire regimes 4 mech. Fuel treatment, harv No reduction Rx Fire 5 Harv. prohib. in green zone, Limited new constr., Natural fire and Rx Fire not grey zone Decom in RA 6 Harv. and mech. Fuel Limited reduction Rx Fire treatment restr. in RA 7 Harv. prohib. in RA fuels Limited reduction Rx Fire treatment limited 8 Harv. prohib. in RA Decom. in CAR Rx Fire Mod 8 Harv. and mech. Fuel Limited reduction Rx Fire treatment in RA if support AMS goals

*ADA=Aquatic Diversity Areas, CAR= Critical Aquatic Refuges, EW=Emphasis Watershed, AMS=Aquatic Management Strategy

Summary of Consequences to coast horned lizard All alternatives except Alternative 1 would employ a noxious weed strategy. Restoration of natural fire regimes would promote native plant and animal populations, and therefore be beneficial to lizard populations. Alternatives 3, 6, and 8 propose the highest amounts of prescribed fire to meet fire hazard reduction and ecological goals, and lesser amounts of mechanical treatments that may be detrimental.

FEIS Volume 3, Chapter 3, part 4.4, page 275 – Affected Environment and Environmental Consequences