Environmental Assessment Suppression of HWA Infestations, DBNF

United States Department of Agriculture Environmental

Forest Service Assessment February 2011

Suppression of Hemlock Woolly Adelgid Infestations

Daniel Boone National Forest

In Kentucky counties: Bath, Clay, Estill, Harlan, Jackson, Knox, Laurel, Lee, Leslie, McCreary, Menifee, Morgan, Owsley, Perry, Powell, Pulaski, Rockcastle, Rowan, Wayne, Whitley, Wolfe

Responsible Official: LIZ AGPAOA Regional Forester 1720 Peachtree Road, NW Atlanta, GA 30309-2449

For Information Contact: Amos Stone Forest Silviculturist Daniel Boone National Forest 1700 Bypass Road Winchester, KY 40391 (859) 745-3122

The U.S. Department of Agriculture (USDA) prohibits discrimination in all its programs and activities based on race, color, national origin, gender, religion, age, disability, political beliefs, sexual orientation, or marital or family status. (Not all prohibited bases apply to all programs.) Persons with disabilities who require alternative means for communication of program information (Braille, large print, audiotape, etc.) should contact USDA's TARGET Center at (202) 720-2600 (voice and TDD). To file a complaint of discrimination, write USDA, Director, Office of Civil Rights, Room 326-W, Whitten Building, 14th and Independence Avenue, SW, Washington, DC 20250-9410 or call (202) 720-5964 (voice and TDD). USDA is an equal opportunity provider and employer. Environmental Assessment Suppression of HWA Infestations, DBNF

Table of Contents

1. Purpose & Need for Action ...... 5 1.1. Proposed Action ...... 5 1.2. Purpose and Need for the Proposed Action ...... 9 1.3. Project Objectives ...... 12 1.4. Decision to be made ...... 12 1.5. Public Involvement...... 13 1.6. Issues related to the Proposed Action ...... 13 1.7. Issues dismissed from further consideration ...... 14 2. Alternatives Including the Proposed Action ...... 16 2.1. Alternatives Considered in Detail ...... 16 2.2. Alternatives Considered But Not Evaluated In Detail ...... 18 2.3 Comparison of Alternatives Considered in Detail ...... 20 3. Affected Environment and Environmental Consequences ...... 22 3.1. Resources considered and unaffected by the alternative actions ...... 22 3.2. Hemlock Forests ...... 23 3.3. Management Indicator Species (MIS) ...... 26 3.4. Aquatic Habitat ...... 30 3.5 Proposed, Endangered, Threatened and Sensitive (PETS) Species ...... 33 3.6 Federally Designated Critical Habitat ...... 39 3.7 Wilderness Resources ...... 42 3.8 Human Health and Safety ...... 45 4. Preparers & Persons Consulted ...... 47 APPENDIX A – Hemlock Occurrence and Distribution ...... 48 APPENDIX B – Hemlock Conservation Strategy ...... 49 APPENDIX C – Wilderness Considerations ...... 53 APPENDIX D – Conservation Area Treatment Priority...... 57 APPENDIX E – Hemlock Conservation Area Maps ...... 62 APPENDIX F – Cost Estimates for HWA Suppression ...... 68 Works Cited ...... 70

Environmental Assessment Suppression of HWA Infestations, DBNF

BLANK Environmental Assessment Suppression of HWA Infestations, DBNF

1. Purpose & Need for Action 1.1. Proposed Action

The Daniel Boone National Forest (DBNF) is proposing treatments to suppress infestations of the non-native insect pest Hemlock Woolly Adelgid (Adelges tsugae) (HWA), which attacks and kills both of the hemlock tree species found in the eastern United States. While the rare Carolina hemlock (Tsuga caroliniana) exists only in the higher mountains of Tennessee, North Carolina, and Virginia; the eastern (or Canadian) hemlock (Tsuga canadensis) is found throughout much of eastern Canada and the United States, including the DBNF. Hereafter in this document, “hemlock” refers to eastern hemlock.

The proposed action includes biological as well as chemical control treatments, except in the two federally-designated Wildernesses. Seventy-four areas1 containing a significant component of hemlock (at least 30 percent stocking) would be designated as Hemlock Conservation Areas (HCA). These areas would be designated across the DBNF, generally within 3 to 5 miles of at least one adjacent HCA, to meet the hemlock conservation area design explained in Appendix B. HCA selection considered distance to nearest HCA, the stocking and area of adjacent stands containing hemlock, accessibility for treatment crews, and unique prescription areas identified in the DBNF Land and Resource Management Plan (Forest Plan) (USDA-FS 2004b). Also considered were additional hemlock stands identified from other sources as having important ecological and/or cultural values including Kentucky State Nature Preserve’s Natural Heritage sites. Other known hemlock areas were added as needed to fill gaps in the HCA design.

Insecticide treatment would occur to small groups of approximately 60 infested hemlocks within each HCA. For these groups of trees the treatment would normally be the systemic insecticide imidacloprid injected into the soil around the base of the tree. Injection would occur during the spring or fall season, where soils are moist but not saturated. Figure 1-1 depicts imidacloprid being injected into the soil at the base of an infested hemlock. Although more expensive, imidacloprid may also occasionally be used in pellet form in remote locations. Pellets would be inserted and covered 6-8 inches under soil at the same rate of active ingredient as in soil injection. The chemical is injected or inserted in evenly spaced sites, 6-18 inches from the base of the trunk.

Once successfully taken up by the tree, the insecticide is generally effective for at least two years. Treatments would be repeated after effectiveness declines if evidence of repeated infestation occurs. Treatment of hemlocks with a systemic insecticide would kill the adelgids feeding on those trees, allowing the trees to recover from the attack (Webb, Frank and Raupp 2003). Treatment would cease when effective biological control agents become established or the HWA threat is otherwise diminished, based on annual situation reports from Forest Health Protection (USDA-FS 2010a).

Biological control treatment would include the release of organisms that specifically target HWA, and that have little to no effect on other organisms. Biological control agents would meet

1 The maps in the scoping document correctly display seventy-four HCAs; whereas the scoping document incorrectly states ninety-four areas.

5 Environmental Assessment Suppression of HWA Infestations, DBNF

USDA risk assessment criteria for release (Hennessey 1995) (Salom, S. 1998) (Zilahi-Balogh, G.M.G. 2001) (Montgomery, et al. 1997) (Lu and Montgomery 2001) (Butin, et al. 2002). Current biological control agents for HWA include non-native predatory predator beetles Sasajiscymnus tsugae, Scymnus sinuanodulus, Scymnus ningshanensis, and Laricobius nigrinus (Figures 1-2, 1-3, 1-4, 1-5). L. nigrinus was discovered preying on HWA in the Pacific Northwest in 1934.

Biological control could occur anywhere on the DBNF where infestations are found, except in the two wildernesses on the forest: Clifty Wilderness located on the Cumberland Ranger District; and Beaver Creek Wilderness located on the Stearns ranger District. The two wildernesses on the DBNF each have large populations of hemlock. However, treatment of HWA in wilderness constitutes a trammeling action and therefore this proposal does not include treatment in wilderness.

Biocontrol of HWA within the 74 HCAs would be the priority. Each HCA would be prioritized annually for biocontrol based on criteria such as degree of HWA infestation and the health of the trees. Appendix D is a guide to help prioritize treatment areas, when resources are limited. The number of beetles released in an area would vary by species, ranging from several hundred to several thousand. The predator beetles are expected to take several years to establish widespread reproducing populations.

6 Environmental Assessment Suppression of HWA Infestations, DBNF

Figure 1-1. Imidacloprid injected into soil at base of infested hemlock

7 Environmental Assessment Suppression of HWA Infestations, DBNF

Figure 1-2. Sasajiscymnus tsugae attacking HWA Figure 1-3. Scymnus sinuanodulus on hemlock

Figure 1-5. Laricobius nigrinus larva Figure 1-4. Laricobius nigrinus adult attacking attacking HWA

8 Environmental Assessment Suppression of HWA Infestations, DBNF

1.2. Purpose and Need for the Proposed Action

1.2.1. HWA could potentially kill all Eastern Hemlocks across the DBNF over the next 10 to 20 years. The HWA has the potential to infest the entire range of eastern and Carolina hemlock in the next 30 years. Thirteen years of monitoring HWA in New Jersey indicate tree mortality began within 5 to 6 years after a stand became heavily infested, with more than 90% mortality in about 10 to 12 years (Mayer, et al. 2002).

Dispersal and movement of HWA occurs primarily during the spring. Eggs and crawlers (nymphs) are dispersed by birds, deer, and other forest-dwelling mammals (including humans) that come in contact with the sticky ovisacs and crawlers2. Winged adults are dispersed by wind. However, isolated infestations and long-distance movement of HWA most often occur as the result of people transporting infested nursery stock.

The HWA was first observed in western North America in 1924 in British Columbia and in the eastern United States near Richmond, Virginia in 1951. HWA was collected in Pennsylvania in 1969 and in Maryland in 1973. Since then it has established itself as a serious pest of hemlock and continues to spread throughout the range of hemlock in the eastern United States (Figure 1- 6).

HWA was first detected on the DBNF in February of 2008, in Clay County on the Redbird District of the DBNF. They may have been present but undetected for several years before this. At the time of this analysis, HWA is known to occur on all four Ranger Districts of the DBNF. Based on observations of the high rate of hemlock mortality in neighboring states including West Virginia, Virginia, North Carolina, and Tennessee, quick action is needed to initiate control measures to maintain a genetic representation of the eastern hemlock in Kentucky.

1.2.2. Action is needed because hemlock trees and associated plant communities are an important component of riparian and cove forest ecosystems. Forest Inventory and Analysis data (USDA-FS 2006a) indicated that eastern Kentucky contained over 72 million hemlocks, one inch or greater in diameter. Approximately 26 million hemlock trees, or about 36%, were on National Forest System (NFS) lands. NFS lands make up about 11% of the land area of eastern Kentucky. Therefore, hemlocks make up a larger share of the trees on NFS than on other ownerships in eastern Kentucky. In addition, a disproportionate share of the larger hemlocks (40% of those at least 20 inches in diameter) is on NFS land (Appendix A). Approximately 6% of all the trees in the DBNF are hemlocks (USDA-FS 2006a).

2 Many references which describe HWA and its lifecycle are available on the web (e.g. http://www.invasivespeciesinfo.gov/animals/hwa.shtml ).

9 Environmental Assessment Suppression of HWA Infestations, DBNF

Figure 1-6.

Some of the most notable examples of old growth hemlock in Kentucky occur on the DBNF, typically in poorly accessible drainages such as in Beaver Creek Wilderness. Loss of much if not all of the hemlock component could change the nature and characteristic of these areas.

1.2.3. Introduced non-native predator beetles would have a greater potential to control HWA than native beetles. In its native range in China and Japan, HWA causes little damage to hemlock trees since natural enemies and possible tree resistance have evolved with this insect pest. In the absence of these natural control elements in eastern North America, this introduced insect pest attacks hemlock causing damage and mortality within a few years.

Studies of HWA-consuming predator beetles native to the Eastern United States have not shown the capacity to eat enough HWA to reduce hemlock mortality effectively. In part, this may be due to life-cycles non-synchronous with the HWA life-cycle (USDA-FS 2006b). In contrast, predatory beetles from other areas have demonstrated the potential to reduce HWA populations significantly. These beetles were collected from China and Japan where HWA originated, and from the Pacific Northwest where western hemlock has survived HWA infestation. These

10 Environmental Assessment Suppression of HWA Infestations, DBNF

species demonstrate a life-cycle more synchronous with and more dependent upon HWA (Cheah, et al. 2004).

Studies have shown that the beetles are able to reduce HWA populations on release trees and adjacent trees, reproduce in the wild, overwinter, and disperse to new areas. Some infested hemlocks, especially those on productive sites, show recovery when HWA numbers decline (Cheah and McClure 2002).

The purpose of the introduction of a suite of biological controls would be to slow expansion of current infestations of HWA and to establish reproducing populations of these predators to eventually come into a balance with HWA populations. Long-term, the hope is that these beetles would gain a permanent foothold in the forest, continue to eat large quantities of HWA, and thereby help bring about an overall reduction in hemlock death.

1.2.4. Soil-injected insecticide would ensure survival of individual trees. Even if all environmental impacts could be avoided, it is not feasible to individually chemically treat all of the millions of hemlock trees across the DBNF. Therefore, we are proposing limited use of insecticide, as well as the establishment of biological agents to control HWA populations and reduce long-term impacts to hemlock.

Separate trials demonstrated that soil injection of the insecticide imidacloprid was 99.9% and >98% effective in eliminating HWA from individual trees (Steward and Horner 1994) (Steward, Braness and Gill 1998). One study (Webb, Frank and Raupp 2003) found that hemlocks recovered dramatically with new growth once HWA were reduced following imidacloprid treatment. While large scale or long-term use of this chemical would be impractical, short-term use would allow more trees to survive longer. Delaying mortality for several years could allow enough time for biological controls to take effect. Then these trees would be in good condition to contribute to the regeneration and restoration of the hemlock community. This would ensure that a broad gene pool of hemlocks remain alive in their native environment until biological control takes effect.

Stem injection of imidacloprid is currently regarded as less effective in controlling HWA and not as long lasting as soil injection. Repeated stem injection can also damage the tree itself. It would be a tool for trees in situations where soil injection is inappropriate such as in highly permeable soils.

1.2.5. Widespread hemlock death would degrade Aquatic habitats. Hemlock is common in riparian areas and plays an important role in maintaining cool stream water temperatures, preventing erosion on steep banks, and providing shelter to wildlife by moderating temperatures in cold winter months and during hot summer days. The average number of aquatic macro invertebrate taxa found in hemlock-bordered streams was 37% greater than that found in hardwood-bordered streams (Evans 2002). The implication is that widespread hemlock loss could result in a loss of species diversity and changes in species distribution.

1.2.6. Widespread hemlock death would reduce habitat for Neotropical migratory birds. Hemlock forests increase bird diversity at a landscape scale (Ross, Redell and Bennett 2002).

11 Environmental Assessment Suppression of HWA Infestations, DBNF

While bird species richness is typically lower in hemlock forests than hardwood forests, several species common in hemlock forests are rare in hardwood forests. Birds common in hemlock forests and that use hemlock for breeding habitat include the Acadian flycatcher (Dendroica fusca), and the black-throated green warbler (Dendroica virens). Populations of these species might decline as hemlock forests decline (Ross, Redell and Bennett 2002).

1.2.7. Widespread hemlock death would result in a loss of scenic and other aesthetic values, and would degrade some recreation experiences. Hemlock trees are an outstanding component of the cove and riparian forest scenery in Eastern Kentucky. In addition to providing a year-round scenic backdrop for many viewpoints and a more scenic overall driving experience, they may enhance the scenery by blocking less pleasing areas, providing site and sound buffers between recreationists and distractions such as roads or other developments, and providing welcome shade for outdoor activities. Widespread hemlock loss would result in undesirable changes to many recreation experiences. The presence of dead hemlocks would also increase the danger of fallen snags, and of roads and trails being blocked by fallen trees, thus making some recreational activities more hazardous. 1.3. Project Objectives

1. Reduce hemlock mortality from HWA across the DBNF by establishing reproducing populations of predator beetles that feed on HWA. 2. Maintain a genetic representation of eastern hemlock across the DBNF. HCAs would be sources of genetic diversity and potential sources of natural regeneration for future hemlock restoration, should that become necessary. 3. Support survival of certain ecologically and culturally important groups of hemlock. The long-term goal of this proposal is to maintain reproducing populations of hemlock throughout its historical geographic range across the DBNF. This goal reflects requirements in the National Forest Management Act which gives the Forest Service statutory responsibility to maintain the diversity of tree species. December 1982 regulations (36 CFR 219.19) state that “habitat must be provided to support, at least, a minimum number of reproductive individuals and that habitat must be well-distributed so that those individuals can interact with others in the planning area”. Goal 2.3 of the Forest Plan is to “Reduce outbreak populations of invasive species, or eradicate isolated infestations of invasive species to prevent them from becoming established.” Forest Plan objective 1.1.E. states “Maintain existing white pine/ hemlock habitat associations in riparian and cove positions.” 1.4. Decision to be made

The decision to be made by the Responsible Official is: The responsible Forest Service official will decide: 1) whether to take no action, implement the proposal, or implement an alternative to the proposal; 2) whether the selected action is consistent with the Forest Plan or an amendment is needed; and 3) whether to prepare a “finding of no significant impact or an environmental impact statement.

12 Environmental Assessment Suppression of HWA Infestations, DBNF

1.5. Public Involvement

In June 2009, after internal scoping by a Forest Service interdisciplinary team, a scoping letter was sent to about 170 individuals and organizations on the DBNF project planning mailing list. The scoping letter was also posted to the DBNF website. A news release was distributed to over 70 radio stations and newspapers in eastern Kentucky.

Twelve responses were received. Most responses expressed support for the project, including the use of both beetle for release and cautious use of insecticide. Other respondents recommended a cautious approach for both beetle release and insecticide use. However, some responses, while supporting chemical treatment, were not supportive of using non-native beetles. One commenter recommended use of only one beetle species since “two [of the other beetle species mentioned in the scoping document] had no information regarding success outside of the laboratory ...” Other comments suggested monitoring of water quality during and following the period of chemical application. Other points of view included one organization that recommended more widespread use of insecticide until biological control was proven effective; and one organization recommended treatment within wildernesses. The Responsible Official approved which issues were relevant and which were not. An analysis of scoping comments is in the project record.

The proposed action, issues, and alternatives were made available for a 30-day public review period consistent with direction in 36 CFR 215. A legal notice was published in the Lexington Herald-Leader newspaper on 12/17/2010. This information was also posted to the DBNF website. Two comments were received; both were supportive of taking action.

After the responsible official makes a decision, a legal notice will be published in the paper-of- record and a copy of the decision will be sent to the individuals that responded during scoping. The decision (and EA) will also be posted on the DBNF web site at: http://www.fs.fed.us/r8/boone/planning/project/nepa_project.shtml?project=26551. 1.6. Issues related to the Proposed Action

1. Many of the Kentucky State Nature Preserve Commission (KSNPC) old-growth and high quality A, B or AB hemlock areas were not included in the HCAs and would not receive treatment resulting in a higher probability of hemlock death within these stands.

This issue was addressed by developing Alternative C which includes all the KSNPC recommended sites, including those within wildernesses.

2. Use of insecticide in riparian areas could contaminate streams. Those who commented cite concerns specific to the use of imidacloprid insecticide, when injected into soil around the base of the tree near streams. Respondents are concerned about the potential for impacts to aquatic invertebrates, and the potential for leaching into water. There is also a concern about insecticide leaching from needles that fall into streams.

13 Environmental Assessment Suppression of HWA Infestations, DBNF

The option of not using insecticide is addressed as an alternative considered, but was eliminated from further consideration. Water quality protection requirements are included in the two action alternatives where insecticide use is proposed.

3. Hemlocks in the two wildernesses will not be protected. Those who commented cite concerns that the two wildernesses were not included in the Proposed Action and a change in character could occur if not treated.

Alternative C was developed to include biological control treatment anywhere in the wildernesses, with priority given to the seven HCAs in the wildernesses. 1.7. Issues dismissed from further consideration

1. Released beetles might prey on non-targeted insects. History shows that the introduction of non-native species is not always a good practice, since unforeseen consequences have occurred. In this case, the USDA studied these beetles for years before approving them for release. These beetles are all extremely adelgid-specific in their nutritional requirements and require adelgid eggs to grow to maturity. They will occasionally consume adelgids other than HWA in laboratory tests. They have not shown any ability to survive by eating adelgids that are considered an important food source for any species at risk.

2. Released beetles could become a human pest. The multicolored Asian lady beetle, Harmonia axyridis, was introduced into the US to serve as a biological control agent for arboreal insect pests and has now become well established in many parts of the US. Although H. axyridis has proven to be highly effective at controlling aphids and other soft-bodied insect pests in many agricultural systems, it is recognized as a serious nuisance pest due to its habit of aggregating in very large numbers on and in houses and building while searching for viable overwintering sites. In contrast, the tiny predator beetles that feed on HWA are of different species which are not gregarious, and are not attracted to the warmth or color of human dwellings in winter.

3. Only the Sasajiscymnus tsugae beetle should be used since the others are not proven to be successful. All three beetle species proposed for use are approved by the USDA and have been shown to be effective in reducing HWA populations. The use of a suite of predators that have slightly different development periods and cycles should be more effective and is likely the best strategy to match the HWA during its life cycles.

4. The insecticide imidacloprid is strongly associated with Colony Collapse Disorder (CCD) of bees and could also adversely affect native insects. CCD has occurred in broadcast foliar spray operations in cropland conditions. However, isolated soil injections of small amounts of insecticide would pose little risk to bees or other insects which don’t feed on the treated hemlocks.

5. Soil injection of imidacloprid has been shown to be highly toxic to earthworms and other soil invertebrate communities; repeated application could change the soil community. Trunk injection may be a better option on some sites. Chemical soil treatment would have some localized impacts on terrestrial invertebrates near the treated trees. Imidacloprid is highly toxic to terrestrial invertebrates that directly contact it, but exhibited low toxicity to other beneficial

14 Environmental Assessment Suppression of HWA Infestations, DBNF

invertebrates such as ants and spiders (Kunkel, Held and Potter 1999). Chemical treatments would result in some localized mortality of terrestrial invertebrates, but recovery from individuals moving in from adjacent untreated areas would be expected as the chemical breaks down. However, the small area of soil treated around approximately 7000 trees would harm a relatively small portion of the invertebrate population for a short period of time. Once the insecticide has moved into the tree and/or degraded in the soil, invertebrates will eventually reoccupy that soil.

Trunk injection of imidacloprid could be used to protect hemlock immediately adjacent to streams. However, this method necessitates that crews carry extra equipment that is infrequently used, and causes mechanical wounding to the tree. This method has also proven to be less effective since the chemical often only partially protects portions of the tree. Although this method can be useful in certain urban forestry situations, few forest managers are using this method extensively.

6. Lack of control measures in large hemlocks could result in greater potential for unwanted invasive species. Although it is hoped that beetle release should eventually protect all size- classes of hemlock, invasive species may become a problem in some locations, as they are in many areas of the Forest. The issue of invasive plant species is being addressed through additional project planning and is out of the scope of this project.

7. Long-term treatment using insecticides could result in unacceptable levels of toxic chemicals being added into the ecosystem. Field studies show that imidacloprid can persist in soil, with a half-life ranging from 27 to 229 days (Miles Inc. 1993). Since Imidacloprid insecticide could degrade (worst case) to 30% of initial potency within two years in soil, it could accumulate, in the small, widely spaced treatment areas, although uptake by vegetation would reduce this effect. Even with the potential of imidacloprid to persist in soil, the high photodegradation tendency and high water solubility indicate there is low potential for bioaccumulation in the environment (Bacey 2000). In addition, studies have shown that imidacloprid had relatively little impact on most groups of beneficial invertebrates (Kunkel, Held and Potter 1999).

8. Soil injection should be the first line of defense in order to provide trees with the best chance of survival until mass rearing capabilities raise beetle release numbers to effective levels. Soil injection of insecticide will be “the first line of defense” in non-wilderness HCA’s. We have been assured by the Forest Health Protection unit of the Forest Service that predatory beetles will be available for release as hemlock stands become sufficiently infested with the HWA. Beetles would not be released on or near infested trees treated with insecticide.

9. Future biological control measures not yet approved may be more effective. Other predator insects or pathogens that attack HWA may be shown to be effective that are currently unknown or unapproved. For example, an HWA-specific fungal pathogen may soon be field-tested in Pennsylvania. Mycotal is a commercial formulation of the insect-killing Lecanicillium muscarium (Verticillium lecanii) that is registered in seven European Union countries, Turkey and Japan. Importation is specifically being requested by a biocontrol company for further testing of Mycotal in small-scale trials (<10 acres total area) in PA that target the HWA. Once thorough testing and approval by all appropriate agencies has occurred, this biocontrol may

15 Environmental Assessment Suppression of HWA Infestations, DBNF

become available as an aerial spray for control of HWA. If proven to be more effective, such control methods could supplement or replace this proposed action. Until then we can only use those methods that are approved.

10. Include insecticide treatment in wilderness to protect hemlock. This issue is addressed in an alternative not considered in detail in section 2.2.2 below. 2. Alternatives Including the Proposed Action

This chapter presents a detailed description of the alternatives, including implementation requirements and design criteria. These alternatives were developed by the interdisciplinary team in response to issues identified for this project. Six alternatives were considered.

Three of these alternatives were considered, but eliminated from detailed study. These include 1) an alternative that included the use of insecticidal soap; 2) an alternative that included use of chemical in wildernesses; and 3) an alternative that included biological treatment only.

Three alternatives are analyzed in detail: (A) No Action; (B) the Proposed Action; and (C) the Proposed Action plus 18 additional HCAs and biological control treatment in the two wildernesses. 2.1. Alternatives Considered in Detail 2.1.1. Alternative A – No Action This alternative proposes no activities to meet the objectives presented in Chapter 1. This alternative serves to provide a baseline by which to compare anticipated impacts from implementing suppression actions in alternatives B and C.

2.1.2. Alternative B – Proposed Action Release of Beetles to Establish Long-Term HWA Population Control This alternative is the proposed action also described in section 1.1 above. This alternative does not propose action in wildernesses. Beetle species Sasajiscymnus tsugae, Laricobius nigrinus, Scymnus sinuanodulus and/or Scymnus ningshanensis that prey on the HWA could be released in infested hemlock stands anywhere on the Forest, except in wildernesses. While large populations of hemlocks do exist in wilderness, wilderness populations were not proposed for treatment due to the trammeling nature of the action.

Seventy-four HCAs would have first priority for treatment when they become infested. The HCAs that would form the conservation network were selected from a list of outstanding hemlock areas including Natural Heritage sites, hemlock stands containing rare species, and additional hemlock areas identified through internal and external scoping as having important ecological and/or cultural values. In a few instances, hemlock stands not recognized as ecologically or culturally important were added to the network to fill a gap in the design.

The number of releases each year would depend on the available supply of beetles as well as the availability of employees and volunteers. Each year infested hemlock stands would be

16 Environmental Assessment Suppression of HWA Infestations, DBNF prioritized for beetle release with first priority being infested HCAs. If beetles are available for infestations other than in the HCA, then release priority would depend on degree of infestation, geographic distribution, and location of previous releases. The desire would be to release beetles in all areas that have trees sufficiently infested (showing evidence of adelgids at the base of a majority of observable needles). The number of beetles released in an area would vary by species according to established release protocols developed by Forest Health Protection and university researchers who study the insects. Current protocols call for several hundred to several thousand beetles to be released per area.

Specific infested hemlocks within the release areas would be evaluated as suitable for beetle survival. These would be trees that are infested with HWA to the degree that evidence of adelgids can be seen at most leaflet nodes. The trees themselves, as well as nearby trees, should still be healthy enough to be putting on new growth. The objective is to find a spot with enough HWA so the beetles can successfully feed and reproduce, and where other similarly infested hemlocks are nearby so it is possible for the beetles to disperse.

A representative sample of release areas would be monitored at six months and one year to determine if the beetles are still present, if they have successfully overwintered, and if and how far they have dispersed. The condition of the release trees would also be noted.

Chemical Treatment to Maintain Genetic Reserves At least 60 trees in each of seventy-four HCA would be selected for treatment with insecticide as they become infested. For these groups of trees the insecticide imidacloprid would be injected into the soil at the base of the tree (soil injection), the insecticide is then drawn into the roots and up to the needles. Larger trees require greater amounts of insecticide. Treatment would be repeated when evidence of new infestation occurs; soil injection is generally effective for at least two years. Although more expensive, imidacloprid slow-release pellets may be used in remote locations.

Treatment would cease when effective biological control agents become established or the HWA threat is otherwise diminished, based on annual situation reports from Forest Health Protection (USDA-FS 2010a).

Implementation Requirements and Design Criteria: 1. Chemical treatment would not occur on unsuitable sites including areas with high water table, saturated soils, or highly permeable sandy or gravelly soils. a. Soil would be sampled to determine the presence of sandy or gravelly (highly permeable) soils. The presence of highly permeable soils would disqualify the site for soil injection. b. The area would be scouted for the presence of any surface water or water bodies (springs, creeks, ponds, bogs, etc.). Any tree with a direct vegetative connection to surface water would be eliminated from soil injection treatment. 2. Mixing and/or equipment cleaning would occur at least 100 feet from a water body. 3. Water from a public water source is preferred, although clean creek water may be used. When creek water is used, a container labeled “Water only” must be used to carry water to

17 Environmental Assessment Suppression of HWA Infestations, DBNF

the mixing container. Extreme care would be taken to prevent chemical contamination of the “Water Only” container.

4. A certified applicator would supervise application of the insecticide, and all label and material safety data sheet requirements would be followed. Figure 2-1 lists recommended directions for soil injection of the insecticide (Silcox 2002).

Figure 2-1. Kioritz® Injector Directions for Use

1) Dilute 2 ozs. of Merit® 75WP in 30 to 60 fluid ozs. of water 2) Place up to 3 qts. of this dilution into the Kioritz® reservoir 3) Use the high setting on the Kioritz (5 = 5 ml per stroke) 4) Apply 1 fluid oz./in. of tree diameter at breast height (Six strokes deliver approximately 1 fluid oz.) 5) Use 1 to 2 injection sites per inch of tree diameter at breast height (Use a minimum of 4 injection sites per plant)

2.1.3. Alternative C – Proposed Action Plus Additional HCAs and Wilderness Treatment Alternative C would include those activities and Implementation Requirements and Design Criteria described in Alternative B, more HCA would be added, and beetle release could occur in infestations within the two wildernesses.

Eighteen hemlock conservation areas (HCA) would be added to the originally proposed seventy- four HCAs (totaling 92) at the locations of the Kentucky Nature Preserves Commission Natural Heritage Sites. Seven of these HCA would be added to wilderness: five HCA would be added to Clifty Wilderness; and two HCA added to Beaver Creek Wilderness. The wilderness HCA would be priority biological control areas; however no chemical treatment would occur in the wilderness HCA. See Appendix C for prioritization and mapped locations of HCA.

Treatment of HWA in wilderness would be accomplished to protect the natural qualities of the wilderness. Hemlocks constitute a significant component of both of the wildernesses and the loss of the hemlocks would negatively affect wilderness character. The trammeling action of releasing beetles may also negatively affect wilderness character but it is felt that this action would be the minimum tool for protecting the hemlock component. These additional HCAs will also be important to the continued success of the HCA design. 2.2. Alternatives Considered But Not Evaluated In Detail 2.2.1. Inclusion of Treatment by Spraying Insecticidal Soaps and Horticultural Oils Insecticidal soaps and horticultural oils can be sprayed on hemlocks when the objective is immediate knock down of an insect pest. If complete coverage is achieved, these agents act by smothering all invertebrates on the tree at the time of treatment. There is no residual effect, so HWA could reinfest the tree immediately. With this method there is an increased risk of

18 Environmental Assessment Suppression of HWA Infestations, DBNF

applicator contamination and increased concern with drift, since the product is sprayed. This treatment method is appropriate for smaller, more accessible trees that could be treated frequently. It would not be appropriate for treating large or inaccessible trees. It would not meet the project objective of keeping HWA suppressed for months or years, as would be necessary to ensure tree survival.

2.2.2. Inclusion of Chemical Treatments in Wildernesses The objective of any treatment in wilderness would be to protect the character of wilderness that is threatened by loss of hemlock, which is a large component of both wildernesses (appendix A, Table A1).

Chemical treatment of a small group of trees in a stand of hemlock (an HCA) is intended to sustain a small genetic representation, and is not intended to protect all hemlocks in a stand. The purpose of HCA in wilderness would be to designate priority areas for monitoring and for beetle release if and when HWA infestation occurs.

Chemical treatment of a small number of hemlocks in a wilderness HCA would not meet the objective of protecting the character of wilderness. Additionally, chemical treatment of large numbers of hemlocks on a bi-annual schedule, in remote wilderness locations, would be labor- intensive and cost prohibitive.

Lastly, chemical treatment requires that temporary signs be posted at each treatment site. Treated trees would be marked with paint or metal tags for retreatment every 2-3 years. Chemical treatments, and associated tree marking would result in a diminishing of the “naturalness” of the wilderness with the direct intervention of technology.

2.2.3. Biological Treatment Only Selected populations of hemlock trees would be treated across the DBNF, including wildernesses, by releasing predator beetles only. This alternative was considered in response to concerns raised during public scoping related to the use of insecticides.

In the forest, many factors conspire to affect HWA and predator populations such as fluctuating winter warm and cold spells and minimum low temperatures. Other predators including native and non-native species can also affect HWA-specific predator survival.

This alternative was not considered in detail because the quantity of beetles needed to control the adelgid is extremely large. Beetle production could not be assured and the demand for these beetles spans the range of infestation across numerous jurisdictions all competing for beetles. Survival of selected trees could not be assured with initial releases of predatory beetles. It is essential to the goal of genetic conservation that trees selected for genetic linkage remain in a healthy growing condition.

19 Environmental Assessment Suppression of HWA Infestations, DBNF

2.3 Comparison of Alternatives Considered in Detail

Table 2-1 summarizes the management activities for each alternative considered in detail. See Appendix D for a table listing specific areas, and Appendix E for maps showing the approximate locations of HCAs.

Table 2-1. Activity Summary for Alternatives Considered in Detail

Activities (in infested hemlock stands) Alt. A Alt. B Alt. C Biological control (using 4 approved beetle species) NO YES YES forest-wide, outside wildernesses Insecticide treatment of groups of infested hemlocks in Hemlock Conservation Areas, outside NO YES YES wildernesses Biological control (using 4 approved beetle species), NO NO YES within wildernesses

Hemlock Conservation areas potentially having 0 74 85 chemical treatments (HCA outside of wildernesses)

Total number of Hemlock Conservation Areas (HCA) 0 74 92

Table 2-2 is a summary of the Measures used to compare the analysis of alternatives done in chapter 3. We acknowledge that there is some uncertainty in making projections concerning the response of a non-native species and its predators coming into lands never before occupied. However, our estimates are based on current scientific knowledge of mortality and survival of HWA-infested hemlock and the effects of chemical and biological treatments in other locations in the eastern United States.

20 Environmental Assessment Suppression of HWA Infestations, DBNF

Table 2-2. Comparison of Alternatives Considered in Detail

Resource Measure Alternative A Alternative B Alternative C 1) Percent of hemlock trees 00% 40-60% 50-70% surviving by 2030. Hemlock Forest 2) Maintenance of adequate hemlock genetic NO YES YES representation.

1) Acres age 80 plus hemlock Management habitat available for beetle 00 14,600 acres 17,800 acres Indicator Species release in 2011. (black-throated green warbler & 2) Estimated acres of age 80 Acadian flycatcher) plus hemlock habitat 00 11,600 acres 14,100 acres remaining in 2030.

1) Miles of perennial streams within hemlock stands, 00 126 miles 151 miles potentially influenced by Aquatic Habitat beetle release. (includes macro- invertebrates) 2) Number of HCAs to be treated with insecticide, and 00 72 (of 74) 77 (of 85) containing perennial or intermittent streams. 1) Risk of death or injury to Proposed, N/A None None Endangered, PETS from predator beetles. Threatened and 2) Risk of death or injury to Hazard quotient Hazard quotient N/A Sensitive Species PETS from insecticide. < 1 < 1 Federally 1) Effects of Action on Designated Critical N/A None None Primary Constituent Elements Habitat 1) Naturalness (% of forest 1) 31% 1) 31% 1) 12 % types in wildernesses changing to other types.) 2) Unique components (age 2) 3863 2) 3863 2) 1545 Wilderness 100+ hemlock acres lost) Resources 3) Trammeling (Wildness) 3) no change 3) no change 3) Unnoticeable 4) Solitude 4) no change 4) no change 4) Possibility of encounter with workers, every 2 years 1) Snag Hazard (estimated 1) 30,200 1) 15,500 1) 12,400 acres of hemlock stands Human Health and containing HWA-created Safety snags) 2) Chemical exposure risk 2) none 2) slight 2) slight 3) Tree Climbing risk 3) none 3) slight 3) slight

21 Environmental Assessment Suppression of HWA Infestations, DBNF

3. Affected Environment and Environmental Consequences

The scope of this analysis is confined to those hemlock-dominated stands proposed for treatment on National Forest System lands of the DBNF. The period of time used in this analysis is up to 20 years. Analysis is also based on currently proposed treatments with the knowledge that newly discovered treatments could result in the better survival of hemlock. The following assumptions were made in order to evaluate effects:

1. The application of systemic insecticides would effectively protect treated trees for approximately two years; however repeated applications would be necessary until biological controls become effective.

2. Biological control agents (predator beetles) would disperse up to ¼ mile from release sites, thus covering approximately 125 acres.

3. Limited or no control of HWA would occur on privately-owned lands adjacent to or in close proximity to the DBNF.

Chapter 3 describes the environmental components of the area that could be affected by the alternatives under consideration. It provides the analytic basis for comparison of the alternatives, and describes direct, indirect, and cumulative impacts of the alternatives. Chapter 3 is organized around each potentially affected resource. 3.1. Resources considered and unaffected by the alternative actions

The following resources and conditions were considered, but detailed consideration did not occur because the alternatives are not expected to affect them: • National Parks. Big South Fork National River and Recreation Area and nearby state parks currently have no plans to treat HWA. There’s a remote chance that hemlock within these parks would benefit from release of beetles on DBNF lands. Chemical treatment would not impact parks; • Prime farmlands. The project area is riparian and forested hillsides with no farming activity; • Highways. No highways or roadways will be impacted; • Structures. No structures will be impacted; • Wetlands. No wetlands will be affected by the proposal; • Climate change. Since the forest will naturally regenerate and fill gaps created by death and decay of hemlock, carbon will again be sequestered by growth of other tree species. Variations between the alternatives would have minor influence on this process. Long-term warming of climate could negatively impact hemlock since it prefers cooler drainages at the south end of its range in Kentucky and Tennessee. Effects of changing climate on the HWA and its predators are not fully understood in the southern U.S. None of the actions would contribute greenhouse gas emissions.

22 Environmental Assessment Suppression of HWA Infestations, DBNF

• Environmental justice (Executive Order 12898). A full analysis of human health, economic and social effects on minority and low-income communities was not needed for this proposal. Scoping resulted in no comments relative to environmental justice concerns.

• Heritage resources. This proposed project has no potential for effect, adverse or beneficial, to a heritage resource, and therefore is an Exempt Undertaking; no further Section 106 compliance documentation is required. In June 2009, details of the project were shared with the Cherokee Nation, Eastern Band of Cherokee Indians, United Keetoowah Band of Cherokee, Shawnee Tribe, Eastern Shawnee, and Absentee-Shawnee Tribe of Indians. One comment from the United Keetoowah Band of Cherokee stating no objection was received in July 2009. In January 2011, the locations of additional treatment areas in alternative C were sent to the same tribes. No further comments were received. 3.2. Hemlock Forests

Affected Environment All forest stands containing hemlock trees on lands of the DBNF are in jeopardy of becoming infested by the spreading HWA population. The effect of the infestation would be generally proportional to the amount of hemlock within each individual stand. Hemlock trees are also often found in hardwood forest types which can contain up to 29% hemlock. Table 3-1 displays the four hemlock forest type classifications, and Appendix A lists the occurrence and distribution of hemlock types on the DBNF.

Table 3-1. Description of Hemlock Forest Types Forest Type (type code) Percent stocking within the Stand “Hemlock” (05) 70 - 100% hemlock “Hemlock – Hardwood” (08) 50 - 69% hemlock “Cove Hardwood – White Pine – Hemlock” (41) 10 - 49% hemlock “White Pine – Hemlock” (04) 50 - 100% white pine + hemlock; and at least 10% of each species

The existing condition of the hemlock resource becoming increasingly infested with HWA may be exacerbated by other factors which may contribute to the death of the hemlock trees. A secondary pest Fiorinia externa, an introduced scale insect, and the hemlock borer, Melanophila fulvoguttata can both cause additional stress. The longer or more times a stand is heavily infested with HWA and secondary pests, the greater the chance of tree mortality occurring. Tree mortality typically occurs 5 to 6 years after a stand has been heavily infested and surviving trees may begin to recover if HWA populations decline, but hemlocks are not likely to survive a second heavy infestation (Mayer, et al. 2002).

The loss of hemlock on the DBNF will affect associated plant species. Some species will benefit from the reduced competition for light, moisture, and nutrients while other species that thrive in deep shade may be harmed. As hemlock in the canopy dies and falls, dominance within the community would likely shift to opportunistic species already present on site. It is likely that

23 Environmental Assessment Suppression of HWA Infestations, DBNF species such as rhododendron (Rhododendron maximum), tulip-poplar (Liriodendron tulipifera), red maple (Acer rubrum) and white pine (Pinus strobus) would increase on moist eastern hemlock sites. Actual shifts in species composition would be based upon local site conditions and species present.

Still, some reduced vigor and mortality of shade-dependent forbs and non-vascular plants is likely where hemlock mortality is high. However, there are no plant species (e.g., non-vascular plants) that depend entirely upon hemlock for substrate, and there are no plant species that depend entirely on hemlock for shade.

The loss of hemlock would result in many changes for local wildlife populations. Short-term impacts would include drying out of the forest floor making areas less hospitable for salamander species, an increase in early successional habitat resulting in a flush of herbaceous and soft-mast forage, and an increase in snags for cavity dependent species. Long-term impacts would include replacement of hemlock with deciduous hardwoods and white pines, and an increase in the availability of down woody debris and some younger forest within the landscape. Species closely associated with cool, moist forested conditions would experience short-term declines, and potentially long-term loss of habitat depending upon the future forest conditions. Negative impacts of hemlock mortality due to HWA would be greatest on species with limited mobility that require hemlock forest for nesting or foraging habitat, such as invertebrates and land snails.

Indicators (On NFS lands on the DBNF) 1) Estimated percent of hemlock trees surviving by 2030

2) Maintenance of adequate hemlock genetic representation

Direct and Indirect Effects to Hemlock Forests (Alternative A) With no action, HWA infestation would continue to spread across the DBNF and potentially result in the loss of most if not all hemlock within ten to twenty years. Eventually, no local seed source would remain for propagation of hemlock as seedlings die from further infestation. Local plant and animal populations associated with hemlock forests may also suffer losses dependent upon their ability to adapt to the changing environmental conditions. Potential losses would be highest for those non-mobile species dependent upon the moist and shaded conditions typically found in eastern hemlock stands.

Cumulative Effects to Hemlock Forests (Alternative A) Since the effects in this alternative are the result of uncontrolled events and not associated with agency action, there would be no cumulative effects.

Direct and Indirect Effects to Hemlock Forests (Alternative B) Under this alternative, chemical control is proposed within 74 hemlock conservation areas3 (HCA) outside of wilderness across the DBNF; biological control is proposed anywhere infestation occurs, with the exception of the two wildernesses. Any effects to hemlock trees and

3 These 74 areas were shown on maps attached to the public scoping letter, and are also on maps in Appendix E.

24 Environmental Assessment Suppression of HWA Infestations, DBNF

associated plant species from the control treatments would be negligible as the insecticide imidacloprid is non-toxic to plants and the beetles do not feed on plants.

Timely repeated chemical treatment of hemlock trees within the 74 HCA is expected to protect these trees from the adelgid and preserve a genetic base for future artificial reforestation. If an average of about a two acre area of hemlock trees is treated with insecticide within each HCA, then about 148 acres of hemlock trees would be protected through insecticide treatment.

HWA infestation may eventually be controlled, although not eliminated, by the release of predator beetles if the beetles, HWA, and hemlock trees come into a balanced existence. By introducing the beetles as stands of hemlock become infested, especially at the leading edge of infestation, there may be enough time for this balance to occur.

With the biological control proposed it is expected that many well-distributed groves of hemlock would persist. However, unless beetles spread into wilderness from adjacent release areas, it is possible that all hemlocks in the two wildernesses could eventually become infested and die. Given no limitations in personnel (employees and volunteers) and no limitations in beetle availability, we estimate that 40-60% of hemlock trees on the DBNF could be alive by the year 2030, although at a reduced level of vigor. This is a very rough estimate based on current research of one beetle species in 22 localities from Georgia to Massachusetts (Mausel and etal. 2010). This assumes 100% mortality of hemlock in wildernesses.

Cumulative Effects to Hemlock Forests (Alternative B) A small amount of imidacloprid insecticide treatment has occurred to protect hemlocks in developed recreation areas on the DBNF. This periodic treatment will likely continue to occur as needed. However, there would be no overlap of insecticide treatment sites since HCAs are not coincident with these recreation areas.

Control of HWA on the DBNF would be affected by the condition of hemlock forests and HWA infestations on adjacent private and public lands. HWA treatments in neighboring forests could slow the spread of HWA into the DBNF; and treatments on the DBNF could slow the spread to adjacent ownerships. HWA suppression efforts on other public lands and a few private ownerships are already underway. Limited beetle releases as well as insecticide treatments began in Kentucky in 2006, and are likely to continue in the future. The Kentucky State Nature Preserves Commission, several State Parks, and the University of Kentucky in particular have implemented treatments on their properties. It is hoped that these actions would add to the effectiveness of this proposed action in maintaining hemlock community and genetic diversity across the landscape.

Direct and Indirect Effects to Hemlock Forests (Alternative C) Under this alternative, chemical control is proposed within 85 HCA outside of wilderness across the DBNF; biological control is proposed anywhere infestation occurs, including the two wildernesses. Seven HCA are identified in wildernesses as priority beetle release sites. Any effects to hemlock trees and associated plant species from the control treatments would be negligible as the insecticide imidacloprid is non-toxic to plants and the beetles do not feed on plants.

25 Environmental Assessment Suppression of HWA Infestations, DBNF

Timely repeated chemical treatment of hemlock trees within the 85 HCA would protect these trees from the adelgid and preserve a genetic base for future artificial reforestation. If an average of about a two acre area of hemlock trees is treated with insecticide within each HCA, then about 170 acres of hemlock trees would be protected through insecticide treatment.

With the biological control proposed it is expected that many well-distributed groves of hemlock would persist. Given no limitations in personnel (employees and volunteers) and no limitations in beetle availability, we estimate that (50-70%) of hemlock trees on the DBNF could be alive by the year 2030, although at a reduced level of vigor. This is a very rough estimate based on current research of one beetle species in 22 localities from Georgia to Massachusetts (Mausel and etal. 2010).

Cumulative Effects to Hemlock Forests (Alternative C) A small amount of imidacloprid insecticide treatment has occurred to protect hemlocks in developed recreation areas on the DBNF. This periodic treatment will likely continue to occur as needed. However, there would be no overlap of insecticide treatment sites since HCAs are not coincident with these recreation areas.

The analysis of effects to flora and fauna which are relative to this project follows the framework and uses MIS of the 2004 Final Environmental Impact Statement for the Land and Resource Management Plan (USDA-FS 2004a). Use of this framework is designed to ensure comprehensive consideration of project effects to fish, wildlife, and vegetation and their habitats. Treatments are limited to eastern hemlock-dominated stands and may occasionally include smaller hemlock-dominated pockets within mesic deciduous forests. Community assessments of aquatic macro-invertebrates are discussed in section 3.4, since they are used in the Forest Plan in lieu of specific aquatic MIS.

26 Environmental Assessment Suppression of HWA Infestations, DBNF

Affected Environment Fifteen (15) species are identified as management indicator species in the Forest Plan for the DBNF (Table 3-2). Thirteen (13) of these species will not receive detailed consideration due to a lack of species’ key habitat requirements within the project area (Table 3-3).

Table 3-2. Management Indicator Species Considered in Group Species Common Name further detail Mammal Odocoileus virginianus White-tailed deer No Bird Empidonax virescens Acadian flycatcher Yes Bird Dendroica virens Black-throated green warbler Yes Bird Dendroica cerulea Cerulean warbler No Bird Spizella passerina Chipping sparrow No Bird Pipilo erythrophthalmus Eastern towhee No Bird Spizella pusilla Field sparrow No Bird Colinus virginianus Northern bobwhite quail No Bird Cardinalis cardinalis Northern cardinal No Bird Seiurus aurocapillus Ovenbird No Bird Dendroica pinus Pine warbler No Bird Dendroica discolor Prairie warbler No Bird Piranga rubra Summer tanager No Bird Icteria virens Yellow breasted chat No Plant Pinus rigida Pitch pine No

Table 3-3. Forest MIS not considered in detail, with Reasoning

MIS Species Reason Classified in the Reasons Not Considered in Detail in this Analysis Name Forest Plan as an MIS White-tailed deer uses riparian areas containing hemlock, but is not dependent White-tailed Demand Species upon hemlock. Browse is often limited under highly shaded hemlock stands. deer Treatments would tend to maintain this hemlock shade. MIS for early successional Prairie Prairie warblers are not normally associated with conditions where eastern forest communities and warbler hemlock is dominant. habitat; yellow-pine Eastern Since only about 2000 acres of stands are classified as having greater than 70% MIS for early successional towhee & component of hemlock, the potential for early succession created by hemlock forest communities and yellow- loss would be minor (0.3%) at the forest level. Treatments would reduce this habitat breasted chat potential for increased early succession. Hemlock mortality could result in a minor expansion of the pine and pine oak MIS for pine and pine-oak communities where hemlock is growing off-site in uplands. Expansion of the Pine warbler communities; 70-90 ft2 pine and pine oak community type would provide additional habitat for Pine basal area (BA); >age 40 warblers allowing a few local populations to increase. Treatments would reduce this potential for increased pine and pine-oak habitat. There will be a temporary loss of habitat where old hemlocks die in stands having a minor component of old hardwoods. These forests will again move MIS for older forest interior toward older forest as other tree species become established and mature. No net Ovenbird conditions loss of older forest habitat expected over long-term. Treatments would reduce the potential for loss of some old trees. Hemlock forests are not preferred by the ovenbird. A slight expansion of such communities will likely occur when scattered upper- MIS for upland hardwood slope hemlock succumbs and stocking is reduced below 60 BA in some stands. Summer and oak-pine communities, Any expansion of this community type would provide additional habitat for tanager 30-60 ft2 BA, age > 50 tanagers, which would favor an increase in local populations. The establishment years of beetles could reduce this effect.

27 Environmental Assessment Suppression of HWA Infestations, DBNF

MIS Species Reason Classified in the Reasons Not Considered in Detail in this Analysis Name Forest Plan as an MIS A slight reduction of such communities will likely occur when scattered upper- slope hemlock succumbs and stocking is reduced below 60 BA in some stands. MIS for upland hardwood Any reduction of this community type would reduce breeding habitat for this Cerulean and oak-pine communities, bird, which could have a negative effect on populations. At the same time, warbler > 60 ft2 BA, age > 40 opened stands may provide additional foraging areas for a short period of time years potentially resulting in slight increase in local populations. Hemlock forests are not preferred by this warbler. MIS for the recovery of this Pitch pine populations are often found in xeric soils along rock cliffs and along Pitch Pine uncommon southern NW dry-mesic slopes. No notable impacts are expected. yellow-pine tree species MIS for upland hardwood Chipping or hardwood-yellow pine, sparrow & Hemlock mortality will not reduce upland hardwood stands below 30 BA. No <30 ft2 BA, age >50; Northern notable impacts are expected. cardinal (shrubs); sparrow cardinal (grass) Field sparrow MIS for woodland & & Northern Hemlock mortality will not reduce upland forests below 30 BA. No notable wooded grassland (quail) & bobwhite impacts are expected. grassland (sparrow) quail

MIS Considered in Detail Acadian flycatcher (Empidonax virescens) is a MIS which helps indicate management effects within the mature riparian forest community. Although not abundant, this bird is a widespread statewide summer resident within Kentucky (Palmer-Ball 1996). Acadian flycatchers require riparian forest within relatively large blocks of forest for breeding, and are not often found outside of these habitats during the breeding season. Its presence indicates riparian forests with relatively high levels of canopy cover and low levels of management disturbance, conditions required or preferred by many other riparian species. Acadian flycatchers are known to occupy many hemlock-dominated stands associated with streams on the DBNF.

Black-throated Green Warbler (Dendroica virens) is a MIS which helps indicate effects of forest structure within mature mesic deciduous cove forests (LaSorte, et al. 2007). Survey data on the DBNF indicate that this warbler reaches peak frequencies and densities in a wide variety of hardwood and eastern hemlock/white pine communities, particularly in older age classes.

Indicators: (Only NFS lands on the DBNF) 1) Acres of age 80 plus hemlock habitat available for beetle release.

2) Estimated acres of age 80 plus hemlock habitat remaining in 2030.

Direct and Indirect Effects to MIS (Alternative A) With the no-action alternative, Acadian flycatcher & black-throated green warbler habitat within older hemlock-dominated stands would slowly change as the existing hemlock forest dies from hemlock woolly adelgid. Local Acadian flycatcher & black-throated green warbler populations may experience local shifts in these stands as mature hemlock dominated forests are lost. Any nesting that is occurring within hemlock trees would also be lost over time.

28 Environmental Assessment Suppression of HWA Infestations, DBNF

Acadian flycatchers and black-throated green warblers would continue to utilize older deciduous forest and mixed deciduous-pine forests within the local range. The forest types that replace hemlock-dominated forests would eventually be suitable for these bird species upon maturity as these birds are not hemlock obligate species. Populations within adjacent mature cove hardwoods and mature cove hardwood white pine stands would persist ensuring the Acadian flycatcher and black-throated green warbler should not experience loss of local viability on the DBNF. Since hemlock-dominated & hardwood-hemlock forest older than age 80 represents 7.2% of all forest older than age 80 on the DBNF, there could temporarily be a minor decrease of local populations of these birds.

Cumulative Effects to MIS (Alternative A) Under this alternative there are no known past present or reasonably foreseeable actions that would make eighty year old and older riparian forests unavailable to black-throated green warbler and Acadian flycatchers during any point in their life cycle.

Direct and Indirect Effects to MIS (Alternative B) There would be no direct or indirect effects from predatory beetle release to black-throated green warblers and Acadian flycatchers. Although these birds may eat a small portion of released predatory beetles, this would not harm nor appreciably help the birds. Beetle release would have no effect on the birds’ normal diet since these beetles are specific predators of only HWA, and HWA are not the preferred or primary food of these birds.

Since imidacloprid will be applied through the soil and is taken up through the roots, these two species will not come into contact with the insecticide and there would be no direct effect. There would be no measurable effect from these two bird species bioaccumulating imidacloprid from feeding on adelgids that have fed on treated trees. Imidacloprid will be applied at a rate at which the hazard quotient4 is less than 1 for birds.

Under this alternative it is projected that there would be up to 14,600 acres of associated hemlock habitat treated and 11,600 acres available in twenty years.

Cumulative Effects to MIS (Alternative B) Under this alternative there are no known past present or reasonably foreseeable actions that would make older riparian forests on DBNF lands unavailable to black-throated green warbler and Acadian flycatchers during any point in their life cycle. Actions on other properties where beetles could be released could result in their spread onto the DBNF resulting in a slight increase in hemlock survival. Hemlock survival would promote long-term associated habitat for these two species. However, it is not possible to measure or predict the amount of increase due to unknown suppression efforts on other properties.

4 A measure of pesticide risk as rated by USDA (Anatra-Cordone and Durkin 2005), which meets Forest Plan standard DB-VEG-8.

29 Environmental Assessment Suppression of HWA Infestations, DBNF

Direct and Indirect Effects to MIS (Alternative C) The direct and indirect effects in this alternative are the same as Alternative B. The only difference is that there are an additional 4,250 acres of hemlock stands in wildernesses that are age 80 or older that could potentially be treated with predatory beetles.

Cumulative Effects to MIS (Alternative C) Under this alternative there are no known past present or reasonably foreseeable actions that would make older riparian forests on DBNF lands unavailable to black-throated green warbler and Acadian flycatchers during any point in their life cycle. Actions on other properties where beetles could be released could result in their spread onto the DBNF resulting in a slight increase in hemlock survival. Hemlock survival would promote long-term associated habitat for these two species. However, it is not possible to measure or predict the amount of increase due to unknown suppression efforts on other properties. 3.4. Aquatic Habitat

Environmental consequences to the aquatic habitat are evaluated based on impacts to aquatic macro-invertebrate communities (used in lieu of MIS), fish, and other aquatic organisms. Water quality is also discussed in relation to these organisms.

Affected Environment – Aquatic Habitat Aquatic habitats on the DBNF include large reservoirs, warm-water ponds, ephemeral streams, intermittent streams, and perennial streams with or without fish. All of these streams and water bodies likely support communities of macro-invertebrates. These organisms predominately derive their energy from organic material provided by riparian vegetation (e.g., leaves, sticks and large woody debris) (Pond and McMurray 2003). Abundance of aquatic macro-invertebrates is often influenced by riparian forests. An inventory of macro-invertebrates was made in Kentucky eight years ago. The dataset consisted of 106 wadeable sites containing 286 genera, and 92 headwater sites represented by 235 genera. The average number of organisms collected per sample event was 590 (Pond and McMurray 2003). Although many of these species have been found within streams under hemlock shade, it is not clear as to how directly dependent these species are on hemlock. Although stands containing varying amounts of hemlock may currently be present along some streams, other tree species often dominate riparian stands. Because of this, other trees and shrubs, such as rhododendron, may provide shade along many sections of the streams. There are currently 151 miles of streams that are influenced by hemlock stands.

Indicators (on DBNF lands) 1) Miles of perennial streams within hemlock stands, potentially influenced by beetle release.

2) Number of HCAs to be treated with insecticide, and containing perennial or intermittent streams.

30 Environmental Assessment Suppression of HWA Infestations, DBNF

Direct and Indirect Effects to Aquatic Habitat (Alternative A) There are no effects to aquatic habitat and macro-invertebrates from this alternative. If there is an increase in temperature due to the loss of hemlocks, water temperature would recover downstream as shade from hardwood as well as underground water percolation diminish the cooling effects of hemlock stands. Furthermore, the loss of hemlocks would allow for succession of other tree or shrub species within a relatively short amount of time as these opportunist species would provide shade. In larger streams and rivers, the overall amount of hemlock is relatively low in comparison to other tree species (Taylor 2010). In these systems, an increase in water temperature from the loss of hemlock would not be measurable.

It is expected that there will be an increase of organic materials from dead and dying hemlocks to include large woody debris (LWD) that will enter streams which generally benefit aquatic systems. Studies clearly demonstrate that increased habitat diversity results in more diverse, stable fish communities from LWD (USDA-FS 2004a). Aquatic macro-invertebrate communities will respond accordingly with potential species shifts favoring those that favor LWD. As with many groups of organisms, aquatic community dynamics have been proven cyclic and adaptable to surrounding conditions.

In this alternative, habitat for fish may likely be improved as LWD enters the system through hemlock tree mortality. The influx of LWD would be likely to allow for increases in aquatic invertebrate abundance that fish eat. Also the increase of LWD would provide habitat for many fish species and would likely result in measurable improvements in fish population levels-- especially for species whose habitat requirements are structurally-oriented (e.g. trout, bass). Such improvements would continue until LWD is decomposed or flushed downstream by high flows (Lisle 1986).

Long-term survival of the aquatic habitats and organisms would continue despite the loss of hemlock across the DBNF. Fish populations across the DBNF would continue to fluctuate depending on other environmental factors e.g. floods and droughts.

Water quality could be affected by increased biological oxygen demand (BOD) from large deposits of decomposing plant material. This material (fine organic matter) can potentially affect fish spawning success. In most cases, this fine organic matter is flushed downstream before problems with BOD reach problem levels (USDA-FS 2004a). Also, high water velocity, waterfalls, and cascades keep dissolved oxygen levels at good levels in headwater streams.

At the landscape level across the DBNF, long-term measurable effects on the aquatic community and habitat from this alternative may not be measurable, except locally within stands that may no longer contain hemlock and currently have a dense hemlock canopy with little riparian understory. In these locations, indirect effects would consist of shifts in aquatic invertebrate and fish communities as described above.

Cumulative Effects to Aquatic Habitat (Alternative A) In the no action alternative it can be expected that there will be an influx of LWD from dying hemlocks. There may also be local site specific spikes “fluctuations” in water quality and water temperature in areas that are influenced by pure hemlock stands. There are no known past

31 Environmental Assessment Suppression of HWA Infestations, DBNF

present or reasonably foreseeable actions that would contribute to LWD and shifts in water quality.

Direct & Indirect Effects to Aquatic Habitat - Alternatives B & C There would not be any changes in stream sedimentation from these alternatives. Therefore, effects are within thresholds estimated by the Forest Plan EIS (USDA-FS 2004a).

Aquatic habitat, including macro-invertebrates, will not be affected by predatory beetle release as these beetles will be released on hemlock trees that are infested with HWA. Predator beetles may occasionally fall into streams and be consumed by an array of aquatic organisms including macro-invertebrates. If the biological control within alternative C is completely successful, all 151 miles of aquatic habitat within hemlock stands would still have much of the beneficial integrity of the existing condition. In alternative B, which excludes biological treatment in the wildernesses, 126 miles of streams would benefit from treatment outside of wildernesses.

Imidacloprid is moderately toxic to fish (LC505 of 211 parts per million for rainbow trout) and highly toxic to aquatic invertebrates. However it has a half-life in water and sunlight of less than three (3) hours (Wamhoff and Schneider 1999) (Moza, et al. 1998) and is highly soluble in water (Bacey 2000). The chemical is moderately mobile in permeable soils. Pesticide label restrictions prohibit the use of the chemical where surface water is present. The label further states that use of the chemical where the water table is shallow may contaminate groundwater. Implementation Requirements and Design Criteria of sampling for highly permeable soils, and surveying for presence of water would greatly minimize or eliminate the potential for aquatic resources to be affected by imidacloprid. Mixing and transporting procedures for the chemical are designed to avoid any possibility of imidacloprid accidentally entering a stream. By following these criteria and label restrictions it is unlikely that there would be a negative effect to aquatic resources from the use of imidacloprid. In alternative B, all seventy-four HCAs would contain hemlock trees treated with insecticide; and seventy of these HCAs contain streams. In alternative C, eighty-five HCAs would be treated with insecticide; and seventy-nine of these contain streams.

In the unlikely event than an accident does occur during application that results in imidacloprid entering a stream, the chemical would be rapidly diluted to a level of no observable effect. Even so, local populations of fish would likely be affected. The local aquatic invertebrate community could be lost entirely. It is important to note that the loss of a local population of fish or one local aquatic invertebrate community will not affect overall species’ trends across the Forest. This is because the local stream reach that would be affected by an accidental spill would be quickly re-colonized by fish from upstream or downstream following rapid dilution. Invertebrate populations would recover more slowly.

Cumulative Effects to Aquatic Habitat (Alternatives B & C) There are no known past, present, or reasonably foreseeable actions within the HCAs that would impact aquatic habitat and macro-invertebrate communities.

5 LC50 = lethal concentration for 50% of those tested.

32 Environmental Assessment Suppression of HWA Infestations, DBNF

3.5 Proposed, Endangered, Threatened and Sensitive (PETS) Species 3.5.1. Affected Environment Federally Listed and Proposed Species (Endangered Species Act) Twenty-one (21) federally listed species were identified as potentially occurring on or adjacent to the DBNF (USDI-FWS 2010a). Nine (9) of those species will not receive detailed consideration as described below and as listed in Table 3-4. Two (2) species proposed for federal listing were also considered. On June 24, 2010, the FWS proposed to list the Cumberland darter (Etheostoma susanae) as endangered under the Endangered Species Act (USDI-FWS 2010b). The Cumberland darter was added to the DBNF list of species identified as potentially occurring on or adjacent to the DBNF (USDI-FWS 2010a). On November 2, 2010 the service proposed to list the Snuffbox (Epioblasma triquetra) as endangered under the Endangered Species Act (USDI-FWS 2010c). The snuffbox was added to the DBNF list of species identified as potentially occurring on or adjacent to the DBNF (USDI-FWS 2010a).

Table 3-4. DBNF Federally Listed and Proposed Threatened and Endangered Species Considered in Group Species Common Name Status further detail Mammal Myotis grisescens Gray bat E No Myotis sodalis Indiana bat E No Plecotus townsendii virginianus Virginia big-eared bat E No Fish Etheostoma percnurum Duskytail darter E Yes Notropis albizonatus Palezone shiner E Yes Chrosomus cumberlandensis Blackside dace T Yes

Etheostoma susanae Cumberland darter P Yes

Mussel Alasmidonta atropurpurea Cumberland elktoe E Yes Cyprogenia stegaria Fanshell E Yes Epioblasma brevidens Cumberlandian combshell E Yes Epioblasma capsaeformis Oyster mussel E Yes Epioblasma florentina walkeri Tan riffleshell E Yes Epioblasma torulosa rangiana Northern riffleshell E Yes Epioblasma triquetra Snuffbox P Yes Lampsilis abrupta Pink mucket E Yes Pegias fabula Little-wing pearlymussel E Yes Cumberland bean Villosa trablis E Yes pearlymussel Plant Arenaria cumberlandensis Cumberland sandwort E No Conradina verticillata Cumberland rosemary T No Schwalbea americana American chaffseed E No Solidago albopilosa White-haired goldenrod T No Spiraea virginiana Virginia spiraea T No Trifolium stoloniferum Running buffalo clover E No Status ‘E’ means the species is listed as ‘Endangered’ by USFWS. Status ‘T’ means the species is listed as ‘Threatened’ by USFWS. Status ‘P’ means the species is listed as ‘Proposed for Federal Listing’ by USFWS.

33 Environmental Assessment Suppression of HWA Infestations, DBNF

Federally Listed T&E Species Not Considered in Detail There are three federally listed bats on the DBNF. Of the three, Gray bat and Virginia big eared bat are cave obligate species. Indiana bats do use the general forest as summer habitat. Bats could possibly consume some of the released predatory beetles but they are not a primary food source as the majority of a bats diet is made up of nocturnal flying insects such as moths. Bats will not come into contact with the insecticide as imidacloprid will be injected into the soil and taken up through the roots.

Although hemlock is generally considered to represent poor summer habitat for many species of bats, the death of hemlock could increase potential summer roosting habitat. Many bats will use conifer snags for summer roosting habitat, and hemlock snags could provide good roosting habitat, especially since hemlock snags persist on the landscape for an extended period of time (USDI-FWS 2007). However, it is reasonable to believe that all infested hemlocks across the forest will not respond to treatment, thus some snags will still be created that are potential roost trees for a variety of bats.

Federally listed plant species would not be affected because plants are not affected by imidacloprid and/or the proposed biocontrol agents, since imidacloprid is not toxic to plants and predatory beetles are specific to HWA. The loss of hemlock on the DBNF may have an impact some plant populations. Some species will benefit from the reduced competition for light, moisture, and nutrients while other species that thrive in deep shade may be harmed. There are no plant species that depend entirely upon hemlock for substrate, e.g. non-vascular plants and there are no plant species that depend entirely on hemlock for shade. Cumberland sandwort grows in rock shelters (USDI-FWS 1996) while Cumberland rosemary and Virginia spirea live in stream cobble (USDI-FWS 1997a) (USDI-FWS 1997b). American chaffseed is associated with upland dry sites (USDI-FWS 1995) and the last known record is from the Stearns ranger district in 1933. Running buffalo clover is locally associated with the inner and outer bluegrass regions of Kentucky and prefers open sun (USDI-FWS 2007).

Regional Forester’s Sensitive Species Seventy-two (72) species are identified as sensitive by the Regional Forester (R8, Southern Region) or are candidate species and occur or are likely to occur on the DBNF (Table 3-5). Etheostoma sagita spilotum, the Kentucky arrow darter is listed as a candidate species at this time; therefore it is included in this list. Forty-one (41) of the seventy-two species in Table 3-5 did not receive detailed consideration as explained in the following section.

Table 3-5. Regional Forester’s Sensitive Species for the DBNF Considered in Group Species Common Name Detail Mammal Corynorhinus rafinesquii Rafinesque's big-eared bat No Myotis austroriparius Southeastern myotis No Myotis leibii Eastern small-footed bat No Sorex dispar blitchi Long-tailed shrew No Bird Aimophila aestivalis Bachman's sparrow No Falco peregrinus Peregrine Falcon No Thryomanes bewickii altus Appalachian Bewick's wren No Fish Ammocrypta clara Western sand darter Yes

34 Environmental Assessment Suppression of HWA Infestations, DBNF

Considered in Group Species Common Name Detail Ammocrypta pellucida Eastern sand darter Yes Etheostoma susanae Cumberland Johnny darter Yes Etheostoma cinereum Ashy darter Yes Etheostoma maculatum Spotted darter Yes Etheostoma sagitta spilotum Kentucky Arrow Darter Yes Etheostoma tippecanoe Tippecanoe darter Yes Ichthyomyzon greeleyi Mountain brook lamprey Yes Noturus stigmosus Northern madtom Yes Percina burtoni Blotchside logperch Yes Percina macrocephala Longhead darter Yes Percina squamata Olive darter Yes Typhlichthys subterraneus Southern cavefish Yes Mussel Anodontoides denigratus Cumberland papershell Yes Cumberlandia monodonta Spectaclecase Yes Epioblasma triquetra Snuffbox Yes Fusconaia subrotunda subrotunda Long-solid Yes Pleurobema oviforme Tennessee clubshell Yes Pleurobema rubrum Pyramid pigtoe Yes Ptychobranchus subtentum Fluted Kidneyshell Yes Quadrula cylindrica cylindrica Rabbitsfoot yes Simpsonaias ambigua Salamander mussel Yes Toxolasma lividus Purple lilliput Yes Plethobasus cyphyus Sheepnose Yes Gastropod Paravitrea placentula Glossy supercoil No Pleurocera curta Shortspire hornsnail No Rhodacme elatior Domed ancylid No Vertigo bollesiana Delicate vertigo No Vertigo clappi Cupped vertigo No Crustacean Cambarus bouchardi Big South Fork crayfish No Insect Cheumatopsyche helma Helma's net-spinning caddisfly No Manophylax butleri Cliffline caddisfly No Ophiogomphus howei Pygmy snaketail No Pyrgus wyandot Appalachian grizzled skipper No Speyeria diana Diana fritillary No Speyeria idalia Regal fritillary No Vascular Plant Aster saxicastellii Rockcastle aster No Aureolaria patula Spreading yellow false foxglove No Berberis canadensis American barberry No Carex juniperorum Juniper sedge No Cleistes bifaria Small spreading pogonia No Collinsonia verticillata Stoneroot No Cypripedium kentuckiense Kentucky Lady's slipper No Dodecatheon frenchii French's shooting star No Hexastylis contracta Mountain heartleaf No Juglans cinerea Butternut No Lesquerella globosa Short’s (Globe) bladderpod No Marshallia grandiflora Large-flowered Barbara's buttons No Monotropsis odorata Sweet pinesap No Paxistima canbyi Canby's mountain-lover No Platanthera integrilabia White fringeless orchid No Schisandra glabra Bay starvine No

35 Environmental Assessment Suppression of HWA Infestations, DBNF

Considered in Group Species Common Name Detail Scutellaria arguta Hairy skullcap No Scutellaria saxatilis Rock skullcap No Shortia galacifolia var. galacifolia Southern Oconee bells No Silene ovata Blue Ridge catchfly No Silene regia Royal catchfly No Thalictrum mirabile Little Mountain meadowrue No Thaspium pinnatifidum Cutleaved meadow parsnip No Vitis rupestris Sand grape No Nonvascular Hygrohypnum closteri Closter's brook-hypnum No Plant Plagiochila austinii A liverwort No Plagiochila sullivantii var sullivantii Sullivant's leafy liverwort No Radula sullivantii A liverwort No Scopelophila cataractae Agoyan cataract moss No

Regional Forester’s Sensitive Species Not Considered in Detail There are three Forest sensitive species of bats. Rafinesque’s big-eared bats, Southeastern Myotis and Eastern small-footed bats will not be affected by beetle release or chemical treatment of hemlock infested with hemlock woolly adelgid. This will not affect bats as there would be no contact with the insecticide since it is applied through soil treatments and taken up into the tree systemically. Also, the above listed bats feed on moths and other night time flying insects. Although hemlock is generally considered to represent poor summer habitat for many species of bats, the death of hemlock could increase potential summer roosting habitat. Many bats will use conifer snags for summer roosting habitat, and hemlock snags could provide good roosting habitat, especially since hemlock snags persist on the landscape for an extended period of time. However, it is reasonable to believe that all infested hemlocks across the forests will not be treated, thus some snags will still be created that are potential roost trees for a variety of bats.

The long-tailed shrew occurs in high elevation boulder fields. This habitat does not exist in the project area. Therefore, this species will not be considered in further detail.

Habitat supporting the Bachman’s sparrow, Peregrine falcon and Appalachian Bewick’s wren are not associated with hemlock stands in the project area; therefore, these species will not be considered in further detail.

Due to the limited range and lack of supporting habitat and the fact that none of these species are specific to hemlock stands; the glossy supercoil, shortspire hornsnail, domed ancylid, delicate vertigo and cupped vertigo will not be considered further in this document.

Due to its limited range, and no treatments are proposed on the South Fork of the Cumberland River, the Big South Fork crayfish will not be considered further in this document.

Due to their limited range and or life cycle requirements, the Helma’s net-spinning caddisfly, cliffline caddisfly, Pygmy snaketail and Appalachian grizzled skipper will not be considered further in this document.

36 Environmental Assessment Suppression of HWA Infestations, DBNF

Plants will not be affected by imidacloprid application since imidacloprid is an insecticide, not an herbicide, and can be safely taken up through the roots. Biocontrol agents will have no affect as the released beetles feed exclusively on HWA. Therefore the following Forest Sensitive plant species will not be considered in further detail: Rockcastle aster, spreading yellow false foxglove, American bayberry, juniper sedge, small spreading pogonia, stoneroot, Kentucky lady’s slipper, French’s shooting star, mountain heartleaf, butternut, Short’s bladderpod, large- flowered Barbara’s buttons, Canby’s mountain-lover, white fringeless orchid, bay starvine, hairy skullcap, rock skullcap, southern Oconee bells, Blue Ridge catchfly, royal catchfly, little mountain meadowrue, cutleaved meadow parsnip, sand grape, Sullivant’s leafy liverwort, Plagiochila austinii, and Radula sullivantii. Site specific populations of these plants may exhibit short term fluctuations if they are currently influenced by hemlock overstory.

Federally Listed, Proposed, and Regional Forester’s Sensitive Species Considered in Detail The fish and Mussels listed with “Yes” in Table 3-4 and Table 3-5 are considered in the effects section below. The affected environment for these species would be the streams running through HCAs where nearby hemlock trees would be periodically treated with insecticide.

Indicators (Only DFNF lands) 1) Risk of death or injury to PETS from release of predator beetle. Four risk classes used are “High, Medium, Low, None”.

2) Risk of death or injury to PETS from insecticide, based on SERA risk assessment. Risk assessment indicating a central Hazard Quotient less than1.0 is an acceptable level of risk.

3.5.2 Effects to Federally Listed, Proposed, and Forest Sensitive Species Direct and Indirect Effects to Federally Listed and Forest Sensitive Species - Alternative A This alternative does nothing to change the existing and expected condition. The loss of streamside hemlock-dominated communities will impact the aquatic system. Increased surface runoff of rainfall and resulting soil erosion could occur on slopes where dense hemlock canopies have prevented development of an understory. Year-round shading of streams will be reduced or eliminated in some places. Temperature and hydrologic regimes of streams may become more variable (Evans 2002).

The heavy shade that hemlock provides along streams keeps streams cooler. In a stream temperature monitoring study at Delaware Water Gap National Recreation Area, a several degree Fahrenheit decrease in summer stream temperature was documented as the stream flowed through a hemlock-dominated section of the stream (Evans, Johnson, et al. 1996). Similar results can be expected on the DBNF in areas where streams are influenced by pure hemlock stands. In areas that are mixed hemlock hardwood stands, there may not be as much impact from hemlock mortality.

Stream temperature directly affects dissolved oxygen content. Certain vertebrate and invertebrate species are sensitive to dissolved oxygen content and may respond accordingly as hemlocks along streams loose needles, thereby allowing more sunlight to reach streams.

37 Environmental Assessment Suppression of HWA Infestations, DBNF

Under this no action alternative, changes in water quality may occur as a result of hemlock loss. Likely changes to the aquatic community include increased water temperatures, altered species composition, and changes in biotic densities and diversity. Streams and the aquatic fauna that inhabit the streams can be expected to adjust to any impacts that may occur due to the loss of the hemlock component in the long term.

Listed fish and mussel populations may experience short-term fluctuations depending on what effects the increased daylight, increased woody debris, and possibly altered stream pool or run has on each species specific requirements. Long-term population trends, following the growth of vegetation that replaces the hemlock component and stabilization of stream reach after the breakdown and placement of LWD following a series of rain events is expected. These changes are largely speculative, and will be site specific. Effects to listed fish and mussel species may be undetectable and immeasurable.

Cumulative Effects to Federally Listed and Forest Sensitive Species - Alternative A No cumulative effects would occur to listed fish and mussel species from any past, present or reasonably foreseeable actions within the streams that run through HCAs.

Direct and Indirect Effects to Federally Listed and Forest Sensitive Species - Alts B & C Potential direct and indirect effects are discussed together in this analysis because it is not only difficult to separate these effects in writing, but it is also very difficult for the reader to synthesize how effects on habitat affect species (and vice versa) if the discussions are separated. Therefore, in this analysis, effects are defined as the results of the action on the quality and quantity of aquatic habitat as well as individual species.

Predator Beetles: There would be no risk of death or injury to PETS species. The use of predator beetles will have no measurable direct or indirect effects on any of the PETS species considered or their habitat. Some beetles could fall into the streams and be eaten by the fish, but it would have no effect on fish species. Listed mussels feed on detritus and algae, zooplankton and phytoplankton. Until the beetles become established in riparian hemlock stands, it is assumed that the effects above (reference No Action) could occur, although at a slower rate, until hemlock components of riparian vegetation are maintained. It is likely that treatment of riparian hemlock stands with predator beetles would reduce effects of hemlock loss on aquatic resources and the short-term benefits of increased LWD. Long-term effects would not be as noticeable if the beetles are successful at controlling the HWA.

Insecticide: Imidacloprid is moderately toxic to fish and highly toxic to aquatic invertebrates. There is currently little data on the effects of Imidacloprid on freshwater mussels. Zheng & Liu (1999) found that imidacloprid was stable in acidic and neutral water (representative of the water pH across the DBNF), but more readily hydrolyzed in alkaline water. Hydrolysis of this pesticide is greater than 30 days at pH7 and 25 degrees C (Bacey 2000). Soil half-life for imidacloprid ranged from 40 days in un-amended soil to up to 124 days for soil recently amended with organic fertilizers (Rouchaud, Gustin and Wauters 1994). The pesticide is moderately mobile in some soils, particularly where the pesticide label restrictions prohibit the use of the chemical where surface water is present. It further states that use of the chemical where the water table is shallow may contaminate groundwater.

38 Environmental Assessment Suppression of HWA Infestations, DBNF

The Proposed Action includes Implementation Criteria (section 2.1.2) to aid in the determination of whether it is safe to use imidacloprid to treat individual trees. This process was developed to minimize or eliminate the potential for contact between imidacloprid and aquatic resources. By following the clearance process and label restrictions and Implementation Requirements and Design Criteria, impacts to fish and mussel species from the use of imidacloprid are not expected. The hazard quotient is less than 1.0 for aquatic invertebrates, unless an accident was to occur and chemical spilled into a stream.

In the unlikely event an accident does occur during application that results in imidacloprid entering a stream, local populations of listed fish and mussel species (as well as other fish and mussel species occurring in the affected stream) would likely be affected. The chemical would then be rapidly diluted to a level of no observable effect. It is important to note that the loss of one population listed fish or listed mussel or one local aquatic invertebrate community will not affect overall species’ trends across the Forest. This is because the local stream reach that could be affected by an accidental spill would be quickly recolonized by fish populations from upstream or downstream following rapid dilution. For mussel populations, reestablishment would take longer due to their life history.

This loss of habitat should be slowed down or eliminated in the HCAs depending on the success of the treatment(s). Water and soil chemistry changes could also occur as a result of the loss of the hemlock component. Aquatic wildlife resources may experience short term moderate to major impacts.

Cumulative Effects to Federally Listed and Forest Sensitive Species (Alternatives B & C) No cumulative effects would occur to listed fish and mussel species from any past, present or reasonably foreseeable actions as no other actions were identified within the HCAs containing streams. 3.6 Federally Designated Critical Habitat 3.6.1 Affected Environment Four stream segments are designated by the USFWS as critical habitat6 under the Endangered Species Act (Table 3-6). Table 3-7 displays the acreage of hemlock-dominated stands and hemlock conservation areas (HCA) within the four critical habitat watersheds.

Table 3-6. Federally Designated Critical Habitat under the Endangered Species Act Stream Administrative Associated Species Designated Segment Name Location Cumberlandian Buck Creek main stem from State Buck Creek combshell and London District Road 192 bridge, upstream to the oyster mussel State Road 328 bridge

6 In the two recent listings, the USFWS did not propose to designate critical habitat in their June 24, 2010 listing package for the Cumberland darter nor did they in their November 18, 2010 letter listing package for Snuffbox.

39 Environmental Assessment Suppression of HWA Infestations, DBNF

Stream Administrative Associated Species Designated Segment Name Location Marsh Creek main stem from its Marsh confluence with the Cumberland Cumberland elktoe Stearns District Creek River, upstream to State Road 92 bridge Rock Creek main stem from its Rock Creek Cumberland elktoe Stearns District confluence with White Oak Creek, upstream to River Mile 15.9 Sinking Creek main stem from its Sinking confluence with the Rockcastle Cumberland elktoe London District Creek River, upstream to its confluence with Laurel Branch

Table 3-7. Hemlock Stands within Critical Habitat Watersheds Acres of Hemlock- Total Acres of Hemlock Conservation Stream Name dominated7 stands Watershed Areas within Watershed within Watershed Buck Creek 3,251 00 (None) Marsh Creek 51,386 983 (1.9%) S12, S14, S15 Rock Creek 23,400 370 (1.6%) S18, S19 Sinking Creek 22,541 1202 (5.3%) L08, L09, L11, L12

Hemlock Conservation Areas (HCA) L09 and L12 are on tributaries that are bordering Sinking Creek; S14 borders along the east bank of Marsh Creek; and S18 and S19 are along tributaries of Rock Creek.

Indicators The effects analysis for designated critical habitat is based upon anticipated effects of the project action upon five “Primary Constituent Elements” (page 53147 of the Federal Register) associated with the stream segment designated as critical habitat. These primary constituent elements are:

1. Permanent, flowing stream reaches with a flow regime (i.e. the magnitude, frequency, duration, and seasonality of discharge over time) necessary for normal behavior, growth, and survival of all stages of the listed mussel and their host fish; 2. Geomorphically stable stream driver channels and banks; 3. Stable substrates consisting of mud, sand, gravel, and/or cobble/boulder, with low amounts of fine sediments or attached filamentous algae; 4. Water quality (including temperature, turbidity, oxygen content, and other characteristics) necessary for the normal behavior, growth, and survival of all life stages of the five mussels and their host fish; and 5. Fish hosts with adequate living, foraging, and spawning areas.

7 These are stands consisting of more than 50% hemlock stocking.

40 Environmental Assessment Suppression of HWA Infestations, DBNF

3.6.2 Effects to Federally Designated Critical Habitat Direct and Indirect Effects to Federally Designated Critical Habitat - Alternative A It can be anticipated that in the short term, primary constituent elements 1, 2, 3, and 5 could be impacted by the increased abundance of LWD. This could alter pools and stream reaches, making some areas less suitable and some more suitable for the above listed mussels and their fish hosts. The point in time, the extent of degree and exactly where in hemlock stands adjacent to designated critical habitat will be impacted by HWA is not known at this time. Primary constituent element number 4 may not only be impacted by the increased LWD, but also a temporary increase in water temperature from solar exposure in broader valleys. In the long term, affected portions of designated critical habitat will become stabilized as other plant species grow and mature, thus filling the niche and impacts that the hemlock mortality imposed on these stream segments.

Cumulative effects to Federally Designated Critical Habitat- Alternative A: No cumulative effects would occur to federally designated critical habitat from any past, present or reasonably foreseeable actions.

Direct and Indirect Effects to Federally Designated Critical Habitat (Alternatives B & C) The nine HCA located within designated critical habitat are included in both alternatives. In addition, none of the designated critical habitat is in wilderness, therefore for this resource there is no difference in effects between alternatives B and C.

Hemlock stands which impact designated critical habitat that are treated with imidacloprid, predatory beetles, or a combination of the two would not impact the five primary constituent elements because released beetles that may fall into the creek would not have traces of imidacloprid as they only feed on live adelgids. As a design criterion, no hemlock will have soil application where the root zone extends to the stream bank, to lessen the likelihood that imidacloprid comes into contact with the stream water. Also there are chemical mixing criteria imposed to eliminate any accidental spills.

As a goal of the project is to preserve a network of stands of hemlock across the forest for the retention of future genetic diversity, the overall effect of hemlock mortality and the impacts to Designated Critical Habitat will almost be immeasurable and specifically unpredictable.

Cumulative Effects to Federally Designated Critical Habitat (Alternatives B & C) Only minor cumulative effects would occur to federally designated critical habitat from any past, present or reasonably foreseeable actions. An off-site thirty acre white pine plantation bordering Rock Creek is currently proposed to be removed and regenerated with native hardwood and wetland vegetation. Cumulative impact to water quality would be immeasurable and undetectable.

41 Environmental Assessment Suppression of HWA Infestations, DBNF

3.7 Wilderness Resources

The dilemma of how wilderness can be “protected and managed so as to preserve its natural conditions” while at the same time remaining “untrammeled by man” is recognized as a key topic in the Forest Service’s Wilderness Agenda: Thinking like A Mountain (USDA-FS 2000). Exotic pests, both plants and other organisms, have the potential to drastically alter the natural processes in wilderness. The question of whether or not to take management action to counteract an exotic pest or other unnatural influence is a difficult one. In Naturalness and Wildness: The Dilemma and Irony of Managing Wilderness (Landres, et al. 2000) this is described as an emerging dilemma between managing for these two values which are both central to the concept of wilderness. It is important to manage for naturalness as well as wildness; however, large- scale ecological changes caused by unnatural influences such as exotic pests present difficult choices for the management of wilderness. A decision either to act or not to act will have consequences for the natural or wild conditions of wilderness.

Wilderness is a unique and valuable resource. In addition to offering primitive recreation opportunities, it is valuable for its scientific and educational uses, as a benchmark for ecological studies, and for the preservation of historical and natural features (FSM 2320.1) (USDA-FS 2010).

Wilderness is defined in Section 2 (c) of the Wilderness Act of 1964: “A wilderness, in contrast with those areas where man and his works dominate the landscape, is hereby recognized as an area where the community of life is untrammeled by man, where man himself is a visitor who does not remain.”

An area of wilderness is further defined in this Act to mean an area of undeveloped Federal land retaining its primeval character and influence, without permanent improvements and habitation, which is protected and managed so as to preserve its natural conditions and which (1) generally appears to have been affected primarily by the forces of nature, with the imprint of man’s work substantially unnoticeable; (2) has outstanding opportunities for solitude or a primitive and unconfined type of recreation; (3) has at least five thousand acres of land or is of sufficient size as to make practicable its preservation and use in an unimpaired condition; (4) may also contain ecological, geological, or other features of scientific, educational, scenic, or historic value.”

The National Wilderness Preservation System now includes over 105 million acres in 44 states. Individual wildernesses range in size from some that are millions of acres in Alaska and the Western United States to others that are much smaller in the eastern United States. Wildernesses also vary greatly in the amount of human influences surrounding them and affecting them.

Affected Environment - Wilderness On the DBNF, 17,400 acres have been congressionally designated as wilderness. This includes Clifty (12,550 acres) and Beaver Creek (4,850 acres).

Hemlock is present in both Clifty and Beaver Creek Wildernesses. The hemlock population and associated plant communities are a distinct esthetic attribute contributing to wilderness character and are a large component of the two wildernesses on the DBNF. Hemlock forest types comprise approximately 20% of Clifty Wilderness, and 25% of Beaver Creek Wildernesses.

42 Environmental Assessment Suppression of HWA Infestations, DBNF

Eastern Hemlock is long-lived, reaching ages of 800 years or more in some cases. It is an especially important component of wilderness character because of its distinct visual appearance. Hemlocks are thought of by many visitors as “icons” of the wilderness because of the size and old growth character of many individual hemlock trees and stands of trees.

Hemlock is found primarily in acidic coves and related slope areas and riparian areas and is a dominant feature in a distinct plant community. It is associated with a variety of plant communities that contain species such as red and white oak, yellow-poplar, rhododendron, and others. This plant community is important ecologically in supplying shade for riparian areas as well as contributing significantly to wilderness character as a distinct visual and esthetic attribute.

Table 3-8. Number of HCA in Wilderness, by Alternative, and Treatment Type

Alternative A Alternative B Alternative C Treatment Type Treatment Type Treatment Type Biological Biological Biological Wilderness Chemical Control Chemical Control Chemical Control Clifty 0 0 0 0 0 5

Beaver Creek 0 0 0 0 0 2

Indicators 1) Naturalness. 2) Unique components 3) Trammeling or wildness. 4) Solitude

One other indicator is required to be considered when proposing actions in wilderness. Since no structural development or motorized activity is proposed, the development indicator was not considered relevant to this action.

Direct and Indirect Effects to Wilderness – Alternative A The overall naturalness of the wildernesses will be negatively affected in this Alternative by the disruption of natural processes caused by the HWA. The naturalness and value of wilderness as a biological benchmark is threatened if most or all of the hemlock population and associated plant communities are extirpated, which would likely occur in this alternative. As the hemlock dies, it is likely that non-native invasive species will replace them in some locations. This is likely to occur with aggressive species such as Pawlonia tomentosa in areas where dead hemlocks tip over leaving disturbed ground. Over time this will further degrade the naturalness of the wilderness resource and create problems in removing these additional non-native invasive species.

The high concentration of old hemlock within wildernesses on the DBNF, make this a unique component of these wildernesses. With this no-treatment alternative, there could be losses of

43 Environmental Assessment Suppression of HWA Infestations, DBNF

hemlock of over 90% to occur within the next ten years, and perhaps 100% within twenty years, based on observations of HWA infested forests in the eastern U.S. All the older groups of hemlocks that now exist (some approaching or surpassing 500 years old) would likely be lost. Although as of this writing, infestations of HWA have not been detected in the two wildernesses.

The current level of trammeling of the wilderness resource would remain unaltered because of the lack of direct human intervention to mitigate the effects of HWA. Visitors will continue to be able to find solitude in these areas.

Cumulative Effects to Wilderness – Alternative A Since there would be no action in this alternative, there would be no cumulative effects.

Direct and Indirect Effects to Wilderness – Alternative B The direct and indirect effects are similar to Alternative A, since no treatment would occur in wildernesses. However, since beetle release would occur near the edges of the wildernesses, there would be a slight chance for beetles to spread to adjacent hemlock stands within wildernesses from release points near the edge of wilderness. This could slightly reduce the negative effect of HWA on naturalness & uniqueness within the wildernesses. However, it is unlikely that beetles would have much effect on HWA infestation of hemlock within the interior of wildernesses. There would be no direct or indirect effect on trammeling or solitude within wilderness by this action.

Cumulative Effects in Wilderness – Alternative B Since workers would not be present to treat hemlocks in wilderness, there would be no associated cumulative effect on trammeling or solitude. There would be no cumulative effect on naturalness and uniqueness of wilderness by this action.

Direct and Indirect Effects to Wilderness – Alternative C In this alternative, beetle release in wilderness would occur as infestations are discovered. The seven HCAs located in wildernesses would be priority for monitoring and release of predatory beetles.

The biological control to check the spread of the non-native HWA involves a compromise of choosing the non-native beetles to combat the non-native adelgid. The placing of the predator beetles on HWA in the wildernesses is a human intervention or trammeling of wilderness. There will be some reduction in the wildness of wildernesses because of the release of beetles and monitoring of beetles, HWA, and hemlocks. However, the effects to wildness from these actions would be of very short duration and unnoticeable once the activity is complete. The presence of personnel dispensing beetles or conducting monitoring would be a direct effect on wildness (trammeling) and solitude in this alternative, but such would be subtle, and encounters with visitors unlikely.

Where HWA infestations are detected and beetles are released, the existing naturalness of those hemlock stands may be protected as well as protection of the uniqueness of some older hemlock stands where treated. Alternative C thus should provide a level of protection to the naturalness

44 Environmental Assessment Suppression of HWA Infestations, DBNF

of the wilderness resource. Hemlock mortality will be controlled by the effectiveness of the predatory beetles.

Cumulative Effects in Wilderness – Alternative C Ongoing human influences within wildernesses that tend to increase trammeling and reduce solitude include recreational activities such as trail use impacts, trail maintenance, bridge repair, campsite impacts, and the sights and sounds of humans. In this alternative, personnel would be present to release beetles, minimally adding to such ongoing human influence and presence. 3.8 Human Health and Safety

Affected Environment The following section looks at possible impacts to the health and safety of users of the forest (visitors) and personnel implementing the alternatives.

Indicators: 1) Snag hazard. As hemlock trees die and decompose along roads and trails, they can become a hazard to humans as limbs or entire trees could fall, especially in windy conditions. 2) Insecticide exposure (hazard quotient less than 1.0). There is always some risk to humans when pesticides are applied. This risk is minimized when label directions are followed. 3) Tree Climbing Risk. Some tree climbing may be necessary to distribute beetles on hemlock branches where HWA has become established. Tree climbing would be done only by experienced climbers.

Direct and Indirect Effects on Human Health & Safety – Alternative A This no-action alternative would not prevent the formation of hemlock snags. Dead hemlock snags would be a threat to human safety in the years to come as HWA kills trees along trails and roads. Weakened hemlocks are very susceptible to wind storms and are frequently broken off and uprooted. The threat exists that broken or uprooted hemlocks could fall on humans using or working in the Forest. This alternative would eliminate the possibility of exposure to imidacloprid insecticide or tree climbing accidents related to treatment of HWA.

Cumulative Effects on Human Health & Safety – Alternative A Since this is a no-action alternative, there would be no cumulative effects on human health and safety.

Direct and Indirect Effects on Human Health & Safety - Alternatives B & C In both alternatives, the threat from weakened and dead hemlocks falling would be reduced due to treatments that would keep many hemlocks alive in the forest outside of the wildernesses. Reduction of this hazard in wildernesses would occur in alternative C.

45 Environmental Assessment Suppression of HWA Infestations, DBNF

Tree-climbing may be needed to implement the release of predatory beetles to combat HWA. To accomplish this activity, employees would be trained, or other qualified climbers would be used to do this work. The risk of tree-climbing accidents would be greater in alternative C, since a larger area would be treated.

A risk assessment for imidacloprid is available in the project file (Anatra-Cordone and Durkin 2005). Application rate and work time will not exceed levels that pose an unacceptable level of risk to human or wildlife health, and hazard quotient8 will not exceed 1.0. The potential for the imidacloprid used in this project to impact human health under any circumstances is minimal as its toxicity to mammals in general is very low. The U.S. Environmental Protection Agency’s Office of Pesticide Programs has evaluated imidacloprid and determined it to be non-cancer- causing (US-EPA 1994). The potential human safety issues include accidental spills of the product and accidental contamination of skin or clothing of the applicators. The risks of pesticide application would not be noticeably different in either action alternative, although more days would be spent in the forest applying insecticide in alternative C.

Cumulative Effects on Human Health & Safety - Alternatives B & C Soil injection of imidacloprid occurred at four developed recreation sites and administrative sites in 2009 and 2010. This small amount of activity is keeping hemlock trees alive in high-use areas, thus reducing the number of potential snags. Application of insecticide for this project would not contribute cumulatively other than increasing the workload for applicators. There would be no cumulative effect on human health and safety from beetle release.

8 Although Forest Plan standard DB-VEG-8 applies to herbicide, insecticide application in this project considers this standard as well (USDA-FS 2004b).

46 Environmental Assessment Suppression of HWA Infestations, DBNF

4. Preparers & Persons Consulted

Interdisciplinary Team: Amos Stone Silviculturist (Team Leader) Joe Metzmeier Wildlife Biologist Pam Martin Fisheries Biologist David Taylor Botanist / Ecologist Jon Walker Hydrologist Claudia Cotton Soil Scientist Myra Williamson Recreation Forester Paul Finke Planning Forester Wayna Adams Heritage Resources Jeff Lewis Silviculturist (Cumberland District) Robbie Sitzlar Silviculturist (London District) Tony Martoglio Silviculturist (Stearns District) Elizabeth Robinson Silviculturist (Redbird District)

Agencies/Persons Consulted: Rusty Rhea Entomologist, Forest Health Protection, Asheville, NC Wes Nettleton Director, Forest Health Protection, Atlanta, GA Mike Floyd Wildlife Biologist (Recovery) USDI-FWS, Frankfort, KY Carrie Lona Wildlife Biologist (Consultation) USDI-FWS, Frankfort, KY

47 Environmental Assessment Suppression of HWA Infestations, DBNF

APPENDIX A – Hemlock Occurrence and Distribution

Table A-1. Number of live trees (thousands) in the Hemlock Species group, on forest land by Ownership and Diameter, Eastern Kentucky, 2006

Tree diameter at breast height (inches) Land owner 1.0-4.9 5.0-8.9 9.0-12.9 13.0-16.9 17.0-20.9 21.0 + total

National Forest 19,500 4,800 1,100 930 221 145 26,067

Other Owners 36,705 5,939 2,010 828 177 329 45,990

Total – Eastern KY 56,205 10,739 3,110 1,758 398 474 72,057 (FIA data 2006 – three eastern survey units; diameter at breast height is measured at 4.5’ above ground)

Table A-2. Hemlock Forest, Acres by Forest Type and Age Class, DBNF

Age class Forest Type 00-19 20-39 40-59 60-79 80-99 100-119 120-139 140 + total

Hemlock (05) 00 00 424 47 198 492 406 475 2042

Hemlock- 70 706 1561 1542 5460 5579 3028 1636 19582 Hdwd.(08) Hdwd.-W.P. - 125 352 559 834 2251 2623 850 582 8176 Hemlock (41) White Pine – 00 00 77 00 60 256 00 00 393 Hemlock (04)

Total 195 1058 2621 2423 7969 8950 4284 2693 30193 (FSveg data February 2010)

Table A-3. Hemlock Forest Types, Acres within Wildernesses, DBNF Hemlock Stands Mixed Stands (>50 % hemlock) (30-50% hemlock Other Stands Total Wilderness Wilderness Area Acres (%) Acres (%) Acres (%) Acres Clifty Wilderness 2,839 (23%) 671 (5%) 9,038 (72%) 12,548 (Cumberland District) Beaver Creek Wilderness 1,863 (38%) 0 2,991 (62%) 4,855 (Stearns District) Total 4,702 (27%) 671 (4%) 12,029 (69%) 17,403 (FSveg data February 2010)

48 Environmental Assessment Suppression of HWA Infestations, DBNF

APPENDIX B – Hemlock Conservation Strategy

The Interdisciplinary Team used a logical process to identify and prioritize areas for Hemlock Woolly Adelgid (HWA) control. A network of conservation areas would be laid out across the forest to serve as genetic reserves in order to maintain a diverse hemlock gene pool in situ. Also, within these areas, important hemlock ecosystem functions such as stream shading, large woody debris production, and substrate for non-vascular species may be maintained if control measures are effective. In short, the strategy for controlling the HWA is to identify critical hemlock conservation areas, up front, and to ‘make a stand’ in these priority areas using available treatments. Included within our primary goal to maintain hemlock viability across the Daniel Boone National Forest (DBNF) are genetic and species goals. Genetic conservation goals include maintaining genetic variation within and among populations of hemlock, and assuring that processes such as genetic differentiation and gene flow continue. Without genetic variation and gene flow, populations are less adaptable and their extinction is more probable, all else being equal (Noss 1992). Maintaining genetic diversity within and among the forest-level hemlock populations should be a primary consideration in any conservation area design (Crane 2004). The species conservation goal is to maintain viable populations of eastern hemlock on the Daniel Boone National Forest in cooperation with other ownerships in eastern Kentucky on the Cumberland Plateau and Mountains. Conservation Area Location: The major challenge of locating conservation areas for hemlock and its associated species on the DBNF is determination of the proper distribution and number of areas needed to support reproducing hemlock populations and the management actions needed to protect these habitats. Distribution of Conservation Areas: A primary consideration in location of conservation areas was the silvics of hemlock which influenced the distance between areas (Table B-1). Because within-population genetic variation is often as great as between-population genetic variation in forest trees, several small genetic reserves or subpopulations spread over a large geographical area may conserve the greatest amount of genetic diversity more effectively than a single, large reserve (National Academy of Sciences 1991) and will be the model used here. Number of Conservation Areas: The number of conservation areas to establish across the DBNF is primarily dependent upon how many trees and subpopulations are required to maintain hemlock genetic diversity and the diversity of plant communities where hemlock is a primary component. We considered metapopulation and Minimum Viable Population (MVP) concepts in evaluating the number of conservation areas and number of hemlock trees needed for genetic reserves. There are no “magic numbers” but there are three general approaches to address MVP. They include: Estimating the effective population based on the number required to maintain the evolutionary potential of the population. It has been estimated that a population of 500 individuals is not

49 Environmental Assessment Suppression of HWA Infestations, DBNF likely to lose genetic variance due to drift and can retain enough variation to respond to altered selection pressures (Franklin 1980). For out crossing, monoecious species such as hemlock, the number is 1500-2000 individuals. Calculate the population size that will minimize the sampling loss of alleles that occur in low frequency. It has been estimated that in species with known levels of inbreeding and population structure, a sample size of 1000 will keep the probability of loss of an allele to an acceptable frequency (Namkoong 1984).

Table B-1. Silvics of Eastern Hemlock flowering period April-June seed production cones mature in one year, good crops every 2-3 years; one of the most frequent cone producers among the eastern conifers seed dissemination wind dispersed up to 4,500 ft., wide dispersal favored by hygroscopic cone scales pollen dissemination wind dispersed up to 5 miles seedling periodic, mostly during years of greater than normal rainfall establishment moisture best @ 32-> 56” /year requirements site affinity acidic coves (streamside zones) reaction to most shade tolerant of all species, can survive in suppressed position for decades competition genetics heritable differences in characteristics of seed germination and development from different regions / sites

The population size necessary to avoid loss of genetic variation due to genetic drift is much larger than needed to avoid inbreeding depression, at least for animals. It was first proposed that an effective size of 500 individuals would allow maintenance of long-term viability, because loss of genetic variation from drift would be offset by the creation of new variation through natural mutation (Franklin 1980). Recently, however, this number has been a topic of some debate (Lande 1995) (Franklin and Frankham 1998) (Lynch and Lande 1998) (Allendorf and Ryman 2002). One scientist indicated that only populations with an effect size of over 5000 individuals can be expected to maintain viability in the absence of immigration, because not all mutations are beneficial (Lande 1995). Others argue that an effective population size of 500 to 1000 individuals is sufficient (Franklin and Frankham 1998). At issue are the potential effects of harmful mutations: some consider these effects negligible (Franklin and Frankham 1998), but others have suggested that slightly deleterious mutations are capable of causing population extinction even at effective sizes of several hundreds (Lynch, Conery and Burger 1995) (Lynch and Lande 1998). The debate will likely continue, but a reasonable conclusion is that only populations with actual sizes in the thousands, rather than hundreds, can maintain long-term viability and evolutionary potential in the absence of immigration (Allendorf and Ryman 2002). Metapopulation biology is concerned with the effects of local population dynamics on the regional persistence of a species and thus has achieved its widest application in conservation biology. A metapopulation is a set of local populations (subpopulations) within some larger area that are connected via migration; where a population is a collection of individuals, a

50 Environmental Assessment Suppression of HWA Infestations, DBNF

metapopulation is a collection of local populations. This approach is useful, and sometimes necessary, in studying the dynamics of species that occur in patchy or frequently disturbed habitats. In such cases, the movement of individuals and genes between local populations will make an important difference to species persistence. For species such as hemlock, the viability of single populations may be maintained with effective subpopulation sizes of 50 to 150 reproductive individuals, and it would be possible to contain a total of a few thousand in as few as 50 or so subpopulations (National Academy of Sciences 1991). Maintaining a minimum of 500 individuals in multiple genetic reserves (subpopulations in close proximity) will retain 95% of the species’ genes and should be adequate for maintaining genetic diversity of hemlock on the DBNF, especially when pollen flow between and among areas is considered (Crane 2004). Based upon these references, and the size of the forest area, we have estimated 50 as the minimum number of subpopulations or reserves needed to maintain genetic diversity of eastern hemlock on the DBNF. At least for hemlock, this design will provide for a forest-wide population in the thousands, which should result in a high likelihood for viability (given effectiveness of control treatments). Maintaining the diversity of communities where hemlock is a primary member would require significantly more conservation areas because of the variety of associated species that include known aquatic and presumed terrestrial wildlife members. Viability risk to all species in the community cannot be as easily assessed as viability risk to hemlock because of variable numbers of species occurrences and population densities. It is estimated that the MVP for these multiple species would require at least double (100) that for hemlock reserves to maintain species and community diversity for all species in hemlock communities. Minimum Viable Population (MVP) is defined as the smallest isolated population having a [90% or 95%] chance of surviving for [100, 500, or 1000] years despite the foreseeable effects of demographic, environmental and genetic stochasticity, as well as natural catastrophes. To assess population viability and MVP, generally four threats are considered: (1) demographic stochasticity, (2) environmental stochasticity, (3) catastrophes and (4) genetic drift and inbreeding (Shaffer 1981) (Shaffer 1987). Demographic stochasticity refers to effects of random events on the reproduction and survival of individuals, whereas environmental stochasticity refers to effects of unpredictable events that alter vital rates. For example, if every individual has a 50% probability of annual survival, in a population of 20 individuals 10 will not die each year. Instead some years by chance nine will die, in others eleven and so forth. This is demographic stochasticity, which is analogous to sampling error. It may be that in years with severe winters the probability of survival is only 30%, whereas in years with mild winters the probability of survival is 70%. This is an example of environmental stochasticity. There are two genetic threats to population viability. The first, inbreeding depression, threatens all population sizes of trees and only small populations of animals, whereas the second, genetic drift, can threaten even large populations (Lande 1995). Inbreeding depression reduces the survival and productivity of individuals, and results from the segregation of partially recessive, deleterious alleles. The resulting negative effect on population

51 Environmental Assessment Suppression of HWA Infestations, DBNF

dynamics increases vulnerability of extinction. The amount of inbreeding depression depends on the rate of inbreeding and opportunity for selection to purge recessive lethal and semi-lethal mutations (Lande 1995). Genetic drift results in the loss of genetic variation, which may reduce a species’ ability to adapt and persist in a changing environment, and thereby its viability over long time periods. The rate of loss is inversely related to population size and mutation rate, and viability is achieved when the population size is large enough that loss to drift is in equilibrium with gain from mutation. Cross-pollination or out-crossing is the most desired condition for most plants; because it maintains a level of variability and heterozygosity that provides the opportunity for evolutionary change if environments change (Hartmann and Kester 1983). For many plants (barley, oats, wheat, rice, peanuts, soybeans, many grasses, tomatoes, tobacco) a self-pollination strategy has proven most successful. However, cross-pollination is necessary for trees to reduce the devastating effects of inbreeding depression. The "50/500" rule of thumb initially advanced by Franklin (1980) and Soule (1980) comes the closest of any to attaining "magic number" status. This rule prescribes a short-term effective population size (Ne) of 50 to prevent an unacceptable rate of inbreeding, and a long-term Ne of 500 to maintain overall genetic variability. The Ne=50 prescription (termed "the basic rule" by Soule (1980) corresponds to an inbreeding rate of 1% per generation, approximately half the maximum rate tolerated by domestic animal breeders (Franklin 1980). The Ne=500 prescription is an attempt to balance the rate of gain in genetic variation due to mutation with the rate of loss due to drift (Franklin 1980). Practical applications of the 50/500 rule have been made by several scientists (Foose 1983) (LaCava and Hughes 1984) (Reed, Doerr and Walters 1988). However, these numbers may not be entirely applicable to trees. Relationship of USFS Viability Evaluations to Population Viability Analysis Definitions of population viability in the scientific literature have focused on the probability of a populations’ persistence for a biologically meaningful period of time. For example, Soulé (1980) defines viability in terms of the capacity of the population to maintain itself without significant demographic or genetic manipulation for the foreseeable ecological future centuries with a certain degree of certainty and adaptation. Adaptation requires that the population “maintains a normal level of immediate fitness (individual vigor, fertility, fecundity) and has sufficient genetic variation to adapt by natural selection to changing environmental conditions within the predicted range of frequency and amplitude of disturbance and change of a species or population” (Soule 1980). Population vulnerability analyses are used to estimate minimum viable populations (MVPs). These analyses consider the effects of demographic stochasticity, genetic stochasticity, environmental stochasticity, and catastrophes. They evaluate life history of the species (population), the temporal and spatial distribution of its resources, its level of genetic variation, and the consequences of habitat fragmentation (Gilpin and Soule 1986). Shaffer (1981) defined MVP as “the smallest isolated population having a 99% chance of remaining extant for 1000 years despite the foreseeable effects of demographic, environmental, and genetic stochasticity, and natural catastrophes”.

52 Environmental Assessment Suppression of HWA Infestations, DBNF

APPENDIX C – Wilderness Considerations Evaluation of Potential Treatment for HWA in Wilderness Areas

Steps in the Evaluation This evaluation follows the logic steps of the Minimum Requirement Decision Guide for wilderness in addressing six “key questions” and two “management questions” involving the potential treatment of Hemlock Woolly Adelgid (HWA) in designated wilderness Areas of the Daniel Boone National Forest. The two management questions are:

(A) Should treatment for the Hemlock Woolly Adelgid be considered in any wilderness and would this action be consistent with the Wilderness Act?

(B) If treatment for HWA is considered, which type(s) and amount of treatment will have the least impact to the wilderness resource while accomplishing the purpose of the proposed action?

Current Situation and Description of Significance of Hemlock in Wilderness Areas Eastern Hemlock (Tsuga canadensis) is present in both Clifty and Beaver Creek Wildernesses.

This species is long-lived, capable of reaching ages of 800 years or more. Eastern hemlock is found primarily in riparian areas and adjacent north and east slopes and is a dominant feature in a distinct plant community. It is associated with woody species such as northern red and white oak, black walnut, black birch, bigleaf magnolia, yellow-poplar, rhododendron, and occasionally sycamore, ash, and sweetgum. This plant community is important ecologically in supplying shade for riparian communities as well as contributing significantly to wilderness character as a distinct visual attribute.

These Eastern Hemlock community associations comprise approximately 31% of Clifty and Beaver Creek Wilderness (Appendix A, Table A-3). Infestations of HWA have not yet been confirmed in either wilderness. The infestation and resulting mortality could eventually overcome 100% of the hemlock trees.

Wilderness Act and Forest Service Manual Guidance for Control of Insects and Disease Section 4(d)(1) of the Wilderness Act of 1964 states that…”measures can be taken as may be necessary in the control of fire, insects, and diseases, subject to such conditions as the Secretary deems desirable.”

Forest Service Directives (Manuals and Handbooks) are available online at http://www.fs.fed.us/im/directives/ (USDA-FS 2010). Section 2324.12 of the Forest Service Manual (FSM) states “Do not control insect or plant disease outbreaks unless it is necessary to prevent unacceptable damage to resources on adjacent lands or an unnatural loss to the wilderness resource due to exotic pests.”

FSM Section 2324.15 states: “When control of insect and disease is necessary in National Forest Wilderness, it shall be carried out by measures that have the least adverse effect on the Wilderness resource and are compatible with Wilderness management objectives. Meet the

53 Environmental Assessment Suppression of HWA Infestations, DBNF requirements in FSM 2324.04, FSM 2151, FSM 3430, and FSM 1950 in carrying out insect and disease control projects in Wilderness. Special care must be taken in the use of chemicals inside Wilderness because of the possible effects on the total biological complex. Consider other alternatives to chemical use in the environmental analysis.”

Hemlock Conservation Areas Using concepts of metapopulation and Minimum Viable Population (MVP), conservation reserve areas have been identified on the Daniel Boone National Forest as the best places in which to maintain genetic reserves of hemlock and associated plant communities. Of 92 conservation areas considered, 7 occur in wilderness Areas (in Alternative C).

Six Key Questions Regarding Need for a Management Action in Wilderness The Minimum Requirement Decision Guide (MRDG) adopted by the Forest Service, Park Service, Bureau of Land Management, and Fish and Wildlife Service outlines six key questions to answer in determining if management action is necessary within wilderness (Wilderness.net 2010). If the answer to several or all of the questions is “yes”, then management action should be considered.

1. If the problem/issue is not resolved, or action is not taken, will the natural processes of any Wilderness within the Daniel Boone National Forest be adversely affected? Yes, very likely. Hemlock stands comprise a significant portion of both wildernesses and are generally considered an important component of wilderness character and an important part of wilderness as a biological “benchmark.” There is a chance that hemlock will be permanently eliminated in Kentucky. Loss of most or all of the hemlock plant community would be considered to adversely affect the natural process through the infestation of an exotic pest.

However, treatment of HWA chemically or with biological control is a human intrusion within wilderness even though the treatment is intended to mitigate the influence of the HWA on the natural processes within wilderness. Treatments would in essence reduce the “wildness” of the wilderness resource by the use of chemicals or biological control to allow natural systems to operate more freely. The benefit of treatments to the operation of natural processes must outweigh the loss of “wildness” for treatments to be undertaken.

2. If the issue/problem is not resolved, or action is not taken, will the values of solitude or primitive and unconfined type of recreation be threatened? Somewhat: Although the naturalness of the wilderness resource would be reduced, primitive and unconfined recreation would still be available. Hemlocks contribute to solitude in the sense they provide all-season vegetative screening. 3. If the issue/problem is not resolved, or action is not taken, will evidence of human manipulation, permanent improvements, or human habitation be substantially noticeable? Only indirectly: There would be no increased evidence of human habitation or manipulation to the casual observer. However, a well-informed person would realize that the inadvertent human importation of HWA in the 1950’s to the Eastern US indirectly caused the infestation and death

54 Environmental Assessment Suppression of HWA Infestations, DBNF

of the hemlocks. In addition, trail-clearing saw-through of large dead hemlocks would eventually be evident.

4. Does addressing the issue/problem or taking action protect the wilderness as a whole as opposed to a single resource? Yes. The objective of taking action would be to protect the integrity of the natural processes of wilderness ecosystems as well as the wilderness character as much as possible.

5. Does addressing this issue/problem or taking action contribute to protection of an enduring resource of wilderness for future generations? Yes. The action would be designed to mitigate the long-term effects of the HWA infestation in wilderness. At a minimum, some older hemlock plant communities would have a much better chance of being in place in the future. Quicker recovery of a significant population of the eastern hemlock species and its associated plant communities in the future would be much more likely if components of these plant communities were maintained until the most effective methods of adelgid control are confirmed. Some treatment of the adelgid offers the best chance of these plant communities being maintained as part of the “biological benchmark” of the wilderness ecosystems without an overwhelming influence of an exotic pest.

6. Is this an issue/problem for reasons other than convenience or cost of administration? Yes. Loss of the all or most of the hemlock populations is a threat to wilderness character and the integrity of the ecosystem.

Answers to Management Questions Management Question (A) Should treatment for the Hemlock Woolly Adelgid be considered in DBNF wildernesses, and would this action be consistent with the Wilderness Act?

Yes. Treatments should be considered to protect the operation of the natural systems within wilderness by mitigating to some degree the overwhelming influence of HWA on the hemlocks and hemlock associated plant communities. The purpose of the treatment is to prevent an unnatural loss of the wilderness resource due to the expanding infestation of an exotic pest as noted in FSM 2324.12. Eastern Hemlock and associated plant communities are significant components of wilderness character and ecosystems, and is threatened with up to 100 percent mortality.

However, the benefit of treatment for HWA for protecting the “naturalness” of the wildernesses must be weighed against the loss of “wildness” from the human intrusion of the treatments. This proposal to treat or not treat in wilderness will be made in the context of the Environmental Assessment.

Management Question (B) - If treatment for HWA is necessary, which type(s) and amount of treatment will have the least impact to the wilderness resource while accomplishing the purpose of the action?

The answer to this question constitutes the selection of the “minimum tool” for treatment in wilderness. If a decision is made to treat for HWA, then the treatment or combination of

55 Environmental Assessment Suppression of HWA Infestations, DBNF treatments (minimum tool) with the least impact on wilderness will be chosen. Following are the treatment options presently available for the control of HWA through treatment including insecticidal soap, biological controls, and chemical treatments.

Soap Applications Soap spraying is the application of insecticidal soap to the entire tree. It requires mechanical equipment and motorized transport of large special spraying equipment. The soap is then sprayed, soaking trees up to 150-feet tall. The soap mixture is only effective in killing the HWA on the tree at the moment applied. Treatment using a soap mixture is virtually impossible unless trees are directly adjacent to a road that allows access of large equipment. This is not a realistic option for wilderness due to the conflicts with mechanized/motorized equipment and the difficulty of wilderness terrain.

Chemical Treatments Four general types of chemical treatments using imidacloprid are effective for HWA control. Ground injection; trunk injection; soil drench; and pellet insecticide which is easily transported and inserted in the soil for slow release. Chemical treatments using imidacloprid offer the surest method of hemlock survival at a smaller scale; however, it is impractical to treat a large number of hemlocks on a recurring basis because of time and expense. Unless needed for genetic reserves, marking trees for repeated chemical treatment in wilderness may not be warranted.

Biological Controls Several species of beetles have been found to be predators of HWA, although the beetles are non-native. Beetles are being raised by several universities for HWA control, but the production of beetles is limited. The production level is expected to increase in the next few years. The types of beetles being raised that are approved by the USDA for release have been found to eat only HWA or other types of adelgid. If no HWA or other adelgid is present, these types of beetles will cease to exist in that location. Once introduced in the landscape, there are still some questions regarding how quickly they can control the HWA infestations. Research is on-going for additional biological controls.

Biological control appears to be the most subtle and potentially effective means of treatment of HWA at a large scale in wilderness.

56 Environmental Assessment Suppression of HWA Infestations, DBNF

APPENDIX D – Conservation Area Treatment Priority

This appendix contains the list of Hemlock Conservation Areas (HCAs) selected for treatment in this environmental assessment, along with a guide for priority for treatment of each HCA if resources and materials become limiting. Over 900 hemlock stands were identified on the DBNF using existing stand exam information. All are listed in the new FSVeg-Spatial database and mapped in the Geographic Information System (GIS). These locations were identified and evaluated in the GIS along with other pertinent information such as topography, land ownership, rare species locations, roads, prescription areas and management areas. Hemlock conservation areas are distributed throughout the Management Areas in roughly the same proportion as the current distribution of hemlock stands within the Management Areas on the Daniel Boone National Forest. Table D-1 displays the distribution of HCAs by the Forest Plan Prescription Areas and Management Areas.

Table D-1. Distribution of 92 Hemlock Conservation Areas, by Prescription Area and Management Area - Number (Percent). Licking Middle Ky. Upper Ky. Cumberland Prescription River River Mgt. River Mgt. Riv. Mgt. Area Total Mgt. Area Area Area Area 1A – Research 1 (1.0%) 1 (1.0%) Natural Area 1E – Riparian 5 (5.4%) 8 (8.7%) 12 (13.0%) 53 (57.6%) 78 (84.7%) Corridor 1C - Cliffline 2A&B – Wilderness 5 (5.4%) 2 (2.2%) 7 (7.6%) Areas 3E – Red River 6 (6.5%) 6 (6.5%) Gorge Geol. Area 5C – Source Water

Protection Area Management Area 5 (5.4%) 20 (21.6%) 12 (13.0%) 55 (59.8%) 92 (100%) Total

Ninety-two conservation areas were selected from potential treatment areas including hemlock stands considered ecologically important, culturally important, or that otherwise filled a hole in the conservation area design. Ecologically Important Hemlock Stands: Stands were identified as ecologically important from the Kentucky State Nature Preserves Commission and from DBNF records. Hemlock communities were considered for the quality of the occurrence, which takes into account factors such as population or community size, high cover abundance, and other biological indicators. Proximity to quality streams was one specific biological indicator.

Other Hemlock Stands: When an insufficient number of ecologically or culturally important stands were identified for the hemlock conservation area layout, stands were added based on

57 Environmental Assessment Suppression of HWA Infestations, DBNF

information from the Field-Sampled Vegetation (FSVeg) database or from the personal knowledge of Forest Service botanists and silviculturists.

Most of the conservation areas are in close enough proximity to allow for natural gene flow among a sufficient number of eastern hemlock trees to improve the probability of genetic diversity of the species. Assuming 60 trees in each chemically-treated area, approximately 5100 trees could be maintained in these areas until predator beetles become established and in equilibrium with HWA populations. In addition, gene flow could eventually be facilitated by manually transporting pollen among the genetic reserves.

Process for selecting conservation areas: A system for prioritization of the HCAs based on weighted factors has been devised by the Planning Team. The following logic was used to help select and prioritize hemlock conservation areas on the Daniel Boone National Forest: 1) Identify hemlock stands across the Forest using existing information from: a. Kentucky Natural Heritage Program element occurrences records, b. FSVeg database for the Forest, c. District personnel and shared specialists. 2) Rank stands to evaluate the significance of each site and its value for inclusion using the following criteria: a. Accessibility to treatment site for treatment and monitoring within 1/10 mile of a road = rank 1. b. Ecological significance as defined by the KY Natural Heritage Program site occurrence rank of A or B = rank 1. c. Stands adjacent to streams containing Endangered or Threatened species = rank 1. d. Stands containing Regional Forester’s Sensitive species with fewer than three occurrences on the Forest = rank 1. e. Stands listed in FSveg as Hemlock (05), Hemlock-Hardwood (08) = rank 2. f. Stands listed in FSveg as White Pine-Hemlock (04), Hardwood-Hemlock = rank 3. g. Significant hemlock sites identified by District personnel and shared specialists not recorded in FSveg = rank 3. 3) Distribute conservation areas using the following guidelines: a. Areas should be no closer than 1,320 feet (1/4 mile) apart to reduce overlap of dispersing predator beetles. Consider the effective predator beetle dispersal distance from the point of release. The control area is approximately 125 acres in size. b. Areas should be no greater than five miles apart to increase the probability of pollen exchange between hemlock trees. Consider each area identified for chemical treatment would include up to three groves of trees with at least 20 trees treated with systemic insecticide in each grove (60 trees) estimating that treatment would be approximately 80% effective. c. Strive to balance the significance of sites identified at step 2 with the genetic requirements of maintaining hemlock trees in close proximity (within 3-5 miles) to maximize inclusion of high-ranking areas and allow for well-distributed hemlock subpopulations.

58 Environmental Assessment Suppression of HWA Infestations, DBNF

HCA Numbers and Treatment Priority Rankings The selected treatment areas, and factors considered for priority of treatment are listed in Table D-2. Mapped locations of the HCAs are on the pages following the table. The headings for the columns in Table A are explained below.

Column 1: HCA Number: HCAs with an asterisk are in wilderness. The alpha-numeric identification corresponds to those on the maps. The alphabetical character represents each district as follows: C = Cumberland Ranger District, L = London Ranger District, S = Stearns Ranger District, R = Redbird Ranger District,

Column 2: “Central stand” = the stand at the geographic center of the HCA; Column 3: “NH” = Natural Heritage site (KSNPC); (5 points); Column 4: “Origin >1890” = stand age greater than 120 years in year 2010; (3 points); Column 5: “T&E Strm” = the HCA is located near a stream containing Threatened or Endangered Species; (5 points); Column 6: “Sens.Spp” = Sensitive Species within the HCA; (5 points); Column 7: “Social Value” = significant human interest in the stand; (1 point); Column 8: “Good Access” = Access road within ¼ mile of potential treatment area, no major obstacles to foot travel; (5 points); Column 9: “HCA 5 mi” = the number of HCAs within 5 miles, where chemical treatment may occur; (4+ = 5 points, 2-3 = 3 points); Column 10: “Acres 05/08” = Acres of Forest Types containing at least 50% hemlock tree stocking within a ¼ mile of a random point in the HCA; Column 11: “Acres 04/41” = Acres of Forest Types containing 30-50% hemlock stocking within a ¼ mile of a random point in the HCA. Column 12: “Crit.Score” = sum of (columns 3-9) + (point value of column 9) + column 10*5/20 + column 11*3/20. Column 13: “Priority” = If column 12 >25= 1, 20-25=2, 15-20=3, 10-15=4, <10=5

59 Environmental Assessment Suppression of HWA Infestations, DBNF

Table D-2. Hemlock Conservation Areas (HCA) and priority for treatment 4) 5) 6) 7) 8) 9) 10) 11) 12) 13) 1) HCA 2) Central 3) NH Origin T&E Sens Social Good HCA Acres Acres Crit. Pri- Number Stand Site >1890 Strm Spp Value Access 5 mi 05/08 04/41 Score ority C1 1065-001 0 0 0 0 0 5 1 0.0 16.2 7.4 5 C2 1109-002 0 0 0 0 0 5 2 0.0 30.8 12.6 4 C3 1106-001 0 0 0 0 0 5 2 8.3 0.0 10.1 4 C4 1104-076 0 0 0 0 0 5 2 0.0 25.5 11.8 4 C5 1124-001 0 0 0 0 0 5 1 13.4 16.8 10.9 4 C6 2002-030 0 0 0 0 0 5 4 39.2 0.0 19.8 3 C7 2012-011 0 0 0 0 0 5 4 0.0 33.2 15.0 4 C8 2033-010 0 0 0 0 0 5 5 0.0 40.9 16.1 3 C9 2021-005 0 0 0 0 0 5 7 0.0 25.0 13.8 4 C10 2060-018 0 0 0 0 0 5 3 18.5 27.9 16.8 3 C11 2063-003 0 0 0 0 0 5 6 57.8 0.0 24.5 2 C12 2032-023 5 0 0 0 0 5 4 55.9 0.0 29.0 1 C13 2099-034 0 0 0 0 0 5 1 31.6 9.6 14.3 4 C14* 2023-016 5 0 0 0 1 5 6 0.0 3.5 16.5 3 C15* 2023-005 5 0 0 0 1 5 5 5.7 0.0 17.4 3 C16 2023-005 5 0 0 0 1 0 1 8.1 0.0 8.0 5 C17 2064-001 5 3 0 0 0 0 7 24.5 0.0 19.1 3 C18* 2024-002 5 0 0 0 1 0 5 0.0 30.8 15.6 3 C19 2031-010 5 0 0 0 1 5 4 15.1 39.9 25.7 1 C20 2036-012 5 0 0 0 0 0 5 0.0 0.0 10.0 5 C21* 2014-005 5 0 0 0 0 0 4 24.7 0.0 16.2 3 C22* 2014-021 5 0 0 0 0 0 4 12.5 0.0 13.1 4 L01 3049-009 0 0 0 0 0 5 2 16.7 0.0 12.2 4 L02 3071-042 0 0 0 0 0 5 1 9.0 0.0 7.3 5 L03 3078-001 0 0 0 0 0 5 2 6.8 0.0 9.7 5 L04 3113-012 0 0 0 0 0 5 2 12.7 0.0 11.2 4 L05 3141-030 0 0 0 0 0 5 2 5.6 0.0 9.4 5 L06 4032-001 0 0 0 0 0 5 4 27.8 0.3 17.0 3 L07 4041-019 0 0 0 0 0 5 7 19.5 8.1 16.1 3 L08 4034-014 0 0 0 0 0 5 6 65.0 0.0 26.3 1 L09 4043-022 0 0 5 0 0 5 9 48.5 0.0 27.1 1 L10 5006-001 0 0 5 5 0 5 8 41.9 0.0 30.5 1 L11 4054-034 0 3 0 0 0 5 11 33.7 0.0 21.4 2 L12 4057-001 0 0 5 0 0 5 6 50.9 0.0 27.7 1 L13 4071-004 0 3 0 0 0 5 5 35.1 0.3 21.8 2 L14 4063-019 0 0 0 0 0 5 8 60.6 0.0 25.2 1 L15 5018-014 0 0 0 5 0 5 8 19.2 12.4 21.7 2 L16 4077-009 0 0 0 0 0 5 9 28.7 0.0 17.2 3 L17 5027-004 0 0 0 5 0 5 9 44.7 0.0 26.2 1 L18 5017-002 0 0 0 0 0 5 7 23.4 0.0 15.8 3 L19 5030-006 0 0 0 5 0 5 8 17.7 0.0 19.4 3 L20 4105-052 5 3 0 0 0 5 10 19.2 3.2 23.3 2 L21 4091-008 0 3 0 5 0 5 9 16.1 0.0 22.0 2 L22 4201-016 0 0 0 5 0 5 9 47.4 0.0 26.8 1 L23 4208-009 0 0 0 0 1 5 1 13.7 0.0 9.4 5 L24 4213-013 0 0 0 0 1 5 2 27.2 0.0 15.8 3

60 Environmental Assessment Suppression of HWA Infestations, DBNF

4) 5) 6) 7) 8) 9) 10) 11) 12) 13) 1) HCA 2) Central 3) NH Origin T&E Sens Social Good HCA Acres Acres Crit. Pri- Number Stand Site >1890 Strm Spp Value Access 5 mi 05/08 04/41 Score ority L25 4216-003 0 0 0 0 0 5 4 16.9 0.0 14.2 4 L26 4214-023 0 3 0 0 0 5 7 40.4 0.0 23.1 2 L27 4221-020 0 3 0 5 0 5 5 33.6 0.0 26.4 1 L28 4231-030 0 3 0 0 0 5 7 29.7 0.0 20.4 2 L29 4235-023 0 0 0 0 0 5 3 9.2 0.0 10.3 4 L30 3138-803 5 3 0 0 1 5 1 0.0 0.0 14.0 4 L31 4038-017 5 3 0 0 1 0 6 55.8 0.0 27.9 1 L32 5012-025 5 3 0 0 1 5 5 0.0 0.0 19.0 3 L33 5030-024 5 3 0 0 1 0 9 18.4 0.0 18.6 3 S01 5044-022 5 0 0 0 1 5 9 44.8 0.0 27.2 1 S02 5065-014 0 0 0 0 0 5 9 36.7 0.0 19.2 3 S03 5076-041 0 0 0 0 0 5 7 10.1 0.0 12.5 4 S04 5071-020 0 0 5 5 0 5 7 39.7 0.0 29.9 1 S05 6007-012 0 0 0 0 0 5 9 23.9 0.0 16.0 3 S06 5086-022 0 3 0 0 1 5 7 0.0 0.0 14.0 4 S07 6083-001 0 0 0 0 0 0 4 0.0 0.0 5.0 5 S08 6017-002 0 0 5 5 0 5 10 11.5 0.0 22.9 2 S09 6014-005 0 3 5 5 1 5 6 0.0 0.0 24.0 2 S10 6037-011 0 3 0 0 0 0 5 49.4 5.5 21.2 2 S11 6039-013 0 0 5 5 0 5 10 17.5 0.0 24.4 2 S12 6046-011 0 0 5 5 0 5 7 36.6 0.0 29.1 1 S13 6237-006 0 0 0 0 0 0 6 42.4 0.0 15.6 3 S14 6239-035 0 0 5 5 1 5 6 14.4 0.0 24.6 2 S15 6061-021 0 0 5 5 1 5 6 4.8 0.0 22.2 2 S16 6059-014 0 0 0 0 1 5 4 31.6 0.0 18.9 3 S17 6069-047 0 0 5 5 1 0 2 0.0 0.0 14.0 4 S18 6117-037 0 0 5 5 0 5 1 20.9 0.0 20.2 2 S19 6128-005 0 0 5 5 0 0 1 35.2 0.0 18.8 3 S20 6264-004 0 3 5 5 0 5 0 23.8 0.0 24.0 2 S21 5033-010 5 0 0 0 1 5 3 5.6 0.0 15.4 3 S22 6019-016 5 3 0 0 1 5 9 39.2 0.0 28.8 1 S23 6008-002 5 0 0 0 1 5 7 3.0 0.0 16.7 3 S24* 5061-008 5 3 0 0 1 0 6 20.2 0.0 19.1 3 S25* 5073002 5 3 0 0 1 0 6 25.7 0.0 20.4 2 R1 3907-003 5 0 0 5 1 5 1 22.1 0.0 21.5 2 R2 1904-001 0 0 0 0 1 5 2 19.1 0.0 13.8 4 R3 2903-006 0 0 0 0 0 5 2 19.1 0.0 12.8 4 R4 0506-003 0 0 0 0 0 5 2 16.5 0.0 12.1 4 R5 0510-018 0 0 0 0 0 5 3 9.0 0.0 10.3 4 R6 0401-003 0 0 0 5 0 5 1 13.0 0.0 13.3 4 R7 0501-008 0 0 0 0 0 5 1 16.9 0.0 9.2 5 R8 0901-003 0 0 0 0 1 5 2 0.0 0.0 9.0 5 R9 3113-011 0 0 0 0 0 5 1 0.0 1.9 5.3 5 R10 0806-016 0 0 0 0 0 5 0 11.0 0.0 7.7 5 R11 3604-003 0 0 0 0 0 5 1 15.6 0.0 8.9 5 R12 3204-041 0 0 0 0 0 5 1 6.6 0.0 6.6 5

61 Environmental Assessment Suppression of HWA Infestations, DBNF

APPENDIX E – Hemlock Conservation Area Maps

This appendix contains five small scale (1:250,000) maps indicating the locations of each Hemlock Conservation Area (HCA). The table below indicates which HCAs are in alternatives B and C. Larger scale (1:48,000) maps are available at the Forest’s website http://www.fs.fed.us/r8/boone/planning/project/index.shtml, and at the DBNF Supervisor’s Office in Winchester, KY.

Table E-1. Hemlock Conservation Areas within Alternatives B and C Map Alternative B Alternative C 1 of 5 – Cumberland District C01 – C13 C01 – C22 2 & 3 of 5 – London District L01 – L29 L01 - L33 4 of 5 – Stearns District S01 – S20 S01 – S25 5 of 5 – Redbird District R01 – R12 R01 – R12 (74 HCAs) (92 HCAs)

62 Environmental Assessment Suppression of HWA Infestations, DBNF

Figure E-1. HCAs on the Cumberland District

63 Environmental Assessment Suppression of HWA Infestations, DBNF

Figure E-2. HCAs on the London District (North end)

64 Environmental Assessment Suppression of HWA Infestations, DBNF

Figure E-3. HCAs on the London District (south end)

65 Environmental Assessment Suppression of HWA Infestations, DBNF

Figure E-4. HCAs on the Stearns District

66 Environmental Assessment Suppression of HWA Infestations, DBNF

Figure E-5. HCAs on the Redbird District

67 Environmental Assessment Suppression of HWA Infestations, DBNF

APPENDIX F – Cost Estimates for HWA Suppression

Detection of Infestations: In advance of treatment of hemlock conservation areas (HCA) with insecticide or releasing beetles in the forest, infested sites must be identified, located on the ground with GPS, and flagged. We estimate that $24,000 will be required per year to check for new infestations. This would cover about 10 days each for two persons on each of the 4 districts. Effectiveness monitoring may be done by FHP and/or other cooperators and is not considered in this cost analysis.

Release of predator beetles in stands of HWA-infested hemlocks: There are approximately 150-160 drainages that contain hemlock on the Daniel Boone National Forest. Within all of these areas, predator beetles (as they become available) could be eventually released. For purposes of this estimate, the horizon for this activity could be 5-10 years, or until effective, or until better control methods are found. Selection of sites for treatment each year would be based on detection results including degree of infestation, abundance of hemlocks, health of trees, and accessibility of live branches. If treatment resources become limited, infested HCA will have priority over other infestations, and appendix D will guide prioritization within HCA.

The actual job of releasing predator beetles would be primarily done by DBNF personnel, or initially trained by Forest Health Protection (FHP) personnel. Cooperators and volunteers may be recruited to help with this task. Depending on availability of beetles, anywhere from several hundred to several thousand beetles may be released in each infestation.

For this estimate, we will assume that 2000 predator beetles would be placed on trees at each of 100 locations each year (200,000 beetles /year). The cost of the beetles is about $1.00/beetle x 200,000 beetles = $200,000 /year. This cost would be covered by FHP.

Placement of beetles on HWA-infested branches and transportation to and from these remote and scattered areas would include time and travel expense. We estimate that three release sites could be visited per day by two or three personnel. Transportation and overhead costs would bring the cost per site to about $245 per site. This would be about $24,500 /year (100 sites/year x $245/site), taking slightly over 30 days.

Total cost for beetle releases at 100 sites per year would be approximately $200,000 for beetles plus $24,500 each year for placing beetles on infested trees, or about $2,245 per release site.

Soil injection with imidacloprid insecticide: Each year, up to half (about 40) of the HCA would be treated with insecticide as hemlock trees become infested with HWA. Approximately 20 hemlock trees on an area up to one acre in size (radius = 118’) would be treated at each of 3 sites within an HCA. A fully-stocked hemlock stand (basal area 130 sq ft/ac) with an average diameter of 13” dbh would have about 140 trees per acre. If 100 of these trees on that acre were hemlocks, and we treat 20 trees per acre, then 20% of the hemlocks on that acre would be treated. If three sites were treated within a (e.g.) 125 acre conservation area, then (3/125) x 0.20 = ½ of one percent of the trees in the 125 acre

68 Environmental Assessment Suppression of HWA Infestations, DBNF

conservation area would be treated. For this estimate, if 20 trees per acre were treated each year on 3 sites within each of 40 infested HCA, we estimate that about 2,400 trees per year would be treated on about 120 acres.

According to label directions, a maximum of 8.6 ounces of the unmixed insecticide can be applied per acre per year (8.6 ounces of powder currently costs about $81). Each packet of (75% imidacloprid) powder is mixed with 1.5 quarts of water, so 8.6 ounces of powder would make 266 ounces of mixture (2.4% imidacloprid). Using soil injection with a Kioritz Soil Injector, one ounce of mixture is applied to the root zone of a hemlock per inch of tree diameter at breast height (dbh). Therefore, a maximum of 266 cumulative dbh inches can be applied to an acre in a given year. For example, this could be 44 six-inch trees, 8 thirty-three-inch trees on an acre, or many other combinations.

Treated trees would be painted and tagged with a tree number. Also, the date treated, HCA#, tree #, diameter, degree of infestation, and crown condition would be recorded.

We estimate that about six hours would be needed by three people to tag, measure, and inject the soil around 20 trees at each of three sites within an HCA. Two hours per day would be needed for transportation, preparation, and cleanup. This would require about 6 gallons of insecticide in water to be mixed, carried, and loaded into the injector. At the end of a day, equipment would need to be cleaned, calibrated, and properly stored for use the next day. Other costs include additional equipment, leave and admin time, supervision, and Forest overhead. Total cost for soil injecting about 3 acres within each of 40 HCA would be approximately $662.50 for each HCA treated. This would be approximately $26,500 per year.

Summary of annual suppression costs: HWA detection costs would be about $24,000 per year. The funds needed for predator beetle releases for HWA suppression by the DBNF in a given fiscal year would be $24,500. This excludes the $200,000 for beetles, and other FHP overhead costs. The funds needed for chemical suppression of HWA using soil injection on the DBNF in a given fiscal year would be $26,500. Total cost for the DBNF would be about $75,000 per year.

69 Environmental Assessment Suppression of HWA Infestations, DBNF

Works Cited Allendorf, F.W., and N. Ryman. "The role of genetics in population viability analysis." In Population Viability Analysis, by Steven R. Beissinger and Dale R. McCullough, 50-85. Chicago, London: The University of Chicago Press, 2002. Anatra-Cordone, Michelle, and Patrick Durkin. Imidacloprid - Human Health and Ecological Risk Assessment - Final Report. Fayetteville, New York: Syracuse Environmental Research Associates, Inc., 2005, 283. Bacey, Juanita. Environmental Fate of Imidacloprid. Environmental Monitoring & Pest Management Branch, Sacramento, CA: Department of Pesticide Regulation, 2000, 8. Butin, Elizabeth, Michael Montgomery, Nathan Havill, and Joseph Elkinton. "Pre-release Host Range Assessment for Classical Biological Controls: Experience with." Proceedings: Hemlock Woolly Adelgid in the Eastern United States Symposium, February 5-7, 2002. East Brunswick, New Jersey, 2002. Cheah, C.A.S., and M. S. McClure. "Pseudoscymnus tsugae in Conneticutt forests: the first five years." Proceedings, hemlock wooly adelgid in the Eastern United States symposium. East Brunswick, NJ: Rutgers University, 2002. 150-165. Cheah, Carole, Mike Montgomery, Scott Salem, Bruce Parker, Margaret Skinner, and Scott Costa. Biological Control of Hemlock Woolly Adelgid. Morgantown, West Virginia: USDA- FS Forest Health Technology Enterprise Team, 2004. Crane, Barbara, interview by Ed Brown. US Forest Service Geneticist, Region 8 Atlanta, GA, (2004). Evans, R. A. "An ecosystem unraveling?" Edited by B. Onken, R. Reardon and J. Lashomb. Proceedings, hemlock woolly adelgid in the eastern United States symopsium. East Brunswick, NJ: Rutgers University, 2002. 22-33. Evans, R. A., et al. "Potential impacts of hemlock woolly adelgid (Adelges tsugae) on eastern hemlock (Tsuga canadensis) ecosystems." Eastern U.S. Proceedings of the First Hemlock Woolly Adelgid Review. Charlottesville, VA: U.S. Department of Agriculture, Forest Service, 1996. 129. Foose, T.J. "The relevance of captive populations to the conservation of biotic diversity." In Genetics and Conservation, by C.M. Schonewald-Cox, S.M. Chambers, B. MacBryde and W.L. Thomas, 374-401. Menlo Park, CA: Benjamin/Cummings Publishing, 1983. Franklin, I.R. "Evolutionary change in small populations." In Conservation biology: an evolutionary-ecological perspective, by M.E. Soule and B.A. Wilcox, 135-139. Sunderland, MA: Sinauer Associates, 1980. Franklin, I.R., and R. Frankham. "How large must population be to retain evolutionary potential?" Animal Conservation 1 (1998): 69-70. Gilpin, M. E., and M. E. Soule. "Minimum viable populations: Processes of species extinction." In Conservation Biology: The Science of Scarcity and Diversity, by M. E. Soule, 19-34. Sunderland: Sinauer & Associates, 1986. Hartmann, H.T., and D.E. Kester. Plant Propagation, Principles and Practices. Englewood Cliffs, NJ: Prentice-Hall, Inc., 1983.

70 Environmental Assessment Suppression of HWA Infestations, DBNF

Hennessey, R. Field Release of a Nonindigenous Lady Beetle, Pseudoscymnus sp. (Coleoptera:Coccinellidae), for Biological Control of Hemlock Woolly Adelgid, Adelges tsugae (Homoptera: Adelgidae). Environmental Assessment, Riverdale, MD.: USDA- APHIS, 1995. Kunkel, Brian A., David W. Held, and Daniel A. Potter. "Impact of Halofenozide, Imidacloprid, and Bendiocarb on Beneficial Invertebrates and Predatory Activity in Turfgrass." Journal of Economic Entomology 92, no. 4 (1999): 922-930. LaCava, J., and J. Hughes. "Determining minimum viable population levels." Wildl. Soc. Bull., 1984: 370-376. Lande, R. "Mutation and conservation." Conservation Biology 9 (1995): 782*791. Landres, Peter B., Mark W. Brunson, Linda Merigliano, Chrisse Sydoriak, and Steve Morton. "Naturalness and Wildness: The Dilemma and irony of Managing Wilderness." Proceedings RMRS. 2000. 15. LaSorte, Frank A, Frank R Thompson, Margret K Trani, and Tim J Mersman. Population Trends and Habitat Occurence of Forest Birds on Southern National Forests, 1992-2004. GTR, Newton Square: USDA Forest Service, 2007. Lisle, T. E. "Stabilization of a gravel channel by large streamside obstructions and bedrock bends in Jacoby Creek, northwestern California." Geological Society of America Bulletin 97 (1986): 999-1011. Lu, Wenhua, and Michael E. Montgomery. "Oviposition, development, and feeding of Scymnus (Neopullus) sinuanodulus (Coleoptera: Coccinellidae): a predator of Adelges tsugae (Homoptera: Adelgidae)." Annals of the Entomological Society of America 94, no. 1 (January 2001): 64-70. Lynch, M., and R. Lande. "The critical effective size for a genetically secure population." Animal Conservation 1 (1998): 70-72. Lynch, Michael J., John Conery, and Reinhard Burger. Mutation accumulation and the extinction of small populations. Vols. 146-4, in The American Naturalist. Chicago, IL: The University of Chicago Press, 1995. Mausel, D. L., and etal. "Establishment of the Hemlock Woolly Adelgid Predator, Laricobius nigrinus (Coleoptera: Derondontidae), in the Eastern United States." Environmental Entomology 39 (2010): 440-448. Mayer, Mark, Robert Chianese, Thomas Scudder, Jeffrey White, Khamkeo Vongpaseuth, and Robert Ward. "Thirteen Years of Monitoring the Hemlock Woolly Adelgid in New Jersey Forest." Proceedings: Hemlock Woolly Adelgid in the Eastern United States Symposium 2002. New Jersey Department of Agriculture, Division of Plant Industry, 2002. Miles Inc. "Imidacloprid: Pesticide leaching potential model. Report No. 105008." 1993. Montgomery, M.E., S.M. Lyon, W. Lu, D. Yao, and H. Wang. "Developing biological control technology: evaluation of the feeding range of predaceous beetles." Resource technology 1997: Beijing international symposium proceedings. Beijing, China: China Forestry Publishing House, 1997. 28-33. Moza, P. N., K. Hustert, E. Feicht, and A. Kettrup. "Photolysis of imidacloprid in aqueous solution." Chemosphere 36, no. 3 (1998): 497-502.

71 Environmental Assessment Suppression of HWA Infestations, DBNF

Namkoong, G. "Genetic structure of forest tree populations;." Edited by V.L. Chopra, B.C. Joshi, R.P. Sharma and H.C. Bansal. Proceedings of the 15th International Congress of Genetics. New Delhi, India: Mohan Primlani, Oxford, and IBH, 1984. 351-360. National Academy of Sciences. Managing Global Genetic Resources: Forest Trees. Edited by Board on Agriculture and National Research Council. Washington, DC: National Academy Press, 1991. Noss, Reed F. Biodiversity in the Blue Mountains: a Framework for Monitoring and Assessment. Paper produced for distribution at eh 1992 Blue Mountains Biodiversity Conference, Walla Walla, WA: Whitman College, 1992. Palmer-Ball, Brainard. The Kentucky Breeding Bird Atlas. Lexington, KY: The University Press of Kentucky, 1996. Pond, Gregory, and S. E. McMurray. A Macroinvertebrate Bioassessment Index for Headwater Streams of the Eastern Coalfield Region, Kentucky. Division of Water, Water Quality Branch, Frankfort, KY 40601: Kentucky Department for Environmental Protection, 2003. Reed, J.M., P.D. Doerr, and J.R. Walters. "Minimum viable population size of the red-cockaded woodpecker." J. Wildl. Manage. 52 (1988): 385-391. Ross, R.M., L.A. Redell, and R.M. Bennett. "Mesohabitat use of threatened hemlock forests by breeding birds of the Delaware Water Gap National Recreation Area." Edited by B. Onken, R. Reardon and J. Lashomb. Proceedings: hemlock woolly adelgid in the eastern United States symposium. East Brunswick, NJ, 2002. 353. Rouchaud, J., F. Gustin, and A. Wauters. "Soil Biodegradation and Leaf Transfer of Insecticide Imidacloprid Applied in Seed Dressing in Sugar Beet Crops." Bull. Environ. Contam. Toxicol. 53 (1994): 344-350. Salom, S. "Evaluation of Laricobius nigrinus as a Potential Candidate for Biocontrol of HWA." Hemlock Woolly Adelgid Newsletter, September 1998. Shaffer, M.L. "Minimum population sizes for species conservation." Bioscience 31 (1981): 131- 134. Shaffer, M.L. "Minimum viable populations: coping with uncertainty." 1987: 69-86. Silcox, Charles A. Using Imidacloprid to Control Hemlock Woolly Adelgid. Perkasie, PA: Bayer Corporation, 2002. Soule, M.E. "Thresholds for survival: Maintaining fitness and evolutionary potential." In Conservation biology: An evolutionary-ecological perspective, by M.E. Soule and B.A. Wilcox, 151-169. Sunderland, MA: Sinauer Assoc., 1980. Steward, V.B., and T.A. Horner. "Control of hemlock woolly adelgid using soil injections of systemic insecticides." Journal of Arboriculture 20 (1994): 287-288. Steward, V.B., G. Braness, and S. Gill. "Ornamental pest management using imidacloprid applied with the Kioritz Soil Injector." Journal of Arboriculture 24 (1998): 344-346. Taylor, David, interview by Joe Metzmeier. Forest Botanist/Ecologist Winchester, KY, (January 13, 2010). USDA-FS. A Wilderness Agenda: Thinking Like a Mountain. Washington, DC: Recreation, Heritage, and Wilderness Resources Program, 2000.

72 Environmental Assessment Suppression of HWA Infestations, DBNF

USDA-FS. Final Environmental Impact Statement for the Land and Resource Management Plan for the Daniel Boone National Forest. Winchester, KY: USDA Forest Service, Daniel Boone National Forest, 2004a. —. Forest Health Protection - Hemlock Woolly Adelgid. 2006b. http://na.fs.fed.us/fhp/hwa/newsletters/news06.pdf (accessed August 2010). —. Forest Health Protection. 2010a. http://na.fs.fed.us/fhp/hwa/ (accessed 2010). —. Forest Inventory Data Online. 2006a. http://fiatools.fs.fed.us/fido/index.html (accessed July 2010). —. Forest Service Directives. 2010. http://www.fs.fed.us/im/directives/ (accessed August 2010). USDA-FS. Land and Resource Management Plan for the Daniel Boone National Forest. Winchester, Kentucky: Management Bulletin R8-MB 117A, 2004b. USDI-FWS. Daniel Boone National Forest Federally Listed Species and Designated Critical Habitats. Frankfort, KY: Kentucky Ecological Services Field Office, 2010a. —. "Endangered and Threatened Wildlife and Plants; Listing the Cumberland Darter, Rush Darter, Yellowcheek Darter, Chucky Madtom, and Laurel Dace as Endangered Throughout Their Ranges." Federal Register, June 24, 2010b: 36035-36057. —. "Endangered and Threatened Wildlife and Plants; Listing the Rayed Bean and Snuffbox as Endangered." Federal Register, November 2, 2010c: 67552-67583. USDI-FWS. Indiana Bat (Myotis sodalis) Draft Recovery Plan: First Revision. Fort Snelling, MN: U.S. Fish and Wildlife Service, 2007, 258. USDI-FWS. Running Buffalo Clover (Trifolium stoloniferum) Draft Recovery Plan: First Revision. Fort Snelling, MN: U.S. fish and Wildlife Service, 2007, 6. USDI-FWS. "Species profile for American chaffseed (Schwalbea americana)." 1995. USDI-FWS. "Species profile for Cumberland rosemary (Coradina verticillata)." 1997a. USDI-FWS. "Species profile for Cumberland sandwort (Arinaria cumberlandensis)." 1996. USDI-FWS. "Species profile for Virginia spiraea (Spiraea virginiana)." 1997b. US-EPA. Imidacloprid Pesticide Fact Sheet. Washington, DC: U.S. Environmental protection Agency, 1994. Wamhoff, H., and V. Schneider. "Photodegradation of imidacloprid." Journal of Agriculture Food Chemistry 47, no. 4 (1999): 1730-1734. Webb, Ralph E., J. Ray Frank, and Michael J. Raupp. "Eastern Hemlock Woolly Adelgid Damage Following Imidacloprid Therapy." Journal of Arboriculture 29, no. 5 (2003): 298- 302. Wilderness.net. 2010. http://www.wilderness.net/index.cfm?fuse=MRDG (accessed August 2010). Zheng, W., and W. Liu. "Kenetics and mechanism of the hydrolysis of imidacloprid." Pesticide Science 55 (1999): 482-485. Zilahi-Balogh, G.M.G. Biology of Laricobius nigrinus Fender (Coleoptera: Derodontidae) and its potential as a biological control agent of the hemlock woolly adelgid, Adelges tsugae Annand (Homoptera: Adelgidae) in the eastern United States. Ph.D. dis. Ph.D. dissertation, Blacksburg, Virginia: Virginia Polytechnic Institute and State University, 2001.

73 Environmental Assessment Suppression of HWA Infestations, DBNF