u.s. PIS!I & WILDLIFI': SE.IlV.lCE United States Department of the Interior

FISH AND WILDLIFE SERVICE ~nr'f1'"- ~ · Bend Field Office 63095 Deschutes Market Rd. Bend, Oregon 97701 (541) 383-7146 FAX: (541) 383-7638

File Name: Lakeview B LM Integrated Pest Management BO 20 15 TAILS: 0 I EOFW00-20 16-F-00 18 Document type: Final December 18, 2015

Memorandum

To: District Manager, Lakeview District Office, Bw-eau of Land Management, Lakeview, Oregon

From: Field Supervisor Bend Field Office, Bend, Oregon

Subject: Fonnal Section 7 Consultation for the Lakeview BLM Integrated Pest Management Project for effects to the Warner sucker (Catastorn.us warnerensis) and Foskett speckled dace (Rhinichlhys osculus ssp.)

This docwnent represents the U.S. Fish and Wildlife Service's (Service) biological opinion and responds to your letter of April 16, 2015, requesting consultation for effects on Warner sucker (Catastomus warneren.sis) and its designated critical habitat, and Foskett speckled dace (Rhinichthys osculus ssp.) from the Bureau of Land Management (BLM) proposed Integrated Pest Management Project (Proposed Action), located in Lake County, Oregon. This biological opinion considers the effects of the Proposed Action on Warner sucker and its designated critical habitat, and Foskett speckled dace in accordance with section 7 ofthe Endangered Species Act of 1973, as amended (16 U.S. C. 153 1 et seq.) (ESA). Critical habitat is not designated for Foskett speckled dace. This biological opinion is based on information provided in the "Biological Assessment fo r Integrated Pest Management, Lakeview Resource Area, Lakeview District BLM" (Biological Assessment) (U.S. BLM 2015). Yow· request was received in the Service's Bend Field Office on April22, 2015. A complete administrative record ofthis consultation is on file at the Service's Bend Field Office.

CONSULTATION IDSTORY

The following is a summary ofimpmiant events associated with this consultation: 2

• The BLM provided a draft Biological Assessment to the Service on December 10, 2014

• The Service provided comments on the draft Biological Assessment to the BLM on February 19, 2015

• The BLM provided a draft Biological Assessment to the Service on March 13, 2015

• The Service provided comments on the draft Biological Assessment to the BLM on April 2, 2015

• The BLM provided a draft Biological Assessment to the Service on April 7, 2015

• The Service provided comments on the draft Biological Assessment to the BLM on April 8, 2015

• The BLM provided a final Biological Assessment to the Service dated April 16, 2015 and received on April 22, 2015.

BIOLOGICAL OPINION

DESCRIPTION OF THE PROPOSED ACTION

The Proposed Action would make three changes to the existing Lakeview Resource Area Noxious Weed Management Program (U.S. BLM 2010): 1) the addition of 10 herbicide active ingredients; 2) the addition of non‐native invasive plants to the types of plants that can be treated with herbicides; and 3) an expansion of directed livestock grazing as an invasive plant treatment. The Lakeview Resource Area Noxious Weed Management Program was included in the “Informal Consultation on the BLM Vegetation Treatments Using Herbicides on BLM Lands in Oregon” informal consultation completed August 27, 2010 (U.S. BLM 2010). Other elements, such as prevention, detection, and education would stay the same, and are not addressed in this biological opinion.

The BLM has identified noxious weeds and other invasive plants on approximately 15,700 acres which will be treated over the next 10 to 15 years. Invasive annual grasses are estimated to infest more than 50,000 acres of the action area. The spread of infestation from these inventoried sites is estimated to be approximately 12 percent per year. Up to 10,000 high priority acres will be treated per year, including greater sage-grouse habitat which has been identified for protection 3

and restoration, anticipated post fire emergency stabilization, and rehabilitation of invasive annual grass sites.

Integrated Pest Management Direct control treatments include manual (e.g., pulling, grubbing), mechanical (e.g., chainsaws, mowing, cutting with a string trimmer), directed livestock grazing (including early grazing of cheatgrass), biological controls (usually insects), prescribed fires, planting and seeding, and herbicides applied with hand-held sprayers, booms, and aircraft. The selection of a treatment method is guided by Department of the Interior policy which includes “Bureaus will accomplish pest management through cost‐effective means that pose the least risk to humans, natural and cultural resources, and environment” and requires bureaus to “[e]stablish site management objectives and then choose the lowest risk, most effective approach that is feasible for each pest management project” (U.S. BLM 2015).

Herbicide application methods could include wiping or wicking, spot spraying using backpacks or vehicles with hand wands, vehicles with booms, or aerial. A Pesticide Application Record is completed within 24 hours of the application documenting environmental conditions at the time of treatment as well as actual herbicide use. Objectives of invasive plant management include control of the highest priority noxious weeds with a goal of eradication, and control of widespread invasive plants to reduce spread, contain existing infestations, and reduce their size and density”.

The Proposed Action herbicide treatment description also includes sections describing: Monitoring; Treatment Categories (describe various needs for treatment); Treatment Sites and Priority Setting; Herbicides and Constraints; and Selection of the Treatment Method. Tables in the Proposed Action show which herbicides are available to the BLM to use, applicable buffer widths and formulated application rates (See Tables 1, 2, and 3 in the Biological Assessment (U.S. BLM 2015). Additional invasive plant control methods include directed livestock grazing used to positively direct plant community changes away from invasive species and toward native communities; biological controls such as insects, pathogens and diseases that target the unwanted vegetation; and seeding and planting to restore native vegetation.

The Proposed Action includes “Project Design Features, Standard Operating Procedures, Mitigation Measures, Conservation Measures, Prevention Measures, and Best Management Practices” (see Appendix A in the EA) (U.S. BLM 2015a). The Project Design Features includes measures to protect Water from fire use, directed livestock grazing, and herbicide use. The Standard Operating Procedures and Mitigation Measures have been identified to reduce adverse effects to environmental and human resource, including water resources, wetland and riparian areas, fish and other aquatic resources, and threatened and endangered species, from vegetation 4

treatment activities based on guidance in BLM manuals and handbooks, regulations, and standard BLM and industry practices. Conservation measures s are presented for use with Special Status species as needed, and includes guidance on the use of protective buffer zones, and protective measures related to various treatment methods. Prevention measures are designed to prevent the spread of invasive plants by minimizing the amount of existing non‐target vegetation that is disturbed or destroyed during project or vegetation treatment actions Best Management Practices are designed to maximize beneficial results and minimize negative impacts of management actions, primarily with regards to water quality.

Annual treatment plans and implementation monitoring reports will be developed prior to the beginning of control treatments each year. Implementation and effectiveness monitoring reports will be submitted to the Service annually, and relevant information, including actual versus proposed treatment acres and methods for the previous year will be described. See the Biological Assessment for more detailed descriptions of the Proposed Action (U.S. BLM 2015).

Action Area The action area is defined as all areas to be affected directly or indirectly by the federal action and not merely the immediate area involved in the action (50 CFR 402.02). In delineating the action area, we evaluated the farthest reaching physical, chemical, and biotic effects of the action on the environment. The action area for the Lakeview BLM Integrated Pest Management Program is from the upper extent of the watersheds within BLM administered lands downstream to the Warner Lakes and Coleman Lake which includes occupied Warner Sucker habitat, Warner sucker designated critical habitat, and occupied Foskett speckled dace habitat (see Figure 1 in the Biological Assessment, BLM 2015).

STATUS OF THE SPECIES

Warner Sucker

The Warner sucker is a slender-bodied sucker species that attains a maximum recorded fork length (the measurement on a fish from the tip of the nose to the middle of the tail where a V is formed) of 456 millimeters (17.9 inches). Pigmentation of sexually mature adults can be striking. The dorsal two-thirds of the head and body are blanketed with dark pigment, which borders creamy white lower sides and belly. During the spawning season, males have a brilliant red (or, rarely, bronze) lateral band along the midline of the body, female coloration is lighter. Breeding tubercles (small bumps usually found on the anal, caudal and pelvic fins during spawning season) are present along the anal and caudal fins of mature males and smaller tubercles occasionally occur on females (Coombs et al. 1979).

5

Sexes can be distinguished by fin shape, particularly the anal fin, among sexually mature adults (Coombs et al. 1979). The anal fin of males is broad and rounded distally, whereas the female anal fin is narrower in appearance and nearly pointed or angular. Bond and Coombs (1985) listed the following characteristics of the Warner sucker that differentiate it from other western species of Catostomus: dorsal fin base short, its length typically less than, or equal to, the depth of the head; dorsal fin and pelvic fins with 9 to 11 rays; lateral line (microscopic canal along the body, located roughly at midside) with 73-83 scales, and greater than 25 scales around the caudal peduncle (rear, usually slender part of the body between the base of the last anal fin ray and the caudal fin base); eye small, 0.035 millimeter (0.0013 inch) Standard Length (straight-line distance from the tip of the snout to the rear end of the vertebral column) or less in adults; dark pigmentation absent from lower third of the body; in adults, pigmented area extends around snout above upper lip; the membrane-covered opening between bones of the skull (fontanelle) is unusually large, its width more than one half the eye diameter in adults.

Taxonomy The Warner sucker (Catostomas warnerensis) was first described as a distinct species in 1908. Cope (1883 p. 134) collected suckers he referred to as Catostomus tahoensis from the “third Warner lake” (presumably Hart Lake) although he noted differences in the size of scales between the Warner Lake suckers and C. tahoensis from Pyramid Lake, Nevada. The Warner sucker was recognized as distinct and described as a new species by Snyder (1908 pp. 81-82) based on specimens collected from the Warner Valley in 1897 and 1904. He reported the species from Warner Creek (now Deep Creek), sloughs south of Warner Creek, and Honey Creek. Relationships of the new sucker to existing species were not precisely defined, but Snyder (1908) noted affinities to C. tahoensis of the Lahontan Basin, and C. catostomus of wide distribution in northern North America. The distinctiveness of the Warner sucker as a species was confirmed by additional collections (Andreasen 1975, Bond and Coombs 1985). Relationships of the Warner sucker are clearly within the subgenus Catostomus (Smith 1966), although identification of the closest relative has remained elusive. Morphologically, all these species are similar and probably the result of speciation due to geographic isolation (U.S. Fish and Wildlife Service 1998).

Current Legal Status and Critical Habitat Description The Service listed the Warner sucker as a threatened species and designated critical habitat on September 27, 1985 (U.S. Fish and Wildlife Service 1985a). Warner sucker critical habitat includes the following areas: Twelvemile Creek from the confluence of Twelvemile and Twentymile Creeks upstream for about six stream kilometers (four stream miles); Twentymile Creek starting about 14 kilometers (nine miles) upstream of the junction of Twelvemile and Twentymile Creeks and extending downstream for about 14 kilometers (nine miles); Spillway Canal north of Hart Lake and continuing about three kilometers (two miles) downstream; Snyder 6

Creek, from the confluence of Snyder and Honey Creeks upstream for about five kilometers (three miles); Honey Creek from the confluence of Hart Lake upstream for about 25 kilometers (16 miles). Warner sucker critical habitat includes 16 meters (50 feet) on either side of these waterways.

Constituent elements of designated critical habitat include streams 15 to 60 feet wide with gravel-bottom shoal and riffle areas with intervening pools. Streams should have clean, unpolluted flowing water and a stable riparian zone, and should support a variety of aquatic insects, crustaceans, and other small invertebrates for food.

Reproduction The distribution of Warner sucker is well known, but limited information is available on stream habitat requirements and spawning habits. Relatively little is known about feeding, fecundity, recruitment, age at sexual maturity, natural mortality, and interactions with introduced game fishes. In this account, "larvae" refers to the young from the time of hatching to transformation into juvenile (several weeks or months), and "juvenile" refers to young that are similar in appearance to adults. Young of year refers to members of age-group zero, including transformation into juvenile until January 1, of the following year.

Spawning usually occurs in April and May in streams, although variations in water temperature and stream flows may result in either earlier or later spawning. Temperature and flow cues appear to trigger spawning, with most spawning taking place at 14-20 degrees Celsius (57-68 degrees Fahrenheit) when stream flows are relatively high. Warner sucker spawn in sand or gravel beds in slow pools (White et al. 1990, 1991, Kennedy and North 1993). Allen et al. (1996) surmise that spawning aggregations in Hart Lake are triggered more by rising stream temperatures than by peak discharge events in Honey Creek.

Tait and Mulkey (1993b) found young of year were abundant in the upper Honey Creek drainage, suggesting this area may be important spawning habitat and a source of recruitment for lake recolonization. The warm, constant temperatures of the source springs at the headwaters of Snyder Creek (a tributary of Honey Creek) may provide an especially important rearing or spawning site for Warner sucker (Coombs and Bond 1980).

In years when access to stream spawning areas is limited by low flow or by physical in-stream blockages (such as beaver dams or irrigation diversion structures) Warner sucker may attempt to spawn on gravel beds along the lake shorelines. In 1990, Warner sucker were observed digging nests in 40+ centimeters (16+ inches) of water on the east shore of Hart Lake at a time when access to Honey Creek was blocked by extremely low flows (White et al. 1990). Warner sucker larvae are found in shallow backwater pools or on stream margins where there is no current, 7

often among or near macrophytes. Young of year Warner sucker are often found over deep, still water (from midwater to the surface) but also move into faster flowing areas near the heads of pools (Coombs et al. 1979).

Warner sucker larvae venture near higher velocities during the daytime to feed on planktonic organisms but avoid the mid-channel water current at night. This aversion to downstream drift may indicate that spawning habitats are also used as rearing grounds during the first few months of life (Kennedy and North 1993). None of the studies conducted thus far have succeeded in capturing Warner sucker younger than two years old in the Warner lakes, and it has been suggested that Warner sucker do not migrate down from the streams for two to three years (Coombs et al. 1979). The absence of young Warner sucker in the Warner lakes, even in years following spawning in the lakes, could be due to predation by introduced game fishes (White et al. 1991).

Juvenile suckers (one to two years old) are usually found at the bottom of deep pools, or in other habitats that are relatively cool and permanent, such as near springs. As with adults, juvenile Warner sucker prefer areas of the streams that are protected from the higher velocities of the main stream flow (Coombs et al. 1979). Larval and juvenile mortality over a two month period during the summer has been estimated at 98 percent and 89 percent, respectively, although accurate larval Warner sucker counts were hampered by dense macrophyte cover (Tait and Mulkey 1993b).

Population Structure A population estimate of Warner sucker in streams was conducted in 1993 on the Honey Creek and Twentymile Creek drainages (Tait and Mulkey 1993b). Approximately 20 percent of available stream habitat in the Honey Creek drainage was sampled. The population within the area sampled was estimated at 77 adults, 172 juveniles, and 4,616 young of year. Approximately 60 percent of the available stream habitat in the Twentymile Creek drainage was also sampled. The population estimates within this area sampled was 2,563 adults, 2,794 juveniles, and 4,435 young of year.

In 1996, the Hart Lake Warner sucker population was estimated at 493 spawning individuals (95 percent confidence intervals of 439-563) (Allen et al. 1996). The population estimate for Hart Lake is likely well below the abundances prior to the drought in the early 1990’s (Allen et al. 1996). In 1997, Bosse et al. (1997) documented the continued existence, but reduced numbers, of Warner sucker in the Warner Lakes. The number of Warner sucker, as measured by catch per unit effort, had declined 75 percent over the 1996 results. The reduction in sucker numbers was offset by a sharp increase in the percentage composition of introduced game fish, especially white crappie and brown bullhead. 8

Hartzell and Popper (2002) indicated a continued reduction of Warner sucker numbers and an increase of introduced fish in Warner Lakes. The greatest number of Warner sucker captured was in Hart Lake (96% of total Warner sucker catch) with only a few Warner sucker captured in the other Warner Lakes, including Crump Lake. Suckers represented a greater percentage of the catch in relation to introduced and other native fish compared to the efforts of 1997, although a smaller total number of sucker were captured than in 1997. This was the first year since 1991 that native fish made up a smaller percentage of the catch than introduced fish.

In 2006 Oregon Department of Fish and Wildlife (ODFW) conducted investigations to quantify the abundance of Warner suckers, to document recruitment, and to estimate abundance of Warner suckers relative to nonnative fish abundance. In addition, Warner suckers were tagged with Passive Integrated Transponders (PIT) to determine growth rates and movements, and radio tagged to document seasonal spawning migration. The 2006 surveys focused on lake dwelling Warner sucker and documented similar distribution of Warner sucker described in earlier investigations by Coombs and Bond (1980), and White et al. (1990 and 1991). Additionally, ODFW attempted to estimate the population of Warner sucker in Crump Lake and Hart Lake using a statistically-based sampling procedure. The number of Warner sucker recaptured in the lakes was not sufficient for a valid mark-recapture population estimate (Scheerer et al. 2006). The 2006 surveys were dominated by large sized, older suckers, and ODFW did not find evidence of substantial recruitment of suckers (Scheerer et al. 2006).

The ODFW found that: (1) minimal recruitment is believed to have occurred in recent years because the majority (91%) of the suckers captured was larger than 250 mm (8.84 inches) fork length (FL); and (2) the average length of suckers has increased substantially since 1994 (Scheerer et al. 2006). The ODFW concluded that the population of Warner sucker in Crump and Hart Lakes is severely depressed (Scheerer et al. 2006).

In 2007, ODFW conducted investigations of Warner basin tributaries to describe the current distribution of stream resident populations of Warner sucker and to quantify their abundance. Access through private lands to sample the streams was limited and only a fraction of the desired sample sites were actually surveyed. The 2007 population estimate for the streams of the entire Warner basin was 6,852 Warner sucker for fish larger than 60 mm (2.36 inches) fork length (FL) (95% CI: =/-92%) (Scheerer et al. 2007). Because of the patchy distribution of Warner sucker in the streams and the presence of a few sites with high sucker densities, the abundance estimates had low levels of precision (Scheerer et al. 2007).

Samples also showed an irregular distribution of age classes. This is not uncommon for stream fishes where the younger age classes are very abundant. Warner sucker ranged in size from 22 9

mm (0.87 inches) to 330 mm (12.99 inches) FL with the majority (85 %) less than 100 mm (3.94 inches) FL and only a few (2%) larger than 200 mm (7.87 inches) FL (Scheerer et al. 2007). Warner sucker mature at 3 to 4 years at a length of 130 mm (5.12 inches) to 160 mm (6.30 inches) FL (Coombs et al. 1979). Most of the suckers collected were estimated to be less than one year old (Scheerer et al. 2007).

In 2007, the ODFW also obtained a mark-recapture population estimate of adult Warner suckers at the Summer Lake Wildlife Area. This population resulted from natural production of adult suckers that were moved to the irrigation canal at the Wildlife Area when the Warner Lakes desiccated during the 1992 drought. The population estimate was 142 fish (95% CI: 91-218) in a section of the ditch from the well head to a gated culvert (Scheerer et al. 2007).

In 2008, ODFW resumed survey work in the lake habitats of Crump and Hart Lakes (Scheerer et al. 2008). An additional attempt at a mark-recapture estimate was conducted but was not completed because only one Warner sucker was recaptured. However, an estimate of Warner sucker in Hart Lake was completed based on the recapture of Warner sucker marked in 2006. ODFW estimated 565 Warner suckers larger than 155 mm (6.10 inches) FL (95% CI: 250-1,114; 56-97%), and assumed a 33 percent mortality rate over the two year period between marking and recapture (Scheerer et al. 2008). Additionally, ODFW did not find evidence of substantial recruitment of suckers.

In 2009, the mark-recapture population estimate for 21.3 km (13.23 miles) of streams within the Twentymile Creek drainage was 4,612 Warner suckers (95% CI: 3,820-5,567) (fish larger than 59 mm, 2.32 inches FL). The 2009 estimate of adult Warner suckers (>159 mm, 6.23 inches FL) was 1,169 fish (95% CI: 969-1,412) (Richardson et al. 2010 and Scheerer et al. 2011).

The ODFW also obtained a mark-recapture population estimate of adult Warner suckers at the Summer Lake Wildlife Area in 2009. The population estimate was 660 fish (95% CI: 421- 1,024), up substantially from the 2007 estimate of 142 fish. Suckers ranged in size from 40-270 mm (1.57-10.63 inches) FL. The 2009 size distribution was broader than the distribution in 2007 (90-220 mm, 3.54 -8.66 inches FL) (Richardson et al. 2010). The presence of smaller, presumably younger fish recruited into the population, may explain the larger population size observed in 2009. An additional population estimate was conducted by ODFW in 2012 to estimate numbers of Warner sucker in Hart Lake and survival in Honey Creek (Scheerer et al. 2012).

Ecology and Habitat Characteristics A common phenomenon among fishes is phenotypic plasticity (the ability of different individuals of the same species to have different appearances despite identical genotypes) 10

induced by changes in environmental factors (Wooton 1990, Barlow 1995). This is most easily seen by a difference in the size of the same species living in different but contiguous, and at times sympatric (occurring in the same area) habitats for a portion of their lives (Healey and Prince 1995, Wood 1995). The Warner Basin provides two generally continuous aquatic habitat types; a temporally more stable stream environment and a temporally less stable lake environment (e.g., lakes dried in 1992 and in the early 1930's).

Observations indicate that Warner sucker grow larger in the lakes than they do in streams (White et al. 1990). The smaller stream morph (development form) and the larger lake morph are examples of phenotypic plasticity within metapopulations of the Warner sucker. Expressions of these two morphs in Warner sucker might be as simple as the species being opportunistic. When lake habitat is available, the stream morph migrates downstream and grows to become a lake morph. These lake morphs can migrate upstream to spawn or become resident populations while the lake habitat is available. Presumably, when the lake habitat dries up the lake morph is lost but the stream morph persists. When the lakes refill, the stream morph can reinvade the lakes to again become lake morphs. The lake habitat represents a less stable but more productive environment than the stream populations of Warner sucker use on an opportunistic basis (Dehaan and Von Bargen 2011). The exact nature of the relationship between lake and stream morphs is still not fully understood.

The lake and stream morphs of the Warner sucker probably evolved with frequent migration and gene exchange between them. The larger, presumably longer-lived, lake morphs are capable of surviving through several continuous years of isolation (e.g., drought or other factors) from stream spawning habitats. Similarly, stream morphs probably serve as sources for recolonization of lake habitats in wet years following droughts, such as the refilling of the Warner Lakes in 1993 following their desiccation in 1992. The loss of either lake or stream morphs to drought, winter kill, excessive flows and a flushing of the fish in a stream, in conjunction with the lack of safe migration routes and the presence of predaceous exotic fishes, may strain the ability of the species to rebound (White et al. 1990, Berg 1991).

Lake morph Warner sucker occupy the lakes and, possibly, deep areas in the low elevation creeks, reservoirs, sloughs and canals. Recently, only stream morph suckers have exhibited frequent recruitment, indicated by a high percentage of young of year and juveniles in Twelvemile and Honey Creeks (Tait and Mulkey 1993a,b). Lake morph suckers, on the other hand, were skewed towards larger, older adults (8-12 years old) with no juveniles and few younger adult fish (White et al. 1991) before the lakes dried up in 1992. After the lakes refilled, the larger lake morph suckers reappeared. Captured lake suckers averaged 267 millimeters (10.5 inches) SL in 1996 (Allen et al. 1996), 244 millimeters (9.6 inches) SL in 1995 (Allen et al. 11

1995a) and 198 millimeters (7.8 inches) SL in 1994 (Allen et al. 1995b). Stream caught fish averaged 138 millimeters (5.4 inches) SL in 1993 (Tait and Mulkey 1993b).

Warner sucker recovered from an ice induced kill in Crump Lake were aged to 17 years old and had a maximum fork length of 456 millimeters (17.9 inches) (White et al. 1991). Lake resident suckers are generally much larger than stream residents, but growth rates for adults are not known for either form. Sexual maturity occurs at an age of three to four years (Coombs et al. 1979), although in 1993, captive fish at Summer Lake Wildlife Management Area, Oregon, successfully spawned at the age of two years (White et al. 1991).

Coombs et al. (1979) measured Warner sucker larval growth and found a growth rate of approximately 10 millimeters (0.39 inch) per month during the summer (i.e., when the larvae were 1-4 months old). Sucker larvae at Summer Lake Wildlife Management Area grew as large as 85 millimeters (3.3 inches) in three months during the summer of 1991, but this was in an artificial environment (earthen ponds) and may not reflect natural growth patterns.

The feeding habits of the Warner sucker depend to a large degree on habitat and life history stage, with adult suckers becoming more generalized than juveniles and young of year. Larvae have terminal mouths and short digestive tracts, enabling them to feed selectively in midwater or on the surface. Invertebrates, particularly planktonic (having weak powers of locomotion) crustaceans, make up most of their diet. As the suckers grow, they develop subterminal mouths, longer digestive tracts, and gradually become generalized benthic (living on the bottom) feeders on diatoms (small, usually microscopic, plants), filamentous (having a fine string-like appearance) algae, and detritus (decomposed plant and animal remains). Adult stream morph suckers forage nocturnally over a wide variety of substrates such as boulders, gravel, and silt. Adult lake morph suckers are thought to have a similar diet, though caught over predominantly muddy substrates (Tait and Mulkey 1993a, b).

White et al. (1991) found in qualitative surveys that, in general, adult suckers used stretches of stream where the gradient was sufficiently low to allow the formation of long (50 meters [166.6 feet] or longer pools. These pools tended to have undercut banks, large beds of aquatic macrophytes (usually greater than 70 percent of substrate covered), root wads or boulders, a surface to bottom temperature differential of at least two degrees Celsius (at low flows), a maximum depth greater than 1.5 meters (5 feet), and overhanging vegetation (often Salix spp.). About 45 percent of these pools were beaver ponds, although there were many beaver ponds in which Warner sucker were not observed. Warner sucker were also found in smaller or shallower pools or pools without some of the above mentioned features. However, they were only found in such places when a larger pool was within approximately 0.4 kilometer (0.25 mile) upstream or downstream of the site. 12

Submersed and floating vascular macrophytes are often a major component of Warner sucker- inhabited pools, providing cover and harboring planktonic crustaceans which make up most of the young of year Warner sucker diet. Rock substrates such as large gravel and boulders are important in providing surfaces for epilithic (living on the surface of stones, rocks, or pebbles) organisms upon which adult stream resident Warner sucker feed, and finer gravels or sand are used for spawning. Siltation of Warner sucker stream habitat increases the area of soft stream bed necessary for macrophyte growth, but embeds the rock substrates utilized by adult Warner sucker for foraging and spawning. Embeddedness, or the degree to which hard substrates are covered with silt, has been negatively correlated with total Warner sucker density (Tait and Mulkey 1993).

Habitat use by lake resident Warner sucker appears to be similar to that of stream resident Warner sucker in that adult Warner sucker are generally found in the deepest available water where food is plentiful. Not surprisingly, this describes much of the habitat available in Hart, Crump, and Pelican Lakes, as well as the ephemeral lakes north of Hart Lake. Most of these lakes are shallow and of uniform depth (the deepest is Hart Lake at 3.4 meters (11.3 feet) maximum depth), and all have mud bottoms that provide the Warner sucker with abundant food in the form of invertebrates, algae, and organic matter.

Tait and Mulkey (1993a,b) investigated factors limiting the distribution and abundance of Warner sucker in streams above the man-made stream barriers. The detrimental effects of these barriers are well-known, but there may be other less obvious factors that are also affecting the suckers in streams. These studies found that general summertime stream conditions, particularly water temperature and flows, were poor for most fish species.

Kennedy and North (1993) and Kennedy and Olsen (1994) studied sucker larvae drift behavior and distribution in streams in an attempt to understand why recruitment had been low or nonexistent for the lake morphs in previous years. They found that larvae did not show a tendency to drift downstream and theorized that rearing habitat in the creeks may be vital to later recruitment.

Historical Status and Distribution The Warner sucker (Catostomus warnerensis) is endemic to the Warner Valley in southeast Oregon, an endoreic (closed) sub-basin of the Great Basin area. The valley contains a dozen lakes and many potholes during wet years, but only the three southernmost lakes are semi- permanent. In addition, three permanent creeks drain into the valley (Honey Creek, Deep Creek, and Twentymile Creek).

13

The probable historic range of the Warner sucker includes the main Warner Lakes (Pelican, Crump, and Hart), and other accessible standing or flowing water in the Warner Valley, as well as the low to moderate gradient reaches of the tributaries which drain into the Warner Valley. Warner sucker historic distribution in tributaries includes Deep Creek (up to the falls about 5 kilometers (3.1 miles) west of Adel), the sloughs south of Deep Creek, the Honey Creek drainage, and the Twentymile Creek drainage (Snyder 1908). In Twelvemile Creek, a tributary to Twentymile Creek, the historic range of Warner sucker extended through Nevada and back into Oregon (Allen et al. 1994). Andreasen (1975) reported that long-time residents of the Warner Valley described large runs of suckers in the Honey Creek drainage, even far up into the canyon area.

Stream resident populations of Warner sucker are found in Honey Creek, Snyder Creek, Twentymile Creek and Twelvemile Creek. Intermittent streams in the drainages may support small numbers of migratory suckers in high water years. Surveys conducted in Deep Creek in 1984 and 1994 found no stream resident Warner sucker (Smith et al. 1984, Allen et al. 1994). However, a lake resident female apparently trying to migrate to stream spawning habitat was captured and released in Deep Creek in 1990 (White et al. 1990). Their distribution appears to be discontinuous and centered around low gradient areas that form deep pools with protective cover. In the lower Twentymile Slough area on the east side of the Warner Valley, White et al. (1990) collected adult and young suckers throughout the slough and Greaser Reservoir. This area dried up in 1991, but because of its marshy character, may be important sucker habitat during high flows. Larval, young-of-year, juvenile and adult Warner sucker captured immediately below Greaser Dam suggest either a slough resident population, or lake resident suckers migrating up the Twentymile Slough channel from Crump Lake to spawn (White et al. 1990, Allen et al. 1996).

Current Status and Distribution Most of the habitat occupied by Warner sucker is located on BLM administered lands. Additional Warner sucker habitat is located on private lands, State lands, and bordered by Hart Mountain National Antelope Refuge. Warner sucker inhabit lakes, sloughs, and potholes in the Warner Valley, including the canal north of Hart Lake, Hart Lake, Crump Lake, Anderson Lake, Swamp Lake, Mugwump Lake, Greaser Reservoir, Honey Creek, Snyder Creek, Twentymile Creek and Twelvemile Creek. A majority of Warner sucker habitat is located in waterways managed by the Lakeview BLM.

Between 1987 and 1991, five consecutive drought years prompted resource agencies to plan a Warner sucker salvage operation and establish a refuge population of Warner sucker at the Service’s Dexter National Fish Hatchery and Technology Center (Dexter), New Mexico. Salvage operations consisted of intensive trap netting in Hart Lake to collect Warner sucker, then 14

transportation of the captured fish to a temporary holding facility at ODFW's Summer Lake Wildlife Management Area (Summer Lake). The suckers were held at Summer Lake until September 1991, when 75 adults were recaptured and transported to Dexter.

While being held at Summer Lake, Warner sucker spawned successfully, leaving an estimated 250+ young in the Summer Lake holding ponds. The young suckers survived, growing approximately 85 millimeters (3.3 inches) during their first summer and reaching sexual maturity at the age of only two years. Warner sucker larvae were observed in the ponds during the summer of 1993, just over two years after the original wild suckers from Hart Lake were held there. Approximately 30 of the two year-old suckers were captured and released in Hart Lake in September 1993. In June 1994, over 100 10-17.5 centimeter (4-7 inch) Warner sucker were observed in the Summer Lake ponds. In 1996, nine adult fish were observed in these ponds along with approximately 20 larvae.

The suckers taken to Dexter were reduced from 75 to 46 individuals between September 1991 and March 1993, largely due to Lorna (anchor worm) infestation. In March 1993, the 46 survivors (12 males and 34 females) appeared ready to spawn, but the females did not produce any eggs. Between March 1993 and March 1994, Lorna further reduced the population to 20 individuals (5 males and 15 females) (U.S. Fish and Wildlife Service 1998). In May 1994, the five males and seven of the females spawned, producing a total of approximately 175,000 eggs. However, for reasons that are not clear, none of the eggs were successfully fertilized. The remaining 20 fish at Dexter died in 1995 (U.S. Fish and Wildlife Service 1998). In November of 1995, approximately 65 more suckers from Summer Lake were transferred to Dexter for spawning purposes but no additional attempts to spawn Warner sucker were conducted.

Between 1977 and 1991, eight studies examined the range and distribution of the Warner sucker throughout the Warner Valley (Kobetich 1977, Swenson 1978, Coombs et al. 1979, Coombs and Bond 1980, Hayes 1980, White et al. 1990, Williams et al. 1990, White et al. 1991). These surveys have shown that when adequate water is present, Warner sucker may inhabit all the lakes, sloughs, and potholes in the Warner Valley. The documented range of the sucker extended as far north into the ephemeral lakes as Flagstaff Lake during high water in the early 1980's, and again in the 1990's (Allen et al. 1996). The Warner sucker population of Hart Lake was intensively sampled to salvage individuals before the lake went dry in 1992.

While investigating the distribution of Cowhead Lake tui chub, Scoppettone and Rissler (2003) discovered a single juvenile Warner sucker in West Barrel Creek. West Barrel Creek is a tributary to Cow Head Slough that eventually enters Twelvemile Creek at the known upper extension of suckers in the Twelvemile drainage. This discovery of a Warner sucker in the Cowhead Lake drainage is a significant range extension for Warner sucker. 15

Threats Warner sucker were listed due to reductions in the range and numbers, reduced survival due to predation by introduced game fishes in lake habitats, and habitat fragmentation and migration corridor blockage due to stream diversion structures and agricultural practices. Since the time of listing, it has been recognized that the status and viability of the Warner sucker has declined due to habitat modification, from both stream channel degradation and overall reduced watershed function and from the introduction of exotic game fish into Crump and Hart Lake.

The first large scale human impact to migration of the Warner sucker within the Warner Basin was the construction of irrigation diversion structures in the late 1930s (U.S. Fish and Wildlife Service 1998). These structures hamper or block both upstream and downstream migrations of various life stages of Warner sucker. Few irrigation diversions have upstream fish passage. Adult suckers that have spawned and are moving downstream can be diverted from the main channel to become lethally trapped in unscreened irrigation canals. Larval, post larval, young of year, and juvenile suckers are probably also diverted into unscreened irrigation canals. In high water years, the amount of water diverted from Warner Valley streams may be only a small portion of the total flow, but in drought years, total stream flows often do not meet existing water rights, and so entire streams may be diverted. Over a series of drought years, reduced flows can cause drops in lake levels and sometimes, especially in conjunction with lake pumping for irrigation, cause complete dry-ups, as was the case with Hart Lake in 1992.

The 1985 listing rule noted that water diversions exist on all streams occupied by the Warner sucker and that these barriers and diversion structures have blocked the movement of suckers into spawning streams, particularly during periods of low flow. During periods of low flow all water is often diverted (FR 50 no. 188, pp 39119-39120).

There are water diversion structures on all major tributaries (Twentymile, Honey and Deep Creeks). Stern (1989) assessed the water diversion structures in Warner Valley; these included four cement dams with permanent retaining walls, approximately nine wooden weirs, five outlets running underneath the roadway along the north end of Hart Lake, and two spillways (one adjacent to the Greaser Reservoir dam, and one adjacent to the Hart Lake outflow dam/weir).

In 2007, a screen was installed on one of the Deep Creek irrigation diversions; a small screen was installed on upper Honey Creek above Warner sucker distribution. From 2005 through 2007, Natural Resource Conservation Service (NRCS) assisted local landowners to install several spray type screens on lateral withdrawal points along the “main drain” irrigation canal in the Twentymile Creek watershed. From 2009 to 2010, two passage and screening projects were designed and constructed by the Service’s Partners for Fish and Wildlife Program in cooperation 16

with ODFW and private landowners for two of the upstream irrigation diversions on Honey Creek.

An additional irrigation diversion on the lower end of Honey Creek has been retrofitted with a passage structure. Passage and screening has been constructed for Warner sucker passage on the Dyke diversion on Twentymile Creek. ODFW conducted a swim performance analysis of Warner sucker to aid in passage design (Scheerer and Clements 2013) and continues to monitor effectiveness of fish passage projects.

Although the native species composition in the Warner basin included some piscivorus fishes like the Warner Valley redband trout (Oncorhynchus mykiss sp.), the introduction of exotic game fish disrupted this balance and the native ichthyofauna has declined. In the early 1970s, ODFW stocked white crappie (Pomoxis annularis), black crappie (P. nigromaculatus), and largemouth bass (Micropterus salmoides) in Crump and Hart Lakes. Prior to this, brown bullhead (Ameiurus nebulosus) and non-native rainbow trout were introduced into the Warner Valley. The adults of all five piscivorus fish species feed on Warner sucker to varying degrees (U.S. Fish and Wildlife Service 1998).

The presence of the introduced game fishes may also threaten Warner sucker through competitive interactions. Brown bullhead are bottom oriented omnivores (Moyle 1976) that may compete directly with Warner sucker for the same food sources. Bullhead may also prey on sucker eggs in the lower creek or lake spawning areas, as well as on sucker larvae and juveniles. Young crappie probably eat many of the same zooplankton and other small invertebrates that young suckers depend on. Habitat use by young Warner sucker remains poorly understood, but there may be competition between suckers and other fishes for what scarce cover resources are available.

White et al. (1991) found signs of stream channel and watershed degradation was common in the Warner Valley, and included fences hanging in mid-air because stream banks had collapsed beneath them, high cut banks on streams, damaged riparian zones, bare banks, and large sagebrush flats where there were once wet meadows. The BLM responded to the listing of Warner sucker by modifying grazing allotments with potential to impact Warner sucker or its designated critical habitat in recognition that livestock trampling of stream side riparian vegetation could have a negative impact on Warner sucker. The BLM modified grazing allotments to either preclude cattle by building exclosure fences, or to set standards and monitor to ensure those standards are met in riparian areas of streams tributary to occupied habitat. After implementation of changes to the grazing management in the Warner basin, the BLM completed a ten year long monitoring program of stream conditions of the Warner basin from 1997-2007. The report summarizes stream conditions over a ten year period and concludes that exclusion of 17

livestock and management prescriptions has been effective at improving vegetative and channel conditions, although it stated that “there is still room for improvement” (Munhall 2007).

With few exceptions, designated Warner sucker critical habitat is excluded from grazing and other land use authorizations analyzed in the Lakeview Resource Area Resource Management Plan (U.S. BLM 2003). The one exception is on the Deppy Creek/ Honey Creek confluence where a water gap allows stock access. The other exception is in the 0207 allotment on Twentymile Creek. This area is not occupied by Warner sucker and is an intermittent, rock- armored channel.

Recovery of riparian conditions has been hindered in some reaches of stream due to unauthorized use of allotments and excluded areas, and exceedence of authorized use periods, numbers, and rangeland utilization standards (Munhall 2007). The report also identifies roads and road maintenance as impacting stream conditions by confining and cutting off floodplains and contributing excess sediment to streams. Many remaining degraded stream conditions are the result of watershed level affects from water withdrawal, channelization, logging, and road construction (Munhall 2007). The report provided management recommendations including the need for instream work along some reaches to stabilize banks, rock placement to create pools, and stabilization of the channel around the Twelvemile O’Keeffe diversion to prevent a head cut. The report also stated that the Nevada reach of Twelvemile Creek could benefit from bank stabilization or stream habitat improvement projects (Munhall 2007).

The Fremont-Winema National Forest (Forest) implements cattle grazing management via allotments in tributaries upstream of occupied Warner sucker habitat. The Forest modified grazing practices to implement standards that limit the utilization of riparian vegetation along streams tributary to Warner sucker inhabited streams or streams designated as critical habitat (U.S. Forest Service 2007). In 2007, after following the terms and conditions of a biological opinion and monitoring their activities for ten years, the Forest re-evaluated the effects to Warner sucker and determined that the grazing activities currently are “not likely to adversely affect Warner sucker, their habitat, or designated critical habitat” (U.S. Forest Service 2007). The conclusion of the assessment was based on current stream habitat assessments and the past use of strict utilization standards on allotments previously determined to negatively affect Warner sucker habitat downstream of the allotment (U.S. Forest Service 2007). The Service concurred with the Forests’ determination based on improved habitat quality, key elements in the grazing program to ensure use levels that maintain a proper functioning condition on an upward trend, and annual effectiveness monitoring.

18

Foskett Speckled Dace

The Foskett speckled dace (R. osculus ssp.) is represented by a single population that inhabits Foskett Spring on the west side of Coleman Lake in Lake County, Oregon (U.S. Fish and Wildlife Service 1998). A refuge sub-population has been established at Dace Spring 0.75 miles south of Foskett Spring. Size ranges up to 4" (10 cm). Foskett speckled dace is described as elongate, rounded with a flat belly. Color of its back is dusky to dark olive; sides are grayish green, with dark lateral stripe, often obscured by dark speckles or blotches; and the fins are plain. Breeding males are reddish on lips and fin bases. The snout is moderately pointed; the eyes and mouth are small, and ventral barbels are present. Foskett speckled dace have 8 dorsal fin rays; 7 anal fin rays and the caudal fin is moderately forked. The lateral line is complete with 60–90 scales (U.S. Fish and Wildlife Service 1998).

Taxonomy The Foskett speckled dace is considered to be an undescribed subspecies of R. osculus (Girard) 1857. R. osculus (speckled dace) have a large geographic range throughout major drainages in the western United States, and populations show high degrees of endemism and exhibit large differences in morphological traits (Pfrender et al. 2003). Pfrender et al. (2003) stated that our understanding of the relationships among populations in this complex is limited, and there is no clear consensus regarding the number of distinct evolutionary lineages within R. osculus. Foskett speckled dace can be distinguished from other speckled dace by external characteristics, such as: a much reduced lateral line with about 15 scales with pores; about 5 lateral line scales; a large eye; the dorsal fin is positioned well behind the pelvic fin but before the beginning of the anal fin; and barbells are present on most individuals (U.S. Fish and Wildlife Service 1998). However, Bond did not provide a formal description or a scientific name for this subspecies, nor was his work peer reviewed. No changes to the taxonomic classification of Foskett speckled dace has occurred since the time it was listed in 1985.

A genetic analysis was conducted by Ardren et al. (2009), and compared Foskett speckled dace to other populations of dace in the Warner basin as well as the adjacent Goose Lake Basin. The results indicate that Foskett dace and other populations of dace in the in the Warner Basin are approximately equally diverged from one another evolutionarily, suggesting similar times of divergence since the late Pleistocene. Ardren et al. (2009) indicates further studies are needed to determine if rapid evolution of novel traits have occurred in dace inhabiting the unique ecological setting of Foskett Spring during the past 10,000 years.

In 2013, additional analysis of the genetic and morphometric variation of several dace in Oregon’s Great Basin regions, including Foskett speckled dace, was conducted by Hoekzema (2013). Based on the apparent lack of genetic distance between Foskett speckled dace and the 19

rest of Warner Basin speckled dace, Hoekzema concluded that the Foskett speckled dace have not been isolated long enough to diverge significantly. But Foskett speckled dace exhibit diagnostically different morphology from other dace in Oregon’s Great Basin region (Hoekzema 2013). Hoekzema and Sidlauskas (2014) found no genetic evidence that speckled dace from Foskett Spring warrant subspecies or species status.

It is unlikely the differences between spring and stream resident dace in the Warner Basin resulted from randomly accumulated mutations over time, but rather resulted from either ecophenotypic (non-heritable modifications of an organism’s physical characteristics, produced in response to factors in the environment or habitat) induction or rapid local adaptation in certain genes to the unusual habitat of Foskett Spring (Hoekzema 2013). The effects of rapid genetic adaptation and phenotypic plasticity are difficult to distinguish without the use of genomic, next- generation sequencing to identify functional genes under selection (Hoekzema 2013).

Although Sidlauskas and Hoekzema (2014) provide new information on the uniqueness of Foskett speckled dace, they state that the Foskett speckled dace would not currently be classified as a separate species or subspecies according to criteria by Zink (2004). However, they suggest that the Foskett speckled dace may fit as a distinct population segment (DPS) (Sidlauskas and Hoekzema 2014) based on the Service’s DPS policy (U.S. Fish and Wildlife Service 1996). Sidlauskas and Hoekzema also present their application of the Service’s DPS policy as summarized below. Service has not made a determination designating Foskett speckled dace as either a DPS pursuant to our Policy Regarding the Recognition of Distinct Vertebrate Population (U.S. Fish and Wildlife Service 1996), nor has there been a determination designating Foskett speckled dace s a subspecies (U.S. Fish and Wildlife Service 2015).

Analysis conducted by Hoekzema and Sidlauskas (2014) shows that Foskett speckled dace are related to other populations of speckled dace in Oregon. However, both a population genetic analysis (Hoekzema and Sidlauskas 2014) and a morphometric analysis (Hoekzema, 2013) indicate Foskett speckled dace are a discrete entity relative to dace elsewhere in Oregon’s Great Basin. Their analysis shows that Foskett speckled dace are a separate genetic entity, implying reproductive isolation, from surrounding Warner Valley populations of speckled dace for the past 10,000 years (Hoekzema and Sidlauskas 2014). A morphometric study, based on detailed measurements of body proportions and counts of scales and fin elements, conducted by Hoekzema (2013) indicates that Foskett speckled dace differ in mean morphology when compared to other speckled dace in surrounding basins. Foskett speckled dace have significantly fewer scales in the lateral line scale-series, shorter caudal peduncles (the narrow part of the body nearest to where the tail fin is attached), dorsal fins located further back on the body toward the tail, and larger heads compared to other speckled dace in surrounding basins in southeastern Oregon (Hoekzema, 2013). 20

Sidlauskas and Hoekzema (2014) also state that Foskett Spring itself represents a unique habitat within the Warner Valley. The spring is isolated in the Coleman Subbasin, water temperature is constant at approximately 65 degrees F year round, and has higher mineral concentrations than other springs in Warner Valley (Mauger 2000). Speckled dace occur in several other springs in southeastern Oregon, but none of the springs have the same physical characteristics as Foskett Spring, and the dace residing in other springs do not match Foskett speckled dace morphologically (Sidlauskas and Hoekzema 2014). The differing morphologies could represent either a phenotypic response or genetic adaptation based on the various spring conditions. Current data do not permit a formal test of either hypothesis, but either scenario suggests a substantial role of the unusual habitat of Foskett Spring in shaping the morphology of the Foskett speckled dace living there (Sidlauskas and Hoekzema 2014).

Sidlauskas and Hoekzema also present differences of Foskett speckled dace from other speckled dace populations based on microsatellite analysis. Their analysis indicates that Foskett speckled dace have been reproductively isolated and on an evolutionary trajectory since the end of the Pluvial period, approximately 10,000 years ago representing a significant contribution to the genetic diversity of the speckled dace taxonomic group (Hoekzema and Sidlauskas 2014).

Current Legal Status and Critical Habitat Description The Foskett speckled dace is endemic to one spring on the western margin of Coleman Lake, Lake County, Oregon and was listed as threatened March 28, 1985 (U.S. Fish and Wildlife Service 1985). Special rules concerning "take" for this subspecies can be found in 50 CFR 17.44(j). The “Recovery Plan for the Threatened and Rare Native Fishes of the Warner Basin and Alkali Subbasin”, which includes Foskett speckled dace, was finalized April 27, 1998 (U.S. Fish and Wildlife Service 1998).

No critical habitat has been designated or proposed for the Foskett speckled dace and therefore, no Primary Constituent Elements have been described for Foskett speckled dace.

Reproduction Breeding behavior has not been observed. Presumably Foskett speckled dace have habits similar to other dace and require rock or gravel substrate for egg deposition (Sigler and Sigler 1987). Foskett speckled dace are believed to spawn between late May and early July and apparently reproduce in their second year of age.

An abundance estimate of Foskett speckled dace conducted at Foskett Spring by ODFW in 2009, included dace ranging from 18-76 mm TL. A length-frequency analysis suggests the presence of multiple age-classes, with two apparent peaks, one approximately 25 mm the other 21

approximately 45 mm. The presence of fish < 25mm in all three sampling years (2005, 2007, and 2009) suggests that successful reproduction occurs annually. Presence of young of the year fish (<25 mm) provides evidence of recent recruitment (Scheerer and Jacobs 2009).

Population Structure Bond (USDI 1985) estimated the population of Foskett speckled dace in Foskett Spring to be approximately 1,500 individuals. In 1997, ODFW obtained mark-recapture population estimates at both Foskett and Dace springs (Dambacher et al. 1997). The Foskett Spring estimate was 27,787 fish (95% CI: 14,057-41,516). The majority of the fish (97%) were found in the downstream open water pool located outside the cattle exclosure. This shallow pool was dry in 1989 (Dambacher et al. 1997). In 2005, 2007, and 2009, ODFW obtained population estimates of 3,147 (95% CI: 2,535-3,905); 2,879 (95% CI: 2,319-3,573); and 2,830 (95% CI: 2,202-3,633) dace, respectively (Scheerer and Jacobs 2009).

Although the three estimates were statistically valid, there is a great discrepancy between the sizes of the population present in 1997 compared to the number that was estimated in 2005; the distribution of the fish was also substantially different. Scheerer and Jacobs (2007) postulated that the lower population abundance in 2005 and 2007 compared to 1997 was probably due to the reduction of open water habitat in the cattail marsh. General observations made during the population surveys of 2005 and 2007 included the presence of multiple age-classes, and evidence of recent recruitment as indicated by presence of young-of-the-year.

Additional population estimates were obtained by ODFW from 2005 through 2014 (Table 1.). Population estimates obtained between 2005 and 2012 were done using the Lincoln‐Petersen model. However, with recent habitat restoration efforts, the dace population increased fourteen fold since 2011 (Table 1).

In 2011, ODFW added an additional model to adjust the population estimates to be more accurate. The Huggins model was used along with the Lincoln-Petersen model for sample years 2011 through 2012 to compare results of the two methods. The comparison revealed that the Lincoln-Petersen method underestimated the number of individuals. In 2013 and 2014, just the Huggins model for estimating the population was used.

22

Table 1. Foskett speckled dace population estimates from ODFW reports 1997 through 2014 (Dambacher et al. 1997, Scheerer and Jacobs 2005, 2007, and 2009; Scheerer 2011; Scheerer et al. 2012, 2013; and 2014). Model Year Population 95% Confidence Interval Lincoln‐Petersen 1997 27,787 (14,057 ‐ 41,516) Lincoln‐Petersen 2005 3,147 (2,535 ‐ 3,905) Lincoln‐Petersen 2007 2,984 (2,403 ‐ 3,702) Lincoln‐Petersen 2009 2,830 (2,202‐3,633) Lincoln‐Petersen 2011 751 (616 ‐ 915) Huggins 2011 1,728 (1,269‐2,475) Lincoln‐Petersen 2012 988 (898‐1,098) Huggins 2012 1,848 (1,489‐2,503) Huggins 2013 13,142 (10,665‐16,616) Huggins 2014 24,888 (19,250-31,510)

Experimental habitat enhancement work by BLM in 2012, consisted of a controlled burn to reduce vegetation in the tule and cattail marsh, and excavation of eight 2.7 cubic yard pools, resulted in a threefold increase in the open water habitat in 2013. Additional hand excavation of the lower spring brook and marsh sections of the habitat was conducted in 2014. The population at Foskett Spring in 2013 was observed to increase by over 700 percent from the previous population estimate in 2012 and increased another 190% from 2013 to 2014. Large increases in the abundance of the dace population was observed in 2014 in the spring brook and tule marsh areas where BLM excavated open water pools in 2013 and early 2014. This population increase appears to be the result of habitat enhancement by excavating pools to increase open-water habitat. Based on observations of an increase in population from 1,848 to 24,888 in just two years, it is apparent that this type of habitat enhancement is a great benefit to the Foskett speckled dace.

In comparing the estimates from 1997 to 2005, 2007, 2009, 2011, 2012, 2013, and 2014 by habitat type, the ODFW estimated that there was an increase in the number of Foskett speckled dace inhabiting pool habitat and a decrease in the number inhabiting the stream or marsh type habitat. The results of the estimates conducted in 2013 and 2014 show an increase in abundance of Foskett speckled dace in areas where open-water habitat was expanded, but also showed an increase where open-water habitat was already in existence at the “spring pool” (Scheerer et al. 2013). This supports the hypothesis that an increase in open-water habitat will result in higher numbers of dace occupying that habitat and that Foskett Spring habitat relies on active vegetation and sediment removal to assure conservation of the habitat, by maintaining and increasing the amount of open-water in the long-term.

23

The 2015 estimate at Foskett Spring was 16,340 adult fish (95% CI: 10,980-21,577). This estimate was lower than, but not significantly different from the 2014 estimate of 24,888 adult fish (95% CI: 19,250-31,510). Substantial reductions in dace numbers occurred in the marsh habitats where aquatic vegetation was encroaching on open water habitat (Scheerer Pers. Comm. 2015).

Dace Spring Dace Spring is located within one mile of Foskett Spring. No Foskett speckled dace were found in a 1970 survey of Dace Spring. An attempt was made to establish a refuge population at Dace Spring starting in 1979 and 1980. In each year, 50 fish were removed from Foskett Spring and transplanted into Dace Spring. An estimated 300 fish were present in 1986 (Williams et al. 1990). In 1997, only 19 fish were estimated to remain in Dace Spring (Dambacher et al. 1997). Fish observed in Dace Spring were larger than those in Foskett Spring, probably due to the older age classes present and an indication that reproduction at Dace Spring was no longer occurring. The population persisted for at least 17 years; the last observation of fish from the 1979-80 translocation in Dace Spring was made in 1997.

The loss of Foskett speckled dace from Dace Spring was likely due to the limited area of habitat; the open-water filled in with sediment and vegetation, and habitat conditions for reproduction were not adequate. It is suspected that the Dace Spring habitat was not adequate for fish to persist in the long-term. The outflow of Dace Spring terminated in a metal cattle trough. The fish were probably transported with the water flow to the trough, and were unable to return to the spring. The ODFW recommended in their 2005 Progress Report that restoration of Dace Springs and introduction of Foskett speckled dace could reduce the risk of extinction and aid in recovery.

The BLM and the Service worked together to construct two ponds connected to the outlet channel of Dace Spring to provide additional habitat. Construction was completed in 2009, and 49 Foskett speckled dace were transferred to the constructed ponds in 2010. The population at Dace Spring ponds was estimated to be 34 in 2011 (survival from 2010 was 69 percent). The ODFW transferred an additional 75 Foskett speckled dace into the two ponds in 2011. Only 13 Foskett speckled dace were captured in the ponds in 2012 (survival of individuals from 2010 to 2011 was 11 percent). Large algal blooms and poor dissolved oxygen levels were observed at the ponds and the low survival was attributed to these factors (Scheerer et al. 2012).

In 2013, the BLM reconfigured the inlet and outlet to the two ponds allowing greater water flow which has improved the water quality. Also in 2013, ODFW estimated there were 34 individuals (95% confidence level: 17 to 62) in the ponds and inlet at Dace Spring (Scheerer et al. 2013). After a few adjustments to the pond inlets to assure adequate water quality (reduction of algal 24

blooms and improvement of dissolved oxygen levels), ODFW transplanted 200 Foskett speckled dace from Foskett Spring into the two ponds (100 fish in each pond) in October 2013.

The 2014 population estimate at Dace Spring was 552 individuals (95% confidence level: 527- 694). The presence of smaller fish, and more fish than the number transplanted, indicates recent successful recruitment (Scheerer et al. 2014). Substantial recruitment was observed in 2014, and many of these recruits grew rapidly with most (66%) were in the medium size category. The two constructed pools at Dace Spring appear to be successful at providing additional habitat to serve as a refuge population for Foskett speckled dace.

The 2015 estimate at Dace Spring (introduced population) was 876 adult fish (95% CI: 692- 1,637). This estimate was higher than, but not significantly different from the 2014 estimate of 552 adult fish (95% CI: 527-694). Larval and juvenile dace were abundant in 2015 at Dace Spring and not included in this estimate (Scheerer Pers. Comm. 2015).

Ecology and Habitat Characteristics The Foskett speckled dace became isolated in Foskett Spring at the end of the Pluvial Period about 9,000-10,000 yrs. ago. Its main natural habitat is the small, shallow pool at the spring source. Foskett Spring is a cool-water spring with temperatures recorded at a constant temperature of approximately 18 degrees Celsius (Scheerer and Jacobs 2009). The source pool has a loose sandy bottom and is choked with macrophytes. The spring brook (outflow channel) eventually turns into a marsh and dries up before reaching the bed of Coleman Lake. Foskett speckled dace occur naturally in the main spring pool, outflow channel, and tiny outflow rivulets that are at times only a few inches wide and deep. The fish find cover under overhanging bank edges, grass, exposed grass roots, and filamentous algae. Dace are the only fish species present in Foskett Spring (Scheerer et al. 2012).

The wet areas at the spring, along the course of the rivulets, and at the edge of the playa supports growth of grasses and some aquatic vegetation, including bull rush and cattails. The main population is in the spring-hole, which is about 6 feet in diameter and mostly 6 to 12 inches deep. Water in the spring is clear, the water flow slow but significant. The bottom is primarily mud.

Foskett speckled dace appear to be non-territorial and is known to form small aggregations. The individuals are found in restricted habitats including the small spring pool, narrow rivulets, and small depressions, including cow tracks, so that home range and total range might coincide. Extensive migration is not known, but larval and early juvenile dace have been observed only in the marsh at the edge of the lake bed, so there is either a migration of adults downstream to spawn, or a migration of the hatched larvae from the spring hole or rivulets to the marsh (a distance of about 6-12 feet). 25

Baseline water quality and vegetation monitoring at Foskett and Dace springs were initiated by BLM in 1987. The following data collected on 28 September 1988 from Foskett Spring and Dace Spring, respectively, exemplify the two habitat similarities: air temperature 19 and 17 C, water temperature 17 and 16 C, dissolved oxygen 5.3 and 5.9 mg/I, conductivity 350 and 250 mohs/cm, pH 8.1 and 8.2, alkalinity 114 and 99 mg/l CaC03, hardness 40.0 and 24.7 mg/l, and turbidity 1.4 and 1.8 NTU. The spring temperatures measured in Foskett Spring from 14 August 2007 through 16 August 2009, were a constant 18.2o C, similar to temperatures recorded previously (Scheerer and Jacobs 2009).

In 2005, 2007, and 2009, the ODFW considered the Foskett speckled dace habitat to be in good condition, but limited in extent (Scheerer and Jacobs 2005, 2007, and 2009). They noted that encroachment by aquatic macrophytes may be limiting population abundance and that the decline in abundance of Foskett speckled dace since 1997 is probably due to the reduction in open water habitat. Dambacher et al. (1997) noted that past habitat enhancement efforts to increase open water habitat have been unsuccessful due to sediment infilling and growth of macrophytes. Little information is available on water quality or flows. Deeper water with moderate vegetative cover would presumably be better habitat, judging from conditions under which other populations of speckled dace live.

Historical Status and Distribution Foskett speckled dace were probably distributed throughout prehistoric Coleman Lake of the Warner Basin during times that it held substantial amounts of water. The timing of the isolation between the Warner Lakes Subbasin and the Coleman Subbasin is uncertain although it might be as recent as 10,000 years ago (Bills 1977). Foskett speckled dace were probably distributed throughout prehistoric (approximately 12,000 years ago) Coleman Lake during times that it held substantial amounts of water. As the lake dried, the salt content of the lake water increased. Suitable habitat would have been reduced from a large lake to any spring systems that provided enough suitable habitat for survival. Springs that remain within the vicinity of Coleman Lake include Foskett Spring and Dace Spring. Both springs are extremely small and shallow with limited habitat for fish. Foskett Spring has the only known native population of Foskett speckled dace. The Recovery Plan describes Foskett Spring as originating in a pool about 5 meters across. The outflow channel is approximately 5 centimeters deep and it gradually transitions to marshland, drying up before reaching the dry bed of Coleman Lake (U.S. Fish and Wildlife Service 1998).

Dace Spring is approximately 0.8 kilometer (0.5 mile) south of Foskett Spring. Dace Spring is smaller than Foskett Spring and the spring outflow terminates in a cattle watering trough where fewer than 20 Foskett speckled dace were seen in 1996 (Dambacher 1997). A population of 26

Foskett speckled dace was initially established in Dace Spring in November 1979, from an introduction of 100 fish from Foskett Spring (Williams et al. 1990); however this population failed due to habitat loss (vegetative succession) and lack of successful recruitment. Foskett speckled dace appeared to have persisted in the trough for several additional years, but none were detected during surveys in 2005

Surveys of Foskett Spring conducted in 2005 and 2007 document Foskett speckled dace in the spring pool, outflow stream, and the tule and cattail marshes of Foskett Spring. The ODFW estimated approximately 722 m2 of wetted habitat in the spring pool, spring brook, tule marsh, cattail marsh, and sedge marsh (Scheerer and Jacobs 2005). In 2005 and 2007, approximately half of the population of Foskett speckled dace was located in the 33 m2 spring pool. The open water habitat at Foskett Spring has undergone changes resulting in significantly reduced open water area due to vegetation encroachment between 1997 and 2012.

Current Status and Distribution In 1987 the BLM acquired Foskett Spring and the surrounding 65 ha, of which approximately 28 ha were fenced to exclude cattle. The dace population at Foskett Spring has since expanded to the spring pool, its outflow, and downstream marsh (Williams et al.1990). Current management of the Foskett and Dace spring systems excludes livestock use from a majority of its habitat. An area outside the Foskett Spring exclosure is at times occupied when water flows and lack of encroaching vegetation provide enough habitat for the dace. Surveys of Foskett Spring conducted from 2005 to 2014 document Foskett speckled dace in the spring pool, outflow stream, and the tule and cattail marshes of Foskett Spring. The ODFW estimated approximately 722 m2 of wetted habitat in the spring pool, spring brook, tule marsh, cattail marsh, and sedge marsh (Scheerer and Jacobs 2005, Scheerer et al. 2014).

In 2009, the BLM and the Service completed a habitat enhancement project by excavating two spring-fed pools at Dace Spring. Foskett speckled dace were translocated into the ponds. In 2010-2011, Oregon Department of Fish and Wildlife (ODFW) introduced 124 dace from Foskett Springs into these ponds; however survival of these fish was low, due to frequent prolonged algal blooms and resultant anoxic conditions (Scheerer et al. 2012; 2013). In September 2013, BLM excavated flow-through channels to improve water circulation in the Dace Spring ponds and saw immediate improvement in water clarity (algal bloom subsided) and water quality (Scheerer et al. 2013). In October 2013, ODFW transferred an additional 200 speckled dace from Foskett Spring into the Dace Spring ponds (100 fish ea.). Additionally, since the translocation of dace into Dace Spring, there have been some observations of dace in the cattle trough outside the exclosure (Scheerer et al. 2014).

27

Threats Threats identified in the final rule listing Foskett speckled dace as threatened includ actual or potential modification of habitat; restricted distribution; and pumping of ground water with concomitant lowering of the water table. Mechanical modification of the aquatic ecosystem had occurred in the past evidenced by the remnant rock structure. The spring also had been used for livestock watering resulting in negative affects to Foskett speckled dace. The State of Oregon has listed the Foskett speckled dace as threatened under the Oregon Endangered Species Act, which prohibits taking the fish without an Oregon scientific collecting permit, but does not protect the habitat (U.S. Fish and Wildlife Service 1985b).

The field surveys conducted in 2005 through 2014 at Foskett Spring did not reveal any sign of artificial channeling of water or mechanized impacts beyond the remnants of the historical activities (two small rock cribs and side-casting of material around the spring). The habitat at Foskett Spring is limited in extent, and encroachment by aquatic vegetation has reduced the area of open-water in the past.

Mechanical modification and livestock watering uses are no longer considered a threat since BLM acquired the property containing both Foskett and Dace springs and constructed an exclosure fence to exclude cattle from most of the habitat.

A query of the Oregon Water Resources Department database for water rights did not reveal any wells within five miles of Foskett Spring. The closest well listed on the Oregon Water Resources Department database is 5.9 miles away and located along Twentymile Creek. No other wells were located closer to Foskett Spring. There are no Oregon Water Resources Department records of established water rights in the vicinity of the springs. Any development of water resources and filing of water rights on BLM affected lands would require a permit from the BLM (U.S. BLM 2003).

The outflow of the spring at one time apparently formed a small rivulet, which prior to listing was used heavily for cattle grazing and is now occupied by Foskett speckled dace. At the time of listing, trampling of the habitat by cattle was perceived as the main reason for diminution of the habitat. The wetland on the edge of normally dry Coleman Lake may have formerly afforded some habitat, but is now either occupied by cattails and other vegetation. Therefore, a new threat would be encroachment of vegetation (cattails and rushes) into the open water habitat occupied by Foskett speckled dace.

No exotic fish introduction or of vandalism has occurred since the time of listing. The Foskett speckled dace is vulnerable to invasive or nonnative species (aquatic plants, invertebrates, or fish species). However, this vulnerability is reduced in part due to the remoteness of the site and the 28

lack of recreational or other reasons for the public to visit the area. It is also reduced by the potential establishment of a refuge population in Dace Spring. The risk of such invasions occurring through human caused mechanism may be low, but the potential magnitude of the impact is great due to the highly restricted distribution of this species.

ENVIRONMENTAL BASELINE

Regulations implementing the Endangered Species Act (50 CFR 402.02) define the environmental baseline as the past and present impacts of all Federal, State, or private actions and other human activities in the action area. Also included in the environmental baseline are the anticipated impacts of all proposed federal projects in the action area that have undergone section 7 consultation, and the impacts of state and private actions which are contemporaneous with the consultation in progress.

Warner Sucker

Federal land management. The Federal agencies responsible for management of the habitat in the Warner Basin have consulted on activities that might impact the Warner sucker. Since the listing of Warner sucker as threatened in 1985, the Lakeview Resource Area has completed numerous consultations on BLM actions affecting Warner sucker. The following lists the subject and year the consultation was completed: Habitat Management Plan for the Warner Sucker 1985; Fort Bidwell-Adel County road realignment 1987; Warner Wetlands Habitat Management Plan 1990; relocation of Twentymile stream gauge 1993; Lakeview BLM grazing program 1994; reinitiation of consultation on grazing program 1995; Noxious Weed Control Program 1996; reinitiation of consultation on grazing program 1996; informal consultation on guided fishing activities 1997; reinitiation of consultation on grazing program and consultation on a number of small non-grazing projects 1997; reinitiation of consultation on grazing program 1999; informal consultation on Long Canyon Prescribed Fire 1999; grazing permit renewal concurrence 1999; reinitiation of consultation on grazing program 2000; reinitiation of consultation on grazing program 2001; consultation on the Lakeview Resource Management Plan in 2003; consultation on the Hart Lake Pump and Screen project in 2006; consultation on the Twelvemile Juniper Management Project in 2006; and consultation on the South Warner Juniper Removal Project in 2012; and informal consultation on the Sagehen Crossing Project in 2013.

Effects of most BLM management activities are short term in nature with very little effect on Warner sucker habitat. Planning of management activities include consideration for management of Warner sucker habitat and would result in benefits to Warner sucker. Road management projects would have a short term impact to the habitat, and result in proper stabilization of roadways which is expected to improve watershed and stream habitat conditions. 29

Cattle grazing is a persistent activity that occurs year after year and can result in chronic effects to the habitat without causing direct impacts to Warner sucker.

The Forest Service has conducted numerous timber sales and vegetation treatments in the Warner watershed. In 2015, the Forest Service completed informal consultation on an approximately 3,000 acre vegetation thinning and harvest project within the Warner watershed which will include s approximately ten stream miles of habitat restoration in the Honey Creek watershed. Most of the stream restoration work entails repair or replacement of failing road culverts. Other recent vegetation treatment and timber harvest projects include the Grassy Fire Salvage project consulted on in 2004 and the Burnt Willow vegetation treatment project consulted on in 2007.

Cattle Grazing Grazing of cattle has been an ongoing activity in the Warner Valley since settlement by ranchers in the 1800’s. Present grazing prescriptions and monitoring protocols are in accordance with biological opinions issued by the Service, and results of grazing monitoring reports are submitted annually to the Service. The BLM has reduced the impact to occupied and potential Warner sucker habitat by taking actions such as excluding occupied stream reaches from cattle grazing and prescribing light early season grazing to stream reaches that may affect Warner sucker or it’s designated critical habitat. Proper Functioning Surveys of riparian conditions were conducted by BLM in 1996. More recent surveys have not been conducted. Following is a summary of the Proper Functioning Condition riparian and stream reach conditions within the action area for this consultation.

Twentymile Creek Primary streams in this watershed are Twelvemile Creek, Fifteenmile Creek, Twentymile Creek, and Horse Creek. The BLM-administered portions of the streams in the Twentymile Creek Watershed were surveyed in 1996; a portion of Horse Creek was re-surveyed in 2000. These surveys found over 65% of the streams in Proper Functioning Condition, and over 20% Functioning at Risk with an Upward Trend. The primary reaches not at Proper Functioning Condition or Functioning at Risk with an Upward Trend was the lower-most 1.05 mile BLM administered reach of Twelvemile Creek, and the 2.5 mile reach of Twelvemile Creek that begins with the upper-most one mile of the Nevada reach and extends upstream. These reaches were determined to be non-functional in 1996. Most of the occupied Warner Sucker habitat on BLM-administered lands in this watershed has been excluded from cattle grazing since 1995 or earlier (U.S. BLM 2015).

30

Deep Creek

Primary streams in this watershed are Deep Creek, Parsnip Creek, Drake Creek, and Camas Creek. The BLM-administered portions of the streams in the Deep Creek Watershed were surveyed in 1996; a portion of Drake Creek was re-surveyed in 2013. Over 75% of the streams were rated as Proper Functioning Condition, and over 20% Functioning at Risk with an Upward Trend. Most of the occupied Warner Sucker habitat on BLM-administered lands in this watershed has generally been excluded from grazing since 2000 or earlier (U.S. BLM 2015).

Honey Creek

Primary streams on in this watershed are Honey Creek, Snyder Creek, and Twelvemile Creek. The BLM- administered portions of the streams in the Honey Creek Watershed were surveyed in 1996. All of the streams surveyed were rated in Proper Functioning Condition. Most of the occupied Warner Sucker habitat on BLM-administered lands in this watershed has been excluded from grazing since 1985 or earlier (U.S. BLM 2015).

In addition to BLM Grazing allotments, most of the Forest Service and private lands within the Warner watershed is also grazed by cattle seasonally. The Forest Service allotments are managed on a deferred rotation. This type of management is not as beneficial to riparian area management as rest rotation systems. Rest rotation allows riparian areas to recover and are preferred for most riparian management because it allows for seasonal plant regrowth, stream and riparian rehabilitation, conserves stream bank stability, and brushy species condition (Platts 1991). The Forest Service uses standards and guidelines to meet vegetation use criteria which are intended to retain riparian vegetation at a level that will allow for the “near natural rate of recovery” of the designated riparian habitat conservation areas. Large areas of the private lands are managed as hay fields or pastures used during seasons in which Forest Service or BLM permitted lands are not available, usually during the winter months.

Fish passage improvements: The Dyke diversion structure is a 1.2 meter (4 feet) high irrigation diversion that was impassable to Warner sucker and redband trout before a fishway was installed in 1991. It blocked all migration of fish from the lower Twentymile Creek, Twentymile Slough and Greaser Reservoir populations from moving upstream to spawning or other habitats above the structure. In 1991, BLM installed a modified steep-pass Denial fish passage structure on the Dyke diversion on lower Twentymile Creek. The fish-way was intended to re-establish a migration corridor, and allow access to high quality spawning and rearing habitats. In 2015, the passage structure at the Dyke diversion was reconstructed by BLM to accommodate Warner sucker passage. Monitoring of the fish-way documented Warner sucker, and other fish, passing through the structure shortly after completion.

Irrigation water is diverted out of Twentymile Creek and into a series of canals which are then diverted out onto agricultural fields for forage and livestock. Warner sucker likely disperse downstream into the irrigation canals. Larvae stage fish are most vulnerable to be affected by 31

the diversion structures and irrigation pumps. In 2007, the NRCS sponsored a program to assist landowners with irrigation improvements. As a part of that program three irrigation pump diversions were equipped with screens to prevent impingement of Warner sucker. Although surveys have not been conducted indicating Warner sucker presence in the canal, the Service assumes Warner sucker would be present in the vicinity of the irrigation diversion structures because no surveys have been conducted and the waterway is connected to occupied Warner sucker habitat without any known barriers to their downstream movement. Spray type screens were installed by the irrigation water users, on the three irrigation diversion points.

An evaluation of fish passage alternatives has been done for diversions on Honey Creek which identifies the eight dams and diversions on the lower part of the creek that are barriers to fish migration (Campbell-Craven Environmental Consultants 1994). In May 1994, a fish passage structure was tested on Honey Creek. It consisted of a removable fish-way and screen. The ladder immediately provided passage for a small redband trout.

In 2008 through 2010, two passage and screening projects were completed on Honey Creek on the Taylor Ranch. These two passage structures are located on the upper limits of Warner sucker known distribution, with the anticipation that Warner sucker would be able to expand their distribution without the hindrance of blockage from the irrigation structures. The passage structures also allow for passage of redband trout.

In 2014, a fish passage structure was completed at the Rookery diversion on Honey Creek. A screen is scheduled to be installed during the low flow period in 2015-2016. To date, this screen has not been installed. The passage structure has not been monitored for effectiveness due to drought induced low stream flows since it was installed.

Foskett Speckled Dace:

Federal Land management In 1985 BLM acquired the site of the only known location of Foskett speckled dace and constructed an exclosure fence around the Foskett Spring and Dace Spring habitat soon after acquisition. Cattle have not been allowed to graze within the exclosure since then, except for occasional stray cattle trespassing into the excluded area.

Current Actions In 2009, the Service and BLM implemented a project to create two spring-fed ponds in the outflow of Dace Spring. The purpose of this project was to establish a second population of Foskett speckled dace in the outflow of Dace Spring for Foskett speckled dace. In September 2010, 49 Foskett speckled dace were transferred from Foskett Spring to the two Dace Spring 32

ponds. The objectives of the translocation of Foskett speckled dace to Dace Spring are to provide more open water habitat and to establish a refuge sub-population in addition to the Foskett Spring population.

In order to expand the population numbers at Foskett Spring, the BLM implemented a habitat enhancement project which excavated a total estimated 825 to 1,100 cubic feet of material from the stream and marsh habitat at Foskett Spring. The increase in open-water habitat had an immediate effect on the Foskett speckled dace with an observed 1,247 percent increase in the population resulting in an estimated 24,000 fish (Scheerer et al. 2014). The BLM is planning an additional project to excavate several larger pools in order to increase the open water habitat at Foskett Spring and its outflow.

In 2015, the Service in cooperation with ODFW and BLM, completed and signed the “Foskett Speckled Dace Foskett Speckled Dace Cooperative Management Plan”. The purpose of the Cooperative Management Plan is to provide for the long term persistence of the Foskett speckled dace through the management and protection of Foskett speckled dace habitat at Foskett and Dace springs. The Cooperative Management Plan defines actions needed for long term conservation of the Foskett speckled dace and identifies the roles and responsibilities of the Cooperators (U.S. Fish and Wildlife Service 2015).

Foskett Speckled Dace Research Recent studies and observations have begun to help understand growth rates, age of reproduction and behavioral patterns. The ODFW learned from monitoring the Dace Spring translocation that Foskett Speckled Dace can mature in one year. This was evident in 2014 when the population more than doubled in abundance from 2013 and recruits grew to adult size in one year. The ODFW also learned from monitoring this translocation, that these dace live at least three years, as adults stocked in 2011 were still present in 2013 and 2014. From studies at Foskett Spring, the ODFW documented annual recruitment and a broad size distribution, and noted that spawning occurs, as evidenced by presence of larval dace, beginning in early spring (March- April) and extending into July (Scheerer 2014, pers. comm.). Regarding habitat use and preference, the ODFW described speckled dace preferring open water habitat and noted that the population increased rapidly in response to habitat restoration/creation of open water habitat (Scheerer et al. 2013). The ODFW also found that young-of-the-year dace are more common in the shallow marsh habitats.

ODFW is expected to continue population monitoring at Foskett Spring, every 2-3 years, into the foreseeable future. Fish are captured (some up to three times), handled, and marked (fin clip), during this process. Up to 10 percent of the population may be moved to Dace Spring in study 33

years. Up to 500 total fish are expected to be moved to Dace Spring over the life of the Dace Spring Population Re-establishment Project.

EFFECTS OF THE ACTION

The effects of the action refers to the direct and indirect effects of an action on the species or critical habitat, together with the effects of other activities that are interrelated or interdependent with that action, that will be added to the environmental baseline (50 CFR 402.02). Direct effects occur simultaneously with, or immediately after implementation of the Proposed Action. Indirect effects are those that are caused by the Proposed Action and are later in time, but still are reasonably certain to occur. Future actions with Federal involvement are not covered under this document but are subject to separate section 7 analyses and review. All of these actions are evaluated against, and subsequently added to, the current environmental baseline.

Direct and Indirect Effects

Integrated Pest Management activities are likely to cause short-term displacement or disturbance to Warner sucker and Foskett speckled dace that are feeding, or resting in occupied habitat within the action area. The Proposed Action includes “Standard Operating Procedures and Mitigation Measures” identified in the Biological Assessment (U.S. BLM 2015) to reduce the impacts to listed fish as well as other resources.

Mechanical and herbicidal treatments of invasive plant species in riparian areas are not likely to substantially decrease shading of streams. Most invasive plants are low growing streamside vegetation that do not provide a significant amount of streamside shade and, will eventually be replaced by native vegetation. The loss of shade would persist until native vegetation reaches and surpasses the height of the invasive plants that were removed. Shade recovery may take one to several years, depending on the success of invasive plant treatment, stream size and location, topography, growing conditions for the replacement plants, and the density and height of the invasive plants when treated. However, short-term shade reduction may occur due to removal of riparian weeds, which could slightly affect stream temperatures or dissolved oxygen levels, which could cause short-term impacts to fish adults, juveniles and eggs.

Manual and Mechanical Treatments Manual and mechanical treatments will result in low level effects to stream habitat conditions. Hand-pulling of emergent vegetation will result in localized suspension and mobilization of sediments. Treatment of streamside invasive plant species with herbicides (by stem injection or spot spray) is likely to result in short-term releases of suspended sediment when treatment of 34

locally extensive streamside monocultures occurs. These treatments are likely to produce minor damage to riparian soil and vegetation.

Under some applications, stream shade will decrease, suspended sediment and temperature in the water column will increase, organic inputs (e.g., insects, leaves, woody material) will be reduced and streambanks and the composition of stream substrates will be altered. These affects are likely to occur when applying treatment to invasive plant monocultures that encompass the stream channel. This effect would vary depending on extent, site aspect, elevation, and amount of topographic shading.

Herbicide Applications The effect analysis of herbicide applications can be analyzed under three different modes of exposure: 1) runoff from riparian application; 2) application within perennial stream channels; and 3) runoff from intermittent stream channels and ditches. Stream margins often provide shallow, low-flow conditions, have a slow mixing rate with mainstem waters, and are the site at which subsurface runoff is introduced. Studies of Warner sucker and Foskett speckled dace are not available for this analysis. Analysis have been done for other species and are used as surrogates in this analysis. Juvenile fish, particularly recently emerged fry, often use low-flow areas along stream margins. For example, wild Chinook salmon rear near stream margins until they reach about 60 mm in length. As juveniles grow, they migrate away from stream margins and occupy habitats with progressively higher flow velocities. Stream margins continue to be used by larger fish for a variety of reasons, including nocturnal resting, summer and winter thermal refuge, predator avoidance, and flow refuge (U.S. Fish and Wildlife Service 2013). Warner sucker and Foskett speckled dace use habitats with low stream flow early in their development and shift to higher velocities as they mature in much the same way.

Spray and vapor drift are important pathways for herbicide entry into aquatic habitats. Several factors influence herbicide drift, including spray droplet size, wind and air stability, humidity and temperature, physical properties of herbicides and their formulations, and method of application. For example, the amount of herbicide lost from the target area and the distance the herbicide moves both increase as wind velocity increases. Under inversion conditions, when cool air is near the surface under a layer of warm air, little vertical mixing occurs. Spray drift is most severe under these conditions, since small spray droplets will fall slowly and move to adjoining areas even with very little wind. Low relative humidity and high temperature cause more rapid evaporation of spray droplets between sprayer and target. This reduces droplet size, resulting in increased potential for spray drift. Vapor drift can occur when the herbicide particles volatilize. The formulation and volatility of the compound will determine its vapor drift potential. The potential for vapor drift is greatest under high air temperatures and low humidity and with ester formulations. For example, ester formulations of triclopyr are very susceptible to vapor drift, 35

particularly at temperatures above 80°F. When temperatures go above 75˚F, 2, 4-D ester chemicals evaporate and drift as vapor. Even a few days after spraying, ester-based phenoxytype herbicides still release vapor from the leaf surface of the sprayed weed (DiTomaso et al. 2006). The herbicides 2,4-D and triclopyr, as well as many other herbicides and pesticides are detected frequently in freshwater habitats (U.S. Fish and Wildlife Service 2013).

When herbicides are applied with a sprayer, nozzle height controls the distance a droplet must fall before reaching the weeds or soil. Less distance means less travel time and less drift. Wind velocity is often greater as height above ground increases, so droplets from nozzles close to the ground would be exposed to lower wind speed. The higher an application is made above the ground, the more likely it is to be above an inversion layer that will not allow herbicides to mix with lower air layers and will increase long distance drift. Several proposed Standard Operating Procedures address these concerns by ensuring that aerial application will be buffered 100 feet from water and will not be applied when wind velocity exceeds six miles per hour.

Standard height of spray aircraft is 30 to 45 feet above ground level. Spray droplet sizes range from 200 to 800 microns. A 500 micron droplet is will likely fall approximately 6.6 feet per second (http://hypertextbook.com accessed 2015). With a cross wind of six miles per hour or 8.8 feet per second, any given droplet will hit the ground from a 45 foot level approximately 60 feet from the release point. The BLM has determined that the 100 foot buffer size is adequate to prevent spray drift from entering adjacent waterbodies. If Herbicide is intentionally sprayed directly onto or in close proximity to waterbodies, the aquatic approved herbicides shall be used.

Herbicide treatments made using ground equipment or by hand, will be conducted under calm conditions, preferably when humidity is high and temperatures are relatively low. Ground equipment reduces the risk of drift, and hand equipment nearly eliminates it.

Surface water contamination with herbicides can occur when herbicides are applied intentionally or accidentally into ditches, irrigation channels or other bodies of water, or when soil-applied herbicides are carried away in runoff to surface waters. Direct application into water sources is generally used for control of aquatic plant species. Accidental contamination of surface waters can occur when irrigation ditches are sprayed with herbicides or when buffer zones around water sources are not wide enough. In these situations, use of hand application methods will greatly reduce the risk of surface water contamination.

The contribution from runoff will vary depending on site and application variables, although the highest pollutant concentrations generally occur early in the storm runoff period when the greatest amount of herbicide is available for dissolution (Stenstrom and Kayhanian 2005). Lower exposures are likely when herbicide is applied to smaller areas, when intermittent stream 36

channel or ditches are not completely treated, or when rainfall occurs more than 24 hours after application. Under the Proposed Action, some formulas of herbicide can be applied within the bankfull elevation of streams, in some cases up to the water’s edge. Any juvenile fish in the margins of those streams will be exposed to herbicides as a result of overspray, inundation of treatment sites, percolation, surface runoff, or a combination of these factors.

Groundwater contamination is another important pathway. Most herbicide groundwater contamination is caused by “point sources,” such as spills or leaks at storage and handling facilities, improperly discarded containers, and rinses of equipment in loading and handling areas, often into adjacent drainage ditches. Point sources are discrete, identifiable locations that discharge relatively high local concentrations. In soil and water, herbicides persist or are decomposed by sunlight, microorganisms, hydrolysis, and other factors. Triclopyr and 2,4-D are detected frequently in freshwater habitats within the four western states (U.S. Fish and Wildlife Service 2013). Proposed Standard Operating Procedures minimize these effects by ensuring proper calibration, mixing, and cleaning of equipment. Non-point source groundwater contamination of herbicides is relatively uncommon but can occur when a mobile herbicide is applied in areas with a shallow water table. Proposed Standard Operating Procedures minimize the potential for non-point pollution by restricting the formulas used, and the time, place and manner of their application to minimize offsite movement.

The BLM concludes that no measurable effects to sediment, temperature, or dissolved oxygen are expected from herbicide treatments in riparian areas given the treatments planned and the limited extent of current infestations. Application rates, levels of toxicity, and protection measures, defined as part of the Proposed Action, will result in levels below lethal concentration for aquatic organisms due to most of the herbicides exhibiting no risk to fish under any application (U.S. BLM 2015a) coupled with the Standard Operating Procedures, Mitigation Measures, and other measures (Appendix A of the EA) designed to keep non‐aquatic herbicides from getting into waterways.

Ecological Risk Assessments of chronic (long‐term) and acute (short‐term) exposures models for chlorsulfuron, clopyralid, dicamba, dicamba + diflufenzopyr, hexazinone, imazapic, imazapyr, metsulfuron methyl, picloram, and sulfometuron methyl showed no risk to any fish under any planned application (Appendix D of the EA, Herbicide Risk Tables). Glyphosate, 2, 4-D, and triclopyr had risk ratings of low to moderate for susceptible fish and no risk to aquatic invertebrates under any application. Fluridone had low to moderate risk to fish, and high risk to aquatic invertebrates in a pond setting. See Table 2 for a summary of risk to fish and aquatic invertebrates for those herbicides that had risk levels above zero.

37

Table 2 Summary of risk to fish and aquatic invertebrates from herbicides that had risk levels above zero based on Ecological Risk Assessments (U.S. BLM 2015) Herbicide 2,4-D glyphosphate triclopyr fluridone Typ1 Max Typ Max Typ Max Typ Max Fish Exposure L2 L L M 0 L 0 L-M Aquatic Invertebrate 0 0 0 0 0 0 0 L-H exposure 1Typ = Typical application rate; and Max = Maximum application rate. 2 Risk categories: = 0 = No risk (majority of RQs < most conservative LOC); L = Low risk (majority of RQs 1‐10x most conservative LOC); M = Moderate risk (majority of RQs 10‐100x most conservative LOC); and H = High risk (majority of RQs >100 most conservative LOC); Risk Category is based on risk level of the majority of risk quotients observed in any scenario for a given exposure group and receptor type.

Under the Proposed Action, Standard Operating Procedures, Mitigation Measures, and other measures are designed to keep non‐aquatic herbicides from getting into waters. Standard Operating Procedures and Mitigation Measures including limitations on the herbicides, adjuvants, carriers, handling procedures, application methods, drift minimization measures, and riparian buffers greatly reduce the likelihood that significant amounts of herbicide would be transported to aquatic habitats, although some herbicides may enter streams through aerial drift, in association with eroded sediment in runoff, and dissolved in runoff, including runoff from intermittent streams and ditches. Herbicide applications could result in herbicide drift or transportation into streams that would harm listed species by chemically impairing normal fish behavioral patterns related to feeding, rearing, and migration. Adverse effects to Warner Sucker and Foskett Speckled Dace are anticipated, particularly from use of the herbicides listed in Table 2 above (2,4-D, glyphosate, triclopyr, and fluridone), which have more risk to fish than zero (see Appendix D of the EA (U.S. BLM 2015a), Herbicide Risk Tables). Fluridone in particular has a low to moderate risk to fish, a low to high risk to aquatic invertebrates, and is an aquatic herbicide designed to be directly applied to aquatic habitats. Refer to the Environmental Consequences portions of the Soil Resources and Water Resources sections of the EA for a discussion of environmental fate and persistence of the proposed herbicides (U.S. BLM 2015a).

Herbicide toxicity. Herbicides included in the invasive plant project were selected due to their low to moderate aquatic toxicity to fish. The risk of adverse effects from the toxicity of herbicides and other compounds present in formulations to listed aquatic species is minimized in this programmatic activity by reducing stream delivery potential by restricting application methods. Only aquatic labeled herbicides are to be applied within wet stream channels. Aquatic glyphosate and aquatic imazapyr can be applied up to the waterline using spot spray or hand selective application methods in both perennial and intermittent channels. Triclopyr TEA (triethylamine salt) and 2,4-D amine can be applied up to the waterline, but only using hand selective techniques. The associated application methods were selected for their low risk of contaminating soils and subsequently introducing herbicides to streams. However, direct and indirect exposure and toxicity risks are inherent in some application types. Generally, herbicide 38

active ingredients have been tested on a limited number of species and mostly under laboratory conditions.

While laboratory experiments can be used to determine acute toxicity and effects to reproduction, cancer rates, birth defect rates, and other effects to fish and wildlife, laboratory experiments do not typically account for species in their natural environments and little data is available from studies focused specifically on the listed species in this opinion. This leads to uncertainty in risk assessment analyses. Environmental stressors increase the adverse effects of contaminants, but the degree to which these effects are likely to occur for various herbicides is largely unknown. The effects of the herbicide applications to various representative groups of species have been evaluated for each proposed herbicide.

The effects of herbicide applications using spot spray, hand/select, and broadcast spray methods were evaluated under several exposure scenarios: (1) runoff from riparian (above high water mark) application along streams, lakes and ponds, (2) runoff from treated ditches and dry intermittent streams, and (3) application within perennial streams (dry areas within channel and emergent plants). The potential for herbicide movement from broadcast drift was also evaluated. Risks associated with exposure and associated effects were also evaluated for terrestrial species.

Although the Standard Operating Procedures and Protective Measures would minimize drift and contamination of surface and ground water, herbicides reaching surface waters could result in mortality to fish during incubation, or lead to altered development of embryos if the protective buffers were not implemented. Stehr et al. (2009) found that the low levels of herbicide delivered to surface waters are unlikely to be toxic to the embryos of ESA-listed salmonids. However, mortality or sub-lethal effects such as reduced growth and development, decreased predator avoidance, or modified behavior are likely to occur. Herbicides may also impact the food base for listed salmonids and other fish, which includes terrestrial organisms of riparian origin, aquatic macroinvertebrates and forage fish if introduced to their habitats.

Adverse effect threshold values for each species group were defined as either 1/20th of the LC50 (standard measure of the toxicity of a surrounding medium that will kill 50 % of the sample population of a specific test animal in a specified period through exposure) value for listed salmonids, 1/10th of the LC50 value for non-listed aquatic species, or the lowest acute or chronic “no observable effect concentration,” whichever was lower, found in Syracuse Environmental Research Associates, Inc. (SERA) risk assessments that were completed for the USFS; i.e., sulfometuron-methyl (SERA 2004d) imazapic (SERA 2004a), chlorsulfuron (SERA 2004b), dicamba (SERA 2004c), 2,4-D (U.S. Forest Service 2006), glyphosate (SERA 2011a), imazapyr (SERA 2011b), Picloram (SERA 2011c), and triclopyr (SERA 2011d). These assessments form 39

the basis of the analysis in U.S. Forest Service et al. (2013) and ARBO II (U.S. Fish and Wildlife 2013).

Generally, effect threshold values for listed salmonids were lower than values for other fish species groups, so values for salmonids were also used to evaluate potential effects to other ESA- listed fish. In the case of sulfometuron-methyl, threshold values for fathead minnow (surrogate) were used to evaluate effects to ESA-listed fish. Data on toxicity to wild fish under natural conditions are limited and most studies are conducted on lab specimens. Adverse effects could be observed in stressed populations of fish, and it is less likely that effects would be noted in otherwise healthy populations of fish. Long-term studies on fish egg-and-fry are seldom conducted. Risk characterizations for both terrestrial and aquatic species are limited by the relatively few animal and plant species on which data are available, compared to the large number of species that could potentially be exposed. This limitation and consequent uncertainty is common to most if not all ecological risk assessments. Additionally, in laboratory studies, test animals are exposed to only a single chemical. In the environment, wildlife may be exposed to multiple toxicants simultaneously, which can lead to additive or synergistic effects.

Hazard quotients evaluations are summarized for the herbicides chlorsulfuron, clopyralid, glyphosate, imazapyr, metsulfuron methyl, and sulfometuron methyl. Hazard quotients were calculated by dividing the expected environmental concentration by the effects threshold concentration. Adverse effect threshold concentrations are 1/20th (for ESA listed aquatic species) or 1/10th (all other species) of LC50 values, or “no observable adverse effect” concentrations, whichever concentration was lower. The Water Contamination Rate values are categorized by herbicide, annual rainfall level, and soil type. Variation of herbicide delivery to streams among soil types (clay, loam, and sand) is displayed as low and high Water Contamination Rate values. All Water Contamination Rate values are from risk assessments conducted by SERA. Given that there are hazard quotient values >1 adverse effects are likely to occur. Hazard quotient values were calculated for fish, aquatic invertebrates, algae, and aquatic macrophytes. Adverse effect threshold values for each species group were defined as either 1/20th of the LC50 value for listed salmonids, 1/10th of the LC50 value for non-listed aquatic species, or the lowest “no observable effect concentration,” whichever was lower, found in available literature (U.S. Fish and Wildlife Service 2013).

For dicamba, diflufenzopyr + dicamba, imazapic, picloram, triclopyr, and 2,4-D, which were added to list, ARBO II referred to the NMFS’s opinions, SERA reports, various other literature sources, and the 2013 ARBO II BA (U.S. Fish and Wildlife Service 2013) to characterize risk to ESA-listed fish species. Each type of Integrated Pest Managenment treatment could affect fish and aquatic macrophytes through a combination of pathways, including disturbance, chemical 40

toxicity, dissolved oxygen and nutrients, water temperature, sediment, instream habitat structure, forage, and riparian and emergent vegetation (Table 3).

Table 3. Potential pathways of effects of Integrated Pest Management. Pathways of Effects

Treatment ed Methods Water Forage habitat toxicity nutrients Instream structure emergent Dissolv Chemical vegetation oxygen and temperature Disturbance* and turbidity and Riparian and Fine sediment

Herbicides X X X X X X X Manual and X X X X X X Mechanical Directed Livestock X X X X X X X Prescribed Fire X X X X X X X *Stepping on redds, displacing fish, interrupting fish feeding, or disturbing banks.

Manual and Mechanical Certain manual and mechanical treatments within riparian areas that disturb soil, such as grubbing and pulling, carried out over a large area, may lead to increased erosion and stream sedimentation. Resultant sedimentation may adversely affect fish by covering eggs or spawning gravels, reducing prey availability, or directly harming fish gills. The risk of harm to aquatic ecosystems due to fine sediment /production from manual treatment or use of motorized hand tools is low, and short‐term, resulting in effects likely to be localized and minor. Depending on the scale of treatment, pulling significant numbers of plants or treating large riparian areas with motorized hand tools could increase the risk to fish. Cut vegetation left on the treatment site can reduce the potential for erosion and subsequent sediment delivery to water bodies (U.S. BLM 2015).

The risk of harm to fish from use of wheeled or tracked machinery would vary, depending on the extent of treatment area and proximity to aquatic environments; vehicle tracks can compact soils and divert water. Fish are temporarily affected when water is affected by increased turbidity, excessive sedimentation, and local increases in surface water runoff. However, all wheeled equipment (including off‐highway vehicles and other herbicide application equipment) would be kept away from riparian areas to minimize effects and the risk of spills. Power‐tool use near water can cause water contamination with minor amounts of chainsaw oil or minor fuel spill. An oil skim on water, while highly unlikely, can deplete oxygen levels and cause injury to fish (U.S. BLM 2015). 41

Directed Livestock Use The impacts of livestock grazing to stream habitat and fish populations can generally be separated into acute and chronic effects. Acute effects are those which contribute to the immediate loss of individual fish and loss of specific habitat features (undercut banks, spawning beds, etc.) or localized reductions in habitat quality (sedimentation, loss of riparian vegetation, etc.). Chronic effects are those which, over a period of time, result in loss or reductions of entire populations of fish, or widespread reductions in habitat quantity and/or quality. No chronic effects are expected as the treatments proposed in this Biological Assessment as the proposed treatments are extremely infrequent, and short-term in nature. Directed livestock would not be conducted within occupied Warner sucker habitat (U.S. BLM 2015).

Potential Direct Effects Fish spawning periods and associated incubation of eggs within streambeds are very important to the success of a healthy fish population. Spawning fish can be disturbed by livestock during spawning periods. Trampling of spawning sites can also occur if grazing occurs when fish are spawning or while eggs are incubating within streambeds. Roberts and White (1992) studied the effects of rainbow trout redd disturbance by anglers and showed a significant reduction in the survival of rainbow trout eggs and pre-emergent fry due to disturbance of redd sites by angler wading. Similar impacts could occur due to livestock grazing in Foskett speckled dace spawning areas.

Incubating Foskett Speckled Dace generally emerge from streambeds by early July (U.S. BLM 2015). The potential for direct impacts (trampling) to spawning steelhead has been studied and can be used as a surrogate to analyze potential impacts to Foskett speckled dace. During the early season of use, cattle prefer uplands and not riparian areas, greatly reducing the risk of direct effects. Ballard and Krueger (2005) studied late summer (August) cattle distribution and habitat use in northeast Oregon. The study was conducted in a pasture grazed to 70% utilization. Cattle spent 94% of their time in terrestrial habitats and 6% of their time in stream habitats (5% gravel bars and less than 1% in aquatic habitats). Gregory and Gamett (2009) found highly variable levels of cattle trampling on simulated bull trout redds (12-78%). Gregory and Gamett (2009) also found that impacts to redds were higher in pastures where stocking intensity was higher, but impacts were also determined by site conditions adjacent to the simulated redds. Therefore, disturbance to spawning substrates from cattle grazing in stream channels where fish spawn is expected to harm or harass Foskett speckled dace throughout the length of stream occupied by Foskett speckled dace.

42

Potential Indirect Effects Acute effects of grazing to habitat can include compacting stream substrates, collapsing of undercut banks, destabilizing streambanks, and localized reduction or removal of herbaceous and woody vegetation along streambanks and within riparian areas (Platts 1991 ). Increased levels of sediment can result from the re-suspension of material within existing stream channels as well as increased contributions of sediment from adjacent streambanks and riparian areas. The effects of sedimentation occur both within and downstream of treatment areas. Impacts to stream and riparian areas resulting from grazing are dependent on the intensity, duration, arid timing of grazing activities (Platts 1989) as well as the capacity of a given watershed to assimilate imposed activities, and the pre-activity condition of the watershed (Odum 1981).

When riparian areas are grazed, increased sediment levels resulting from cattle use can be expected to impact stream reaches within, and downstream of grazed areas. Vulnerability of Foskett speckled dace to acute effects of grazing is greatest during early development stages. During early phases of their life cycle, fish have little or no capacity for mobility, and large numbers of embryos or young are concentrated in small areas. Cattle entering spawning areas can trample, destroy, or dislodge embryos and early larvae. Embryo and larval mortality can also result from localized sedimentation of spawning beds (Bjornn and Reiser 1991). Accumulations of silt, if delivered in sufficient quantity, can fill interstitial spaces in streambed material, impeding water flow through gravels, reducing dissolved oxygen levels, and restricting removal of wastes from spawning areas. As fish development progresses, vulnerability to direct mortality from acute effects decreases. Impact distance is a function of channel slope, stream water level, and the sequence of habitat conditions downstream of the impact area.

Prescribed Fire for Invasive Plant Control The risk of harm to fish from prescribed fire for invasive plant control depends on fire intensity, timing, and landform, among other factors. Prescribed burning has the potential to bare large areas of soil, and thus increase both surface erosion and sedimentation of streams. Heavy runoff from burned areas can increase water pH, indirectly affecting aquatic biota. Implementation of Standard Operating Procedures would prevent prescribed fire from being used where it may result in adverse effects to streams (U.S. BLM 2015).

Summary of Direct and Indirect Effects

The use of herbicides has no to moderate risk to fish and aquatic invertebrates if applied to directly to occupied waterways. With the use of the proposed application buffers, the risk drops to no to low risk. Directed livestock grazing may affect Foskett speckled dace due to direct effects from grazing riparian areas occupied by Foskett speckled dace. Effects would include impacts related to vegetation removal from grazing cattle, including streambank impacts from 43

hoof action, increased input of sediment instream, potential increase in water temperature, and impacts to aquatic invertebrates, resulting in reduction of primary productivity thereby affecting feeding behavior of Foskett speckled dace. Also, some direct effects to Foskett speckled dace could occur, including the disturbance of spawning fish and trampling of incubating eggs, as directed livestock grazing may occur in occupied Foskett speckled dace habitat.

Implementation of the Integrated Pest Management Project may adversely modify Warner sucker designated critical habitat through introduction of chemicals which have a low to moderate toxicity to fish. The BLM has designed the Proposed Action to minimize potential for introduction of chemicals to occupied habitat and include Standard Operating Procedures which will further limit the introduction of chemicals to streams. Constituent elements defined for the designated critical habitat include “streams 15 feet to 60 feet wide with gravel-bottom shoal and riffle areas with intervening pools. Streams should have clean, unpolluted flowing water and a stable riparian zone. The streams should support a variety of aquatic insects, crustaceans, and other small invertebrates for food” (U.S. Fish and Wildlife Service 1985a).

Stehr et al. (2009) studied developmental toxicity in zebrafish (Danio rerio), which involved conducting rapid and sensitive phenotypic screens for potential developmental defects resulting from exposure to six herbicides (picloram, clopyralid, imazapic, glyphosate, imazapyr, and triclopyr) and several technical formulations. Available evidence indicates that zebrafish embryos are reasonable and appropriate surrogates for embryos of other fish, including salmonids. The absence of detectable toxicity in zebrafish screens is unlikely to represent a false negative in terms of toxicity to early developmental stages of threatened or endangered salmonids. Their results indicate that low levels of noxious weed control herbicides are unlikely to be toxic to the embryos of ESA-listed fish. Those findings do not necessarily extend to other life stages or other physiological processes (e.g., disease susceptibility, behavior, etc.) (U.S. Fish and Wildlife Service 2013).

The BLM has designed the protective buffers for herbicide application specifically to avoid application to occupied Warner sucker and Foskett speckled dace habitat. The protective buffers prevent the entry of toxic levels of herbicides into waterways occupied by the species under consideration in this biological opinion. Therefore, no harm or injury expected from the proposed herbicide application.

The Proposed Action’s standard operating procedures and protection measures, including limitations on the herbicides, adjuvants, carriers, handling procedures, application methods, drift minimization measures, and riparian buffers, will reduce the likelihood of significant amounts of herbicide will be transported to aquatic habitats. The BLM also identifies the potential for some herbicides to enter streams through aerial drift, in association with eroded sediment in runoff, or 44

dissolved in runoff (BLM 2015). Although contamination form runoff may be an associated effect from herbicide application, the protective buffers defined in the Proposed Action will prevent runoff of significant amounts of herbicide from entering habitat occupied by Warner sucker of Foskett speckled dace.

Interrelated and Interdependent Actions

No interrelated or interdependent actions associated with the Integrated Pest Management Project have been identified.

CUMULATIVE EFFECTS

Cumulative effects are defined in the implementing regulations of section 7 of the Act as those effects of future State, Tribal, local, or private actions not involving Federal activities, that are reasonably certain to occur within the action area of the Federal action subject to consultation (50 CFR 402.2). Future Federal actions that are unrelated to the Proposed Action are not considered in this section because they require separate consultation pursuant to section 7 of the Act.

Warner Sucker Privately owned land parcels are interspersed with BLM and Forest Service land in the Honey, Deep, and Twentymile creeks watersheds. These private lands are mostly wet meadows and riparian areas along streams, although some uplands are also private. Many of the private lands are grazed season-long every year and have moderately to severely degraded riparian and stream conditions. In some areas, streams are diverted to irrigate the private pastures, resulting in fish in canals, barriers to upstream and downstream passage at the diversions, dewatering of streams, and further degradation of water quality. These practices are expected to continue in the foreseeable future.

The State owns several parcels of land in the Warner Basin, including two allotments near the confluence of Snyder Creek and Honey Creek, and lakeshore riparian areas. These lands are managed primarily for livestock grazing. Some aquatic habitat condition issues (bank stability and channel downcutting) were identified, particularly on a one mile reach of Honey Creek near its confluence with Snyder Creek (U.S.BLM 2015). The private land grazing strategy is unknown.

Other non-Federal activities in the action area include road maintenance by the State and Lake County. Highway 140 is the only State right-of-way in the area, and herbicides have been used in the past to control roadside vegetation near Deep Creek. The only County road that directly 45

affects sucker-inhabited streams is a section of Twentymile Creek road which parallels the creek just before it enters the Warner Valley. The road is occasionally graded and parts of it treated for dust abatement near residences. Roads are graded to maintain the design drainage of the road and help prevent catastrophic road failure. Some fine sediment may become destabilized during grading and eventually reach Twentymile Creek.

Foskett Speckled Dace Non-Federal activities in the action area include road maintenance done by the Lake County Road Department. The Coleman Valley Road is located approximately 200 feet from Foskett Spring and approximately 400 feet from Dace Spring. This road is occasionally graded. No effects to fish or fish habitat from this activity have been observed, or are expected.

CONCLUSION

After reviewing the current status of Warner sucker and Foskett speckled dace, the environmental baseline for the action area, the effects of the Proposed Action, interrelated and interdependent effects, and cumulative effects, it is the Service’s biological opinion that the proposed Lakeview BLM Integrated Pest Management Project is not likely to jeopardize the continued existence of the threatened Warner sucker and Foskett speckled dace nor will it adversely modify the designated critical habitat for Warner sucker. With the protective buffers in place, in combination with the “Project Design Features, Standard Operating Procedures, Mitigation Measures, Conservation Measures, Prevention Measures, and Best Management Practices” (BLM 2015), toxic levels of all herbicides would be prevented from entering occupied Warner sucker and Foskett speckled dace habitat. Critical habitat has not been designated for Foskett speckled dace; therefore, it will not be affected.

INCIDENTAL TAKE STATEMENT

Section 9 of the Act and Federal regulation pursuant to section 4(d) of the Act prohibit the take of endangered and threatened species, respectively, without special exemption. Take is defined as to harass, harm, pursue, hunt, shoot, wound, kill, trap, capture or collect, or to attempt to engage in any such conduct. Harm is further defined by the Service to include significant habitat modification or degradation that results in death or injury to listed species by significantly impairing essential behavioral patterns, including breeding, feeding, or sheltering. Harass is defined by the Service as intentional or negligent actions that create the likelihood of injury to listed species to such an extent as to significantly disrupt normal behavior patterns which include, but are not limited to, breeding, feeding or sheltering. Incidental take is defined as take that is incidental to, and not the purpose of, the carrying out of an otherwise lawful activity. Under the terms of section 7(b)(4) and section 7(o)(2), taking that is incidental to and not 46

intended as part of the agency action is not considered to be prohibited taking under the Act provided that such taking is in compliance with the terms and conditions of this Incidental Take Statement.

The measures described below are non-discretionary, and must be undertaken by the BLM so that they become binding conditions of the project, as appropriate, for the exemption in section 7(o)(2) to apply. The BLM has a continuing duty to regulate the activity covered by this incidental take statement. If the BLM (1) fails to assume and implement the terms and conditions or (2) fails to require contractors to adhere to the terms and conditions of the incidental take statement through enforceable terms that are added to the permit or grant document, the protective coverage of section 7(o)(2) may lapse. In order to monitor the impact of incidental take, the BLM must report the progress of the action and its impact on the species to the Service as specified in the incidental take statement [50 CFR §402.14(i)(3)].

AMOUNT OR EXTENT OF TAKE

The Service anticipates incidental take of Warner sucker and Foskett speckled dace will be difficult to detect due to the inherent biological characteristics of aquatic species. The likelihood of discovering an individual death or recording sub-lethal effects to Warner sucker or Foskett speckled dace attributable to Integrated Pest Management is very small. For example, small size, behavioral modifications before death, presence of aquatic vegetation, stream flow, and rapid rates of decomposition make finding an incidentally taken individual fish extremely unlikely.

The Service anticipates that the following incidental take is reasonably certain to occur due to the Integrated Pest Management activities associated with the implementation of the Proposed Action.

1. No direct harm in the form of injury or mortality is expected from the proposed herbicide application due to the low toxicity of ten of the 14 herbicides proposed for use and the application of protective buffers to prevent application to habitats occupied by Warner sucker and Foskett speckled dace. Four Herbicides have been identified by the BLM as having higher than zero effect on aquatic animals (fish and aquatic insects). These include 2,4-D, glyphosphate, triclopyr, and fluridone. Without the design features and protective buffers defined by BLM, application of herbicide could drift or be transported into streams that could affect Warner sucker or Foskett speckled dace in the form of harm or harassment in streams and wetland areas in which Integrated Pest Management activities are planned to occur, by chemically impairing normal fish behavioral patterns related to feeding, rearing, and migration.

2. No direct harm in the form of injury or mortality is expected from herbicide application to Warner sucker habitat. Two areas of designated critical habitat are 47

typically unoccupied by Warner sucker. Intermittent sections of Twentymile Creek and the irrigation ditch flowing into the Warner Wetlands are designated as critical habitat. Herbicides will not be applied to these areas when Warner sucker are present in that habitat. Affects from herbicides applied during the time when Warner sucker are absent from the designated critical habitat are not expected to persist until wet seasons when Warner sucker would once again be present.

3. The grazing action would be allowed to occur through the entire extent of Foskett Spring and its out flow. Indirect take to Foskett speckled dace occupied habitat would occur from implementation of “directed livestock grazing”. The amount of take would be limited by meeting standards for cattle grazing utilization standards identified in the Proposed Action (4 to 10 inch stubble height and 15 to 25 percent stream bank alteration). Subsequent treatments using directed livestock grazing will not be allowed until after adequate rest which is defined as full streambank recovery to at least pretreatment condition.

4. Application of manual or mechanical, plant controls will result in short-term reduction of vegetative cover or soil disturbance and degradation of water quality which will cause injury to fish in the form of sub-lethal, adverse physiological effects. These sub-lethal effects, described in the effects analysis for this opinion, will include reduced feeding success, and subtle behavioral changes that can result in predation.

EFFECT OF THE TAKE

In the accompanying biological opinion, the Service determined that this level of anticipated take is not likely to result in jeopardy to the species.

REASONABLE AND PRUDENT MEASURES

The Service believes the following reasonable and prudent measure is necessary and appropriate for BLM to minimize take of Warner sucker and Foskett speckled dace:

1. The BLM shall provide a report that documents that all actions are in compliance with the action described within the Biological Assessment including, but not limited to, adherence to “Project Design Features, Standard Operating Procedures, Mitigation Measures, Conservation Measures, Prevention Measures, and Best Management Practices” consistent with the Biological Assessment and referenced environmental assessment (U.S. BLM 2015) and as required by the Terms and Conditions below.

48

TERMS AND CONDITIONS

In order to be exempt from the prohibitions of section 9 of the Act, the Lakeview Resource Area BLM must comply with the following terms and conditions, which implement the reasonable and prudent measure described above and outline required reporting/monitoring requirements. These terms and conditions are non-discretionary.

1. A complete report of Integrated Pest Management activities will be submitted to the Service's Bend Field Office by February 1, each year following project implementation. Reports shall describe annual Integrated Pest Management activities, and protective measures taken to assure minimization of take to listed species, as described in the Biological Assessment and referenced environmental assessment for the Proposed Action.

2. Upon locating dead, injured, or sick Warner sucker or Foskett speckled dace, initial notification must be made within three working days to the Service's Division of Law Enforcement, Senior Resident Agent in charge in Oregon at (503) 682-6131. The Division of Law Enforcement will issue instructions for proper handling and disposition of such specimens. Care must be taken in handling sick or injured fish to ensure effective treatment and care and in handling dead specimens to preserve biological material in the best possible state. In conjunction with the care of sick or injured Warner sucker or Foskett speckled dace, or the preservation of biological materials from a dead fish, the BLM has the responsibility to ensure that information relative to the date, time, and location of the fish when found, and possible cause of injury or death of each fish be recorded and provided to the Service.

The Service concludes that Warner sucker or Foskett speckled dace will be incidentally taken as a result of the Proposed Action. The extent of this take has been quantified through the use of a habitat surrogate and is estimated to include all occupied Warner sucker and Foskett speckled dace habitat in the action area, but will have no lethal take and has a very low potential for causing harm to Foskett speckled dace and Warner sucker. In order to further limit the amount of take, the BLM shall assure that the project’s water quality protective objectives are met, and documented.

The reasonable and prudent measure, with its implementing terms and conditions, is designed to minimize the impact of incidental take that might otherwise result from the Proposed Action. If, during the course of the action, incidental take occurs due to a failure to comply with the protective measures to assure minimization of take to listed species, as described in the Biological Assessment and referenced environmental assessment for the Proposed Action, Terms 49

and Conditions, or the Reasonable and Prudent Measure such incidental take represents new information requiring reinitiation of consultation and review of the reasonable and prudent measure provided. The BLM must immediately provide an explanation of the causes of the taking and review with the Service the need for possible modification of the reasonable and prudent measure.

CONSERVATION RECOMMENDATIONS

Section 7(a)(1) of the Act directs Federal agencies to utilize their authorities to further the purposes of the Act by carrying out conservation programs for the benefit of endangered and threatened species. Conservation recommendations are discretionary agency activities to minimize or avoid adverse effects of a Proposed Action on listed species or critical habitat, to help implement recovery plans, or to develop information.

1) The BLM should work with adjacent landowners to assure Integrated Pest Management of BLM land is coordinated to more effectively control the targeted invasive species on both BLM and adjacent private lands. For example if control of a patch of invasive plants is conducted on BLM land, the continuity of the patch on adjacent private land should be taken into consideration when designing the treatment strategy.

2) The BLM should work with the Oregon Department of Fish and Wildlife to inventory and further refine the known distribution of Warner sucker to aid in defining areas of occupied habitat.

In order for the Service to be kept informed of actions minimizing or avoiding adverse effects or benefiting listed species or their habitats, the Service requests notification of the implementation of any conservation recommendations.

REINITIATION NOTICE

This concludes formal consultation on the Proposed Action described in the Biological Assessment dated April 16, 2015. As provided in 50 CFR §402.16, reinitiation of formal consultation is required where discretionary Federal agency involvement or control over the action has been retained (or is authorized by law) and if: (1) the amount or extent of incidental take is exceeded; (2) new information reveals effects of the agency action that may affect listed species or critical habitat in a manner or to an extent not considered in this opinion; (3) the agency action is subsequently modified in a manner that causes an effect to the listed species or critical habitat not considered in this opinion; or (4) a new species is listed or critical habitat designated that may be affected by the action. In instances where the amount or extent of 50

incidental take is exceeded (in this case, non-compliance with the Reasonable and Prudent Measure and its implementing Terms and Conditions) any operations causing such take must cease pending reinitiation.

If you have questions or require additional information regarding this consultation, please contact Alan Mauer or me at the Bend Field Office, at 541-383-7146.

cc: Chip Dale, ODFW, Bend, OR. Michele Weaver, ODFW, Salem, OR.

51

References

Allen, C., A. Atkins, and M. Stern. 1994. Status and Recolonization of the Warner sucker (Catostomus warnerensis) and other fishes in the Warner Lakes in Southeast Oregon, 1994. Unpublished report to Bureau of Land Management and Oregon Department of Fish and Wildlife. 22 pp.

Allen, C., K. Hartzell, M. Stern and A. Munhall. 1995a. Status of the Warner sucker (Catostomus warnerensis) and other fishes in the Warner Basin in Southeast Oregon, 1995. Unpublished report to the Fish and Wildlife Service. 35 pp.

Allen, C., M. Stern, and A. Munhall. 1995b. Monitoring of the Warner sucker and other fishes in the Warner Basin in 1994 and 1995. Unpublished report to the Fish and Wildlife Service. 4 pp.

Allen, C., K. Hartzell, and M. Stern. 1996. Warner sucker progress report - 1996 findings. Unpublished report to Bureau of Land Management. 55 pp.

Andreasen, J.K. 1975. Systematics and status of the family Catostomidae in southern Oregon. Doctoral dissertation. Oregon State University, Corvallis, Oregon. 76 pp.

Ardren, W. R., J. Baumsteiger , C. S. Allen. 2009. Genetic Analysis of Threatened Foskett Spring Speckled Dace. U.S. Fish and Wildlife Service, Abernathy Fish Technology Center. Longview, WA.

Ballard, T.M. and W.C. Krueger. 2005. Cattle and salmon I: cattle distribution and behavior in a northeast Oregon riparian ecosystem. Rangeland Ecology and Management. 58:267- 273.

Barlow, G.W. 1995. The relevance of behavior and natural history to evolutionarily significant units. Pages 169-175 in J. L. Nielsen editor. Evolution and the Aquatic Ecosystem. American Fisheries Society Symposium 17. Bethesda, Maryland.

Berg, W. J. 1991. Selected observations and interpretations on the life-history of the Warner sucker (Catostomus warnerensis): conservation genetic management. Unpublished report to the Warner Sucker Working Group. 11 pp.

52

Bills, F.T. 1977. Taxonomic status of isolated populations of tui chub referred to Gila bicolor oregonensis (Snyder). Masters thesis. Oregon State University, Corvallis, Oregon. 116 pp.

Bjornn, T. C., and D. W. Reiser. 1991. Habitat requirements of salmonids in streams. Pages 83– 38 in W. R. Meehan (ed.), Influences of forest and rangeland management on salmonid fishes and their habitats. American Fisheries Society Special Publication 19, Bethesda, Maryland.

Bond, C.E.1974. Endangered plants and animals of Oregon: I. Fishes. OR Agric. Exp. Sta. Spec. Rept. 205:1-9.

Bond, C.E. and C.I. Coombs. 1985. Characteristics and life history of Catostomus warnerensis. Proc. Desert Fishes Council. 13:40-53.

Bosse, S.E., D. Hartzel, C. Allen, M. Stern, and A. Munhall. 1997. Warner Sucker Progress Report-1997 Findings. Unpublished Report to Bureau of Land Management. 78pp.

Campbell-Craven Environmental Consultants. 1994. Draft evaluation of fish passage alternatives for water diversions in Honey Creek, tributary to Hart Lake, Lake County, Oregon. Prepared for Oregon Department of Fish and Wildlife. Portland, Oregon. 16 pp.

Coombs, C. I., C. E. Bond, and S. F. Drohan. 1979. Spawning and early life history of the Warner sucker (Catostomus warnerensis). Unpublished report to Fish and Wildlife Service. 52 pp.

Coombs, C.I., and C.E. Bond. 1980. Report of investigations on Catostomus warnerensis, fall 1979 and spring 1980. Unpublished report to Fish and Wildlife Service. 32 pp.

Cope, E.D. 1883. On the fishes of the recent and Pliocene lakes of the western part of the Great Basin, and the Idaho Pliocene Lake. Proc. Acad. Nat. Sci. Philadelphia. 1883:134-166.

Dambacher, J.M., A.G. Talabere, D.L. Hill, and D.F. Markle. 1997. Foskett speckled dace investigations. Fish Research Progress Report. Oregon Department of Fish and Wildlife and Oregon State University.

53

DeHaan, P. and J. VonBargen. 2011. Genetic analysis of threatened Warner suckers. Report Submitted to: U.S. Army Corps of Engineers. U.S. Fish and Wildlife Service, Abernathy Fish Technology Center, Longview, WA.

DiTomaso, Joseph M, et al. 2013. Weed Control in Natural Areas in the Western United States, Hastings 2005. Hastings, Alan, et al. 2005. The spatial spread of invasions: new developments in theory and evidence. Ecology Letters (2005) 8: 91–101. doi: 10.1111/j.1461‐0248.2004.00687.x

Gregory, J.S., and B.L. Gamett. 2009. Cattle trampling of simulated bull trout redds. North American Journal of Fisheries Management 29:361-366.

Hartzell, K. E, and K. Popper. 2002. Warner Sucker Progress Report-2001 Findings. Unpublished Report to Bureau of Land Management.36pp.

Hayes, J.P. 1980. Fish of the Warner Valley. Pages 131-137 in C. Gilman and J.W. Feminella, editors. Plants and animals associated with aquatic habitats of the Warner Valley. Oregon State University, Corvallis, Oregon.

Healey, M.C. and A. Prince. 1995. Scales of variation in life history tactics of Pacific salmon and conservation of phenotype and genotype. Pages 176-184 in J. L. Nielsen editor. Evolution and the Aquatic Ecosystem. American Fisheries Society Symposium 17. American Fisheries Society, Bethesda, Maryland.

Hoekzema, K. 2013. Phylogenetic and morphometric analysis of speckled dace (Rhinichthys osculus) from Oregon's Great Basin. M.Sci. Thesis, Oregon State University.

Hoekzema, K., and B.L. Sidlauskas. 2014. Molecular phylogenetics and microsatellite analysis reveal cryptic species of speckled dace (Cyprinidae: Rhinichthys osculus) in Oregon’s Great Basin. Molecular Phylogenetics and Evolution, Volume 77, August 2014, pp 238- 250.

Hubbs, C.L., and R.R. Miller. 1948. The zoological evidence of correlation between fish distribution and hydrographic history in the desert basins of western United States. Pp. 17-166. In: The Great Basin, with emphasis on glacial and postglacial times. Bull. Univ. Utah 30(2).

Kennedy, T.B. and J.F. North. 1993. 1992 Report: Drift behavior and distribution of Warner sucker (Catostomus warnerensis) and preliminary assessment of stream habitat conditions 54

in the Warner Valley, Oregon. Unpublished report to Bureau of Land Management and Oregon Department of Fish and Wildlife. 25 pp.

Kennedy, T.B. and M. Olsen. 1994. Drift ecology of Warner sucker larvae (Catostomus warnerensis) and lake recolonizations by native and exotic fishes of the Warner Valley, Oregon. Unpublished report to Bureau of Land Management and Oregon Department of Fish and Wildlife. 42 pp.

Kobetich, G. C. 1977. Report on survey of Warner Valley Lakes for Warner suckers, Catostomus warnerensis. Unpublished report to Fish and Wildlife Service. 6 pp.

Mauger, S. 2000. Invertebrate composition and distribution in desert springs of Oregon. Thesis, Oregon State University.

Moyle, P.B. 1976. Inland fishes of California. University of California Press, Berkely, California. 405 pp.

Munhall, A. 2007. Long term monitoring report of stream conditions on the BLM lands of the Warner basin. Lakeview, OR.

Odum, E.P. 1981. The effects of stress on the trajectory. of ecological succession. In Barrett and Rosenberg eds. Stress effects on natural ecosystems. John Wiley and Sons Ltd., New York, NY. 305 p.

Pfrender M.E., J. Hicks, and M. Lynch. 2003. Biogeographic patterns and current distribution of molecular-genetic variation among populations of speckled dace, Rhinichthys osculus (Girard). Molecular Phylogenetics and Evolution, 30(3), 490-502.

Platts, W.S. 1989. Compatibility oflivestock grazing strategies with fisheries. In Gresswell, Barton, and Kershner eds. Practical approaches to riparian resource management. U.S. Bureau of Land Management, Billings, MT. 193 p.

Platts, W.S. 1991. Livestock Grazing. In: W.R. Meehan, Influences of forest and rangeland management on salmonid fishes and their habitats. American Fisheries Society Special Publication 19: 389-423.

Richardson, S.E., P.D. Sheerer, S.A. Miller, S.E. Jacobs, G. Swearingen, B. Berger, J. Deibner- Hanson, and J. Winkowski. 2010. Warner Sucker Investigations (2009). Oregon 55

Department of Fish and Wildlife, Fish Research Project HLP0630009 and 134206M093, Annual Progress Report, Salem.

Roberts, B. C., and R. G. White. 1992. Effects of angler wading on survival of trout eggs and pre- emergent fry. North American Journal of Fisheries Management. 12:450-459.

Scheerer, P.D. and S.E. Jacobs. 2005. Hutton Spring tui chub and Foskett Spring speckled dace investigations. Oregon Department of Fish and Wildlife, Salem, Oregon.

Scheerer, P. D., S. E. Jacobs, and A. V. Munhall. 2006. 2006 Warner Valley fish investigations- Warner suckers. Oregon Department of Fish and Wildlife, Fish Research Project Annual Progress Report, Corvallis.

Scheerer, P.D., S.E. Jacobs, and M.P. Heck. 2007. Warner sucker stream investigations. Oregon Department of Fish and Wildlife, Salem, Oregon.

Scheerer, P.D. and S.E. Jacobs. 2007. Hutton Spring tui chub and Foskett Spring speckled dace investigations. Oregon Department of Fish and Wildlife, Salem, Oregon.

Scheerer, P.D., and S.E. Jacobs, K. Bratcher, G. Swearingen, and S. Kramer. 2008. Warner Valley fish investigations - Warner suckers, annual progress report, fish research project, Oregon Department of Fish and Wildlife, Corvallis, Or.

Scheerer, P.D. and S.E. Jacobs. 2009. 2009 Foskett Spring speckled dace investigations. USFWS cooperative agreement 13420-080J814, Annual Progress Report, Oregon Department of Fish and Wildlife, Salem, Oregon.

Scheerer, P. D. 2011. 2011 Foskett Spring speckled dace investigations. USFWS cooperative agreement 13420-080J814, Annual Progress Report, Corvallis, Oregon.

Scheerer, P. D., S. E. Jacobs, M. Terwilliger, S. A. Miller, S. Gunckel, S. E. Richardson, and M. Heck. 2011. Status, distribution, and life history investigations of Warner suckers, 2006- 2010. Oregon Department of Fish and Wildlife, Information Report #2011-02, Salem.

Scheerer, P. D., S. Clements, and J. T. Peterson. 2012. 2012 Foskett Spring speckled dace investigations. USFWS cooperative agreement F11AC00095, Annual Progress Report, Corvallis, Oregon.

56

Scheerer, P. D., J. T. Peterson, and S. Clements. 2013. 2013 Foskett speckled dace investigations. USFWS cooperative agreement F12AC01064, Annual Progress Report, Corvallis, Oregon.

Scheerer, P. D., J. T. Peterson, and S. Clements. 2014. 2014 Foskett speckled dace Investigations. USFWS cooperative agreement F13AC00622 and BLM L12AC20619, Annual Progress Report, Corvallis, Oregon.

Scheerer, P. D., S. Clements, and J. T. Peterson. 2012. 2012 Warner Valley fish investigations - Warner suckers, annual progress report, fish research project, Oregon Department of Fish and Wildlife, Corvallis, Or.

Scoppettone, G. and P. Rissler. 2003. Information on status, demographics, and life history of five populations of tui chub in the Cow Head Lake Slough basin. U.S. Geological Survey, Biological Resource Division, Reno. Contract number 1448-13420-3-N348

SERA (Syracuse Environmental Research Associates) 2004a. Syracuse Environmental Research Associates, Inc. (SERA). 2004. Chlorsulfuron – Human Health and Ecological Risk Assessment Final Report. SERA TR 04‐43‐18‐01c. Prepared for the U.S. Department of Agriculture Forest Service, Arlington, Virginia. Fayetteville, New York

SERA 2004b. Syracuse Environmental Research Associates, Inc. (SERA). 2004. Clopyralid (Transline) – Human Health and Ecological Risk Assessment Final Report. SERA TR 04 43‐17‐03c. Prepared for the U.S. Department of Agriculture Forest Service, Arlington, Virginia. Fayetteville, New York

SERA 2004c. Syracuse Environmental Research Associates, Inc. (SERA). 2004. Imazapic Plateau and Plateau Dg – Human Health and Ecological Risk Assessment Final Report. SERA TR 04‐43‐ 17‐04b. Prepared for the U.S. Department of Agriculture Forest Service, Arlington, Virginia. Fayetteville, New York

SERA 2004d. Syracuse Environmental Research Associates, Inc. (SERA). 2004d. Imazapyr – Human Health and Ecological Risk Assessment Final Report. SERA TR 04‐43‐17‐05b. Prepared for the U.S. Department of Agriculture Forest Service, Arlington, Virginia. Fayetteville, New York.

SERA 2011a. Syracuse Environmental Research Associates, Inc. (SERA). 2011. Glyphosate – Human Health and Ecological Risk Assessment Final Report. SERA TR 052‐22‐03b. Prepared for the U.S. Department of Agriculture Forest Service, Atlanta, Georgia. 57

SERA 2011b. Syracuse Environmental Research Associates, Inc. (SERA). 2011. Imazapyr– Human Health and Ecological Risk Assessment Final Report. SERA TR 052‐29‐03b. Prepared for the U.S. Department of Agriculture Forest Service, Atlanta, Georgia.

SERA 2011c Syracuse Environmental Research Associates, Inc. (SERA). 2011. Picloram– Human Health and Ecological Risk Assessment Final Report. SERA TR 052‐27‐03b. Prepared for the U.S. Department of Agriculture Forest Service, Atlanta, Georgia.

SERA 2011d. Syracuse Environmental Research Associates, Inc. (SERA). 2011. Triclopyr. Human Health and Ecological Risk Assessment Final Report. SERA TR 052‐25‐03b. Prepared for the U.S. Department of Agriculture Forest Service, Atlanta, Georgia.

Sidlauskas, B.L. and K. Hoekzema. 2014. Molecular phylogenetics and microsatellite analysis reveal cryptic species of speckled dace (Cyprinidae: Rhinichthys osculus) in Oregon’s Great Basin. Molecular Phylogenetics and Evolution 77 (2014) 238–250

Sigler, W.F. and J.W. Sigler. 1987. Fishes of the great basin. University of Nevada press. Reno, Nevada. 425 pp.

Smith, G. R. 1966. Distribution and evolution of the North American catostomid fishes of the subgenus Pantosteus, genus Catostomus. Miscellaneous Publications of Zoology. University of Michigan 129:1-132.

Smith, G.R., T. Steinback, and G. Pampush. 1984. Distribution, foraging relationships and colony dynamics of the American white pelican (Pelecanus erythrorhynchos) in southern Oregon and northeastern California. Oregon Department of Fish and Wildlife Nongame Wildlife Program Technical Report 83-0-04.

Snyder, J.O. 1908. Relationships of the fish fauna of the lakes of southeastern Oregon. Bulletin of the U.S. Bureau of Fisheries vol. 27(1907):69-102.

Stern, M.A. 1989. Draft report Water diversion structures in the Warner Basin. The Nature Conservancy, Oregon Natural Heritage Program. Portland, Or.

Stehr, Carla M., T. L. Linbo, D. H. Baldwin, N. L. Scholz, and J. P. Incardona. 2009. Evaluating the Effects of Forestry Herbicides on Fish Development Using Rapid Phenotypic Screens. North American Journal of Fisheries Management 29:975–984.

58

Stenstrom, M.K., and M. Kayhanian. 2005. “First Flush Phenomenon.” Final report prepared for the California Department of Transportation, Report No. CTSW-RT-05-73-02.6, Sacramento, CA.

Swenson, S.C. 1978. Report of investigations on Catostomus warnerensis during spring 1978. Unpublished report to Fish and Wildlife Service. 27 pp.

Tait, C.K. and E.J. Mulkey. 1993a. Assessment of biological and physical factors limiting distribution of stream-resident Warner suckers (Catostomus warnerensis). Unpublished report to Bureau of Land Management and Oregon Department of Fish and Wildlife. 35 pp. Tait, C.K. and E.J. Mulkey. 1993b. Estimation of stream-resident Warner sucker abundance and total habitat area in two basins using a statistically valid sampling design. Unpublished report to Bureau of Land Management and Oregon Department of Fish and Wildlife. 40 pp.

U.S. BLM (Bureau of Land Management). 2003. Lakeview Resource Management Plan and Record of Decision. Bureau of Land Management, Lakeview, Oregon. 186 pp.

U.S. BLM (Bureau of Land Management). 2007. Vegetation Treatments using Herbicides on Bureau of Land Management Lands in 17 Western States Record of Decision. Available at http://www.blm.gov/wo/st/en/prog/more/veg_eis.html

U.S. BLM. (Bureau of Land Management). 2010. Vegetation Treatments using Herbicides on BLM Lands in Oregon Final Environmental Impact Statement. http://www.blm.gov/or/plans/vegtreatmentseis/

U.S. BLM (Bureau of Land Management). 2015. Biological Assessment for Integrated Pest Management Lakeview Resource Area Lakeview District BLM.

U.S. BLM (Bureau of Land Management). 2015a. Integrated invasive plant management for the Lakeview Resource Area environmental assessment (DOI-BLM- OR-L050-2014-0021- EA).

U.S. Fish and Wildlife Service. 1985a. Endangered and threatened wildlife and plants; Determination that the Warner Sucker is a threatened species and designation of critical habitat. Federal Register 50(188):39117-39123.

59

U.S. Fish and Wildlife Service. 1985b. Endangered and threatened wildlife and plants: Determination of threatened status for Hutton tui chub and Foskett speckled dace. Federal Register 50(60):12302-12306.

U.S. Fish and Wildlife Service. 1986. Recovery plan for the Pacific bald eagle. Portland, Oregon.

U.S. Fish and Wildlife Service. 1998. Recovery plan for the native fishes of the Warner Basin and Alkali Subbasin. Portland, Oregon. OR. 86pp. http://ecos.fws.gov/docs/recovery_plans/1998/980427.pdf

U.S. Fish and Wildlife Service 2015. 2015 Foskett speckled dace (Rhinichthys osculus ssp.) 5- year review: summary and evaluation. Portland, Oregon.

U.S. Forest Service. 1997. Biological assessment grazing and associated activities affecting listed suckers within four watersheds.

U.S. Forest Service. 2006. 2,4‐D human health and ecological risk assessment final report. Forest Health Protection, USDA‐Forest Service. Arlington, Virginia.

U.S. Forest Service. 2007. Biological assessment for the potential effects of managing the Fremont-Winema National Forests in the Warner Lakes section 7 watershed on Warner sucker and its designated critical habitat. Lakeview, OR.

U.S. Fish and Wildlife Service. 2013. Aquatic Restoration Biological Opinion II (ARBO II). Portland, Oregon.

White, R.K., T.R. Hoitsma, M.A. Stern, and A.V. Munhall. 1990. Final report on investigations of the range and status of the Warner sucker, Catostomus warnerensis, during Spring and Summer 1990. Unpublished report to Bureau of Land Management, Oregon Department of Fish and Wildlife, and Fish and Wildlife Service. 66 pp.

White, R.K., T.L. Ramsey, M.A. Stern, and A.V. Munhall. 1991. Salvage operations and investigations of the range and stream habitat characteristics of the Warner sucker, Catostomus warnerensis, during spring and summer 1991. Unpublished report to Bureau of Land Management and Oregon Department of Fish and Wildlife. 44 pp.

Williams, J.E., M.A. Stern, A.V. Munhall, and G.A. Anderson. 1990. Conservation status of the threatened fishes of the Warner Basin, Oregon. Great Basin Naturalist 50(3):243-248. 60

Wood, C. 1995. Life history variation and population structure in sockeye salmon. Pages 195- 216 in J.L. Nielsen, Editor. Evolution and the aquatic ecosystem. American Fisheries Society symposium 17. American Fisheries Society, Bethesda, Maryland.

Wooton, R.J. 1990. Ecology of Teleost Fishes. Chapman and Hall, London, England. 404 pp.

Pers. Comm.

Scheerer P.D. 2015. Email Nov. 30, 2015 to A. Mauer re: Foskett speckled dace population estimate for 2015.

Internet reference: http://hypertextbook.com/facts/2007/EvanKaplan.shtml. Accessed Dec. 9, 20115 at 10:15 AM.