Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

United States Department of Environmental Agriculture

Forest Assessment Service

September 2011 2011 Suppression of Hemlock Woolly Adelgid Infestations

Cherokee National Forest Carter, Cocke, Greene, Johnson, McMinn Monroe, Polk, Sullivan, Unicoi, and Washington Counties, TN

For Project Information Contact: Bob Lewis, HWA Team Leader 2800 Ocoee Street N. Cleveland, TN 37312 423.476.9700

The U.S. Department of Agriculture (USDA) prohibits discrimination in all its programs and activities on the basis of 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 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Table of Contents

Summary ...... i Chapter 1 Introduction ...... 1 Background ...... 1 Purpose and Need for Action ...... 2 Proposed Action ...... 3 Decision Framework ...... 5 Public Involvement ...... 5 Issues ...... 5 Chapter 2 Alternatives ...... 7 Alternatives ...... 7 Alternative 1 -- No Action ...... 7 Alternative 2 -- The Proposed Action ...... 7 Alternative 3 -- No Additional Chemicals ...... 10 Alternatives Considered but Eliminated from Detailed Study ...... 10 Alternative 4: Include additional proposed treatments in Wilderness areas ...... 10 Mitigation Common to All Alternatives ...... 10 Comparison of Alternatives ...... 12 Chapter 3 Environmental Consequences ...... 13 Biological Resources...... 13 Threatened (T), Endangered (E), Sensitive (S), Viability Concern (VC) & Demand Species ..32 Recreation ...... 83 Wilderness Resources ...... 85 Climate Change ...... 86 Road Management ...... 90 Soils/Groundwater ...... 90 Heritage Resources ...... 92 Human Health and Safety ...... 92 Economics ...... 93 Consultation and Coordination ...... 95 Literature Cited and references ...... 98

Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

SUMMARY

In 2005, the US Forest Service Regional Forester issued a decision for The Suppression of Hemlock Woolly Adelgid (HWA) Infestations on the Cherokee National Forest which allowed for treatment of woolly adelgid infestations affecting eastern and Carolina hemlock on 91 sites across the Cherokee National Forest (CNF) using a combination of biological (beetle release) and chemical (imidacloprid) treatments. In 2010 the Cherokee National Forest produced a Supplemental Information Report, which allowed a limited increase in treatment on some of the non- wilderness sites.

In the six years since that decision, HWA populations have continued to spread and decimate hemlocks across the forest. The decline of hemlock is much more rapid than originally anticipated and more aggressive management is needed to protect the remaining stands of hemlock while long term solutions are developed. We currently have the opportunity to protect a number of eastern and Carolina hemlock stands that are still relatively intact, by utilizing additional treatment methods and by broadening the scope of treatment areas. Project objectives are to: (1) Reduce hemlock mortality from HWA by establishing reproducing populations of predator beetles that feed on HWA; (2) Maintain reproducing populations of eastern hemlock and Carolina hemlock throughout the historical geographic and elevational range across the Forests, and; (3) Ensure survival of ecologically and culturally important groups of hemlocks.

Progression of HWA in the Eastern U.S.

ii Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

CHAPTER 1 INTRODUCTION Background ______The USDA Forest Service has statutory responsibility to take steps to preserve the diversity of tree species in the forest. The Hemlock Woolly Adelgid (HWA, Adelges tsugae) is a catastrophic introduced pest of hemlock trees in the eastern United States. Without long-term control of HWA there may be devastating impacts to southern Appalachian ecosystems as eastern and Carolina hemlocks are lost throughout their range. HWA began attacking hemlock along the east coast in the 1950’s and it currently infests about one-half of the area where hemlocks grow in the eastern U.S. It was first detected on the Cherokee National Forest in the early-2000’s. Early in 2004 it became obvious that infestation was widespread and many trees were heavily infested with visible signs of decline. Current monitoring indicates that over half the range of hemlock in the eastern U.S. is now infested and that the entire range of eastern hemlock is at risk. The 2005 Decision by the Regional Forester for The Suppression of Hemlock Woolly Adelgid Infestations (2005 Decision) allows for the treatment of 91 sites using HWA predatory beetles and/or the insecticide imidicloprid. The 91 conservation areas were selected to meet the requirements of a hemlock conservation network designed to represent community diversity within the distribution of known Carolina and eastern hemlock stands. These conservation sites are located across the general forest area including wilderness. Conservation areas are identified as plant communities in which Carolina or eastern hemlock are the predominant tree species and which contain an adequate number of trees to ensure genetic diversity within the population. In 2010 the Cherokee National Forest produced a Supplemental Information Report, which allowed a limited increase in treatment on some of the non- wilderness sites. The HWA predatory beetle species currently approved for release have had their biology and environmental safety thoroughly evaluated, and meet USDA risk assessment criteria for release (Hennessey, R. 1995, Salom, S. 1998, Zilahi-Balogh, G.M.G. 2001, Montgomery et al. 1997, Lu and Montgomery 2001, Butin et al. 2002). In addition to beetle releases, ongoing treatments include the application of the chemical imidacloprid in specified conservation areas, including Wilderness areas. Imidacloprid is a widely used systemic insecticide with applications ranging from agricultural food production to flea control on pets. Unlike common aerosol insecticides that may injure or kill non-target species, systemics are absorbed by the host plant and kill only those insects that feed upon the host. Imidacloprid is classified as a neonicotinoid (nicotine- like) pesticide with remarkably high insecticidal activity against plant-sap-sucking insects, such as HWA. It has proven to be a valuable treatment for ensuring hemlocks remain alive until effective biocontrols become established.

1 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Purpose and Need for Action ______Careful consideration went into the selection of conservation areas during the 2005 analysis to ensure that adequate eastern and Carolina hemlock genetic diversity was conserved. A conservation strategy was designed to maintain genetic, species, community and landscape diversity with consideration for capturing representative communities across different environmental strata and ecological zones. Over the last six years some of the originally identified conservation areas are no longer viable and therefore the Forest is expanding the scope of treatments outside of the original conservation design. Due to the fact that Carolina hemlock is a sensitive species with a limited range, there is additional emphasis on locating and treating Carolina hemlock sites. In addition to ongoing chemical and biological treatments to protect important hemlock areas, the Forest Service is also working with Camcore under a cooperative agreement to collect hemlock cones in an effort to establish ex situ (off-site) seedling establishment. The overall goal of the project is to preserve hemlock gene pools in perpetuity until a time when effective HWA management strategies are in place and conserved seed resources can be utilized to restore hemlock throughout its native range (Jetton et al. 2008). Monitoring over the last six years indicates that infested hemlock areas that have not received chemical treatment have continued to decline and hemlocks in many of the areas have been killed by the HWA infestations. The rate of HWA spread across the forest is more rapid than originally thought, and the loss of hemlocks has had devastating effects on the ecosystem. The effect of the predator beetle treatments on the Cherokee NF are not conclusive at this time. There is a pressing need to supplement the HWA control efforts with other treatments to ensure adequate numbers of trees remain viable within each conservation area. Additionally we need to ensure that enough conservation areas receive effective treatments to maintain hemlocks across the landscape. The purpose of this proposal is to ensure that we continue to use the most effective suite of treatment options to protect as many viable stands of remaining hemlocks as possible. There is a need to: (1) Utilize additional bio-control agents for long-term control of HWA across the landscape; (2) Preserve reproducing hemlock populations across the landscape as seed sources for future re-establishment; (3) Allow greater flexibility with chemical treatments to support the protection of genetically diverse hemlock populations across the landscape; and (4) Increase the number of treatment sites in an effort to conserve viable hemlock populations.

2 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Proposed Action ______The action proposed by the Forest Service to meet the purpose and need is Authorize the use of fungus treatments, Lecanicillium muscarium, as a biocontrol agent for HWA. Include the chemical dinotefuran, SafariTM, as a potential pesticide treatment in hemlock areas identified as having heavy infestation and needing immediate attention for the survival of the trees. Expand the scope and flexibility of potential treatment areas beyond those identified in the original conservation design. Include the use of additional predator beetle species as they become available and are evaluated for effectiveness. Expand the use of imidacloprid as an option on all hemlock treatment areas, consistent with label direction. This proposal to add treatment options does not apply to designated Wilderness areas or wilderness study areas. HWA predator beetle and chemical treatments in Wilderness will continue under the 2005 Decision but no additional treatments are being proposed in Wilderness or wilderness study areas at this time. Any additional treatments in wilderness will be disclosed in a separate environmental assessment in the future.

Monitoring In addition to the specific proposed actions, this proposal incorporates monitoring to ensure that the appropriate and most effective treatments are being used. When responding to a forest threat such as an invasive insect it is essential that treatments adapt to changing conditions in the forest. As new areas of infestation are identified they will be evaluated for treatment within the broader context of the landscape. Monitoring for this project will address the following areas: 1. Establishment, dispersal, and effectiveness of biocontrols; 2. Water quality in areas receiving chemical treatments; 3. Effectiveness of chemical treatments and application methods; and 4. Locations of conservation areas as new critical populations of eastern and Carolina hemlock are identified. While existing and proposed treatments include the use of chemicals it is important to recognize that chemical treatments are an essential short-term treatment to preserve declining hemlock populations. As monitoring indicates that adequate biocontrols have become established in the forest and are effective at controlling HWA populations, there will be less reliance on chemical treatments.

3 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest excluding Wilderness areas. The intent of the fungus treatment is to treat a select number of demonstration sites, with the long-term objective of treating all hemlock conservation areas if treatments in the demonstration area are shown to be effective. Lecanicillium muscarium occurs naturally in Eastern Tennessee and has been shown to cause significant reductions in adelgid populations in small scale field trials. Initial studies show that an application of the fungus in a whey-based solution (cheese by- product) enables the fungi to better persist in the environment and have lasting effects on the HWA population (Costa 2009). The fungus has been labeled for use in the UK as the product Mycotal. An evaluation for European registration that included human health, environmental and ecotoxicological analysis concluded that Mycotal is not considered harmful to the environment. Research conducted in the laboratory and field with L. muscarium found no negative impacts on a predatory beetle being released for hemlock woolly adelgid management (Reardon and Onken 2004). Based on initial field studies with Mycotal, there is little concern associated with the effects of the fungus on non- target insects.

Dinotefuran (SafariTM Treatment) SafariTM is a trade name for an insecticide that has the chemical dinotefuran as the active ingredient. It is a relatively new insecticide that the U.S. Forest Service is considering for use to control HWA. It has been used effectively in the Great Smoky Mountain National Park as well as on private land in the southern Appalachians. Similar to imidacloprid, dinotefuran is a systemic insecticide that is absorbed by the tree and kills only those insects that feed upon the tree. It is highly water soluble and therefore much faster acting than imidacloprid. It has been shown to be effective at treating large heavily infested trees that are on the edge of decimination and may not be responding to the slower acting imidaclopid treatments. Current research documents that dinotefuran has been shown to effectively reduce adelgids on branch tips within two weeks (Cowles and Lagalante 2009). Dinotefuran treatments would be used in areas prioritized for attention due to their advanced level of infestation and need for immediate treatment. When used in combination with imidacloprid, infested trees would receive both the short term benefits of dinotefuran along with the longer lasting protection attributed to imidacloprid. Because of its greater mobility in the soil, there is a greater risk to contamination of aquatic systems from the application of dinotefuran. However, risks to aquatic systems would be minimized through the strict adherence to label direction which sets guidelines regarding application near water.

Expanded Scope of Conservation Areas The 2005 Decision identified 91 eastern and Carolina hemlock areas for potential treatment. Areas were selected to meet the requirements of a hemlock conservation network designed to represent community diversity within the distribution of known hemlock stands on the Cherokee National Forest. Monitoring in recent years has indicated that some of the original sites have already been decimated by the HWA, beyond the point of preserving viable trees. Additional sites continue to be identified

4 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest across the forest and are proposed for treatment to supplement the larger hemlock conservation network.

Expanded Use of Imidacloprid The 2005 Decision identified specific hemlock treatment areas that would receive application of imidacloprid in order to ensure that genetically diverse hemlocks remain alive until biological controls take effect. This proposal would allow greater flexibility of using imidacloprid on any of the identified hemlock treatment areas as necessary. The beetles that have been released on the Cherokee National Forest for HWA control have not demonstrated success at this time, but their populations are not yet high enough to maintain viable trees without the supplemental use of chemical treatments. It is necessary to maintain the option of using chemical treatments until the biocontrol agents become sufficiently established in the ecosystem. Once biocontrols become adequately abundant in the ecosystem and effective at controlling HWA populations, there will be less reliance on chemical treatments. The 2005 Decision specified that soil injection of imidacloprid would not be used in close proximity to water or in highly permeable (sandy or gravelly) soils. Stem injection was previously the preferred method of imidacloprid application in these areas. However, aquatic monitoring studies at Coweeta Research Station show no detectable levels of imidacloprid in streams following soil application. This revision to the EA would allow less restrictive application of imidacloprid in areas near flowing aquatic systems, consistent with label direction, and with continued periodic aquatic monitoring to ensure there are no detectable traces of chemicals in the streams. Decision Framework ______Given the purpose and need, the deciding official reviews the proposed action and the other alternatives in order to make the following decision: whether or not to expand the treatment options and number of areas for suppression of HWA infestation beyond what was identified in the 2005 Decision and the 2010 Supplemental Information Report. Public Involvement ______A detailed proposal with preliminary issues and alternatives was provided to the public and other agencies and tribes for comment beginning April 26, 2011. One comment letter was received from the Eastern Band of the Cherokee Indians. The comment states that proposal does involve ground disturbance which might adversely affect sites which are eligible for inclusion on the National Register of Historic Places. Issues ______The April 26, 2011 scoping and 30-day notice and comment letter that was mailed to the public outlined two potential issues and three alternatives. These issues were based on public comments received during the original 2005 analysis and recent internal scoping. The comment received during the 30-day notice and comment period was evaluated to determine whether there were any additional issues regarding the proposed action. The

5 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Forest Service defined significant issues as those directly or indirectly caused by implementing the proposed action. Non-significant issues were identified as those: 1) outside the scope of the proposed action; 2) already decided by law, regulation, Forest Plan, or other higher level decision; 3) irrelevant to the decision to be made; or 4) conjectural and not supported by scientific or factual evidence. The Council for Environmental Quality (CEQ) NEPA regulations require this delineation in Sec. 1501.7, “…identify and eliminate from detailed study the issues which are not significant or which have been covered by prior environmental review (Sec. 1506.3)…” Based on the comment received, there were no additional issues identified with the proposed action. The two significant issues to be analyzed in this environmental assessment are: Issue 1 The aerial application of fungus may have negative effects on non-target insects in the forest. Issue 2 Due to its relatively high mobility in soil, there is concern regarding the use of dinotefuran as a chemical treatment and the possible environmental effects near aquatic ecosystems.

6 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

CHAPTER 2 ALTERNATIVES

This chapter describes and compares the alternatives considered for the 2011 HWA Suppression project. It includes a description of each alternative considered. These alternatives were developed by the interdisciplinary team in response to the purpose and need and the issues identified for this project. Mitigation measures for activities in each alternative, if any, are also described in this chapter. Alternatives ______

Alternative 1 -- No Action The No Action alternative consists of continuing suppression of the HWA under the existing 2005 Decision and the 2010 Supplemental Information Report. Treatments include application of imidacloprid and beetle releases on identified hemlock conservation areas. This alternative responds to issues 1 and 2 regarding the expanded scope of treatment options. Under this alternative there would be no change in our current management strategy. The Forest would continue to treat conservation areas but we would be missing opportunities to utilize the best treatment strategies to combat the spread of HWA.

Alternative 2 -- The Proposed Action The Cherokee National Forest proposes the following: Authorize the use of fungus treatments, Lecanicillium muscarium, as a biocontrol agent for HWA. Include the chemical dinotefuran, SafariTM, as a potential pesticide treatment in hemlock areas identified as having heavy infestation and needing immediate attention for the survival of the trees. Expand the scope and flexibility of potential treatment areas beyond those identified in the original conservation design. Include the use of additional predator beetle species as they become available and are evaluated for effectiveness. Expand the use of imidacloprid as an option on all hemlock treatment areas, consistent with label direction. This proposal to add treatment options does not apply to designated Wilderness areas or wilderness study areas. HWA predator beetle and chemical treatments in Wilderness will continue under the 2005 Decision but no additional treatments are being proposed in Wilderness or wilderness study areas at this time. Any additional treatments in wilderness will be disclosed in a separate environmental assessment in the future.

7 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Monitoring In addition to the specific proposed actions, this proposal incorporates monitoring to ensure that the appropriate and most effective treatments are being used. When responding to a forest threat such as an invasive insect it is essential that treatments adapt to changing conditions in the forest. As new areas of infestation are identified they will be evaluated for treatment within the broader context of the landscape. Monitoring for this project will address the following areas: Establishment, dispersal, and effectiveness of biocontrols; Water quality in areas receiving chemical treatments; Effectiveness of chemical treatments and application methods; and Locations of conservation areas as new critical populations of eastern and Carolina hemlock are identified. While existing and proposed treatments include the use of chemicals it is important to recognize that chemical treatments are an essential short-term treatment to preserve declining hemlock populations. As monitoring indicates that adequate biocontrols have become established in the forest and are effective at controlling HWA populations, there will be less reliance on chemical treatments.

Fungus Treatment While predator beetles have shown to be effective at reducing HWA populations in localized areas, it is likely that a complex of natural enemies (introduced predatory insects and diseases), rather than a single “magic bullet,” would be needed to maintain hemlock woolly adelgid below damaging levels (Ward et al. 2004). The insect-killing fungus Lecanicillium muscarium is being proposed for aerial application to hemlock conservation areas, excluding Wilderness areas. The intent of the fungus treatment is to treat a select number of demonstration sites, with the long-term objective of treating all hemlock conservation areas if treatments in the demonstration area are shown to be effective. Lecanicillium muscarium occurs naturally in Eastern Tennessee and has been shown to cause significant reductions in adelgid populations in small scale field trials. Initial studies show that an application of the fungus in a whey-based solution (cheese by- product) enables the fungi to better persist in the environment and have lasting effects on the HWA population (Costa 2009). The fungus has been labeled for use in the UK as the product Mycotal. An evaluation for European registration that included human health, environmental and ecotoxicological analysis concluded that Mycotal is not considered harmful to the environment. Research conducted in the laboratory and field with L. muscarium found no negative impacts on a predatory beetle being released for hemlock woolly adelgid management (Reardon and Onken 2004). Based on initial field studies with Mycotal, there is little concern associated with the effects of the fungus on non- target insects.

8 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest use to control HWA. It has been used effectively in the Great Smoky Mountain National Park as well as on private land in the southern Appalachians. Similar to imidacloprid, dinotefuran is a systemic insecticide that is absorbed by the tree and kills only those insects that feed upon the tree. It is highly water soluble and therefore much faster acting than imidacloprid. It has been shown to be effective at treating large heavily infested trees that are on the edge of decimination and may not be responding to the slower acting imidaclopid treatments. Current research documents that dinotefuran has been shown to effectively reduce adelgids on branch tips within two weeks (Cowles and Lagalante 2009). Dinotefuran treatments would be used in areas prioritized for attention due to their advanced level of infestation and need for immediate treatment. When used in combination with imidacloprid, infested trees would receive both the short term benefits of dinotefuran along with the longer lasting protection attributed to imidacloprid. Because of its greater mobility in the soil, there is a greater risk to contamination of aquatic systems from the application of dinotefuran. However, risks to aquatic systems would be minimized through the strict adherence to label direction which sets guidelines regarding application near water.

9 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest imidacloprid in streams following soil application. This revision to the EA would allow less restrictive application of imidacloprid in areas near flowing aquatic systems, consistent with label direction, and with continued periodic aquatic monitoring to ensure there are no detectable traces of chemicals in the streams. This proposal uses the best available science to treat hemlock populations across the forest using and adaptive management strategy to determine the most effective treatment for a specific area.

Alternative 3 -- No Additional Chemicals Alternative 3 is being presented in response to issue 2, the concern over adding dinotefuran as a chemical treatment for HWA. This alternative differs from the proposed action only in that dinotefuran would not be available as a treatment option. Alternatives Considered but Eliminated from Detailed Study ______

Alternative 4: Include additional proposed treatments in Wilderness areas 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). The Forest recognizes the importance of conserving hemlocks as an important component in Wilderness and Wilderness study areas, and as such, treatment with imidacloprid and beetle release would continue in wilderness and wilderness study areas as described in the 2005 HWA Decision. Additional treatment options are not being proposed for Wilderness areas at this time

Mitigation Common to All Alternatives ______1. All chemicals used for the suppression of hemlock woolly adelgid will be applied according to registered label requirements and specifications. 2. All predator beetles proposed for release will have their biology and environmental safety thoroughly evaluated, and meet USDA risk assessment criteria for release (Hennessey, R. 1995, Salom, S. 1998, Zilahi-Balogh, G.M.G. 2001, Montgomery et al. 1997, Lu and Montgomery 2001, Butin et al. 2002). 3. The labeled restriction on using Safari within twenty feet of water will be followed. 4. Any chemical treatments proposed in the following areas will be discussed with the US Fish and Wildlife Service (Cookeville, TN) prior to application: all of the Conasauga River within the Forest boundary; all of the Hiwassee River within the Forest boundary; the Tellico River from the lower Forest boundary upstream to

10 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

where the river road (Forest Service road 210 splits off from the Tellico- Robbinsville road - state highway 165); Citico Creek from the lower Forest boundary upstream to Jake Best Creek (1.5 miles above the Critical Habitat); and the Nolichucky River from the North Carolina state line downstream to the Chestoa bridge. These areas cover all occupied habitat, all Critical Habitats, and all likely sites for occurrence for aquatic T&E species. 5. Susceptible sites, which include the soil map units identified in Appendix B (Groundwater Assessment Technical Report, Table 1) as having seasonally high water tables, should not be treated with either chemical during the months identified in the individual soil map unit descriptions (usually January through May and November through December) unless the trunk spray technique is used in lieu of soil injection, drenching, or placement of granular pesticide. In these areas, treatment by trunk spray instead of soil application should be effective in achieving treatment objectives and would have little probability of affecting shallow groundwater (Cowles, 2010; Griffin, 2010).

11 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Comparison of Alternatives ______

Table 1. Comparison of Alternatives Alternative 1 Alternative 2. Alternative 3 Activities Current Management Proposed Action No Additional Chemicals

Fungus Treatment None Potential to be used on all Potential to be used on all treatment sites* treatment sites * Use of dinotefuran None Potential to be used on all None treatment sites*

Scope of Treatment Approximately 91 conservation Use an adaptive Use an adaptive Areas areas management approach to management approach to identifying new areas identifying new areas Use of imidacloprid Only on sites specifically Potential to be used on all Potential to be used on all identified for chemical treatment sites * treatment sites * treatment in the 2005 Decision *Excluding sites located in Wilderness and Wilderness Study Areas One approach to considering the alternatives in this analysis is to look at their relative potential for successfully conserving hemlocks. The 2005 EA considered the implications of doing no treatment and the potential indirect effects of hemlock loss on the ecological and human environment. The analysis in chapter 3 of this EA includes a comparison of alternatives with varying levels of treatment options. The assumption that this analysis makes is that the more treatment options we have available, the greater the potential is for saving hemlock populations across the landscape. The more hemlocks that remain in the forests, the less indirect adverse effects there will be to the ecosystem. Our current management strategy described in Alternative 1, is the result of the 2005 analysis decision, and was based on our best available treatment options in 2005. Current treatment options include the use of a fungus and additional chemical options. The figure below demonstrates the relative potential for successfully maintaining hemlock populations across the forest based on the suite of treatments proposed for use.

No Treatment Alt. 1 Alt.3 Alt.2

Least likelihood Most likelihood of success. Most of success. Least indirect effects. indirect effects.

12 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

CHAPTER 3 ENVIRONMENTAL CONSEQUENCES

This section summarizes the physical, biological, social and economic environments of the affected project area and the potential changes to those environments due to implementation of the alternatives. It also presents the scientific and analytical basis for comparison of alternatives presented in the chart above. Chapter 3 describes the environmental components of the area that would 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. It should be emphasized here that the No Action alternative, as described in chapter 2, is the continuation of current HWA management as decided in the 2005 Hemlock Woolly Adelgid EA.

The 2011 Environmental Assessment for Suppression of Hemlock Woolly Adelgid Infestations is a follow-up to the 2005 Suppression of Hemlock Woolly Adelgid Infestations on the Cherokee National Forest. The 2011 EA analyzes the effects of chemical and fungus treatments beyond what was analyzed in the 2005 analysis. This analysis thereby incorporates by reference the 2005 EA and effects analysis for the Suppression of Hemlock Woolly Adelgid Infestations.

As required by 36 CFR 219 the best available science was used in this analysis. The project record and in combination with the references and literature citations, demonstrates a thorough review of relevant scientific information, consideration of responsible opposing views, and where appropriate, the acknowledgement of incomplete or unavailable information, scientific uncertainty, and risk. Biological Resources ______BOTANICAL Major Forest Communities Affected Environment Eastern and Carolina hemlocks occur within two distinctly different forested environments on the Cherokee National Forest. Eastern hemlock forests are found primarily in association with north facing coves and slopes and riparian systems. Years of fire suppression have allowed individual hemlocks to creep upslope onto more xeric slopes and ridges where they would not likely exist under a natural fire regime. Carolina hemlock occurrences, on the other hand, are typically restricted to dry, rocky bluff habitats, but can also be found on valley bottoms, gorge slopes, or other protected landforms.

Eastern Hemlock Forests Eastern hemlock (Tsuga canadensis) forests have a broad range from Canada to the southern Appalachians. On the Cherokee National Forest these forests typically occur on acidic soils and often have a dense shrub layer composed of ericaceous species. These communities vary in their level of herbaceous diversity, from relatively high in areas with

13 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest a diverse midstory, to low in areas with dense Rhododendron in the shrub layer. These communities almost always support rich bryophyte communities.

Eastern hemlock forests are an important plant community, integral to the ecology of the Cherokee National Forest as described in the Final Environmental Impact Statement for the Revised Land and Resource Management Plan: “The combination of a largely evergreen canopy and a dense midstory in naturally occurring hemlock forests provide for a variety of benefits, including shading and cooling of riparian systems, thermal cover for wildlife, and nesting and foraging habitat for several species of neotropical migrant birds dependent upon the layered canopy structure and understory thickets (Rhea and Watson 1994). There is some evidence that hemlock-white pine forests provide necessary habitat components for the long-term conservation of red crossbills (Dickson 2001). Eastern hemlock forests may also be important refugia for species typically adapted to higher elevations. Dickson (2000) states that red-breasted nuthatches, winter wrens, and golden-crowned kinglets are found in late successional hemlock forests down to elevations of 2,000 feet, and several species of rare bryophytes that are known to occur primarily within the spruce/fir zone are also found at lower elevations in humid gorges often under a canopy that includes eastern hemlock (Hicks 1992).” (USDA Forest Service 2004a). Thirty-three species of plants and animals with viability concerns are associated with mature eastern hemlock forests across the southern Appalachian region, thirteen of which are of concern on the Cherokee National Forest (USDA Forest Service 2004a & 2004b). See section entitled “Threatened, Endangered, Sensitive, and Viability Concern Species” for more information on these species.

Carolina Hemlock Forests Carolina Hemlock (Tsuga caroliniana) communities have a much more restricted range, occurring primarily in the Southern Blue Ridge, with scattered occurrences in the upper Piedmont and Ridge and Valley of North Carolina, Tennessee, South Carolina, and Virginia. Occurrences on the Cherokee National Forest are typically small and restricted to dry, rocky bluff habitats, but within its range, this species can be found on valley bottoms, gorge slopes, or other protected landforms. As stated in the Final Environmental Impact Statement for the Revised Land and Resource Management Plan: “The ecology of this forest community is poorly understood with some debate in the current literature regarding the role of fire. Much of the species composition would suggest a fire maintained community, yet monitoring studies in North Carolina have shown evidence of stand expansion both following fires and periods of fire exclusion (Schafale and Weakley 1990). Rentch et al. (2000) reported that Carolina hemlock at a site in Montgomery County, VA was long-lived, very tolerant of drought stress, and had reproduced episodically over the past 200 years. There is no mention of fire in that paper. Further description of the life history of Carolina hemlock can be found in Humphry (1989).” (USDA Forest Service 2004a). Four species of plants with viability concerns that are associated with Carolina hemlock forests across the southern Appalachian ecoregion, three of which are of concern on the Cherokee National Forest (USDA Forest Service 2004a & 2004b). See section entitled “Threatened, Endangered, Sensitive, and Viability Concern Species” for more information on these species.

14 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Environmental Consequences Effective treatment methods to prevent infestation of eastern and Carolina hemlocks are limited, and it is not feasible to individually treat the millions of trees that would be necessary to protect the hemlock species as a whole. Based upon this, forest managers have determined that the best long-term potential to save hemlocks within their range lies with protecting areas of significant genetic variation (conservation areas) and the use of biological control agents. While the 2005 decision approved the use of imidacloprid and predator beetles, the current proposal expands treatment methods to include another chemical (dinotefuran) and the insect killing fungus (Lecanicillium muscarium). Genetic conservation areas for the Cherokee National Forest were defined in the 2005 environmental analysis (USDA Forest Service 2005) and are not repeated here. The following assumptions were made in order to evaluate potential effects:

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. Predator beetles would disperse up to ¼ mile from release sites, thus covering approximately 25 acres. Aerial application of the fungus (Lecanicillium muscarium) would be applied directly to identified conservation areas. Limited to no control of HWA would occur on non-federal lands adjacent to the forest.

The following paragraphs summarize the potential environmental consequences to eastern and Carolina hemlock communities from implementing the various alternatives.

Alternative 1 Under this alternative no additional treatments are proposed to suppress HWA infestations on the Cherokee National Forest. Despite implementation of the 2005 decision, the rate of spread of HWA across the forest has been much more rapid than was originally anticipated and treatment efforts have not been sufficient to protect all the sites identified for protection in that environmental assessment. Efficacy of beetle treatments is currently inconclusive and some beetle release sites have seen significant hemlock mortality. Without additional control measures the HWA infestation would continue to spread across the Cherokee National Forest potentially killing any remaining untreated sites of both hemlock species over time. Plant and animal species associated with hemlock forests may also suffer losses dependent upon their ability to adapt to the changing environmental conditions. Potential effects to associated species range from altered fitness due to less than optimal habitat conditions, to local extirpations. Potential losses would be highest for those non-mobile species dependent upon the moist and shaded conditions typically found in eastern hemlock stands.

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 species such as rhododendron (Rhododendron maximum), tulip poplar (Liriodendron tulipifera), red

15 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest maple (Acer rubrum) and white pine (Pinus strobus) would increase on moist eastern hemlock sites, and that yellow pines, particularly Virginia pine (Pinus virginiana), would increase on Carolina hemlock sites in the absence of fire. Actual shifts in species composition would likely be based upon local site conditions and species present. Yorks, et al. (1999) found this to be the case in the Catskill Mountains of New York where propagules of the dominant species in the seed bank rapidly colonized sites after hemlock mortality. Alternatively, Evans (2002) and Orwig and Foster (1998) suggest that species likely to expand in declining hemlock stands include sweet birch (Betula lenta), red maple (Acer rubrum), white pine (Pinus strobus), and a variety of invasive exotics including tree of heaven (Ailanthus altissima), Japanese barberry (Berberis thunbergii), and Japanese stiltgrass (Microstegium vimineum).

Stands dominated by eastern hemlock are often noted for their diverse assemblages of bryophyte species and while declines in bryophyte species dependent upon the cool, moist microclimates would be expected as canopies become more open, Cleavitt et al. (2008) found that species richness actually increased in stands that had been affected by HWA. These increases were particularly noted among common species that utilize coarse woody debris as their primary substrate. No mention is made in that paper on how these increases may affect the presence/absence of rare species however.

Alternative 2 With the additional combined treatments of chemical and biological control proposed beyond the original conservation areas, it is hoped that preservation of more hemlock would be possible across the forest. Even so, overall abundance has already been greatly reduced and many hemlock communities have already been negatively affected. Overall effects would be very similar to those described for Alternative 1, though it is hoped that less impacts to hemlock communities would occur under this alternative,

Alternative 3 The only difference between Alternatives 2 and 3 is the use of the chemical dinotefuran in Alternative 2. Neither chemical has any effect on plant species: rather it is the secondary effects of hemlock mortality that may differ slightly between alternatives. Dinotefuran may provide some additional measure of effectiveness over imidacloprid in preventing mortality from HWA however it is just the method, not the magnitude of treatment that differs between alternatives. Based upon this the effects at the forest scale between to the alternatives would be approximately the same.

Cumulative Effects, all Alternatives In the Environmental Impact Statement for the Revised Land and Resource Management Plan for the Cherokee National Forest, the following statement is made regarding the fate of eastern hemlock communities on the forest:

A thirty-nine percent increase in the acreage of white pine-hemlock forests has been documented across both public and private ownerships in the southern Appalachians since the mid 1970’s (SAMAB 1996). This is largely attributable to an increase in managed stands of white pine (plantations) and upland

16 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

encroachment of both white pine and hemlock into areas where it would not occur under a natural fire regime. The use of prescribed fire in the restoration of upland habitats will likely shrink these communities back to a more natural distribution on the landscape over time. Despite Plan protection and restoration objectives, the current amount and distribution of mature eastern hemlock forests is threatened by the recent emergence of the hemlock wooly adelgid in the southern Appalachians. The fact that this community type is naturally limited in distribution, coupled with the impending threats from the hemlock wooly adelgid which will imapct the species regardless of land ownership, leaves the long-term maintenance of historical distribution and abundance of this community type in question. Future distribution and abundance of this community type compared to historic levels would be poor (see Tables 3 and 4, terrestrial viability section). The fate of associated viability concern species will be dependent upon their ability to adapt to changing environmental conditions associated with the decline of hemlock from within these communities. Species that utilize hemlock forests in addition to other vegetative community types will be more likely to persist than species that are obligates to the hemlock forest community.” (USDA Forest Service 2004a).

Hemlock communities (both Carolina and eastern) are afforded plan protection through various goals, objectives, and standards, yet despite these protections it is projected that future landscape distribution would be poor (see above). Suppression efforts on adjacent lands in neighboring North Carolina are already underway and similar efforts are being planned in Georgia. Unfortunately, despite the best intent in aggressively treating HWA, it is likely that hemlock forests and associated species would be much reduced in the future over current and previous levels.

WILDLIFE Affected Environment Analysis of effects to vegetation, wildlife, and fish follows the organization of the habitat framework used in the 2004 Revised Forest Plan (USDA Forest Service 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 and Carolina hemlock dominated stands. This includes pockets of hemlock- dominated forests within mesic deciduous forests. The proposed treatment areas are the focus of our analysis.

Therefore, the following communities, habitats (Habitat Elements), and associated management indicator species (MIS) are discussed.

. Major Forest Communities: Eastern hemlock and White pine forests (Acadian flycatcher to cover riparian communities) . Rare Communities: Carolina hemlock forests. . Successional Habitats: Discussed in general under effects analysis, but not analyzed by successional class as impacts do not differ substantially by alternative.

17 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

. Habitat Attributes: Aquatic habitats . Threatened and Endangered Species . Demand species: Black bear White-tailed deer, wild turkey, and ruffed grouse are discussed in general, but not analyzed individually. Species are not hemlock-dependent, so impacts at the Forest level do not differ substantially by alternative. Wild trout, stocked rainbow trout, and black bass and bream in streams would be discussed individually. Black bass and bream in ponds would not be affected. . Species of local concern: Restricted to those species within proposed treatment areas for which there is a viability concern.

The loss of hemlock and hemlock dominated stands is expected under all alternatives. Hemlock also occurs as a minor forest component within other forest communities. Therefore, impacts would occur within these forest communities and habitat elements, but the effects at the Forest level within stands where hemlock is a minor component would not differ substantially based upon selection of any of the alternatives because no treatments are planned within those communities. Where mortality occurs, hemlock will be replaced in the canopy by other species currently on site. Therefore, impacts to community composition within these stands will be minor. Forest-level MIS would address these impacts in Table 2.

Table 2. Impacts to Forest-level MIS.

Species Purpose Impacts at Forest Level Name Prairie To help indicate Early successional habitat would temporarily increase due warbler management effects of to hemlock mortality allowing opportunities for creating and maintaining establishment. Prairie warblers are not normally associated early successional forest with conditions where eastern hemlock is dominant. Some communities potential increases at Carolina hemlock sites, but no notable population changes expected. Chestnut- To help indicate Hemlock mortality would result in the creation of early sided warbler management effects of successional habitat resulting in a temporary increase in creating and maintaining habitat. Populations would be expected to increase while early successional forest suitable habitat conditions are present. communities and habitat Pine warbler To help indicate effects of Hemlock mortality is likely to result in an expansion of the management in pine and pine and pine oak communities within suitable habitat; pine-oak communities particularly Carolina hemlock stands. Any expansion of the pine and pine oak community type would provide additional habitat for Pine warblers allowing populations to increase. Pileated To help indicate Hemlock mortality would increase snag availability woodpecker management effects on snag resulting in a temporary increase in Pileated woodpecker dependent wildlife species populations.

18 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Table 2. Impacts to Forest-level MIS.

Species Purpose Impacts at Forest Level Name Acadian To help indicate Hemlock mortality within riparian areas would result in flycatcher management effects within temporary gaps and openings. Hemlock in these areas mature riparian forest would be replaced by other riparian species (white pine, community Tulip Poplar, etc.). No net loss of habitat expected over long-term. Acadian flycatchers would continue to utilize the riparian community, but populations would experience a slight decline, until mature hemlock stands are replaced. Discussed further under Hemlock-dominated stands. Scarlet To help indicate effects of Hemlock mortality is likely to result in an expansion of the tanager management in xeric oak xeric oak and oak pine communities within suitable habitat; and oak pine communities particularly Carolina hemlock stands. Any expansion of the xeric oak and oak pine community type would provide additional habitat for Scarlet tanagers allowing populations to increase. Ruth’s To help indicate Populations are restricted to rock outcroppings within river golden aster management effects on the corridors and is not impacted by any forested stand. No recovery of this T&E plant notable impacts are expected. species Ovenbird To help indicate Temporary loss of habitat where mortality occurs. management effects of Conditions would improve as the new stand matures. No wildlife species dependent net loss of habitat expected over long-term. upon mature forest interior conditions Black bear To help indicate Black bears utilize a variety of habitat types and management effects on successional stages to meet the life requirements (food, meeting hunting demand for shelter, cover). Bears would continue to utilize areas this species impacted by hemlock woolly adelgid to obtain the life requirements that are being provided. Notable impacts to bear populations and the ability to meet hunting demand on the CNF are not expected. Discussed further under demand species. Hooded To help indicate effects of Hemlock mortality is likely to result in an expansion of warbler management on providing mesic deciduous forest communities within suitable habitat; dense understory and particularly eastern hemlock stands. Any expansion of the midstory structure within mesic deciduous forest community type would provide mature mesic deciduous additional habitat for Hooded warblers allowing forest communities populations to increase when conditions become suitable.

Scope of Analysis The scope of this analysis for terrestrial wildlife is confined to those hemlock-dominated stands proposed for treatment on the CNF. Conditions within untreated hemlock- dominated stands are expected to follow the effects and trends described for the No Action Alternative (1), though some of those would be treated under the 2005 decision. Impacts to areas where hemlock is a minor component have already been discussed. The period of time used in this analysis is up to 20 years. At this point in time, hemlock woolly adelgid would be well established throughout the CNF, hemlock mortality would have occurred, and succession to other forest types would be taking place. Analysis is also based on current treatments with the knowledge that new treatments could result in

19 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest the survival of additional hemlock. The following assumptions were made in order to evaluate potential effects: 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. Biological control agents (predator beetles) would disperse up to ¼ mile from release sites, thus covering approximately 25 acres. Limited to no control of HWA would occur on non-federal lands adjacent to the CNF.

Analysis of cumulative effects also includes past, present and reasonably foreseeable activities on the forest, and thus may extend beyond the limits of the defined project areas to include the range of a species or habitat type. Time frames for cumulative effects analysis for terrestrial elements generally include 10 years prior to 50 years post treatment.

Eastern Hemlock and Carolina Hemlock Forests Eastern hemlock forests create distinctive microclimates and provide important habitat for a wide variety of wildlife. Many species of birds, small mammals, amphibians, and insects inhabit hemlock forests. In the the Northeast, 96 bird and 47 mammal species are associated with hemlock forests (Yamasaki et al. 2000). Use of hemlock by various wildlife include habitat considerations such as the distribution of hemlock trees and the variety of structural habitat features, dense patches of regeneration, hard-mast inclusions, cavity trees, coarse wood debris, wetland seeps and inclusions, and suitable cover opportunities (Yamasaki et al. 2000). During the non-breeding season and throughout the winter, eastern hemlocks provide an important seed source for pine siskin, goldfinch, red crossbill and evening grosbeak (DeGraaf and Rudis 1986, Howe and Mossman 1995). Birds such as blackburnian warblers and black-throated green warblers occur almost exclusively in hemlock forests in the Delaware Water Gap National Recreation Area (Evans et al. 1996). Ruffed grouse, yellow-bellied sapsucker, great horned owl, and a number of overwintering forest birds use hemlock for a variety of reasons. Ruffed grouse habitat management guidelines often address the importance of hemlock stands, inclusions, and single trees as high quality fall and winter roosting locations (Edminster 1947, Jordan and Sharp 1967). Affinity for hemlock tree boles by foraging and cavity dwelling primary excavators like the yellow- bellied sapsucker and pileated woodpecker have been recognized by Rushmore (1969). Hemlock tends to be long-lived, develops a number of potential cavity sites and perhaps a higher level of cavity-dwelling and foraging use by an array of woodpeckers and smaller mammals. Many small mammals, such as deer mouse, southern red-baked vole, masked shrew, short-tailed shrew, white-footed mouse, and woodland jumping mouse use hemlock forests. Annual small mammal abundance and species richness can fluctuate dramatically due to variables such as food availability (e.g., prior year’s mast crop) and winter severity (e.g., frozen ground with no snow cover). Important structural habitat features for

20 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest smaller mammal communities include a range of overstory canopy closures. The resulting effects on the midcanopy and shrub layers, and perhaps the patterns of coarse woody debris, contribute to the subsequent accessibility of prey by both avian and mammalian predators such as northern goshawk, barred and great horned owls, and typical forest carnivores like raccoon, red fox, and bobcat (DeGraaf et al. 1992, Powell et al. 1997a,b). Species like gray squirrel, eastern chipmunk, and northern flying squirrel also use hemlock stands and inclusions, especially when hard mast-producing trees such as beech (Fagus grandifolia) and oak (Quercus spp.) are present in the overstory even though hemlock is not their preferred habitat (DeGraaf et al. 1992). Throughout the winter, eastern hemlock provide an important seed source for red squirrels. Cavity trees, both live and dead, provide summer roosting opportunities for forest bats, including the endangered Indiana bat. The hoary bat is known to roost in coniferous foliage (DeGraaf and Rudis 1986). Hemlock is a well-documented habitat element in winter deer range management throughout the northeastern United States and eastern Canada (Mattfeld 1984, Huot et al. 1984, Blouch 1984, Reay et al. 1990). Black bear are known to forage in wetland seeps, swales, and riparian drainages in the spring for ephemeral herbaceous forage (e.g., skunk cabbage, various sedges, grasses, and tubers) present in these habitat conditions (Elowe 1984). Female black bear use coniferous riparian areas in Maine when hard mast crops are marginal (Schooley 1990). Vander Haegen and DeGraaf (1996) found black bear travel along coniferous tributary buffer zones between forested watersheds. Coarse woody debris is a source of grubs and ants especially in the spring, and large hollow trees and logs, and slash piles can be wintering den sites (DeGraaf and Rudis 1986).

Environmental Consequences Potential direct and indirect effects are discussed together in the terrestrial wildlife analysis because it is difficult to separate effects to habitat and effects to species that depend on those habitats. For this analysis, direct effects are defined as the results of the action on the quality and quantitiy of terrestrial habitat, and indirect effects are the result of the action’s effects on terrestrial species due to habitat changes.

Alternative 1 This alternative consists of continuing suppression as outlined in the existing 2005 decision and 2010 SIR. Without additional control measures, the HWA infestation would continue to spread across the CNF and potentially result in the loss of some hemlock stands over time. Plant and animal species associated with hemlock forests may also suffer losses dependent upon their ability to adapt to the changing environmental conditions. Potential effects to associated species range from altered fitness due to less than optimal habitat conditions, to local extirpations. Potential losses would be highest for those non-mobile species dependent upon the moist and shaded conditions typically found in eastern hemlock stands.

21 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest maple (Acer rubrum) and white pine (Pinus strobus) would increase on moist eastern hemlock sites, and that yellow pines, particularly Virginia pine (Pinus virginiana), would increase on Carolina hemlock sites in the absence of fire. Actual shifts in species composition would be based upon local site conditions and species present.

The loss of hemlock would result in many changes for wildlife species. 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 and white pine trees, and an increase in the availability of down woody debris and age class diversity 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 would be greatest on species with limited mobility that require hemlock forest for nesting habitat, such as invertebrates and land snails.

Management Indicator Species Acadian Flycatcher (Empidonax virescens) is a MIS to help indicate management effects within the mature riparian forest community. This bird is a fairly common statewide summer resident of low elevations (Nicholson 1997). Acadian flycatchers require deciduous forests near streams for breeding, and are not often found outside of these habitats during the breeding season (USDA Forest Service 2004a:190). In the eastern mountains, Acadians also occupy streamside hemlock forests (Nicholson 1997). 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 (USDA Forest Service 2004a:190). Acadian flycatchers are known from many hemlock-dominated stands associated with streams on the CNF where treatment may occur.

Point count data on the CNF indicate that this bird reaches peak frequencies and densities in a wide variety of hardwood and eastern hemlock/white pine communities, particularly in older age classes. The overall regional trend (Blue Ridge Mountains) for 1966-2001 is negative (Sauer 2005).

Acadian flycatcher habitat within untreated hemlock-dominated stands would slowly deteriorate as the existing hemlock forest is killed by hemlock woolly adelgid. Individuals would utilize standing snags for perch sites and the periphery of the stands for foraging. Any nesting that is occurring within hemlock trees would also be lost over time. Acadian flycatchers would continue to utilize the deciduous forest and mixed deciduous-pine forests within occupied habitat. The forests types that replace hemlock- dominated forests would be suitable for Acadian flycatchers upon maturity. Therefore, Acadian flycatcher populations are expected to decrease as mature hemlock dominated forests are lost. Recovery would take place at these sites as the new stand matures. Hemlock-dominated forest represents only one percent of the forested acres on the CNF. Populations within treated hemlock stands as well as adjacent cove hardwoods and white

22 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest pine-hardwood stands would persist ensuring the Acadian flycatcher continued presence on the CNF.

Alternative 2 Effects of using predator beetles on terrestrial wildlife would be similar to those described for Alternative 1. Additionally, Alternative 2 includes expanded use of imidacloprid on all hemlock treatment areas. Effects from the use of imidacloprid on terrestrial wildlife would be similar to those described for Alternative 1 but more areas would be treated with imidacloprid. Alternative 2 proposes the use of dinotefuran on trees that are heavily infested and need immediate attention to ensure survival. The chemical compound is highly soluble in water, is relatively stable to hydrolysis at pH 4 to 9 (the range of most soils and water across the Forests), and has an average aerobic half-life of 81 days. The chemical is highly mobile in most soil types. High water solubility combined with high soil mobility indicates low potential for bioaccumulation. The major dissipation route for dinotefuran is aqueous photolysis (half-life 1.8 days) (EPA 2004). Dinotefuran is nontoxic to birds on an acute basis and slightly toxic on a subacute dietary basis. Dinotefuran is nontoxic to mammals on an acute basis. Slight effects on mammalian reproductive characteristics were noted, but these changes did not affect reproductive success. Dinotefuran is highly toxic to honeybees on both oral and contact bases (EPA 2004). Therefore, the primary risk of using dinotefuran on terrestrial wildlife is to pollinating insects. Hemlock is a wind-pollinated species. Therefore, the only contact between pollinating insects and plant material exposed to dinotefuran would be incidental. Dinotefuran is a systemic herbicide and would be applied through soil injection. There would be no aerial or broadcast application of the chemical; therefore, there would be little risk of acute or contact exposure to pollinating insects. Alternative 2 also proposes the use of naturally occurring fungus Lecanicillium muscarium. This fungus attacks the hemlock woolly adelgid by penetrating directly through the body wall, inducing death, and then releasing spores to continue the fungal infection throughout the adelgid population. Although most fungi are generalists within their natural range, a variety of biological, ecological, and behavioral factors limit effects on non-target species. Additionally, certain species and strains of a given species tend to be virulent to insects within a given order (Gouli et al. 2008; Reid 2003). Extensive laboratory studies have been done to verify fungi and isolates that are effective in infecting the hemlock woolly adelgid (Reid 2003). Based on laboratory assays and initial field studies, there is little concern about non-target effects of L. muscarium on non-target insect species, including HWA predator beetles, Sasajiscymnus spp. and Laricobius spp., also used as adelgid biocontrols (Cheah et al. 2002). Acadian flycatcher populations would still undergo a slight population decline as previously described under Alternative 1. However, hemlocks that survive as a result of treatment would continue to provide suitable habitat conditions for this species resulting in fewer impacts on individuals occurring within these areas.

23 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Alternative 3 Alternative 3 is the same as Alternative 2 except that dinotefuran would not be used. Refer to the effects analysis above (Alternative 2) for the expanded use of imidacloprid and treatment with Lecanicillium muscarium. Effects of using predator beetles on terrestrial wildlife would be similar to those described for Alternative 1.

Cumulative Effects Past and present timber harvest and prescribed burning activities on the CNF have affected Eastern and Carolina hemlocks and have affected sites that could support hemlock, such as acidic coves. Many of these activities would continue in the future, but the individual minor effects are insignificant when compared with the magnitude of change to the landscape expected from infestation of the HWA. Therefore, past and present forest management activities are unlikely to add measurably to potential impacts of an expanding HWA population. This project would result in lessening cumulative impacts to hemlock from HWA infestation and all other actions. When hemlock dies, numerous changes will occur across the landscape. Where trees die along streams, cover is eliminated initially from riparian travel corridors (especially for black bears). Once the canopy is opened, sunlight reaches the forest floor, drying out the leaf litter and making those moist areas unsuitable for salamanders (until cover and shade return). A flush of herbaceous and soft mast species growth would occur, creating forage for black bears and ruffed grouse. The increase in the creation of snags would increase potential habitat for several species, especially those that use snags for denning, roosting or foraging. Increased snags would benefit raccoons, black bears, bats, and pileated woodpeckers, and increased downed woody debris would benefit salamanders and ruffed grouse. Treating infested hemlock trees would reduce species mortality and effects on local terrestrial wildlife habitats. The use of chemicals would have no negative effect on any of the Management Indicator Species. It is reasonable to believe that not all infested hemlocks across the CNF would be treated, thus some snags would still be created that could function as potential roost trees for a variety of bats, den trees for raccoons and black bears, and nesting and foraging habitat for pileated woodpeckers, as well as downed woody debris for salamanders and ruffed grouse.

Demand Species Black bear was selected as a MIS to help indicate management effects on meeting hunting demand for this species. They use a wide variety of habitats in the Southern Appalachians, primarily on public lands. Important habitat elements include habitat remoteness and diversity, availability of hard mast, den site availability, large home ranges, and hard mast (USDA Forest Service 2004a and Whitaker and Hamilton 1998). Black bears are not dependent upon hemlock-dominated forests, but utilize the unique habitat qualities when they are available. Activities would include denning, limited foraging and loafing areas, and use as travel corridors. Bear populations throughout the Southeast have increased moderately over the past 30 years, and have increased significantly over the last 20 years on the CNF (USDA Forest

24 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Service 2004a). These trends are due to increases in older oak forests, increased protection, and conservative hunter harvest (SAMAB 1996:1).

White-tailed deer, wild turkey, and ruffed grouse are public demand species identified in the 2004 Revised Forest Plan EIS (USDA Forest Service 2004a) that utilize hemlock- dominated stands or habitat conditions are influenced by their presence. However, they are not dependent upon this forest-type. Hemlocks are utilized for roosting by grouse and turkey, and thermal and winter cover by all three species. Hemlocks are occasionally eaten by grouse and deer. Hemlock-dominated stands are not critical habitat elements for any of these species in the southern Appalachians. White-tailed deer and wild turkey populations have increased over the past 25 years, while grouse populations have declined over this same period (USDA Forest Service 2004a). This decrease is likely due to the reduction in hardwood forest cover in the sapling-pole class (0-20 years of age).

Alternative 1 The loss of hemlock may affect black bear both negatively and positively. Bears use forested riparian corridors, where hemlock is often present and provides important cover, as travel ways. Loss of hemlock would initially open up riparian areas, and cover would be lacking until other trees species grow up to replace the hemlock in the canopy. However, with the opening up of stream-side areas, a flush of growth (both herbaceous and soft mast) would occur, thus providing food for black bears. Hemlock snags, especially larger ones (>36” dbh) could provide denning opportunities for black bears. An increase in coarse woody debris could increase the amount of protein food source (e.g., grubs and ants), especially in the spring. Black bear would continue to utilize hemlock-dominated stands, but how they are utilized to meet their life requirements would change as habitat conditions change within these stands. Mature hemlock stands are utilized by bear for denning, limited foraging, and travel corridors. As stands are killed by this insect, bears would utilize fallen logs and root wads for ground dens, roll logs and large down debris in search for insects, utilize the dense regeneration for cover, and forage upon the fruits and berries within the young stand.

Impacts to eastern wild turkey, ruffed grouse, white-tailed deer are expected to be similar to those described for black bear. Individuals would continue to utilize these areas to meet their needs. However, populations of each of these species are limited to some extent by the availability of early successional habitat. Loss of hemlock trees across the CNF would increase the availability of early successional habitat as well as the distribution of early successional habitat on the landscape. Current populations of turkey, grouse, and deer should respond to the increase in early successional habitat resulting in a population increase for these species on the CNF.

25 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest terrestrial wildlife would be similar to those described for Alternative 1 but more areas would be treated with imidacloprid. Alternative 2 proposes the use of dinotefuran on trees that are heavily infested and need immediate attention to ensure survival. The chemical compound is highly soluble in water, is relatively stable to hydrolysis at pH 4 to 9 (the range of most soils and water across the Forests), and has an average aerobic half-life of 81 days. The chemical is highly mobile in most soil types. High water solubility combined with high soil mobility indicates low potential for bioaccumulation. The major dissipation route for dinotefuran is aqueous photolysis (half-life 1.8 days) (EPA 2004). Dinotefuran is nontoxic to birds on an acute basis and slightly toxic on a subacute dietary basis. Dinotefuran is nontoxic to mammals on an acute basis. Slight effects on mammalian reproductive characteristics were noted, but these changes did not affect reproductive success. Dinotefuran is highly toxic to honeybees on both oral and contact bases (EPA 2004). Therefore, the primary risk of using dinotefuran on terrestrial wildlife is to pollinating insects. Hemlock is a wind-pollinated species. Therefore, the only contact between pollinating insects and plant material exposed to dinotefuran would be incidental. Dinotefuran is a systemic herbicide and will be applied through soil injection. There will be no aerial or broadcast application of the chemical; therefore, there will be little risk of acute or contact exposure to pollinating insects. Alternative 2 also proposes the use of naturally occurring fungus Lecanicillium muscarium. This fungus attacks the hemlock woolly adelgid by penetrating directly through the body wall, inducing death, and then releasing spores to continue the fungal infection throughout the adelgid population. Although most fungi are generalists within their natural range, a variety of biological, ecological, and behavioral factors limit effects on non-target species. Additionally, certain species and strains of a given species tend to be virulent to insects within a given order (Gouli et al. 2008; Reid 2003). Extensive laboratory studies have been done to verify fungi and isolates that are effective in infecting the hemlock woolly adelgid (Reid 2003). Based on laboratory assays and initial field studies, there is little concern about non-target effects of L. muscarium on non-target insect species, including HWA predator beetles, Sasajiscymnus spp. and Laricobius spp., also used as adelgid biocontrols (Cheah et al. 2002). Impacts to black bear, eastern wild turkey, ruffed grouse, and white-tailed deer would be similar to those previously described under Alternative 1. More treatment areas would provide some mature hemlock forest habitat allowing individuals the continued use of these habitat attributes. Increased use of imidacloprid and the addition of dinotefuran and the fungus Lecanicillium muscarium would have negligible impacts on food availability (decrease in invertebrates) for black bear, turkey and grouse. Impacts to black bear, turkey, grouse, and white-tailed deer populations would continue to be negligible. Maintenance of hemlock along streams will help provide cover along riparian travel corridors. It is reasonable to believe that not all infested hemlocks across the CNF would be treated, thus some snags and coarse woody debris will still be created as untreated hemlock die, creating potential den trees and foraging habitat for black bears.

26 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Cumulative Effects, all Alternatives Past and present timber harvest and prescribed burning activities on the CNF have affected eastern and Carolina hemlocks and have affected sites that could support hemlock, such as acidic coves. Many of these activities would continue in the future, but the individual minor effects are insignificant when compared with the magnitude of change to the landscape expected from infestation of the HWA. Therefore, past and present forest management activities are unlikely to add measurably to potential impacts of an expanding HWA population. This project would result in lessening cumulative impacts to hemlock from HWA infestation and all other actions. When hemlock dies, numerous changes would occur across the landscape. Where trees die along streams, cover is eliminated initially from riparian travel corridors (especially for black bears). A flush of herbaceous and soft mast species growth will occur, creating forage for black bears and ruffed grouse. The increase in the creation of snags would increase potential habitat for several species, especially those that use snags for denning, roosting or foraging. Increased snags would benefit raccoons, black bears, bats, and pileated woodpeckers, and increased downed woody debris would benefit ruffed grouse. Treating infested hemlock trees would reduce species mortality and effects on local terrestrial wildlife habitats. The use of chemicals would have no negative effect on any of the Management Indicator Species. It is reasonable to believe that not all infested hemlocks across the CNF would be treated, thus some snags would still be created and utilized as potential roost trees for a variety of bats, den trees for raccoons and black bears, and nesting and foraging habitat for pileated woodpeckers, and downed woody debris for ruffed grouse.

AQUATICS Affected Environment Streams on the Cherokee National Forest drain to the Tennessee and Conasauga River systems. They occur primarily in the Blue Ridge Province but a small number are in the Ridge and Valley Province. The twenty-four watersheds on the forest are, generally, in average condition (USDA, 2004b) with four rated excellent and three rated below average based on sediment. There are approximately 16,495 miles of streams and 125 acres of ponds on the forest (Table 3). All perennial and many intermittent streams support aquatic communities. Intermittent and most perennial streams do not support fish because they are too small in size, too steep in gradient or a combination of these two factors. There are thirteen warmwater ponds that support fish communities and are managed for recreational fishing.

27 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Table 3. Forest Service managed aquatic habitats Aquatic Habitats Miles/Acres Ephemeral Streams 10,800

Intermittent Streams 3,527 Perennial Streams (no fish) 2,168 Fish Supporting Waters Coldwater 517 Coolwater 275 Warmwater Ponds 125 acres

Coldwater streams support aquatic species adapted to tolerate persistent summer water temperatures up to about 72o F. Trout are the most significant coldwater species on the forest. Stream reaches capable of supporting fish and located above 1200 feet in elevation often have trout present year-round (USDA Forest Service 2010a). Hemlocks provide a portion of the tree canopy essential to shading these stream reaches and maintaining the cold water conditions.

Hemlocks commonly occur in riparian areas and provide large woody debris essential for stream habitat diversity and productivity including: fish hiding cover, invertebrate production, and stream channel function.

Environmental Consequences Potential direct and indirect effects are discussed together because habitat changes (direct effects), such as chemical alteration of water quality, result in effects to aquatic animals (indirect effects). No direct effects to aquatic animals are anticipated from the chemical treatments, beetle additions, or fungus.

Alternative 1 Alternative 1 would continue suppression of the HWA without adding additional treatment methods or expansion of treatment areas. Coldwater streams could suffer short term (decade), summer warming effects due to the loss of canopy provided by hemlocks. The range of trout on the forest would move up in elevation and could result in the loss of entire populations within some stream reaches. In the long term (decades) the summer shade that had been provided by hemlocks would eventually be provided by other species such as white pines and rhododendron and the cold water conditions could return.

The large woody debris provided by hemlocks essential to the functioning of the riparian system would increase following the death of the hemlocks but it would take decades for significant amounts of large woody debris to be generated by replacement tree species.

28 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

The Environmental Assessment - Conservation of Native Hemlock by Suppression of Hemlock Woolly Adelgid Infestations (USDA Forest Service. 2005) was implemented to protect selected stands (refuges) of hemlock from destruction by HWA infestation. These refuges were intended to: 1) reduce loss of remnant older hemlocks that contribute to old growth characteristics in the form of large trees and unique wildlife habitat; 2) reduce the loss of hemlock along native trout streams that provide nutrients in the form of large woody debris and protection from temperature increases in the streams; 3) protect rare plant communities that occur in hemlock dominated forest where viability is a concern for at least five species; 4) protect areas where hemlocks contribute to the character and quality of scenery and recreation settings especially where these aesthetic evergreen trees provide cover, shade and screening throughout the year; and 5) provide for a diverse genetic base of hemlock from which future populations could be established in the event wide spread hemlock mortality occurs. These areas represent direct treatment of only about 3100 individual hemlock trees. Based on monitoring (USDA Forest Service 2009) since the inception of this project “The Hemlock Wooly Adelgid continues to severely impact hemlock across the entire Forest. Chemical treatment of refuge areas appears to be very effective in protecting the trees from mortality. The effectiveness of predator beetle releases in protecting the hemlocks in larger land areas is unknown at this time.” Expansion of treatment areas and beetle releases appears to be warranted and should not adversely affect the aquatic habitats or aquatic organisms. Analysis conducted for the suppression of HWA in the Nantahala/Pisgah National Forests (USDA Forest Service, 2010b) and at Royal Blue State Wildlife Management Area (Costa, 2010), determined the naturally occurring fungus Lecanicillium muscarium was effective against HWA but had little effect on none target organisms, including aquatic insects. Application of this fungus is not expected to affect aquatic species or their habitats. Dinotefuran and Imidacloprid both belong to a group of insecticides referred to as Neonicotinoids (derived from nicotine) which act on the central nervous system of insects. Both have low toxicity to fish (Valent USA Corporation, 2011; National Pesticide Information Center, 2011). Neonicotinoids are among the most widely used insecticides worldwide. Dinotefuran treatment is proposed because this chemical is more quickly absorbed by hemlock roots and distributed throughout the tree. It is as effective as Imidacloprid but provides more immediate protection to the hemlocks. Imidacloprid provides a longer period of protection than Dinotefuran. Used in combination and following labeling instructions, these chemicals would greatly improve the efficacy of HWA suppression and would not adversely affect aquatic organisms. Alternative 3 Alternative 3 is the same as Alternative 2 except that Dinotefuran would not be used. Refer to the effects analysis for Alternative 2.

Cumulative Effects, all Alternatives Sedimentation of streams is the most significant adverse effect that contributes to cumulative effects to aquatic habitats on national forest lands. None of the three alternatives proposes any ground disturbance. Compliance with labeling instructions for

29 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest the two chemicals would minimize or eliminate harmful impacts and cumulative adverse impacts to the aquatic habitats and associated animals. Management activities, on National Forest System lands, that are reasonably foreseeable and under the control of the Forest Service would be implemented under the standards for protecting streams listed in the Revised Land and Resource Management Plan for the Cherokee National Forest (USDA Forest Service 2004a). Implemented in conjunction with any of the three alternatives, these activities would not increase the adverse cumulative effects on the aquatic habitats now occurring. Reasonably foreseeable activities that occur on private lands are not controlled by the Forest Service and could have negative impacts on aquatic systems.

Demand Species Affected Environment Table 4 displays the demand (game) fish species found in throughout the Cherokee National Forest (USDA Forest Service 2010a). All of these represent wild populations although rainbow trout are stocked into several streams at a catchable size. The 2008 Monitoring and Evaluation Report (USDA Forest Service 2009) describes the demand fish populations as meeting the public level of expectation with the exception of brook trout which is declining throughout its range.

Table 4. Aquatic demand species Common Name Status Populations rainbow trout Introduced/Desirable 312 brook trout Game 155 brown trout Introduced/Desirable 101 rock bass Game 80 bluegill Game 60 smallmouth bass Game 57 redbreast sunfish Game 43 green sunfish Game 37 largemouth bass Game 26 spotted bass Game 23 redeye bass Game 14 longear sunfish Game 13 redear sunfish Game 12 warmouth Game 3 shadow bass Game 3 Alabama bass Game 3 spotted sunfish Game 1

30 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Environmental Consequences Alternative 1 Alternative 1 would continue suppression of the HWA without adding additional treatment methods or expansion of treatment areas. No new adverse effects would occur to demand species as a result of selecting this alternative. Hemlock trees would continue to die in areas outside the limited treatment zones. Trout species could diminish in numbers do to warming of water. In the long term (decades) the summer shade that had been provided by hemlocks would eventually be provided by other species such as white pines and rhododendron.

Alternative 2 Alternative 2 would add the fungus treatment, add the use of the chemical dinotefuran, and increase the areas where these treatments and the ongoing treatments with Imidacloprid and beetle releases would occur. The Environmental Assessment - Conservation of Native Hemlock by Suppression of Hemlock Woolly Adelgid Infestations (USDA Forest Service. 2005) was implemented to protect selected stands (refuges) of hemlock from destruction by HWA infestation. These refuges were intended to: 1) reduce loss of remnant older hemlocks that contribute to old growth characteristics in the form of large trees and unique wildlife habitat; 2) reduce the loss of hemlock along native trout streams that provide nutrients in the form of large woody debris and protection from temperature increases in the streams; 3) protect rare plant communities that occur in hemlock dominated forest where viability is a concern for at least five species; 4) protect areas where hemlocks contribute to the character and quality of scenery and recreation settings especially where these aesthetic evergreen trees provide cover, shade and screening throughout the year; and 5) provide for a diverse genetic base of hemlock from which future populations could be established in the event wide spread hemlock mortality occurs. These areas represent direct treatment of only about 3100 individual hemlock trees. Based on monitoring (USDA Forest Service 2009) since the inception of this project “The Hemlock Wooly Adelgid continues to severely impact hemlock across the entire Forest. Chemical treatment of refuge areas appears to be very effective in protecting the trees from mortality. The effectiveness of predator beetle releases in protecting the hemlocks in larger land areas is unknown at this time.” Expansion of treatment areas and beetle releases appears to be warranted and should not adversely affect the aquatic demand species. Analysis conducted for the suppression of HWA in the Nantahala/Pisgah National Forests (USDA Forest Service, 2010b) and at Royal Blue State Wildlife Management Area (Costa, 2010), determined the naturally occurring fungus Lecanicillium muscarium was effective against HWA but had little effect on none target organisms, including aquatic insects. Application of this fungus is not expected to affect aquatic demand species. Dinotefuran and Imidacloprid both belong to a group of insecticides referred to as Neonicotinoids (derived from nicotine) which act on the central nervous system of insects. Both have low toxicity to fish (Valent USA Corporation, 2011; National Pesticide Information Center, 2011). Neonicotinoids are among the most widely used insecticides worldwide. Dinotefuran treatment is proposed because this chemical is more

31 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest quickly absorbed by hemlock roots and distributed throughout the tree. It is as effective as Imidacloprid but provides more immediate protection to the hemlocks. Imidacloprid provides a longer period of protection than Dinotefuran. Used in combination and following labeling instructions, these chemicals would greatly improve the efficacy of HWA suppression and would not adversely affect aquatic demand species.

Alternative 3 Alternative 3 is the same as Alternative 2 except that Dinotefuran would not be used. Refer to the effects analysis for alternative 2.

Cumulative Effects, all Alternatives Sedimentation of streams is the most significant adverse effect that contributes to cumulative effects to aquatic demand species on national forest lands. None of the three alternatives proposes any ground disturbance. Compliance with labeling instructions for the two chemicals would minimize or eliminate harmful impacts and cumulative adverse impacts to the aquatic demand species. Management activities, on National Forest System lands, that are reasonably foreseeable and under the control of the Forest Service would be implemented under the standards for protecting streams listed in the Revised Land and Resource Management Plan for the Cherokee National Forest (USDA Forest Service 2004a). Implemented in conjunction with any of the three alternatives, these activities would not increase the adverse cumulative effects on the aquatic demand species now occurring. Reasonably foreseeable activities that occur on private lands are not controlled by the Forest Service and could have negative impacts on aquatic demand species.

Threatened (T), Endangered (E), Sensitive (S), Viability Concern (VC) & Demand Species ______

Affected Environment Threatened and Endangered Species Botanical Resource The Cherokee National Forest tracks 291 rare plant species, seven of which are federally listed, 101 which are regional forester’s sensitive species, and an additional 183 that were determined to be of additional viability concern during the terrestrial viability analysis conducted during the 2004 Forest Plan revision (USDA Forest Service 2004a & 2004b).

In order to determine the sub-set of these species to be evaluated in this environmental assessment, habitat requirements and known occurrences on the Cherokee National Forest were considered for all species. The seven species of T&E plants that are either known to occur, or that could potentially occur, on the Cherokee National Forest are shown below in Table 5.

32 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Table 5. Threatened, Endangered, and Candidate Plants with Known Sites or Potential Habitat on the Cherokee National Forest Life Form Scientific Name Common Name Occurrence* Status* Nonvasc. Plant Gymnoderma lineare Rock gnome lichen P E Vascular Plant Geum radiatum Spreading avens P E Vascular Plant Hedyotis purpurea montana Roan Mountain bluet P E Vascular Plant Isotria medeoloides Small whorled pogonia N T Vascular Plant Pityopsis ruthii Ruth's golden aster P E Vascular Plant Solidago spithamaea Blue Ridge goldenrod P T Vascular Plant Spiraea virginiana Virginia spiraea H T *Occurrence: P=Present; N=Not Present; H=Historic *Status: E=Endangered; T=Threatened

None of the species in Table 5 have known sites that are coincident with hemlock stands on the Cherokee National Forest and none are dependent upon either eastern or Carolina hemlock forests for their continued existence. One species (Gymnoderma lineare) is found associated with eastern hemlock in other portions of its range, but is only known from high elevations (spruce-fir zone) on the Cherokee where it occurs on moist rock substrates that would not be affected by hemlock mortality. Based upon this, there would be no effect to any of these species as a result of this project and they are eliminated from further consideration.

Wildlife Resource In order to determine the sub-set of these species to be evaluated in this environmental assessment, habitat requirements and known occurrences on the Cherokee National Forest were considered for all species. The three wildlife species of T&E are either known to occur, or that could potentially occur, on the Cherokee National Forest are shown below in Table 6.

Table 6: Threatened and Endangered Wildlife Species with Known Occurrences in Stands with a Dominant Hemlock Component Scientific Name Common Name Freq1* Freq2* Status* Mammals Carolina Northern Flying Glaucomys sabrinus coloratus Squirrel 1 3 E Myotis grisescens Gray bat 1 8 E Myotis sodalis Indiana bat 1 4 E *Freq1 = Number of known occurrences within stands that have a dominant hemlock component *Freq2 = Number of known occurrences on the Cherokee national Forest *Status: S = Regional Forester’s Sensitive Species; VC = Viability Concern

Glaucomys sabrinus coloratus (Carolina northern flying squirrel) is endemic to the northern hardwood and spruce-fir forests of North Carolina and Tennessee (USFWS 1990). Carolina northern flying squirrels are mostly nocturnal and feed on a variety of foods including lichens, fungi, seeds, buds, fruits, insects, and other

33 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest animal material. These squirrels are gregarious and apparently live in family groups. Adults produce one litter of young in spring or summer. Carolina northern flying squirrels occupy two types of nests. In the spring and summer these squirrels construct leaf nests, but in fall and winter, seek the shelter of tree cavities and woodpecker holes. Occasionally they will use the dense branches in the tops of spruce and fir trees or underground burrows for shelter. Artificial nest boxes are also used. Life expectancy is generally less than four years (Linzey 1995). Principle predators include: owls, weasel, and bobcat (Choate et al. 1994). Threats to the species include long-term vegetation changes, habitat alteration, and loss of habitat.

Myotis grisescens (gray bat) is found throughout the limestone region of southern middle-western and southeastern United States (Whitaker 1998). They use caves year round for hibernating, maternity colonies, and roosting. Most individuals migrate seasonally between maternity and hibernating caves. About 95 percent of the known population inhabits nine winter caves. Both major hibernacula and Priority 1 maternity caves are known from Tennessee, but neither hibernacula nor maternity caves are known from the CNF.

Gray bats forage for insects over water along riparian areas and shorelines with forest cover (Mitchell 2001). All captures on the CNF occurred on the Unaka Ranger District within the French Broad River and Nolichucky River watersheds (Cochran et al. 2000; Libby et al. 2004). One adult male banded on the CNF was recaptured in a privately owned maternity cave approximately nine miles from the original capture site. Based on these results, CNF provides riparian foraging habitat for nearby nursery colonies.

Myotis sodalis (Indiana bat) distribution is generally associated with limestone caves in the eastern U.S. Within this range, the bats occupy two distinct types of habitat. During summer months, maternity colonies roost under sloughing bark of dead and partially- dead trees of many species, often in forested settings (Callahan et al. 1997). Reproductive females require multiple alternate roost trees to fulfill summer habitat needs. Adults forage on winged insects within three miles of the occupied maternity roost. Swarming of both males and females and subsequent mating activity occurs at cave entrances prior to hibernation. During this autumn period, bats roost under sloughing bark and in cracks of dead, partially-dead and live trees.

Aquatic Resource There are 20 (Table 7) aquatic Threatened and Endangered (T&E) species on or near the Cherokee National Forest and 11 of them have Critical Habitat on the forest (CNF TES List 2001). All of the T&E species and their Critical Habitats are restricted to five large rivers: Conasauga, Hiwassee, Tellico, Citico and Nolichucky. The Biological Evaluation contains the complete list and the explanation for deciding which of these species could be affected by these alternatives.

34 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Table7. Aquatic T&E Species and Critical Habitat on Forest T&E Critical Habitat Fish 8 2 Mussels 12 9

Sensitive and Viability Concern Species Using a Geographic Information System, the database for known locations of rare plant and wildlife species was queried for occurrences that coincide with forested stands that contain a dominant hemlock component. A similar analysis was performed for the 2005 environmental assessment however in some cases there have been changes in mapped boundaries of stands dominated by hemlock. Likewise, revisions have been made to the state rare species database, thus numbers of sites and species considered may vary from the 2005 analysis. The results of this query are shown in Table 8. Frequency of occurrence for each species is shown within stands with a hemlock component and for the entire forest.

Table 8: Botanical and Wildlife Regional Forester’s Sensitive Species and Other Species of Viability Concern with Known Occurrences in Stands with a Dominant Hemlock Component Scientific Name Common Name Freq1* Freq2* Status* Non-Vascular Plants Aneura maxima A Liverwort 1 1 S Hydrothyria venosa A Lichen 1 2 S Megaceros aenigmaticus A Hornwort 5 38 S Porella wataugensis Watauga Porella 1 3 S Jungermannia fossombronioides A Liverwort 1 1 VC Vascular Plants Buckleya distichophylla Piratebush 3 13 S Cardamine clematitis Mountain Bittercress 1 16 S Carex roanensis Roan Mountain Sedge 1 45 S Diervilla rivularis Mountain Bush-honeysuckle 2 11 S Gentiana austromontana Appalachian Gentian 2 102 S Helianthus glaucophyllus White-leaved Sunflower 1 13 S Hypericum mitchellianum Blue Ridge St. John's-wort 1 10 S Juglans cinerea Butternut 2 18 S Scutellaria saxatilis Rock Skullcap 6 56 S Stachys clingmanii Clingman's Hedge-nettle 2 12 S Thermopsis fraxinifolia Ash-leaved Bush-pea 2 31 S Trillium rugelii Southern Nodding Trillium 2 6 S Tsuga caroliniana Carolina Hemlock 12 57 S Acer leucoderme Chalk Maple 3 32 VC

35 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Table 8: Botanical and Wildlife Regional Forester’s Sensitive Species and Other Species of Viability Concern with Known Occurrences in Stands with a Dominant Hemlock Component Scientific Name Common Name Freq1* Freq2* Status* Adlumia fungosa Climbing Fumitory 2 13 VC Cardamine rotundifolia Round-leaf Watercress 3 17 VC Carex ruthii Ruth's Sedge 4 30 VC Clintonia borealis Clinton's Lily 4 18 VC Corydalis sempervirens Pale Corydalis 1 10 VC Cymophyllus fraserianus Fraser's Sedge 11 62 VC Dryopteris carthusiana Spinulose Shield Fern 1 4 VC Glyceria laxa Northern Mannagrass 1 1 VC Goodyera repens Dwarf Rattlesnake-plantain 4 16 VC Heracleum maximum Cow-parsnip 1 13 VC Hieracium scabrum Rough Hawkweed 1 7 VC Hydrophyllum virginianum Appalachian Waterleaf 1 31 VC Lilium canadense Canada Lily 1 4 VC Melanthium latifolium Broadleaf Bunchflower 1 20 VC Platanthera orbiculata Large Round-leaved Orchid 6 80 VC Platanthera psycodes Small Purple Fringed Orchid 1 16 VC Poa palustris Fowl Bluegrass 1 4 VC Prunus virginiana Choke Cherry 1 4 VC Pyrola americana American Wintergreen 2 16 VC Streptopus roseus Rosy Twisted-stalk 2 16 VC Symplocos tinctoria Horse-sugar 2 25 VC Trichomanes petersii Dwarf Filmy-fern 1 10 VC Veronica americana American Speedwell 1 1 VC Woodsia scopulina ssp. appalachiana Alleghany Cliff-fern 2 6 VC Xerophyllum asphodeloides Eastern Turkeybeard 2 28 VC Amphibians Desmognathus aeneus Seepage Salamander 9 32 VC Carolina Mountain Dusky Desmognathus carolinensis Salamander 1 5 S Plethodon aureolus Tellico Salamander 1 17 S Plethodon welleri Weller's Salamander 2 15 S Birds Aegolius acadicus Northern Saw-whet Owl 1 3 VC Aquila chrysaetos Golden Eagle 1 1 VC Limnothlypis swainsonii Swainson's Warbler 4 10 VC Sphyrapicus varius Yellow-bellied Sapsucker 2 7 VC Insects

36 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Table 8: Botanical and Wildlife Regional Forester’s Sensitive Species and Other Species of Viability Concern with Known Occurrences in Stands with a Dominant Hemlock Component Scientific Name Common Name Freq1* Freq2* Status* Speyeria diana Diana fritillary 1 32 S Mammals Myotis leibii Eastern Small-footed bat 1 20 S Napaeozapus insignis Woodland Jumping Mouse 3 38 VC Sorex cinereus Cinereus Shrew 6 54 VC Sorex fumeus Smoky Shrew 7 70 VC Sorex longirostris Southeastern Shrew 2 11 VC Synaptomys cooperi Southern Bog Lemming 2 7 VC Zapus hudsonius Meadow Jumping Mouse 1 9 VC Snails Glyphyalinia pentadelphia Pink Glyph 1 14 VC Inflectarius ferrissi Smoky Mountain Covert 1 1 VC Mesomphix rugeli Wrinkled Button 2 13 VC Paravitrea lamellidens Lamellate Supercoil 1 6 VC Stenotrema altispira Highland Slitmouth 1 10 VC Ventridens decussatus Crossed Dome 2 13 VC *Freq1 = Number of known occurrences within stands that have a dominant hemlock component *Freq2 = Number of known occurrences on the Cherokee national Forest *Status: S = Regional Forester’s Sensitive Species; VC = Viability Concern *Freq1 = Number of known occurrences within stands that have a dominant hemlock component *Freq2 = Number of known occurrences on the Cherokee national Forest *Status: S = Regional Forester’s Sensitive Species; VC = Viability Concern

The species listed in Table 8 all have at least one known occurrence on the Forest associated with a plant community with a dominant hemlock component. In order to determine any other potential species that could be affected, the list of species in Appendix E (Terrestrial Viability Analysis) to the Final Environmental Impact Statement for the Revised Land and Resource Management Plan for the Cherokee National Forest (USDA Forest Service 2004b) was reviewed. This appendix lists 39 plant species as being associated with habitat types that could have a significant hemlock component. Of these species, 27 are not included in Table 8 (i.e. do not have known sites that are coincident with hemlocks) and represent additional species to be considered in this environmental assessment. These additional species are listed below in Table 9. This appendix lists 25 terrestrial wildlife species as being associated with habitat types that could have a significant hemlock component. Of these species, 21 are not included in Table 8 (i.e. do not have known sites that are coincident with hemlocks) and represent additional species to be considered in this environmental assessment. These additional species are listed below in Table 9.

37 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Table 9: Botanical and Wildlife Regional Forester’s Sensitive Species and Other Species of Viability Concern that could be Associated with Habitats with a Significant Component of Eastern Hemlock or Carolina Hemlock but have no Known Occurrences Within Those Habitats (Data from Appendix E to the Final Environmental Impact Statement for the Revised Land and Resource Management Plan for the Cherokee National Forest) Scientific Name Common Name Freq1* Freq2* Status* Non-Vascular Plants Acrobolbus ciliatus Liverwort 0 1 S Cheilolejeunea evansii Liverwort 0 1 S Drepanolejeunea appalachiana Liverwort 0 4 S Fissidens appalachensis Appalachian pocket moss 0 1 S Lejeunea blomquistii Liverwort 0 2 S Lophocolea appalachiana Liverwort 0 2 S Nardia lescurii Liverwort 0 3 S Pellia appalachiana Liverwort 0 3 VC Plagiochila echinata Liverwort 0 5 S Plagiochila sullivantii var. spinigera Sullivant's leafy liverwort 0 1 S Plagiochila sullivantii var. sullivantii Sullivant's leafy liverwort 0 1 S Radula voluta Liverwort 0 1 S Riccardia jugata Liverwort 0 3 S Anastrophyllum helleranum Liverwort 0 1 VC Frullania plana Liverwort 0 1 VC Plagiochila ludoviciana Liverwort 0 1 VC Radula tenax Liverwort 0 4 VC Vascular Plants Fothergilla major Witch alder 0 4 S Carex gracillima Graceful sedge 0 F1 VC Chrysosplenium americanum Golden saxifrage 0 F3 VC Dirca palustris Leatherwood 0 F3 VC Liparis liliifolia Large twayblade 0 F3 VC Lygodium palmatum Climbing fern 0 7 VC Mertensia virginica Virginia bluebell 0 F2 VC Populus grandidentata Large-tooth aspen 0 5 VC Stewartia ovata Mountain camellia 0 F3 VC Trillium undulatum Painted trillium 0 F3 VC Amphibians Ambystoma talpoideum Mole salamander 0 F1 VC Desmognathus ochrophaeus Mountain dusky salamander 0 F3 VC Santeetlah dusky Desmognathus santeetlah 0 4 S salamander Eurycea junaluska Junaluska salamander 0 8 S Eurycea longicauda Longtail salamander 0 F3 VC Plethodon dorsalis Zigzag salamander 0 F2 VC Pseudacris brachyphona Mountain chorus frog 0 F2 VC Pseudotriton montanus Eastern mud salamander 0 F2 VC Birds Accipiter striatus Sharp-shinned hawk 0 F3 VC Dendroica cerulea Cerulean warbler 0 5 VC

38 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Regional Forester’s Sensitive Species Buckleya distichophylla (pirate bush) is known from thirteen locations on the Cherokee National Forest, only three of which overlay with locations identified as hemlock stands. This species is a southern Appalachian endemic, only known to occur at a few, widely scattered locations in the mountains of southern Virginia, western North Carolina, and northeastern Tennessee. The species inhabits open, dry, rocky woods and bluffs where it is a root parasite on hemlock and other tree species (Weakley 2011). The plant's rarity may be explained by its specific habitat requirements, which include both host plant(s) and canopy openings. NatureServe (2011) reports that pirate bush typically occurs in mountain woods at lower elevations (450-1100 m) within a variety of habitats including Virginia pine and white pine and acidic mixed-oak forests. The plants can be found scattered among host trees within openings of hemlock forests, but habitats also include south-facing slopes and chestnut oak forests. Many of the known occurrences contain a dense understory of Rhododendron maximum. Although some of the current element occurrences are found within heavily shaded areas it is noted that the most robust plants occur in areas with more sunlight.

Cardamine clematitis (mountain bittercress) is known from sixteen locations on the Cherokee National Forest, only one of which overlays with locations identified as hemlock stands. This species is a southern Appalachian endemic known from the mountains of southern Virginia, North Carolina, Tennessee, and northern Georgia. It

39 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest occurs at high elevations (generally above 3,500') in wet, rocky habitats including springs, seeps, and streambanks (Weakley 2011).

Carex roanensis (Roan Mountain sedge) is known from forty-five locations on the Cherokee National Forest, only one of which overlays with locations identified as hemlock stands. This species is known from Pennsylvania and Virginia, south through West Virginia, eastern Kentucky, North Carolina, Tennessee, and Georgia (Weakley 2011). Unlike many sedges that are restricted to specific wetland habitats, Roan Mountain sedge is known to occur in a variety of mesic forested habitats including rich cove and northern hardwood forests. NatureServe (2011) reports the species from rich soils of mesic forests, associated with beech and birch at relatively high elevations.

Diervilla. rivularis (mountain bush honeysuckle) is known from eleven locations on the Cherokee National Forest, only one of which overlays with locations identified as hemlock stands. This species is a southern Appalachian endemic, currently known to occur in Tennessee, Georgia, Alabama, and North Carolina. This species usually occurs on bluffs, rock outcrops, or riverbanks, from moderate to high elevations. (Weakley 2011), but is found locally along the Ocoee River at approximately 1,000' elevation.

Fothergilla major (mountain witch alder) is known from four locations on the Cherokee National Forest with none overlaying with locations identified as hemlock stands. Mountain witch alder is distributed from Arkansas east to Tennessee, Alabama, Georgia, and the Carolinas. It typically occurs on dry, ridgetop forests of moderate elevations, especially along the Blue Ridge escarpment (Weakley 2011).

Gentiana austromontana (Appalachian gentian) is known from over one hundred locations on the Cherokee National Forest, only two of which overlay with locations identified as hemlock stands. This southern Appalachian endemic is known from West Virginia and Virginia, south to the mountains of North Carolina and Tennessee. Plants are typically found at medium to high elevations in open forests, or grassy balds (Weakley 2011). Many Cherokee occurrences are along forest roads and trails.

Helianthus glaucophyllus (white-leaved sunflower) is known from thirteen locations on the Cherokee National Forest, only one of which overlays with locations identified as hemlock stands. This species is a narrow southern Appalachian endemic known only to occur in Tennessee and the Carolinas. It is typically found in mesic forests and woodlands at medium elevations. Flowering specimens are usually found in areas with increased light, such as canopy gaps and along roadsides. Sometimes associated with Coreopsis latifolia (Weakley 2011). This species is usually associated with mesic forest types that favor a hardwood component in the overstory.

Hypericum mitchellianum (Blue Ridge St. John’s wort) is known from ten locations on the Cherokee National Forest, only one of which overlays with locations identified as hemlock stands. This southern Appalachian endemic occurs at moderate to high elevations in the mountains of West Virginia, Virginia, Tennessee and North Carolina. Habitat includes grassy balds, seeps, and forest openings. This species is often associated

40 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest with Hypericum graveolens with which it may hybridize and produce populations with intermediate characteristics (Weakley 2011).

Juglans cinerea (butternut) is known from eighteen locations on the Cherokee National Forest, only two of which overlay with locations identified as hemlock stands. This formerly common tree ranges across most of the eastern United States except the deep south. Butternut canker disease has severely depleted populations and the species is considered to be in severe decline. Typical habitat is moist, rich forests, especially along creeks and rivers, in bottomlands and floodplains.

Scutellaria saxatilis (rock skullcap) is known from fifty-six locations on the Cherokee National Forest, six of which overlay with locations identified as hemlock stands. Weakley (2011) describes this species from Delaware, Ohio, and Indiana south to South Carolina and Tennessee, where it inhabits rocky, dry to mesic forests and open areas. NatureServe (2011) describes a slightly broader range and states the primary threats to this species are loss of forest canopy (which affects the moist microclimate of forest floor) and invasion of exotic species such as Microstegium vimineum and Lonicera japonica.

Stachys clingmanii (Clingman’s hedge-nettle) is known from twelve locations on the Cherokee National Forest, only two of which overlay with locations identified as hemlock stands. This narrow southern Appalachian endemic is known only from the mountains of southwestern North Carolina and adjacent eastern Tennessee. Habitat includes cove forests and rich, boulder forests, mostly at high elevations (Weakley 2011). NatureServe (2011) also reports the species from South Carolina and further describes the habitat as clearings and forest meadows at high altitudes in the Appalachian mountains, on calcareous soils of forest edges, meadows, and clearings, often previously disturbed by fire.

Thermopsis fraxinifolia (ash-leaved bush-pea) is known from thirty-one locations on the Cherokee National Forest, only two of which overlay with locations identified as hemlock stands. This southern Appalachian endemic ranges from North Carolina and Tennessee, south to northern portions of Georgia and South Carolina. Habitat includes openings in dry woodlands and ridges (Weakley 2011) and on the Cherokee National Forest most sites have been found along road banks and trails.

Trillium rugelii (southern nodding trillium) is known from six locations on the Cherokee National Forest, two of which overlay with locations identified as hemlock stands. This species ranges from western North Carolina and eastern Tennessee, south to South Carolina, Georgia, and Alabama. Habitat includes rich forests and coves, usually underlain with mafic or calcareous rock (Weakley 2011). NatureServe (2011) reports habitat as rich woodlands and forest over mafic or calcareous rocks, often found near Rhododendron catawbiense. The general habitat is moist, but well drained. Trillium rugelii is found at lower slope elevations, over limestone, dolomite, or marble where forest vegetation is dominated by closed or nearly closed canopy of mesophytic trees

41 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest including calciphilic or basophilic species. Some typical canopy species in this community include Quercus muehlenbergii, Juglans nigra, and Fraxinus Americana.

Tsuga caroliniana (Carolina hemlock) is known from fifty-seven locations on the Cherokee National Forest, only twelve of which overlay with locations identified as hemlock stands. This southern Appalachian endemic ranges from Virginia, south through Tennessee and North Carolina, to northern portions of Georgia and South Carolina. This tree occurs on ridge tops, rocky bluffs, and open forests, generally on drier, rockier sites than Tsuga canadensis, though the two species have been found intermixed in humid gorges (Weakley 2011). Most Cherokee National Forest records are for scattered individuals rather than Carolina hemlock forested communities.

Acrobolbus ciliatus (a liverwort) is currently known from just one location on the Forest (F1), collected at Falls Branch Falls in the Citico Wilderness. This site is not listed as coincident with a stand dominated by hemlock though it is likely there are at least some hemlocks in the canopy. This species is known from the mountains of North Carolina, Tennessee, South Carolina, and Georgia with disjunct populations in Alaska, Japan, Taiwan, and India (Hicks 1992). Habitat in our area includes rock substrates in moist ravines, spray cliffs, and cascading streams. The species is also known to occur in humid, high elevation spruce/fir forests.

Aneura maxima (A Liverwort) is a rare Appalachian endemic known from the mountains of Vermont south to North Carolina and Tennessee. The species typically occurs on humus or gravelly soil at the base of wet outcrops, along streams, and around waterfalls (Hicks 1992, NatureServe 2011). One known site has been documented on the Cherokee National Forest and it occurs in a stand dominated by eastern hemlock.

Cheilolejeunea evansii (A liverwort) is currently known from one location on the Forest (F1), collected at Falls Branch Falls in the Citico Wilderness. This site is not listed as coincident with a stand dominated by hemlock though it is likely there are at least some hemlocks in the canopy. This species is considered to be a rare endemic currently known only from Jackson and Transylvania counties in North Carolina, Oconee and Pickens counties in South Carolina (Hicks 1992), Lawrence county, Alabama (NC Heritage and USFWS 1997), and Monroe county, Tennessee (Collected by Sara Noble, 2001). It is known to occur on the bark of various hardwood species in humid gorges and in relatively open microsites within shaded gorges. The species is typically found from ground level to a height of 3 meters up the trunk on a variety of mesic to dry-mesic hardwoods such as Quercus spp., Liriodendron tulipifera, Nyssa sylvatica, Carya spp., Liquidambar styraciflua, Fraxinus spp., and Ilex opaca (NatureServe 2011).

Drepanolejeunea appalachiana (A liverwort) is currently known from four locations on the Forest (F1) though none are listed as coincident with a stand dominated by hemlock. This species is a southern Appalachian endemic, known from the mountains of Virginia, North Carolina, South Carolina, Georgia, and Tennessee. It is known to occur on the bark of trees in humid gorges (Hicks 1992) and on moist rock and Rhododendron (North Carolina Natural Heritage Program 1999). NatureServe (2011) reports that while the

42 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest species is usually restricted to sites near flowing water it can occur on the bark and twigs of a variety of tree and shrub species and is not necessarily restricted to gorges. Outside of escarpment gorges it is known from the bark of Betula alleghaniensis, B. lenta, and Tsuga canadensis usually near the borders of cascading streams. Substrates at the locations of the four Cherokee collections are moist rock, Quercus alba and Betula alleghaniensis.

Fissidens appalachensis (Appalachian pocket moss) is currently known from one location on the Forest (F1), collected at Falls Branch Falls in the Citico Wilderness. This site is not listed as coincident with a stand dominated by hemlock though it is likely there are at least some hemlocks in the canopy. This species is known from the mountains of North Carolina and Tennessee where it occurs in rock crevices submerged in swift running, shallow water of Appalachian streams (Crum and Anderson 1981).

Hydrothyria venosa (A Lichen) is currently known from two locations on the Cherokee National Forest, one of which is coincident with a stand dominated by hemlock. The species is known from the Sierra Mountains of California, north through the mountains of the Pacific Northwest and western Canada, localized in western Montana and the Idaho panhandle, and from southeastern Canada south along the Appalachian mountain chain to Tennessee (Brodo, et al. 2001). Habitat throughout its range is rock substrates within clear, free-flowing mountain streams.

Lejeunea blomquistii (A liverwort) is currently known from two locations on the Forest (F1) neither of which is coincident with stands dominated by hemlock. This species is a southern Appalachian endemic, known from the mountains of North Carolina, Georgia, and Tennessee. It is known to occur on rock and tree bark in humid gorges (Hicks 1992) and on dead trees or vertical rock faces in the spray zone of waterfalls (North Carolina Natural Heritage Program 1999).

Lophocolea appalachiana (A Liverwort) is known from two locations on the forest neither of which is coincident with stands dominated by hemlock. This rare liverwort is restricted to shady gorges in mountain counties of North Carolina and Tennessee (Hicks 1992). NatureServe (2011) indicates that this species is known from ten extant occurrences and five to seven historical occurrences in North and South Carolina, Tennessee, and Kentucky. Habitat for this species is typically within shaded riparian zones, on wet, non calcareous rock faces and spray zones, dripping rocks, usually by waterfalls or cascades (NatureServe 2011).

Megaceros aenigmaticus (A Hornwort) is known from thirty-eight locations on the Cherokee National Forest, five of which overlay with locations identified as hemlock stands. This species is a rare endemic confined to the southwest corner of NC and adjacent TN. It is typically found on wet, shaded rocks, in and along streams, from 2,000' to 2,800', often in association with Dumortiera hirsuta (Hicks 1992). This species is locally common in Tellico River Gorge and found well outside the range of elevations listed above. According to NatureServe (2011) the species is often found on shaded rocks in small streams, springs, or waterfall spray zones, especially in small shaded streams forming headwaters of rivers. The species apparently requires cool, non-turbid

43 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest waters and is very sensitive to changes in water quality such as sedimentation, run-off from farms, and increased water temperatures due to canopy removal in adjacent forests. Undisturbed streams within the species' small range have thriving populations, while nearby streams in farmed and developed areas lack or have very reduced populations. In some areas Megaceros aenigmaticus appears to be threatened by habitat disturbance, such as intensive farming, forest clearing, or new home development that can create run- off into nearby streams or open up the forest canopy letting in drying and warming sunlight (Hicks and Amoroso 1996).

Nardia lescurii (A liverwort) is currently known from three locations on the Forest (F1), none of which are coincident with stands dominated by hemlock though it is likely that hemlock is at least present in the canopy. This species is an Appalachian endemic, known from Virginia, West Virginia, Kentucky, North Carolina, South Carolina, Georgia, and Tennessee. It is typically found at low elevations in the mountains on peaty soil over rock near well shaded streams (Hicks 1992).

Pellia appalachiana (A liverwort) is currently known from three locations on the Forest (F1), none of which are coincident with stands dominated by hemlock though it is likely hemlock is at least present in the canopy. According to NatureServe (2011), Pellia appalachiana is currently believed to be present in Alabama, Georgia, South Carolina, North Carolina, Tennessee, and Kentucky. It has also been (disjunctly) reported from eastern Minnesota. Collections since 1992 indicate that there are at least 20-80 extant occurrences of this species, and it is believed to be globally secure. This taxon was previously thought to be a hybrid between P. epiphylla and P. neesiana, but recent morphological and genetic evidence supports classification as a distinct species (Self and Crandall-Stotler 2001). The species occurs on soil along shaded creek banks, on soil on the floor of sandstone rock-houses with moist air, and sometimes on rock outcrops or cliffs near waterfalls. Able to colonize newly exposed soil along creeks. In general, occupied sites are permanently damp to wet, where submersion will not occur but where the plants are not subject to drought. Surrounding plant community is typically mid- to late-successional riparian deciduous forest, within the Appalachian (Hemlock) - Northern Hardwood forest ecological system (NatureServe 2011).

Plagiochila echinata (A liverwort) is currently known from five locations on the Forest (F1), none of which are coincident with stands dominated by hemlock though it is likely hemlock is at least present in the canopy. This species is a southern Appalachian endemic known from North Carolina, South Carolina, and Tennessee. Plants are typically found on damp, shaded rock faces and crevices in mountain gorges, above cascades and near waterfalls (Hicks 1992).

Plagiochila sullivantii var. spinigera (A liverwort) is currently known from one location on the Forest (F1), collected at Falls Branch Falls in the Citico Wilderness. This site is not listed as coincident with a stand dominated by hemlock though it is likely there are at least some hemlocks in the canopy. This species is a southern Appalachian endemic known from the mountains of Virginia, West Virginia, North Carolina, South Carolina, and Tennessee. Plants are typically found on moist, shaded rock outcrops,

44 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest under cliff ledges, and in rock crevices (Hicks 1992), on the moist rocks in the spray zones of waterfalls and in spruce/fir forests (North Carolina Natural Heritage Program 1999), ranging in elevation from 2,500 feet to the spruce/fir zone (NC Heritage & USFWS 1997). NatureServe (2011) lists loss of hemlock and resulting microclimatic changes as one of the primary threats to this species.

Sullivant's leafy liverwort (Plagiochila sullivantii var. sullivantii) is currently known from one location on the Forest (F1), collected at Falls Branch Falls in the Citico Wilderness. This site is not listed as coincident with a stand dominated by hemlock though it is likely there are at least some hemlocks in the canopy. This species is a broad Appalachian endemic known from the mountains of Virginia, West Virginia, Kentucky, North Carolina, South Carolina, Georgia, and Tennessee. Plants are typically found on moist, shaded rock outcrops, under cliff ledges, and in rock crevices (Hicks 1992), on the moist rocks in the spray zones of waterfalls and in spruce/fir forests (North Carolina Natural Heritage Program 1999), ranging in elevation from 2,500 feet to the spruce/fir zone (NC Heritage & USFWS 1997). The species is known historically from about thirty sites with only six currently extant (NatureServe 2011).

Porella wataugensis (Watauga porella) is currently known from three locations on the Forest (F1), one of which is coincident with a stand dominated by hemlock. According to Paul Davison (Pers. Comm. 2001) this is the same taxon as Porella japonica ssp. appalachiana which is listed separately on the RF Sensitive Species List. This is a rare southern Appalachian endemic known from vertical rock faces in humid gorges in North Carolina, Tennessee, and South Carolina (Hicks 1992). Known sites for this species occur along waterways within Wilderness areas on the Cherokee National Forest (Bald River and Citico Creek Wilderness) and likely have at least some hemlock in the canopy.

Radula voluta (A liverwort) is currently known from one location on the Forest (F1), collected at Falls Branch Falls in the Citico Wilderness. This site is not listed as coincident with a stand dominated by hemlock though it is likely there are at least some hemlocks in the canopy. This species is a rare disjunct from Europe and South America, known in our area from the mountains of North Carolina and Tennessee. The species is known to occur on shady rock faces in spray areas around waterfalls (Hicks 1992).

Riccardia jugata (A liverwort) is currently known from three locations on the Forest (F1), none of which are coincident with stands dominated by hemlock though it is likely hemlock is at least present in the canopy at these sites. This species is a southern Appalachian endemic known from the mountains of North Carolina and Tennessee. The species is known to occur on moist wood and humus in mesic areas and humid gorges (Hicks 1992).

Corynorhinus rafinesquii (Rafinesque's big-eared bat) ranges widely over the southern states from Virginia, West Virginia, Ohio, Indiana, and Illinois south to the Gulf of Mexico; west to Louisiana, Oklahoma, and eastern Texas. It inhabits forested regions. Hibernation in the north and in mountainous regions most often occurs in caves or similar sites; small caves are selected, and the bats stay near the entrance and are thought to move about in winter. Winter habitat in the south is poorly known. Summer roosts often

45 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest are in hollow trees, occasionally under loose bark, or in abandoned buildings in or near wooded areas, instead of being restricted to caves (NatureServe 2001).

Desmognathus carolinensis (Carolina Mountain Dusky Salamander) is one of the most common species in the Appalachian Mountains (Petranka 1998). It ranges from the Doe River to the Pigeon River. This salamander concentrates near seeps, springs, and streams at lower elevations and during the winter. It may venture into adjacent wooded areas in wet weather and is often found on wet rock faces. It is more terrestrial at higher elevations. It seeks refuge under cover such as rocks and logs in the day (Petranka 1998). It feeds on terrestrial invertebrates and is active both night and day.

Desmognathus santeetlah (Santeetlah dusky salamander) has a small range in the southern Appalachians. Populations are stable, with no known significant existing threats. They can be found at higher elevations of the Unicoi, Cheoah, Great Smoky, and Great Balsam mountain ranges of the southwestern Blue Ridge Physiographic Province in western North Carolina and eastern Tennessee. These salamanders occur at elevations of 2100-5920 feet. Their habitat includes stream headwaters and seepage areas, where ground water percolates to surface through muck, mossy rocks, impatiens, and nettles. Eggs are laid mostly beneath mosses growing on rocks, on logs, or on soil surface. They nest commonly beneath mosses on logs lying in and around seepage areas. There are at least a few dozen distinct occurrences (assuming a separation gap of 1 km between occurrences) (NatureServe 2011).

Eurycea junaluska (Junaluska salamander) have a known range encompassing a portion of the Blue Ridge Mountains in southwestern North Carolina and southeastern Tennessee. No populations are known to have been lost since the species was described, but the population at the type locality on the Cheoah River may be extirpated and the species has not been collected at Snowbird Creek since 1994. Populations are possibly stable in some areas (e.g., Santeetlah Creek), but specific rangewide information on population trends is not available. Tennessee populations appear to be stable and may be larger than those in North Carolina. Rangewide, this species currently appears to be not very threatened. Potential threats include siltation due to logging, road construction for logging activities, urban development, and other activities that would negatively impact water quality. Adults hide under objects in or along streams and can be found on roads on rainy nights (NatureServe 2001).

Myotis leibii (Eastern small-footed bat) is found in rocky mountainous areas from Quebec southwest along the Southern Appalachians to northern Georgia, and west to Oklahoma. Abundance is extremely difficult to assess, and populations and occurrences are relatively scattered and small throughout its range. Several bachelor colonies and two maternity colonies have been observed in bridges, mines and rock crevices during the period 2000-2003 (G. Libby, Personal communication). Summer roosts include rock outcrops and cliffs, rock faults and crevices, bridge expansion joints, and abandoned mines and buildings. Rocky areas or bridges with sun exposure in a forested landscape may be important maternity site features. These bats hibernate singly or in small groups in caves, mines and buildings and are often found under talus and rocks on cave floors or

46 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest wedged into cracks and crevices. Known threats include direct human disturbance of roosts, and landscape changes that alter habitat parameters of roosts or hibernacula. Snag retention is important.

Plethodon aureoles (Tellico salamander) has a small range between the Little Tennessee and Hiwassee rivers on the western slopes of the Unicoi Mountains and nearby lowlands in northeastern Polk and eastern Monroe counties, Tennessee, and in northwestern Graham and northwestern Cherokee counties, North Carolina. Compared to larger ambystomatid salamanders, the home ranges of plethodontids tend to be very small on the order of a few meters to a few dozen meters in diameter. Yet, on occasion, dispersing plethodontids likely travel at least several hundred meters. Microhabitat types utilized by plethodontid salamanders (rocky substrates, downed wood, and leaf litter) help define their distribution (NatureServe 2001). They have been recorded at 17 sites on CNF.

Plethodon welleri (Weller’s salamander) occurs between Mt. Rogers in Virginia just into western North Carolina and northeast Tennessee (Petranka 1998). Weller’s salamanders inhabit spruce-fir, birch-hemlock, and cove hardwood forests, grassy spots, and boulder fields from 2200-5900 feet. The most robust populations are in spruce-fir forests. They tend to be associated with rocky substrates and breeds under moss mats of rotting conifer logs. They retreat to cover under rocks, logs, or leaf litter during the day (NatureServe 2004).

Weller’s salamanders are stable to declining within their small range. Populations are apparently isolated from each other, but the existence of low elevation populations suggests that there are additional populations in between. The small natural distribution and restriction to higher elevations are believed to be the primary limiting factors to populations, although effects of forest clearing, catastrophic fires, or spruce-fir forest decline could have significant effects on isolated populations (NatureServe 2004). In Tennessee 21 locations are known, mostly in Unicoi County. They have been found in at least 13 locations on the CNF.

Sorex palustris punctulatus (Southern water shrew) is distributed in the unglaciated Appalachians in Pennsylvania, Maryland, Virginia, West Virginia, Tennessee, and North Carolina. The range and abundance in the Appalachians have apparently been declining with climatic warming since the last Ice Age. About 27 occurrences are known throughout its range. Only four locations have been documented on the CNF, all in Monroe County. Few individuals have been collected during surveys, suggesting small population sizes; however, the potential riparian habitat is extensive. They are found along mountain streams, especially shaded sections in northern hardwood and subalpine conifer forests; also, peatlands with small streams. Cool, swift running streams with overhanging banks, roots, and rocks provide its preferred habitat (Choate et al. 1994). Other habitats parameters include moss-covered rocks and rhododendron on the banks, and yellow birch as one of the main canopy trees; other trees in the habitat may include hemlock, red spruce,

47 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest red maple, sugar maple, beech, or tulip trees. These shrews eat primarily immature stages of aquatic insects, obtained in the water.

Speyeria diana (Diana fritillary) is primarily found in the mountains from central Virginia and West Virginia through the western Carolinas and eastern Tennessee into extreme north Georgia and Alabama. It is somewhat common in a small area from southwestern Virginia to the Great Smokies region and rare and sporadic elsewhere. In the past it apparently underwent a major range wide decline resulting in loss of substantial portion of historic range. However, some workers believe it is again increasing in some areas where second growth forests are becoming mature, and where gypsy moth spraying is not presently widespread (NatureServe 2004). It is known to occur in all the counties of the northern CNF and has been observed in at least 15 locations in recent years. Habitat for this species includes open areas within rich, moist mountain forests (Glassberg 1999). The Diana fritillary lays eggs in meadow among the grass. The caterpillars hatch, hibernate over the winter, and then crawl to nearby violets in the springtime. Adults are present from June to September with males emerging before females (NatureServe 2004).

Viabilty Concern Species Acer leucoderme (chalk maple) is known from 32 locations on the forest, only three of which overlay with locations identified as hemlock stands. This species is primarily from the southeast Piedmont (North Carolina to Alabama) but has documented occurrences from Oklahoma, Arkansas, Texas, Louisiana, and Mississippi. It typically occurs on rocky slopes and bluffs, particularly over mafic or calcareous substrates (Weakley 2011), and in moist woods along rivers and ravines (Wofford 1989). Acer leucoderme is locally common within the Ocoee and Hiwassee River drainages.

Adlumia fungosa (climbing fumitory) is known from thirteen locations on the Forest, only two of which overlay with locations identified as hemlock stands. This species ranges from Quebec west to Wisconsin and Minnesota, south to Delaware, North Carolina, Tennessee, and Indiana. Habitats include cliffs, talus, rocky slopes, rich stream bottom forests, and cool rocky forests (Weakley 2011). Wofford (1989) also cites rich coves and cliffs habitats, but notes that the species is often found on disturbed sites.

Cardamine rotundifolia (round-leaf watercress) is known from seventeen locations on the Forest, three of which overlay with locations identified as hemlock stands. This species is a broad endemic of the central Appalachians ranging from Pennsylvania and western New York, west to Ohio and Kentucky, and south to the mountains and Piedmont of North Carolina. Habitat is seepages, streambanks, and swampy depressions (Weakley 2011).

Carex gracillima (graceful sedge) is estimated to have one to five locations on the forest (F1) but is not tracked as a rare species by the state natural heritage program thus none are currently recorded in the rare species database. This species occurs from Canada south to Arkansas, Georgia, and Alabama. Habitat is moist ravine and slope forests and floodplains of rivers or large streams. (Weakley 2011)

48 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Carex ruthii (Ruth’s sedge) is known from thirty locations on the Forest, four of which overlay with locations identified as hemlock stands. This southern Appalachian endemic ranges from southwestern Virginia, through western North Carolina and eastern Tennessee, to the northern portions of South Carolina and Georgia. Habitat includes seepage areas in both forested and open areas (Weakley 2011). On the Cherokee National Forest this species seems to be associated with higher elevations, usually above 2,500 feet.

Chrysosplenium americanum (golden saxifrage) is estimated to have twenty-one to one hundred locations on the forest (F3) but is not tracked as a rare species by the state natural heritage program thus none are currently recorded in the rare species database. This species occurs from Canada south through Virginia and Indiana to North Carolina, Tennessee, and Georgia. Habitat is shallow seepage in full shade (Weakley 2011).

Clintonia borealis (Clinton’s lily) is known from eighteen locations on the Forest, four of which overlay with locations identified as hemlock stands. This species occurs in spruce/fir forests, northern hardwood forests, and occasionally high elevation northern red oak forests from Canada south to the mountains of Tennessee, North Carolina, and Georgia (Weakley 2004).

Corydalis sempervirens (pale corydalis) is known from ten locations on the Forest, one of which overlays with a location identified as a hemlock stand. This species is primarily known from rock outcrops in dry to mesic forests, from Alaska, through Canada and south in the mountains to Georgia (Weakley 2011).

Cymophyllus fraserianus (Fraser’s sedge) is known from sixty-two locations on the Forest, eleven of which overlay with locations identified as hemlock stands. This species is primarily known from acidic cove forests associated with Rhododendron maximum and is a central and southern Appalachian endemic ranging from western Maryland and southern Pennsylvania, south through West Virginia and Virginia, western North Carolina, eastern Tennessee, northwestern South Carolina and north Georgia (Weakley 2011).

Dirca palustris (leatherwood) is estimated to have twenty-one to one hundred locations on the forest (F3) but is not tracked as a rare species by the state natural heritage program thus none are currently recorded in the rare species database. The species is widespread in eastern North America, ranging from Nova Scotia and Quebec south to the Florida panhandle, Alabama, and Oklahoma. The species typically occurs in very rich forests, on slopes or bottomlands, limited to calcareous soils (Weakley 2011).

Dryopteris carthusiana (Spinulose shield fern) is known from four locations on the forest, one of which overlays with a location identified as a hemlock stand. The species typically occurs in acidic, organic rich bogs and swamps, or less frequently in moist, rocky ravines, rich forests and sloping rock outcrops. Spinulose shield fern has a

49 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest circumboreal distribution from the Yukon to Quebec and south to North Carolina, South Carolina, Tennessee, Arkansas, Nebraska, Montana, and Washington (Weakley 2011).

Glyceria laxa (northern mannagrass) is only known from one location on the forest, and this site overlays with a location identified as hemlock stand. This species is considered to be primarily an Appalachian species derived perhaps from two more common northern ranging species. Habitat is described as boggy sites (Weakley 2011).

Goodyera repens (dwarf rattlesnake plantain) is known from sixteen locations on the forest, four of which overlay with locations identified as hemlock stands. It typivally occurs in moist forests usually under conifers (often hemlock) and Rhododendron. This species is circumbroeal in distribution reaching its southern limit in North Carolina and Tennessee (Weakley 2011).

Heracleum maximum (cow parsnip) is known from thirteen locations on the Forest, one of which overlays with a location identified as a hemlock stand. This species has a broad range, known from Siberia and the United States where it has circumboreal distribution. Habitat is described as forests, roadbanks, and forest openings (Weakley 2011) and moist woods and floodplains (TN Natural Heritage 2008).

Hieracium scabrum (rough hawkweed) is known from seven locations on the Forest, one of which overlays with a location identified as a hemlock stand. Habitat for this species is described as dry forests, woodland margins, and roadsides. The species ranges from Canada west to Minnesota and south to Virginia, Georgia, Kentucky, and Missouri (Weakley 2011).

Hydrophyllum virginianum (Appalachian waterleaf) is known from thirty-one locations on the Forest, one of which overlays with locations identified as hemlock stands. This species occurs in cove and other moist, rocky forests from West Virginia, Virginia, eastern Kentucky, western North Carolina, and east Tennessee (Weakley 2011).

Lilium canadense (Canada lily) is known from four locations on the Forest, one of which overlays with a location identified as a hemlock stand. This species is known from Canada to North Carolina, Georgia, and Tennessee where it occurs in wet meadows (Weakley 2011), rich woods and seeps (TN Natural Heritage 2008).

Liparis liliifolia (large twayblade) is estimated to have twenty-one to one hundred locations on the forest (F3) but is not tracked as a rare species by the state natural heritage program thus none are currently recorded in the rare species database. The species is widespread in eastern North America ranging from Canada south to Georgia and Mississippi, and east to Oklahoma and Wisconsin (NatureServe 2011). Habitat is poorly described, listed simply as “moist forests and floodplains” (Weakley 2011).

Lygodium palmatum (climbing fern) is known from seven locations on the forest, none of which currently overlay with stands identified as having a major hemlock component, however it is considered to be associated with such habitats (Plan Appendix E). The species is widespread in the eastern United States, from Maine and Michigan south to

50 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Florida and east to Mississippi, but is considered to be rare throughout its range. Habitats are described as bogs, moist thickets, and swamp forests in strongly acid soils (Weakley 2011), yet this species is often found in xeric openings on the Cherokee National Forest. This species was included as a viability concern species for the Cherokee National Forest primarily due to a possible association with wetland habitats. This species is not tracked by the Tennessee Division of Natural Heritage, thus records have not been kept on known occurrences. Populations have been estimated to number between 6 and 20 locations.

Melanthium latifolium (broadleaf bunchflower) is known from twenty locations on the forest, one of which overlays with a location identified as a hemlock stand. This species taxonomic status is currently under review and is provisionally placed in the Genus Veratrum by Weakley (Weakley 2011). Broadleaf bunchflower is Appalachian endemic ranging from Connecticut to South Carolina, and considered rare in the southern Appalachians. Habitat is poorly defined and simply described as moist to dry forests (Weakley 2011).

Mertensia virginica (Virginia bluebells) is estimated to have six to twenty locations on the forest (F2) but is not tracked as a rare species by the state natural heritage program thus none are currently recorded in the rare species database. Based upon habitat associations it could occur within stands dominated by hemlock. The species is widespread in the east occurring from New York west to Wisconsin and Indiana, and south to North Carolina, Georgia, Alabama, and Arkansas, but is considered rare south of Virginia. Typical habitat is nutrient-rich, moist, alluvial soils of floodplain forests and thickets (Weakley 2011).

Platanthera orbiculata (large round-leaved orchid) is known from eighty locations on the Forest, six of which overlay with locations identified as hemlock stands. This species typically occurs in moist hardwood forests, ranging from the Pacific Northwest though Canada and the northern United States to the southern Appalachian mountains (Weakley 2011).

Platanthera psychodes (small purple-fringed orchid) is known from sixteen locations on the Forest, one of which overlays with a locations identified as a hemlock stand. This species is widespread in eastern North America where it is known from northern hardwood forests, bogs, seepages, and moist forests (Weakley 2011).

Poa palustris (fowl bluegrass) is known from four locations on the Forest, one of which overlays with a location identified as a hemlock stand. This species has a circumboreal distribution, extending south in North America to the southern Appalachians and New Mexico. Habitat is described as meadows and moist areas (Weakley 2011).

Populus grandidentata (bigtooth aspen) is previously reported from five locations on the forest though the state natural heritage program stopped tracking this as a rare species in 2007 thus records may not be accurate. Based upon habitat associations from Appendix E of the Forest Plan the species could occur within stands dominated by hemlock. Habitat is described as dry, rocky, upland forests, known from Canada west to

51 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Minnesota and south to North Carolina, Tennessee and Missouri. The species is considered rare south of Virginia (Weakley 2011)

Prunus virginiana (choke cherry) is known from four locations of the forest, one of which overlays with a location identified as a hemlock stand. The species is widespread from Canada south and west to North Carolina, Georgia, Arkansas, and Oklahoma, but considered rare in our area. It is known to form clonal thickets in oak and northern hardwood forests (Weakley 2011).

Pyrola Americana (American wintergreen) is known from sixteen locations on the Forest, two of which overlay with locations identified as hemlock stands. This species is widespread in northeast North America, south to North Carolina and Tennessee. It is known from a variety of habitats including xeric to mesic woodlands (Weakley 2011) and dry woods and bogs (Wofford 1989).

Stewartia ovata (mountain camellia) is not tracked by the state natural heritage as a rare species, thus many records of known locations have not been documented. Populations on the forest have been estimated to number between 21 and 100 locations (F3) and could occur within stands dominated by hemlock. Mountain camellia is known from all the southeastern states from Virginia and Kentucky south through Mississippi to Florida (NatureServe 2011). It typically occurs within openings in mesic forests, especially acidic bluffs or Rhododendron thickets (Weakley 2011).

Streptopus roseus (rosy twisted stalk) is known from sixteen locations on the Forest, two of which overlay with locations identified as hemlock stands. The taxonomic status of the species has been revised and the species is now correctly named Streptopus lanceolatus var. lanceolatus. This species occurs in moist forests at high elevations from Canada south to the southern Appalachians (Weakley 2011).

Symplocos tinctoria (horse sugar) is known from twenty-five locations on the forest, two of which overlay with locations identified as hemlock stands. The species has a widespread, but discontinuous distribution in the eastern and southeastern United States, known from Delaware south through Florida to Texas (Weakley 2011). Habitat is extremely variable, including coastal pocosins, mesic forests, and mountain ridge-top forests. This species is locally abundant within the Ocoee and Hiwassee River drainages on the Cherokee National Forest.

Trichomanes petersii (dwarf filmy fern) is known from ten locations on the forest, one which overlays with a location identified as a hemlock stand. This is a diminutive species that is easily overlooked and often mistaken to be a bryophyte. Weakley (2011) now places this in the genus Didymoglossum. Habitat is described as vertical faces of rock outcrops in humid gorges and sometimes tree bark, ranging from western North Carolina and South Carolina, southwest to Forida, Alabama, Mississippi, and Louisiana, and north to Arkansas and southern Illinois. The species also has disjunct populations in Mexico (Weakley 2011).

52 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Trillium undulatum (painted trillium) is not tracked by the state natural heritage as a rare species, thus many records of known locations have not been documented. Populations on the forest have been estimated to number between 21 and 100 locations (F3) and could occur within stands dominated by hemlock. The species is widespread through eastern Canada and the eastern United States south to North Carolina, South Carolina, Georgia, and Tennessee. Occurs on acidic soils of ridges, slopes, and bog margins, mostly at high elevations and often associated with Rhododendron, Tsuga, Pinus, or Picea (Weakley 2011) This species is relatively common on the north end of the forest and probably should be removed from the viability concern list.

Veronica americana (American speedwell) is known from one location on the forest and this site overlays with a location identified as a hemlock stand. The species is known to occur in bogs, marshes, and along streams and is primarily northern in the east (rare south of Virginia). Occurs across Canada to Alaska, south to California in the west and North Carolina, Tennessee, and Texas in the east. Also known from Asia! (Weakley 2011).

Woodsia scopulina ssp. appalachiana (Alleghany cliff-fern) is known from six locations on the Forest, two of which overlay with locations identified as hemlock stands. This species is endemic to the southern and central Appalachians of Virginia, West Virginia, North Carolina, and Tennessee, and the Ozarks of Arkansas where it occurs on cliffs of sandstone, shale, granite, and gneiss (Weakley 2011).

Xerophyllum asphodeloides (eastern turkeybeard) is known from twenty-eight locations on the forest, two of which overlay with locations identified as hemlock stands. This species ranges in two disjunct areas; the coastal plain of southern New Jersey and Delaware, and mountains from West Virginia south to Tennessee, North Carolina, South Carolina, and Georgia. Habitat includes dry ridges and slopes in the mountains, primarily on sites with a history of fire. The species is often associated with Pinus rigida or Pinus pungens, but also occurs in xeric oak forests and pine-heath communities (Weakley 2011).

Anastrophyllum helleranum (a liverwort) is currently known from one location on the forest (F1). Though this site is not coincident with a stand dominated by hemlock it is likely hemlock is at least present in the canopy. This species range is mostly northern in the United States with only scattered populations in the southern Appalachians. It is also known from Europe and Asia. The species typically occurs in Spruce/Fir forests above 5,000' on bark or decaying logs but has also been reported from humid gorges on the bark of hemlock (Hicks 1992)

Frullania plana (a liverwort) is currently known from one location on the forest (collected at Falls Branch Falls in 2001). Though this site is not typed as a stand dominated by hemlock there is hemlock present in the canopy. The species ranges from New York and Connecticut south to Tennessee and South Carolina and occurs on shaded rock and occasionally bark in moist, mountain coves (Hicks 1992)

53 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Jungermannia fossombronioides (a liverwort) is currently known from one location on the forest and this site is coincident with a stand dominated by hemlock. The species is found on soil and rocks along streams in lower elevations of the mountains and Piedmont from Connecticut south to Georgia (Hicks 1992).

Plagiochila ludoviciana (a liverwort) is currently known from one location on the forest (collected along the Bald River in 1995). Though this site is not typed as a stand dominated by hemlock there is likely hemlock present in the canopy. This species is primarily known from the coastal plain, but is disjunct to escarpment gorges in the mountains. The Blue Ridge populations are usually found along under-ledges, often associated with Trichomanes petersii (Hicks 1992)

Radula tenax (a liverwort) is currently known from four locations on the forest, none of which currently overlay with stands identified as having a major hemlock component, however it is considered to be associated with such habitats (USDA Forest Service 2004b). This species is endemic to the Appalachian region from Maine to South Carolina where it occurs on moist rocks or trees at elevations below the Spruce/fir zone (Hicks 1992).

Aegolius acadicus (Northern saw-whet owl) breeds from Alaska to Saskatchewan and New Brunswick, to southern California, Arizona, Mexico, Texas, Missouri, Michigan, and Maryland (NatureServe 2005). There is also a breeding population in the higher elevations of the Southern Appalachian Mountains including Virginia, Tennessee, and North Carolina. There are currently four known records from the CNF, one of which is in a hemlock stand. Breeding habitat in the southern Appalachian population is usually associated with high elevation spruce-fir forest and the transition area between this forest type and the adjacent northern hardwood forest (Nicholson 1997). On the south end of the CNF, saw-whet owls have been found calling within the hemlock-northern hardwood interface. Coniferous and mixed deciduous-coniferous forests are utilized as breeding habitat further north. Nesting takes place in natural cavities, old woodpecker holes, and nest boxes. In winter, these birds may migrate downslope to lower elevation coniferous and mixed forests. Birds that breed farther north may also occur within these areas. Diet consists mainly of small mammals, but birds and insects are also taken. Populations in the area appear fairly stable, but populations are restricted by habitat availability.

Accipiter striatus (sharp-shinned hawk) can be found in coniferous, mixed, or deciduous forests and open woodlands. Their primary habitat is boreal forest, with the greatest nesting densities occurring in eastern Canada. Young, dense, mixed or coniferous woodlands are preferred for nesting. Where conifers are scarce, as in the prairie regions, cottonwoods, poplars, and other members of the Betulaceae may be used. This hawk migrates through various habitats, mainly along ridges, lakeshores, and coastlines. They usually nest in tree crotch or on branch next to trunk, hidden by thick foliage. They may build new nests, reuse old ones, or modify old bird or squirrel nests. Nests generally seem to be in a stand of dense conifers near a forest opening, though this may reflect observer bias. Pairs apparently remain faithful to nesting areas for several years, although a new nest is usually constructed each season. This species eats mainly small to medium-sized

54 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest birds; occasionally small mammals, insects, lizards, etc. They hunt from inconspicuous perch or by stealthy flights along paths and around bushes and trees.

Ambystoma talpoideum (mole salamander) range from the Coastal Plain of South Carolina through northern Florida, west to eastern Texas and southeastern Oklahoma; north in the Mississippi Valley to southern Illinois and southern Indiana. Disjunct populations occur in Virginia, the Carolinas, Georgia, Alabama, Tennessee, and Kentucky. Their habitat is usually near breeding ponds--in pine flatwoods, floodplains, and bottomland hardwood forests. In South Carolina, they avoided clearcuts and open fields, but occurred in all types of forest. Terrestrial adults live in underground burrows and are sometimes found under logs or other objects in damp places. This salamander breeds in shallow ponds and flooded depressions that are free of fishes and that often have abundant emergent and/or submerged vegetation. Eggs are attached to stems or sticks or to the substrate. Reproductive success is positively correlated with duration of standing water in breeding pond, but not with number of breeding females or number of eggs laid. Adults eat terrestrial arthropods, molluscs, and worms. Larvae eat copepods, cladocerans, and various other aquatic invertebrates, sometimes the larvae of other Ambystoma

Aquila chrysaetos (golden eagle) is a rare migrant, a rare winter resident, and a very rare summer visitor in Tennessee (Nicholson 1997). There is currently one (non-breeding) known record from the CNF in a hemlock stand. In the past, they once occurred regularly in the summer when livestock were raised on mountaintop pastures. Most summer sightings are thought to be sub-adult birds from northern or western populations. They require cliffs for nesting and extensive open country, mainly at high elevations (over 4,000 feet, Hamel 1992). Diet consists of small to medium sized mammals, but carrion and birds are also taken.

Condylura cristata (star-nosed mole) that occur on the CNF are part of the southern subspecies that occurs within the Appalachian Mountains from West Virginia and Virginia to North Carolina and South Carolina (NatureServe 2005). They can be locally common, but are considered uncommon through out their range. There are currently four known records from the CNF, one of which is in a hemlock forest. Star-nosed moles are seldom far from water and are good swimmers and divers. Habitat preferences include wet soils in moist meadows, fields, swamps, and woods (Webster et al. 1985). These moles move about in underground burrows and above ground runways that sometimes lead directly to water. Diet consists of aquatic and terrestrial worms and insects, crustaceans, and small fish. In the Great Smoky Mountain National Park (Smokies), the star-nosed mole is known to occur at elevations from 1,600 to 5,500 feet (Linzey 1995).

Dendroica cerulea (cerulean warbler) occupies a range in southeastern Nebraska across the southern Great Lakes region to southern Ontario, southwestern Quebec, and western New England, south to northern Texas, central Georgia and Alabama, western North Carolina, and Maryland. They are most abundant in the Cumberland Plateau and surrounding regions. Non-breeding habitat is primarily in a narrow elevational zone on the eastern slopes of the Andes from Colombia and Venezuela through Ecuador to Peru.

55 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Relatively few overwinter elsewhere. They inhabit mature deciduous forests on both the breeding grounds in North America and the non-breeding range in the Andes. Breeding areas in the Northeast are often in floodplains or other mesic conditions and are typified by large, mature trees and closed or semi-open forest canopies. Nests are located on the lateral limbs of a tree at considerable distances from both the ground and the bole. Common vegetation characteristics of nest sites are unknown and use of openings and edge requires further study. Populations have declined across the range in the eastern U.S., although the range has expanded, particularly in the Northeast, perhaps in response to large-scale forest maturation.

Desmognathus aeneus (seepage salamander) are spottily distributed from extreme southeastern Tennessee and southwestern North Carolina to central Alabama (Petranka 1998). Populations have not been found north of the Little Tennessee River and are restricted to elevations between 100-4400 feet. There are 31 known records of this species on the CNF, eight of which are in hemlock forests. Seepage salamanders are most frequently encountered within moist or wet leaf litter in and around seepages, or in terrestrial habitats adjoining small streams. They are also occasionally found beneath logs, moss mats, and other surface objects (Martof et al. 1980). Diet consists of tiny invertebrates, springtails, beetle larvae, and mites. Populations are stable or declining within different portions of its range (NatureServe 2005).

Desmognathus ochrophaeus (mountain dusky salamander) occur on the ridges of the Ridge and Valley Physiographic Province, including Brumley, Clinch, Walker, and Potts mountains of southwestern Virginia; Cumberland Mountains and Plateau of southeastern Kentucky, and the Allegheny Mountains and Plateau of West Virginia, Maryland, Pennsylvania, Ohio, and New York through the Adirondack Mountains to southern Quebec and southern Ontario. Populations in the Cumberland Plateau of Tennessee have not been studied electrophoretically and may or may not represent this species (Tilley and Mahoney 1996). At lower elevations and in winter, these salamanders usually concentrate near seepage areas, springs, and small streams. They may range into adjacent wooded areas in wet weather. They are more terrestrial at higher elevations and are characteristic inhabitants of spruce-fir forests. The mountain dusky salamander is often abundant on wet rock faces. Eggs are laid in wet rock crevices or under rocks, logs, or moss in seepage areas or near small streams.

Eurycea longicauda (longtail salamander) ranges from Southern New York to Missouri, south to Arkansas, Tennessee, extreme northeastern Mississippi, northern Alabama, extreme northwestern Georgia, western North Carolina, and northwestern Virginia. They can be found along streamsides, spring runs, cave mouths, abandoned mines as well as ponds in northern New Jersey. They may disperse into wooded terrestrial habitats in wet weather and often hide in rock crevices and under rocks, logs, and other debris. Eggs are laid in underground crevices associated with springs, temporary pools, and streams; under rocks in streams; in woodland ponds; or are attached to objects in or above water in caves.

56 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Glyphyalinia pentadelphia (pink glyph) historically occurs in Blount, Monroe, Polk, and Sevier Counties, Tennessee. In North Carolina, it occurs in southwestern mountains in Cherokee, Clay, Graham, Macon, and Swain Counties..

Inflectarius ferrissi (Smoky mountain covert) has been found on Clingman's Dome in the Great Smoky Mountains National Park, Tennessee. In North Carolina, it is known from the Great Smoky Mountains and Plott Balsams in Haywood, Jackson, and Swain Counties.

Limnothlypis swainsonii (Swainson’s warbler) is a rare and local breeder in the southeastern United States, mostly coastal plain but also southern Appalachians (NatureServe 2004). There are currently 10 known records from the CNF, four of which are in hemlock forests. During the spring and summer in the mountains, they prefer the moist lower slopes of mountain ravines at elevations up to 3000 feet, and a thick shrub layer (rhododendron is most common) (Nicholson 1997). Swainson’s warblers utilize areas with deep shade from both canopy and understory cover and have a preference for early successional forests or disturbance gaps in old growth forests (NatureServe 2004). This species appears to be increasing within the Appalachian region (Sauer et al. 2005).

Lontra canadensis (river otter) occur throughout most of North America north of Mexico, except the extreme southwestern U.S. Habitat includes streams, lakes, ponds, swamps, marshes, estuaries (in some areas), beaver flowages, and exposed outer coast. When inactive, individuals occupy hollow logs, space under roots, logs, or overhangs, abandoned beaver lodges, dense thickets near water, or burrows of other animal. These sites also are used for rearing young. They have the ability to travel long distances overland, particularly in snow. River otters feed opportunistically on aquatic animals, particularly fish, frogs, crayfish, turtles, insects, etc., sometimes birds and small mammals

Mesomphix rugeli (wrinkled button) are usually found among carbonate cliff habitats. It ranges through Georgia, Kentucky, North Carolina, Tennessee and Virginia and has a G4 ranking.

Napeozapis insignis (woodland jumping mouse) occurs in the eastern United States from Maine and eastern Ohio south through the Appalachians to northeastern Georgia, and in northern Michigan and northern Wisconsin (Whitaker, Jr. and Hamilton, Jr. 1998). There are currently 40 known records from the CNF, seven of which occur in hemlock forests. Woodland jumping mice inhabit a variety of woodland types from spruce/fir to mixed conifer/hardwood forests, and can occur at elevations from 1600 to over 6000 feet. They are found near streams with rhododendron or with dense woods with little of no underbrush. Diet consists of seeds, fungi, fruits, and insects, grubs, and worms (Linzey 1995).

Paravitrea lamellidens (lamellate supercoil) is known from eastern Tennessee and western North Carolina (Field Museum 2002, Hubricht 1985, NatureServe 2005). In Tennessee it occurs from Carter and Unicoi Counties, to Polk County (Field Museum

57 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

2002, Hubricht 1985, NatureServe 2005). There are currently five known records from the CNF, one of which is in a hemlock forest. Habitat for this snail is described as pockets of deep, moist leaf litter on wooded hillsides and ravines at moderate elevations with good cover (Caldwell 2005).

Paravitrea reesei (round supercoil) has been documented in Pike County, Kentucky and Johnson County, Tennessee. It also occurs in North Carolina, Virginia and West Virginia.

Paravitrea tridens (white-foot supercoil) is known from northeast Tennessee and southwest Virginia (NatureServe 2005). In Tennessee it occurs from Sullivan to Cocke County (Field Museum 2002, Hubricht 1985). There are currently four known records from the CNF, one of which is in a hemlock forest. Habitat for this snail is described as moist leaf litter on wooded hillsides (Hubricht 1985). It has been found at low and mid elevations in a variety of habitat conditions (Field Museum 2002).

Plestiodon anthracinus anthracinus (Northern coal skink) habitat generally consists of humid wooded areas with abundant leaf litter and loose rocks. Often the lizard occurs in the vicinity of springs, swamps, and bogs, but it also inhabits clearcuts, highway and powerline rights-of-way, rocky bluffs above creek valleys, dry, rocky, south-facing hillsides, and dry shale barrens. Individuals often shelter under logs and rocks near water. Sometimes they take refuge in water.

Plethodon dorsalis (zigzag salamander) occurs in Illinois and Indiana southward through western and central Kentucky to central Tennessee. This salamander can be found in the vicinity of moist rocky crevices in ravines, canyons, rubble, seepages, caves, and wooded slopes. Often it is found under rocks, logs, or leaves during day.

Pseudacris brachyphona (mountain chorus frog) is discontinuously distributed from western Pennsylvania southwest to northeastern Mississippi, central Alabama, and Georgia. It ranges to elevations of at least 3,500 ft. The frog’s habitat is in wooded hilly areas and probably hides under objects or goes underground when inactive. Eggs and larvae develop in pools in or adjacent to woods including spring pools, flooded ditches, pools along streams and woodland ponds. Eggs are attached to leaves, sticks, or other vegetation in water.

Pseudotriton montanus (Eastern mud salamander) occurs in the Gulf Coastal Plain from eastern Louisiana to central Florida, northward in the Coastal Plain and Piedmont of eastern Georgia, the Carolinas, and Virginia, and in the Coastal Plain only of Maryland and southern New Jersey. West of the Appalachians, it occurs in eastern Tennessee, Kentucky, western West Virginia, western Virginia, and southern Ohio (some regard these populations as a distinct species, P. diastichus). There are also isolated populations in east-central Mississippi and south-central Pennsylvania. Muddy springs, slow floodplain streams, and swamps along slow streams; backwater ponds and marshes created by beaver activity are common habitats. Nonlarval forms usually occur beneath

58 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest logs and rocks, in decaying vegetation, and in muddy stream-bank burrows. It occasionally disperses from wet muddy areas. Eggs are attached separately to objects in water (e.g., undersides of leaves in quiet pool).

Sitta canadensis (red-breasted nuthatch) breeds in southern Alaska to northern Saskatchewan and Labrador, south to southern California, southern New Mexico, western South Dakota, southern Saskatchewan, Minnesota, northern Ohio, New Jersey, and southern Appalachians. It also can be found in isolated areas to the south. The nuthatch winters throughout most of its breeding range and irregularly to Gulf Coast. Habitats include coniferous and mixed forest, aspen woodland, deciduous forest, open woodland, parks, scrub, and riparian woodland. It nests usually in excavated cavity in tree stub or branch of dead tree, or dead top of live tree, also in abandoned woodpecker hole, natural cavity, or bird box; average of 12-15 feet above ground.

Sorex cinereus (common shrew) occurs from the Alaska and Canadian south through the Rocky Mountains to Utah and New Mexico, Northern Great Plains, and Appalachian Mountains to Georgia and South Carolina (NatureServe 2005). There are currently 54 known records from the CNF, seven of which are in hemlock forests. The common shrew is a habitat generalist, occuring in moist areas from grassy, fields and marshes to dark, moss-carpeted spruce forests (Whitaker, Jr. and Hamilton, Jr. 1998). In the Smokies, they are most commonly found among rocks and logs in moist woods (Linzey 1995). Local populations can vary greatly (Webster et al. 1985). Diet consists of insects, earthworms, other shrews and small mice, snails and slugs, and some plant material (Webster et al. 1985).

Sorex fumeus (smoky shrew) occurs in the eastern United States from New England and New York south through the Appalachians to northern Georgia (Whitaker Jr. and Hamilton, Jr. 1998). There are currently 72 known records from the CNF, nine of which are in hemlock forests. Habitat for this shrew consists of cool, damp woodlands with a deep layer of leaf mold on the ground (Linzey 1995). Streamsides, both evergreen and deciduous forests at all elevations, are a favored habitat. Bogs and roadside cuts that expose bare rock faces are also inhabited (Webster et al. 1985). Diet consists of earthworms, centipedes, insects, small salamanders, sowbugs, and other small invertebrates (Whitaker, Jr. and Hamilton, Jr. 1998).

Sorex longirostris (Southeastern shrew) occurs in the southeastern United States from Maryland and Virginia, to Florida, Alabama Mississippi, Tennessee, Kentucky, and Arkansas to West Virginia, Illinois, and Indiana (Whitaker, Jr. and Hamilton, Jr. 1998). There are currently 10 known records from the CNF, one of which in a hemlock forest. This species occurs in a variety of habitats from fields to forest (Linzey 1995). Habitat in early stages of succession and disturbed habitats fields with dense ground cover of honeysuckle, grasses, sedges, and herbs seem to be favored. Forested areas are also utilized (NatureServe 2005). Diet consists of small invertebrates, particularly spiders, caterpillars, slugs and snails, crickets, beetles, centipedes, and some plant material. The Southeastern shrew may respond favorably to disturbances that allow dense cover to thrive.

59 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Sphyrapicus varius (Yellow-bellied sapsucker) breeds from Alaska across Canada to Labrador and Newfoundland to North Dakota, Minnesota, and New York (NatureServe 2005). Breeding populations extend down the Appalachian Mountains into northern Virginia, with a disjunct population occurring in the mountains of southwest Virginia, North Carolina, and Tennessee. Wintering birds occur from Missouri to New Jersey south through Central America, the Bahamas, and Antilles (NatureServe 2005). Seven occurrences were previously documented on the CNF, but recent surveys have resulted in several new breeding records from Carter, Unicoi, Johnson, and Monroe Counties. Three of the documented records are in hemlock forests, but none of these new records are. The yellow-bellied sapsucker is one of the rarest breeding birds in Tennessee (Nicholson 1997). The southern Appalachian breeding population is considered a subspecies (Pearson 2005). Nesting occurs in a variety of forest conditions ranging from the edges of clearcuts and old pastures to forest interior and old growth forests at elevations between 3000-5500 feet. Habitat is best described as open, deciduous and northern hardwood forest. The wintering birds most likely remain close to their breeding ranges migrating to lower elevations during severe winter weather. Diet consists of insects and grubs, drilling holes in bark to feed on cambium and sap, and they also pick berries and nuts in the winter (Hamel 1992). Trend data from 1966-2004 in the Appalachian region indicates that the population is increasing within the region (Sauer et al. 2005).

Stenotrema altispira (highland slitmouth) occurs in southwest Virginia, eastern Tennessee and western North Carolina (Hubricht 1985, Field Museum 2002, NatureServe 2005). It occurs from Carter County to Sevier County in Tennessee (Field Museum 2002). There are currently six known records from the CNF, one of which is in a hemlock stand. Habitat for this snail is described as around logs and in leaf litter at the higher elevations (Hubricht 1985).

Stenotrema barbigerum (fringed slitmouth) has been found in McMinn and Polk Counties, Tennessee as well as northwest Alabama.

Synaptomys cooperi (Southern bog lemming) occurs from Quebec to Manitoba, south to Kansas, Arkansas, Tennessee, North Carolina, Virginia and Maryland (NatureServe 2005). The Southern bog lemming is exceedingly variable in the habitats it occupies (Whitaker, Jr. and Hamilton, Jr. 1998). There are currently eight known records from the CNF, three of which are in hemlock forests. Dry hillsides with a growth of bluegrass, fields matted with a canopy of weeds, grassy areas interspersed with brush and small trees, and dense woods of hemlock and beech all may harbor these little mice. Habitat in the Smokies includes grassy openings in woods, mossy boulders in spruce-fir forests, bogs, clearcuts, pastures, and powerline right-of-ways (Linzey 1995). Diet consists almost entirely of green plant material, but fruit (blueberries, blackberries) may be taken in season. Fungi are also eaten occasionally. Home range varies from 0.25 to one acre (NatureServe 2005). Populations are usually scarce and scattered.

Ventridens decussatus (crossed dome) is known from Carter, Monroe, and Unicoi Counties, Tennessee and the entire mountain region of North Carolina.

60 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Ventridens lawae (rounded dome) ranges from Louisiana and Oklahoma east to Florida and North Carolina and north to New York, Michigan, and Quebec

Zapus hudsonius (meadow jumping mouse) occurs from Alaska to Hudson Bay to Labrador, to North Carolina, South Carolina, Georgia, Alabama, Mississippi, northeastern Oklahoma, Montana, Alberta, and British Columbia (NatureServe 2005). There are currently eight known records from the CNF, two of which are in hemlock forests. Habitat consists of grassy fields or fields with mixed grasses and forbs (Whitaker, Jr. and Hamilton, Jr. 1998). These are normally associated with marshes, streams, or ponds, but this mouse can occur well removed from water. It may also occur in woodlands with dense herbaceous growth where the woodland jumping mouse is not present (Webster et al. 1985). Meadow jumping mice feed on insects, seeds, nuts, berries, fungi, and fruits (Webster et al. 1985).

Aquatice Resource Aquatic species with viability concerns (USDA Forest Service 2010a) include all Sensitive species and those species considered Locally Rare (S1 or S2 and documented on the Cherokee National Forest). A total of 47 species meet these criteria (Table 10).

Table 10. The Viability Concern Species on this Forest (USDA Forest Service 2004c) Group Status Common Name Populations Amphibians Sensitive Carolina mountain dusky salamander 8 Amphibians Sensitive Junaluska salamander 0 Amphibians Sensitive Santeetlah dusky salamander 0 Amphibians Sensitive Southern Appalachian salamander 1 Amphibians Sensitive Tellico salamander 0 Amphibians Sensitive wellers salamander 0 Fish Sensitive black sculpin 20 Fish Sensitive blotchside logperch 2 Fish Sensitive bronze darter 3 Fish Sensitive coldwater darter 0 Fish Sensitive fatlips minnow 2 Fish Sensitive frecklebelly madtom 0 Fish Sensitive holiday darter 3 Fish Sensitive lined chub 0 Fish Sensitive mountain brook lamprey 4 Fish Sensitive olive darter 1 Fish Sensitive sharphead darter 1 Fish Sensitive sickle darter 0

61 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Group Status Common Name Populations Fish Sensitive Tennessee dace 38 Fish Sensitive trispot darter 0 Fish Sensitive wounded darter 3 Fish Locally Rare bridled darter 3 Fish Locally Rare burrhead shiner 4 Fish Locally Rare Mobile logperch 3 Fish Locally Rare riffle minnow 3 Fish Locally Rare rosyface shiner 7 Fish Locally Rare southern brook lamprey 2 Fish Locally Rare Tuckasegee darter 2 Insects Sensitive Allegheny snaketail 0 Insects Sensitive Appalachian snaketail 1 Insects Sensitive Cherokee clubtail 0 Insects Sensitive Edmund's snaketail 1 Insects Sensitive green-faced clubtail 0 Insects Sensitive Helma's net-spinning caddisfly 1 Insects Sensitive mountain river cruiser 0 Insects Sensitive William's giant stonefly 0 Mussels Sensitive Alabama creekmussel 2 Mussels Sensitive Alabama rainbow 2 Mussels Sensitive Coosa creekshell 2 Mussels Sensitive green floater 0 Mussels Sensitive slabside pearlymussel 2 Mussels Sensitive Tennessee clubshell 3 Mussels Sensitive Tennessee heelsplitter 1 Mussels Sensitive Tennessee pigtoe 2 Mussels Locally Rare delicate spike 1 Mussels Locally Rare southern rainbow 2 Snails Locally Rare smooth mudalia 1

Environmental Consequences Botanical Resource Potential effects to botanical species from the three alternatives are described below. This analysis differs from many other environmental analyses in that the “No Action” alternative (Alternative 1) actually is the continuation of the implementation of the previously signed 2005 decision and actions approved under the subsequent 2010 Supplemental Information Report. Thus, under all alternatives, some level of HWA suppression is expected.

62 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Effects of imidacloprid on plants species were evaluated in the 2005 assessment (USDA FS 2005). Like imidacloprid, dinotefuran is a chemical insecticide with no known impact to any species of plant (EPA Fact Sheet, 2004, Fossen 2006). The fungus Lecanicillium muscarium naturally occurs in eastern Tennessee and likewise has no known effect on plant species (Costa 2008). Thus, there are no known direct effects to any rare plant species from the current control treatments (Alternative 1) or the expanded treatments (Alternatives 2 and 3).

Indirect effects to rare plant species may be more variable. In this case, rare plants that are dependent upon either the cool, moist microclimate associated with stands dominated by eastern hemlock, or specific habitat conditions found in Carolina hemlock forests, may be variously affected as these habitats are altered through hemlock mortality. A discussion of these effects for groups of species is found below for each alternative.

Wildlife Resource The following discussions below include potential effects on TES terrestrial wildlife species presented in the above tables. Effects on terrestrial Viability Concern species are expected to be similar to TES species, and therefore are specifically discussed only when they are different.

Alternative 1 Botanical Resource Under this alternative no additional treatments are proposed to suppress HWA infestations on the Cherokee National Forest. Despite implementation of the 2005 decision, the rate of spread of HWA across the forest has been much more rapid than was originally anticipated and treatment efforts have not been sufficient to protect all the sites identified for protection in that environmental assessment. Efficacy of beetle treatments is currently inconclusive and some beetle release sites have seen significant hemlock mortality. Without additional control measures the HWA infestation would continue to spread across the Cherokee National Forest potentially killing any remaining untreated sites of both hemlock species over time. Rare plant species associated with hemlock forests may also suffer losses dependent upon their ability to adapt to the changing environmental conditions. Potential effects to associated species range from altered fitness due to less than optimal habitat conditions, to local extirpations. Potential losses would be highest for those species dependent upon the moist and shaded conditions typically found in eastern hemlock stands. Potential effects to TES and VC species are described below.

Threatened and Endangered Species As described above, there would be no effects to Threatened or Endangered species. None of the species in Table 5 are associated or dependent upon hemlock for their continued existence, and no known records of these species are found within hemlock stands on the Forest. Based upon this, there would be no effect to any of these species as a result of this project.

63 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Sensitive Species Thirty-one sensitive plant species have either known occurrences that are coincident with hemlock stands on the Forest or occur in habitats that could have a significant hemlock component. Information on the habitat, range, and the frequency of occurrence within hemlock stands for each species is described above under the section entitled “affected environment”. From those descriptions, impacts can be grouped into the following categories:

Species is dependent upon micro-habitat conditions that may be altered with the loss of hemlock forests. Loss of hemlock could result in the loss of 50% or more of the known sites for the species on the Cherokee National Forest.

Seven sensitive species (Acrobolbus ciliatus, Aneura maxima, Cheilolejeunea evansii, Plagiochila sullivantii var. spinigera, Plagiochila sullivantii var. sullivantii, Radula voluta, and Tsuga caroliniana) fall into this group.

The six non-vascular species are included in this category of effects primarily because each is only known from one site on the forest and based upon the remoteness of these sites they are not likely within stands that were identified for treatment in the 2005 decision. Potential effects to these species may vary based upon the substrate on which they typically occur.

Acrobolbus ciliata, Plagiochila sullivantii var. spinigera, Plagiochila sullivantii var. sullivantii, and Radula voluta all occur primarily on shaded wet rock-faces and spray cliffs. The unique microclimatic conditions of this habitat type are more directly related to water flow than canopy composition, but can be affected by changes in the canopy that result in increased solar radiation.

More likely to be affected are Aneura maxima and Cheilolejeunea evansii which occur on soil and tree bark respectively. All of these non-vascular species are tied to riparian areas and while not all riparian areas have a hemlock component in the canopy, it is very likely that hemlock would occur in some frequency at sites where these species are found.

Tsuga caroliniana is also included within this group, but has the unique situation of being susceptible to the HWA. There are currently fifty-seven recorded sites for this species on the forest, yet most occur as scattered individuals and are not part of a Carolina hemlock forest community. The 2005 decision targeted the best examples of Carolina hemlock forests for treatment, but even those were primarily beetle treatments and the efficacy of that treatment is in question. Any sites for this species that are not effectively treated will be subject to local extirpation.

Species utilizes micro-habitat conditions found within hemlock forests but also has suitable habitat within other forest types. Loss of hemlock could result in the loss of known sites within hemlock forests, however greater than 50% of the known sites for the species would persist at other locations on the Forest

64 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Fourteen sensitive species (Buckleya distichophylla, Cardamine clematitis, Carex roanensis, Scutellaria saxatilis, Trillium rugelii, Drepanolejeunea appalachiana, Lejeunea blomquistii, Lophocolea appalachiana, Megaceros aenigmaticus, Nardia lescurii, Pellia appalachiana, Plagiochila echinata, Porella wataugensis, and Riccardia jugata fall into this group.

Buckleya distichophylla is somewhat shade intolerant and is generally found in forest openings between host plants. Hosts include hemlock (Tsuga spp.) but also many other tree species, thus the fate of the species is ultimately dependent upon the fate of the host trees. If those hosts are hemlock and the trees become infested and die, the Buckleya distichophylla sites would also be lost. Since only three of the thirteen known sites on the forest occur within stands dominated by hemlock it is expected that viable populations would persist on the forest even at this lowest level of HWA suppression.

Cardamine clematitis, Carex roanensis, Scutellaria saxatilis, and Trillium rugelii all require at least mesic habitat conditions, thus known sites within stands dominated by hemlock would likely be lost or at least severely compromised if there is significant hemlock mortality that results in canopy loss. Scutellaria saxatilis, in particular, is susceptible to microclimate changes from the loss of canopy and also invasion of invasive species that often follow such disturbance (NatureServe 2011). Because these species are also found in other forest types that support similar micro-habitat conditions, impacts to the species as a whole should be much less than for those described above in the first category and it is expected that viable populations would persist on the Forest.

Potential effects to the nine non-vascular species included in this category may vary somewhat based upon the substrate on which they typically occur. All of these species are tied to riparian areas and thus may have at least some component of hemlock where they occur, but based upon known site records and community mapping, all species have more known sites on the forest outside of hemlock dominated habitats than inside them. Lophocolea appalachiana, Pellia appalachiana, Plagiochila echinata, and Porella wataugensis are least likely to be affected as they occur primarily on shaded wet rock- faces and spray cliffs. The unique microclimatic conditions of this habitat type are more directly related to water flow than canopy composition, but can be affected by changes in the canopy that result in increased solar radiation. Drepanolejeunea appalachiana and Lejeunea blomquistii occur both on wet rock and tree bark while Nardia lescurii and Riccardia jugata occur on soil, humus, and down wood which is more susceptible to drying if the canopy is compromised. Megaceros aenigmaticus occurs on rock within flowing streams and is dependent upon cool, non-turbid waters. This species is very sensitive to changes in water quality such as sedimentation and increased water temperatures that could result from canopy die-back, though with only five of thirty-eight known sites on the forest occurring within hemlock stands it is expected that viable populations would persist on the Forest.

Species has at least one known site within hemlock forests or occurs in habitats that could have a significant hemlock component, but is not likely dependent

65 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

upon micro-habitat conditions created by the forest canopy. Loss of hemlock will likely have little effect upon the species.

Ten sensitive species (Diervilla rivularis, Fothergilla major, Gentiana austromontana, Helianthus glaucophyllus, Hypericum mitchellianum, Juglans cinerea, Stachys clinmanii, Thermopsis fraxinifolia), Fissidens appalachensis, and Hydrothyria venosa fall into this group.

Diervilla rivularis, Fothergilla major, Gentiana austromontana, Helianthus glaucophyllus, Hypericum mitchellianum, Stachys clinmanii, and Thermopsis fraxinifolia), are all species that thrive in open forests, canopy gaps, or open areas, and based on this, may actually benefit from new openings in the canopy that could be created through hemlock mortality.

Juglans cinerea is typically found in moist, rich forests, especially along creeks and rivers, in bottomlands and floodplains. Since hemlocks can also occur in these habitats, some known sites for Juglans cinerea are listed as occurring within hemlock stands, however, the species existence at the site is not dependent upon the presence of hemlock.

Fissidens appalachensis, and Hydrothyria venosa are an aquatic moss and aquatic lichen respectively. These species occur submerged in the free flowing water of southern Appalachian streams and thus are much more dependent upon the quantity and quality of water than on the surrounding canopy composition.

Determination of Effect for Sensitive Species: This alternative is the continuation of the implementation of the 2005 decision and does not include the use of the new chemical (dinotefuran) or fungus (Lecanicillium muscarium). As was stated above, there are no known effects to any plant species from imidacloprid thus the above analysis is centered on the secondary effects to rare plant species that may occur due to hemlock mortality. This alternative proposes to treat the fewest number or acreage of hemlock trees and thus has the highest potential for secondary effects. Six non-vascular plant species (Acrobolbus ciliatus, Aneura maxima, Cheilolejeunea evansii, Plagiochila sullivantii var. spinigera, Plagiochila sullivantii var. sullivantii, and Radula voluta) have only one known site on the forest and thus are the most susceptible to loss. Of these, the species occurring on substrates (soil, bark, and down wood) where microclimatic conditions are most likely to be affected by openings in the canopy are speculated to be at a greater risk than those occurring on substrates (spray cliffs and wet rocks) that are controlled by water flow, yet Cleavitt et al. (2008) found that common bryophytes have responded to adelgid impacts though an increase in frequency of species occurring on coarse woody debris and to a lesser extent bare soil.

Based upon the above analysis the implementation of the proposed project may affect individuals of the thirty-one sensitive species but would not lead to a trend toward federal listing or a loss of viability for the species.

66 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Viability Concern Species Forty Viability Concern plant species have either known occurrences that are coincident with hemlock stands on the Forest or occur in habitats that could have a significant hemlock component. Information on the habitat, range, and the frequency of occurrence within hemlock stands for each species is described above under the section entitled “affected environment”. From those descriptions, impacts can be grouped into the following categories:

One species (Jungermannia fossombronioides) fits this category, primarily because there is only one known site for this species on the forest and it occurs within a stand that is dominated by eastern hemlock. Significant reduction of hemlock in the canopy at this site could result in loss of viability at this site.

Nine species (Chrysosplenium americanum, Cymophyllus fraserianus, Goodyera repens, Hydrophyllum virginianum, Liparis liliifolia, Anastrophyllum helleranum, Frullania plana, Plagiochila ludoviciana, and Radula tenax) fall into this group. These species all occur within at least a few sites in which the presence of hemlock in the canopy is likely quite important to the current habitat conditions on which these species depend. However, numerous sites in other plant communities will ensure the continued viability of these species on the forest.

Species has at least one known site within hemlock forests or occurs in habitats that could have a significant hemlock component, but is not likely dependent upon micro-habitat conditions created by the forest canopy. Loss of hemlock will likely have little effect upon the species.

The remaining thirty species (Acer leucoderme, Adlumia fungosa, Cardamine rotundifolia, Carex gracillima, Carex ruthii, Clintonia borealis, Corydalis sempervirens, Dirca palustris, Dryopteris carthusiana, Glyceria laxa, Heracleum maximum, Hieracium scabrum, Lilium canadense, Lygodium palmatum, Melanthium latifolium, Mertensia virginica, Platanthera orbiculata, Platanthera pyschodes, Poa palustris, Populus grandidentata, Prunus virginiana, Pyrola americana, Stewartia ovata, Streptopus roseus, Symplocos tinctoria, Trichomanes petersii, Trillium undulatum, Veronica americana, Woodsia scopulina ssp. appalachiana, and Xerophyllum asphodeloides) are all considered to at least potentially co-occur with hemlock species, but are not dependent upon hemlock at their sites for continued persistence.

67 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Wildlife Resource Amphibians Few of the known sites for the salamanders occur within hemlock-dominated stands; none are dependent directly on hemlock forests. They often occur on the lower mountain slopes and adjacent lowlands in forested habitats with abundant leaf litter and rotting logs on the forest floor. Most terrestrial salamanders require shaded woodlands, usually hardwoods but occasionally coniferous areas in the vicinity of bottomlands, with fallen logs, leaf litter, and an organic soil layer. They forage mainly at night on the surface or within their burrows, capturing arthropods and worms. Some canopy closure is necessary to prevent excessive drying of the forest floor. The Junaluska salamander is found at low elevations under logs and rocks along Tellico, Bald and North Rivers, as well as Citico and Slickrock Creeks and potentially the Hiwassee River drainage. It ventures some distance across land during rainy nights. It requires clean, clear streams with abundant rocks and logs, and its diet consists of small arthropods. Siltation or pollution of streams as well as destruction of adjacent forests and their leaf litter would be detrimental to this species. In areas where hemlocks are not treated the loss of hemlock would alter habitat conditions, but suitable habitat conditions are expected to remain at most sites. Some short-term population declines are expected in response to drier conditions as hemlock mortality takes place. Plethodontids require surface moisture for respiration and to rehydrate following terrestrial activity and have specific soil moisture requirements that must be met. The density of forest canopy has considerable impact on soil moisture levels within a forest. Therefore, the density of the forest canopy can show considerable secondary effects on the viability of salamander populations living under that canopy (NatureServe 2011). Sunlight reaching the forest floor increases and dries out leaf litter when hemlocks die. This drying makes the areas less hospitable for terrestrial salamanders that move on the forest floor (through the leaf litter). When these areas are adjacent to streams, drying is expected to be less because of the influence of the adjacent stream on soils and leaf litter moisture. Recovery is expected at most sites as the new stand matures. Additional down woody material would provide improved habitat conditions for this species over the long-term. Some local site extirpations are possible, but this would be a rare occurrence, especially as some sites would likely be treated. Impacts associated with the loss of hemlock are not expected to lead towards extirpation of these species on the Cherokee National Forest.

By treating the hemlock with biocontrol or chemicals, shade would be maintained in the forest, thus making it more suitable for salamander species that require moist conditions. The use of predator beetles would have no effect on the Carolina mountain dusky salamander, Santeetlah dusky salamander, Junaluska salamander, Tellico salamander, or Weller’s salamander. Because imidacloprid is either injected directly into the trunk of the tree, or approximately 8 -10 inches deep into the soil (below the leaf litter), use of this chemical

68 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest would have no effect on Santeetlah dusky salamander, Junaluska salamander, Tellico salamander, Carolina mountain dusky salamander or Weller’s salamander. Birds There are no Federally-listed or Sensitive birds that depend directly on hemlock forests; although several rare terrestrial bird species do show some dependence on coniferous forests, including hemlock. The loss of hemlock would improve bird habitat locally by creating snags used for foraging and nesting for species such as the golden eagle and yellow-bellied sapsucker. It is reasonable to believe that all infested hemlocks across the CNF would not be treated, thus new hemlock snags would be created regardless of the extent of treatments. The use of predator beetles would have no impact on the eagle or sapsucker. Additionally, since imidacloprid is either injected directly into the trunk of the tree, or into the soil, use of this chemical would have no effect on the eagle or sapsucker. Scattered dying hemlock may cause canopy gaps in hardwood forests that may make the habitat more suitable for nesting for species such as the cerulean warbler and Swainson’s warbler. Also, it is reasonable to believe that all infested hemlocks across the CNF would not be treated, thus some canopy gaps would still be created as hemlock die. The Northern saw-whet owl inhabits spruce/fir forests. Although this bird does not depend on hemlock, it would nest in hemlock trees. Death of hemlock would reduce the amount of nesting habitat available for the owl, though nesting areas would remain in treated areas and other habitats.

Insects There are no rare terrestrial insect species directly dependent on hemlock. Scattered dying hemlock may cause canopy gaps in hardwood forests that may make the habitat more suitable for insect species that prefer ecotones between forested and more open areas including Diana fritillary. Also, it is reasonable to believe that all infested hemlocks across the CNF would not be treated, thus some canopy gaps would still be created as hemlock die. In treated areas, the use of predator beetles would have no impact on Diana fritillary. Since imidacloprid is either injected directly into the trunk of the tree, or into the soil, use of this chemical would have no impact on Diana fritillary.

Mammals In Tennessee, the Carolina Northern flying squirrel occurs at elevations above 4,000 feet. The squirrel is found in spruce-fir and northern hardwood forests where beech or yellow birch occurs. Its optimal conditions include cool, moist, mature forest with abundant standing and down snags. These squirrels occupy tree cavities, leaf nests and underground burrows, but they prefer cavities in mature trees as den sites. Carolina Northern flying squirrel diet includes insects, fruit, nuts, roots, and rhizomes, including those from hemlock. While hemlock decline within the squirrel’s habitat would increase the amount of snags available for cavity nesting, the loss of hemlock as a food source may be more significant to the survival of the species.

69 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Myotis grisescens sites are associated with foraging areas and not maternity caves, bachelor caves, or hibernacula. Gray bats are not hemlock dependent, but utilize stream and river corridors for foraging where hemlock-dominated stands can be found. Loss of hemlock would result in changes in forest composition and age-class distribution along foraging corridors, but these habitat changes would not have any notable impacts on gray bat foraging activity or populations in the area.

Indiana bats, Rafinesques big-eared bats and Eastern small-footed bats roost in snags and trees with crevices and exfoliating bark. Indiana bats have been documented in hemlock on the Forest during summer maternity roosting, but are not dependent upon hemlock communities. Gradual loss of hemlock would improve tree-roosting bat habitat locally by creating snags. Most hemlock sites do not provide the type of microhabitat needed by maternity colonies, so their use is limited. Snags in other community types would be sufficient for roosting.

Scattered dying hemlock may cause canopy gaps in hardwood forests that may make the habitat more suitable for foraging bats, including Indiana bat, big-eared bats and small- footed bat. It is reasonable to believe that all infested hemlocks across the CNF would not be treated, thus new hemlock snags and canopy gaps would still be created as hemlocks die even with treatment. The use of predator beetles would have no effect on the Carolina Northern flying squirrel or bats or their habitat. Imidacloprid was found to be slightly to moderately toxic to mammals when directly exposed to the chemical. It has a long half-life in soil (>31 days) at pH 5, 7, and 9, which represents most of the soil conditions across the Forests. Laboratory studies reported reductions in reproductive success from direct exposure to the chemical. Indiana bats, big-eared bats and small-footed bats would not be exposed directly or indirectly to imidacloprid. The use of imidacloprid would have no effect on the bats. The use of imidacloprid would not affect habitat for the bats since hemlock represents a small portion of total forest habitat. The use of imidacloprid may affect summer maternity habitat by slowing snag recruitment into the forest. It is not likely that all hemlock trees would be treated, therefore some hemlocks would continue to degrade, providing suitable roost trees for Indiana bats. Since imidacloprid would be injected into the tree or below the leaf litter into the soil, direct exposure of Carolina Northern flying squirrels to the chemical is not likely. However, squirrels foraging on cones, roots, and rhizomes of trees treated with imidacloprid may be indirectly exposed to low concentrations of the chemical. It is not likely that exposures to such levels of imidacloprid would approach laboratory study levels and result in effects on the Carolina Northern flying squirrel; however many of these questions remain to be answered scientifically. Therefore, the use of imidacloprid may affect, but is not likely to adversely affect, the Carolina Northern flying squirrel. The use of imidacloprid would positively affect habitat for the Carolina Northern flying squirrel by retaining hemlocks within the species known range.

70 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

The Southern rock vole and Southern water shrew are small mammals that are associated with dry and moist rocky areas, respectively. Although neither species is hemlock- dependent, the loss of hemlock around rock outcrops or stream channels would make the habitat less suitable. Also, it is reasonable to believe that not all infested hemlocks across the CNF will be treated, thus some canopy gaps would still be created as hemlock die. The use of predator beetles would not impact Southern water shrews. Similarly, the use of imidacloprid would not impact Southern water shrews or other small mammals. None of the four known sites for Sorex palustris occurs within a hemlock-dominated stand. Loss of hemlock would result in minor habitat alterations for the small mammals (changes in light conditions, added woody debris to stream, etc.). Although the loss of hemlock trees could result in some sedimentation and bank erosion, suitable habitat conditions should remain. Small population fluctuations may occur in response to these changes, but individuals are expected to persist at this site.

Although river otter is a riparian-associated species, it is not dependent on hemlock. Because otter populations tend to be limited by the availability of denning sites (e.g., dens of other species such as the bank dens of beavers or muskrats) or food sources, the death of hemlock is not likely to affect them. Use of predator beetles or imidacloprid would not negatively affect otters since there would be no contact with the beetles or imidacloprid.

Reptiles Northern coal skinks are not dependant on hemlock communities and in fact use a variety of drier habitats. Loss of hemlock would not affect them. The use of predator beetles would not impact the skink. Similarly, the use of imidacloprid would not impact the lizard. Since imidacloprid would be injected into the tree or below the leaf litter into the soil, direct exposure of this species to the chemical is not likely.

Mollusks There are no terrestrial mollusks dependent entirely on hemlock forests; however the snails occur in forested habitats with abundant leaf litter and rotting logs on the forest floor. Most terrestrial mollusks require shaded woodlands to prevent excessive drying of the forest floor. The amount of sunlight on the ground increases drying out leaf litter on the forest floor when hemlock die, This drying makes the areas less hospitable for terrestrial snails. When these areas are adjacent to streams, drying is expected to be less because of the influence of the adjacent stream on soils and leaf litter moisture. By treating the hemlock with biocontrol or chemicals, shade would be maintained in the forest, thus making it more suitable for this salamander species that requires moist conditions. The use of predator beetles would not affect the snails or their habitat.

71 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Because imidacloprid is either injected directly into the trunk of the tree, or approximately 8 -10 inches deep into the soil (below the leaf litter), use of this chemical would not affect the snails.

Based on the proceeding effects analysis, the Determinations of Effect for each species is listed below in Table 11.

Table 11. Determinations of Effect for Endangered, Threatened and Sensitive Species for Alternative 1 Scientific Name Determination of Effect* Reason for Determination Corynorhinus rafinesquii 1 O occurrences in hemlock 1 of 5 sites may be affected, not hemlock 1 Desmognathus carolinensis dependent, minimal impact Desmognathus santeetlah 1 O occurrences in hemlock Eurycea junaluska 1 O occurrences in hemlock Glaucomys sabrinus coloratus 4 1 of 3 sites may be affected 1 of 8 sites may be affected; impacts of hemlock 4 Myotis grisescens mortality on foraging habitat are negligible. 1 of 20 sites may be affected; impacts of hemlock 1 Myotis leibii mortality on foraging habitat are negligible. 1 of 4 sites may be affected; impacts of hemlock 4 Myotis sodalis mortality on foraging habitat are negligible. 1 of 17 sites may be affected, not hemlock 1 Plethodon aureoles dependent, minimal impact 2 of 15 sites may be affected , not hemlock 1 Plethodon welleri dependent, minimal impact Sorex palustris punctulatus 1 O occurrences in hemlock Speyeria diana 1 1 of 32 sites minimally affected *Effects Codes 1 = May impact individuals, but not likely to cause a trend to federal listing or a loss of viability 2 = Likely to result in a trend to federal listing or a loss of viability 3 = No impact 4 = Not likely to adversely affect

Aquatic Resource Alternative 1 would continue suppression of the HWA without adding additional treatment methods or expansion of treatment areas. No new adverse effects would occur to TES and Viability Concern species as a result of selecting this alternative. Hemlock trees would continue to die in areas outside the limited treatment zones. In the long term (decades) the summer shade that had been provided by hemlocks would eventually be provided by other species such as white pines and rhododendron.

Alternative 2 Botanical Resource Under this alternative, treatments could be expanded beyond the boundaries of the previously identified conservation areas (excluding designated wilderness areas) and the use of the additional chemical (dinotefuran) and the fungus (Lecanicillium muscarium) would be used. As described in the introduction to the “Environmental Consequences” section, there are no known effects to any plant species from these treatments.

72 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Threatened and Endangered Species As described for Alternative 1, there would be no effect to any Threatened or Endangered plant species on the forest.

Sensitive Species Indirect effects to rare plant species from the potential loss of hemlock from the canopy are described above under Alternative 1. Since no alternative proposes to treat all hemlocks on the forest, some indirect effects of hemlock loss are expected under all alternatives and it is anticipated that effects would be very similar to those described under Alternative 1. This alternative does provide for the most treatment of hemlocks however, thus anticipated effects would be the least under this alternative.

Determination of Effect for sensitive species Based upon the above analysis the implementation of the proposed project may affect individuals of the thirty-one sensitive species but would not lead to a trend toward federal listing or a loss of viability for the species.

Viability Concern Species As described for sensitive species above, effects would be very similar to those described for Alternative 1, though anticipated effects would be the least under this alternative.

Wildlife Resource Treatments would result in less hemlock mortality in sites currently occupied by Carolina northern flying squirrels. Impacts to squirrels would be similar to those described under Alternative 1, but treatments would reduce impacts on current habitat conditions, which would benefit individuals occupying this site. Imidicloprid has a low toxicity to mammals and treatment methods (tree or soil injection) would prevent any direct exposure to individuals in the area. Treatments in Tennessee would compliment treatments occurring within this mega population in North Carolina. Beetle release and chemical treatments are not expected to have any notable impacts on flying squirrels. No adverse effects are expected as a result of treatment.

Impacts to gray bat would be similar to those described under Alternative 1. Successful beetle treatments would reduce impacts on current habitat conditions at two sites. Treatment is not expected to have any notable impacts on gray bat populations. No adverse effects are expected.

Impacts on water shrew would be similar to those described under Alternative 1 because no known sites have been selected for treatment.

In treated areas, changes in habitat conditions for Weller’s salamander would be reduced as more hemlock trees are retained in the area. This would reduce impacts to soil moisture resulting in better habitat conditions for this species over the long- term. Beetle release is not expected to have any notable impacts on Weller’s

73 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest salamander. Impacts at the untreated site would be similar to those described under Alternative 1.

The only Diana fritillary site within hemlock-dominated forest would be treated with beetles and chemicals under this alternative. This would result in the fewest changes in present habitat conditions and provide for the largest retention of hemlock of the alternatives under consideration. Chemical treatments could result in some minor impacts to Diana fritillary. Caterpillars would be susceptible to Imidacloprid, which could result in some individual mortality. Because chemicals are injected directly into the tree or soil, exposure should be very limited and is not expected to result in notable population declines.

Alternative 2 includes expanded use of imidacloprid on all hemlock treatment areas. Effects from the use of imidacloprid on terrestrial wildlife would be similar to those described for Alternative 1 but more areas would be treated with imidacloprid. More chemical treatments would result in fewer hemlock snags and small canopy gaps within forests supporting hemlock. It is unlikely that all hemlocks would be treated, and these conditions would continue, although to a lesser extent, across the Forest. Refer to the discussion above for potential effects of imidacloprid use on terrestrial wildlife. Alternative 2 proposes the use of dinotefuran on trees that are heavily infested and need immediate attention to ensure survival. Dinotefuran is highly soluble in water and is relatively stable to hydrolysis at pH 4 to 9 (the range of most soils and water across the Forests). The chemical is highly mobile in most soil types, but demonstrates low potential for bioaccumulation (EPA 2004). Dinotefuran is practically nontoxic to birds and mammals (EPA 2004). Dinotefuran is highly toxic to honeybees (EPA 2004). High water solubility combined with high soil mobility indicates that this compound has the potential to leach into area streams and moist habitats. The major dissipation route for dinotefuran is aqueous photolysis (half-life 1.8 days), although even in its parent state, the chemical is practically nontoxic to fish and freshwater aquatic invertebrates (EPA 2004). From this, it is reasonable to assume that dinotefuran and its major degradates would be practically nontoxic to terrestrial species dependent on surface water for all or part of their life cycle (e.g. salamanders). Additionally, the EPA (2004) recommends using fish toxicity data as a surrogate for amphibians. Since dinotefuran is highly toxic to honeybees and remains relatively stable in most soil types, it is also reasonable to assume that the chemical may be toxic to terrestrial species within the leaf litter or top layers of soil (e.g. snails, spiders, etc.). Data on the toxicity of dinotefuran to other insects is sporadic, with some species being more sensitive to the compound than others. SERA (2009) describes the risk to nontarget insects based on exposure (e.g. direct exposure through broadcast spraying, indirect exposure through eating or using leaves or bark following soil or stem injection). Basically, the risk to nontarget insects is based on the amount and type of exposure (i.e. do nontarget insects eat leaves or bark of treated trees, or are they exposed to hemlock pollen). The Forest is not proposing to use broadcast application of dinotefuran; therefore there would be no direct exposure to any nontarget insects. Hemlock is a wind-pollinated species, so it is unlikely that nontarget insects would come in contact with relatively large amounts of pollen exposed to dinotefuran.

74 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

There is no data on the toxicity of dinotefuran to reptiles; however, the EPA recommends the use of bird toxicity data as a surrogate for these species groups (EPA 2004). The forests would utilize a clearance process similar to that for imidacloprid to aid in the determination of whether it is safe to use dinotefuran to treat individual trees, especially near surface water. This process would minimize the potential for direct contact between dinotefuran and nontarget organisms. Alternative 2 also proposes the use of naturally-occurring fungus Lecanicillium muscarium. Based on initial field studies of the fungus, there is little concern associated with effects of the fungus on non-target organisms, including insects.

Amphibians Use of L. muscarium would have no effects on Santeetlah dusky salamander, Junaluska salamander, Tellico salamander, Southern Appalachian salamander, or Weller’s salamander. Because dinotefuran is either applied directly to the trunk of the tree, or approximately 8 -10 inches deep into the soil (below the leaf litter), use of this chemical would have no effect on Santeetlah dusky salamander, Junaluska salamander, Tellico salamander, Southern Appalachian salamander, or Weller’s salamander. This method minimizes direct exposure of amphibians to the chemical. Additionally, dinotefuran is assumed to be nontoxic to amphibians.

Arachnids There are no rare arachnids directly dependent on hemlock. Because dinotefuran is either applied directly to the trunk of the tree, or approximately 8 -10 inches deep into the soil (below the leaf litter), use of this chemical would not affect Hypochilus coylei, H. sheari, Nesticus cooperi, N. crosbyi, N. mimus, N. sheari, or N. silvanus. This method minimizes direct exposure of spiders to the chemical.

Birds There are no Federally-listed or Sensitive birds that depend directly on hemlock forests; although several rare terrestrial bird species do show some dependence on coniferous forests, including hemlock. The use of dinotefuran would not affect rare birds because the compound is practically nontoxic to birds and there would be no direct exposure to the chemical. The use of L. muscarium would not affect birds as it is a fungus that attacks insects.

Insects There are no rare terrestrial insect species entirely dependent on hemlock. The application of dinotefuran on hemlock trunks or directly into the soil could have effects to rare inspect species but these would be minimal and would not result in federally listing of the species. The use of L. muscarium would not affect rare insects as it has been shown

75 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest to have relatively high host-specificity towards adelgid species and low infectivity to non-host species (Costa 2008).

Mammals Dinotefuran was found to be practically nontoxic to small mammals when they were directly exposed to the chemical. It has a long half-life in soil and is relatively mobile in most soil types. Laboratory studies reported reductions in reproductive success from direct exposure to the chemical. Bats would not be exposed directly or indirectly to dinotefuran and therefore the use of dinotefuran would not affect the bats. The use of dinotefuran is intended to reduce hemlock death and thus would slow snag recruitment, indirectly affecting maternity habitat. This may affect, but is not likely to adversely affect summer maternity habitat for the Indiana bat. There is no shortage of snags on the CNF. It is not likely that all hemlock trees would be treated, and therefore some hemlocks would continue to degrade, providing suitable roost trees for Indiana bats. Because dinotefuran would be directly applied to the trunk of the tree or below the leaf litter into the soil, direct exposure of Carolina Northern flying squirrels to the chemical is not likely. However, squirrels foraging on cones, roots, and rhizomes of trees treated with dinotefuran may be indirectly exposed to low concentrations of the chemical. It is not likely that exposures to such levels of dinotefuran would approach laboratory study levels and result in effects on the Carolina Northern flying squirrel; however many of these questions remain to be answered scientifically. The Southern rock vole and Southern water shrew are small mammals that are associated with dry and moist rocky areas, respectively. Dinotefuran is practically nontoxic to small mammals. The use of L. muscarium would not affect mammals as it is a fungus that attacks insects.

Mollusks Because dinotefuran would be directly applied to the trunk of the tree or below the leaf litter into the soil (below the leaf litter), use of this chemical would have no effect on the noonday globe or its habitat, and no impact on black mantleslug, glossy supercoil, or bidendate dome. The use of L. muscarium would not affect mollusks as it is a fungus that attacks insects.

Reptiles Because dinotefuran would be directly applied to the trunk of the tree or below the leaf litter into the soil (below the leaf litter), use of this chemical would not affect the skink or its habitat. The use of L. muscarium would not affect reptiles as it is a fungus that attacks insects.

76 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Based on the proceeding effects analysis, the Determinations of Effect for each species is listed below in Table 12.

Table 12. Determinations of Effect for Endangered, Threatened and Sensitive Species for Alternative 2 Scientific Name Determination of Effect* Reason for Determination Corynorhinus rafinesquii 1 O occurrences in hemlock 1 of 5 sites may be affected, not hemlock 1 Desmognathus carolinensis dependent, minimal impact Desmognathus santeetlah 1 O occurrences in hemlock Eurycea junaluska 1 O occurrences in hemlock Glaucomys sabrinus coloratus 4 1 of 3 sites may be affected 1 of 8 sites may be affected; impacts of hemlock 4 Myotis grisescens mortality on foraging habitat are negligible. 1 of 20 sites may be affected; impacts of hemlock 1 Myotis leibii mortality on foraging habitat are negligible. 1 of 4 sites may be affected; impacts of hemlock 4 Myotis sodalis mortality on foraging habitat are negligible. 1 of 17 sites may be affected, not hemlock 1 Plethodon aureoles dependent, minimal impact 2 of 15 sites may be affected , not hemlock 1 Plethodon welleri dependent, minimal impact Sorex palustris punctulatus 1 O occurrences in hemlock Speyeria diana 1 1 of 32 sites minimally affected *Effects Codes 1 = May impact individuals, but not likely to cause a trend to federal listing or a loss of viability 2 = Likely to result in a trend to federal listing or a loss of viability 3 = No impact 4 = Not likely to adversely affect

Aquatic Resource Alternative 2 would add the fungus treatment, add the use of the chemical dinotefuran, and increase the areas where these treatments and the ongoing treatments with Imidacloprid and beetle releases would occur.

77 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Forest Service 2009) since the inception of this project “The Hemlock Wooly Adelgid continues to severely impact hemlock across the entire Forest. Chemical treatment of refuge areas appears to be very effective in protecting the trees from mortality. The effectiveness of predator beetle releases in protecting the hemlocks in larger land areas is unknown at this time.” Expansion of treatment areas and beetle releases appears to be warranted and should not adversely affect the aquatic species with viability concerns. Analysis conducted for the suppression of HWA in the Nantahala/Pisgah National Forests (USDA Forest Service, 2010b) and at Royal Blue State Wildlife Management Area (Costa, 2010), determined the naturally occurring fungus Lecanicillium muscarium was effective against HWA but had little effect on none target organisms, including aquatic insects. Application of this fungus is not expected to affect aquatic T&E species. Dinotefuran and Imidacloprid both belong to a group of insecticides referred to as Neonicotinoids (derived from nicotine) which act on the central nervous system of insects. Both have low toxicity to fish (Valent USA Corporation, 2011; National Pesticide Information Center, 2011). Neonicotinoids are among the most widely used insecticides worldwide. Dinotefuran treatment is proposed because this chemical is more quickly absorbed by hemlock roots and distributed throughout the tree. It is as effective as Imidacloprid but provides more immediate protection to the hemlocks. Imidacloprid provides a longer period of protection than Dinotefuran. Used in combination and following labeling instructions, these chemicals will greatly improve the efficacy of HWA suppression and will not adversely affect aquatic T&E organisms.

Based on monitoring (USDA Forest Service 2009) since the inception of this project:

“The Hemlock Wooly Adelgid continues to severely impact hemlock across the entire Forest. Chemical treatment of refuge areas appears to be very effective in protecting the trees from mortality. The effectiveness of predator beetle releases in protecting the hemlocks in larger land areas is unknown at this time.”

Expansion of treatment areas appears to be warranted and, based on the research and monitoring discussed above, should not directly or indirectly adversely affect aquatic T&E organisms.

Alternative 3 Botanical Resource Under this alternative, treatments could be expanded beyond the boundaries of the previously identified conservation areas (excluding designated wilderness areas) and the use of the fungus (Lecanicillium muscarium) would be used. The chemical (dinotefuran) would not be used under this alternative. As described in the introduction to the “Environmental Consequences” section, there are no known effects to any plant species from these treatments.

Threatened and Endangered Species As described for Alternative 1, there would be no effect to any Threatened or Endangered plant species on the forest.

78 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Sensitive Species Indirect effects to rare plant species from the potential loss of hemlock from the canopy are described above under Alternative 1. Since no alternative proposes to treat all hemlocks on the forest, some indirect effects of hemlock loss are expected under all alternatives and it is anticipated that effects would be very similar to those described under Alternative 1. This alternative provides for a similar level of treatment of hemlocks compared to Alternative 2, the only difference being the chemical used. Since neither chemical has any known effect to plant species the anticipated effects would be very similar to those for Alternative 2.

Viability Concern Species As described for sensitive species above, effects would be very similar to those described for Alternative 2.

Wildlife Resource Treatment would be identical to Alternative 2, except that dinotefuran would not be used. Impacts on TES would be similar to those described under Alternative 2. Refer to the effects analysis above for the expanded use of imidacloprid and Lecanicillium muscarium. Based on the proceeding effects analysis, the Determination of Effects for Alternative 3 for each species is listed below.

Table 13. Determinations of Effect for Endangered, Threatened and Sensitive Species for Alternative 3

Scientific Name Determination of Effect* Reason for Determination Corynorhinus rafinesquii 1 O occurrences in hemlock 1 of 5 sites may be affected, not hemlock 1 Desmognathus carolinensis dependent, minimal impact Desmognathus santeetlah 1 O occurrences in hemlock Eurycea junaluska 1 O occurrences in hemlock Glaucomys sabrinus coloratus 4 1 of 3 sites may be affected 1 of 8 sites may be affected; impacts of hemlock 4 Myotis grisescens mortality on foraging habitat are negligible. 1 of 20 sites may be affected; impacts of hemlock 1 Myotis leibii mortality on foraging habitat are negligible. 1 of 4 sites may be affected; impacts of hemlock 4 Myotis sodalis mortality on foraging habitat are negligible. 1 of 17 sites may be affected, not hemlock 1 Plethodon aureoles dependent, minimal impact 2 of 15 sites may be affected , not hemlock 1 Plethodon welleri dependent, minimal impact

79 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Scientific Name Determination of Effect* Reason for Determination Sorex palustris punctulatus 1 O occurrences in hemlock Speyeria diana 1 1 of 32 sites minimally affected *Effects Codes 1 = May impact individuals, but not likely to cause a trend to federal listing or a loss of viability 2 = Likely to result in a trend to federal listing or a loss of viability 3 = No impact 4 = Not likely to adversely affect

Aquatic Resource Alternative 3 is the same as Alternative 2 except that Dinotefuran would not be used. Refer to the effects analysis for Alternative 2.

Cumulative Effects

Wildlife Resource Past and present timber harvest and prescribed burning activities on the CNF have affected eastern and Carolina hemlocks and have affected sites that could support eastern hemlock such as acidic coves. Many of these activities will continue in the future but the resulting individually minor effects are insignificant when compared with the major impacts expected from infestation of the HWA. Therefore, these activities are unlikely to add or combine measurably with the impact resulting from an expanding HWA population over time. This project would result in lessening cumulative impacts to hemlock from HWA infestation and all other actions.

The fact that this community type is naturally limited in distribution, coupled with the impending threats from the hemlock wooly adelgid, which would impact the species regardless of land ownership, leaves the long-term maintenance of historical distribution and abundance of this community type in question. The fate of associated viability concern species would be dependent upon their ability to adapt to changing environmental conditions associated with the decline of hemlock from within these communities. Species that utilize hemlock forests in addition to other vegetative community types would be more likely to persist than species that are obligates to the hemlock forest community (USDA Forest Service 2004a).

Hemlock communities (both Carolina and Eastern) are afforded plan protection through various goals, objectives, and standards, yet despite these protections it is projected that future landscape distribution would be poor (see above). Suppression efforts on adjacent lands in neighboring North Carolina are already underway and similar efforts are being planned in Georgia. Additionally, limited individual tree chemical treatments are currently under decision within selected recreational sites on the CNF.

Alternatives 1

Botanical Resource Despite implementation of the 2005 decision and subsequent SIR, the rate of spread of HWA across the forest has been much more rapid than was originally anticipated and treatment efforts have not been sufficient to protect all the sites identified for protection

80 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest in that environmental assessment. Efficacy of beetle treatments is currently inconclusive and some beetle release sites have seen significant hemlock mortality. Without additional control measures the HWA infestation will continue to spread across the Cherokee National Forest potentially killing any remaining untreated sites of both hemlock species over time. Rare plant species associated with hemlock forests may also suffer losses dependent upon their ability to adapt to the changing environmental conditions. Potential effects to associated species range from altered fitness due to less than optimal habitat conditions, to local extirpations. Potential losses would be highest for those species dependent upon the moist and shaded conditions typically found in eastern hemlock stands.

Wildlife Resource Under this alternative the synergistic effects of combined landscape treatments between states are diminished. It is likely that hemlock would be lost as a significant component of the ecosystem. Wildlife species dependent upon hemlock communities would suffer population declines as hemlock is lost from the landscape. Unless species are able to adapt and utilize habitat conditions available within other forest types, these species could be lost in the future within the CNF. Aquatic Resource Sedimentation of streams is the most significant adverse effect that contributes to cumulative effects to aquatic TES and Viability Concern species on national forest lands. The alternative does not propose any ground disturbance. Compliance with labeling instructions for the two chemicals would minimize or eliminate harmful impacts and cumulative adverse impacts to the aquatic TES and Viability Concern species. Management activities, on National Forest System lands, that are reasonably foreseeable and under the control of the Forest Service would be implemented under the standards for protecting streams listed in the Revised Land and Resource Management Plan for the Cherokee National Forest (USDA Forest Service 2004a). Implemented in conjunction with the alternatives, these activities would not increase the adverse cumulative effects on the aquatic TES and Viability Concern species now occurring. Reasonably foreseeable activities that occur on private lands are not controlled by the Forest Service and could have negative impacts on aquatic TES and Viability Concern species.

Alternative 2 Botanical Resource This alternative maximizes the effort to protect hemlock on the forest in conjunction with similar efforts in adjacent states. By linking conservation areas across landscape positions where hemlock is of the most ecological importance, the best long-term distribution of the species is achieved. Despite this however, HWA would continue to infest and kill untreated hemlocks across the forest and region, especially within designated wilderness where additional treatments are not approved, and indirect impacts to rare plant species would occur as habitats are altered due to hemlock mortality.

Wildlife Resource This alternative maximizes the effort to protect hemlock on the CNF in conjunction with similar efforts in adjacent states. By linking conservation areas across landscape

81 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest positions where hemlock is of the most ecological importance despite geopolitical boundaries, the best long-term distribution of the species is achieved. Wildlife species dependent upon hemlock communities would suffer population declines as hemlock availability is reduced on the landscape. Successful adelgid treatment would provide refugia allowing for some of these species to persist on the Forest. Given the availability and distribution of remaining hemlock habitat, some losses should be expected unless species are able to adapt and utilize habitat conditions available within other forest types. It is still possible that some hemlock losses would be sustained. Even with slight changes in the canopy, some micro-site conditions related to temperature, moisture, and sunlight may be altered. Under this alternative no detrimental effects are anticipated to these species, however some impacts to the hemlock canopy and resultant changes in micro- site conditions could result in reduced fitness for some species’ populations. Viable populations of all other species of local concern would remain on the landscape under this alternative, however the reductions would leave these species more susceptible to future impacts.

Aquatic Resource Sedimentation of streams is the most significant adverse effect that contributes to cumulative effects to aquatic TES and Viability Concern species on national forest lands. The alternative does not propose any ground disturbance. Compliance with labeling instructions for the two chemicals would minimize or eliminate harmful impacts and cumulative adverse impacts to the aquatic TES and Viability Concern species. Management activities, on National Forest System lands, that are reasonably foreseeable and under the control of the Forest Service would be implemented under the standards for protecting streams listed in the Revised Land and Resource Management Plan for the Cherokee National Forest (USDA Forest Service 2004a). Implemented in conjunction with the alternatives, these activities would not increase the adverse cumulative effects on the aquatic TES and Viability Concern species now occurring. Reasonably foreseeable activities that occur on private lands are not controlled by the Forest Service and could have negative impacts on aquatic TES and Viability Concern species.

Alternative 3 Botanical Resource This alternative differs only in method of treatment, not scope, thus is negligibly different from Alternative 2 for cumulative impacts to rare plant species

Wildlife Resource This alternative would achieve limited landscape protection for hemlock in fragmented locations on the CNF. Cumulative impacts to the above species would be similar to those previously discussed under Alternative 2. However, fewer sites would be protected from hemlock mortality due to smaller range of treatments. Wildlife species dependent upon hemlock communities would have fewer refugia remaining on the landscape, so greater population declines and species losses should be expected when compared to Alternative 2.

82 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Aquatic Resource (Alternatives 1, 2, & 3) Sedimentation of streams is the most significant adverse effect that contributes to cumulative effects to aquatic TES and Viability Concern species on national forest lands. The alternative does not propose any ground disturbance. Compliance with labeling instructions for the two chemicals would minimize or eliminate harmful impacts and cumulative adverse impacts to the aquatic TES and Viability Concern species. Management activities, on National Forest System lands, that are reasonably foreseeable and under the control of the Forest Service would be implemented under the standards for protecting streams listed in the Revised Land and Resource Management Plan for the Cherokee National Forest (USDA Forest Service 2004a). Implemented in conjunction with the alternatives, these activities would not increase the adverse cumulative effects on the aquatic TES and Viability Concern species now occurring. Reasonably foreseeable activities that occur on private lands are not controlled by the Forest Service and could have negative impacts on aquatic TES and Viability Concern species.

Recreation ______Affected Environment The forested mountains and diversity of water resources in Cherokee National Forest (CNF) are natural attractions pulling visitors to developed recreation areas and remote backcountry settings for sightseeing, trails, camping, fishing, hunting, boating and other outdoor recreation activities. Hemlock trees are an established part of this attraction and directly contribute to the aesthetic appeal of landscape settings across forest, especially water-based settings.

Along creeks and rivers, hemlock trees are prevalent and also function to provide shade, cover and screening for fish and wildlife habitat, which further enhances recreation opportunities. Many trails, scenic drives, developed and undeveloped recreation sites across the forest are characterized by the presence of hemlock trees. Alternative 1 Beetle releases and imidacloprid would continue to be used to control Hemlock Woolly Adlelgid (HWA) in select developed recreation areas and hemlock conservation areas throughout the Cherokee National Forest (CNF) as described in the 2005 Decision and 2010 Supplemental Information Report (SIR). Hemlock trees located outside of identified conservation areas would remain susceptible to HWA. Decline and mortality of hemlocks would be expected across the CNF landscape. As described for the 2005 Environmental Assessment, the visual impacts of dead and dying hemlock would be noticeable as the distinctive evergreen character of hemlocks continues to fade from the forested landscape. These effects would be perceived as negative deviations from the desired landscape character and would lower the scenic integrity of affected landscapes. In areas viewed from a distance less than a half-mile (foreground scenery) the negative impacts of dead and dying trees may dominate the scenery for several years. From

83 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest distances greater than a half-mile (middleground and background scenery), the dead and dying hemlock trees would appear in patterns similar to those caused by other landscape level disturbances like wildfire or pine beetle epidemics. At this larger scale, alterations to the landscape would appear more subordinate to surrounding hardwood forests and vary during leaf-on and leaf-off seasons.

The effects to recreation opportunities across the forest would extend beyond declining scenic integrity. Similar to the effects of the southern pine beetle, standing dead and fallen trees would become frequent scenes across the CNF especially in affected riparian corridors. As dead hemlock trees begin to fall, the cover, shade and screening desired for fishing, hunting, camping, hiking and enjoyment of other recreation activities would also decrease.

The threat of falling trees would increase over the next 5 to 10 years creating a greater concern for personal safety. Infested hemlock trees may be deliberately fallen and/or removed to address safety concerns along travel corridors including trails, waterways and road right-of-ways and in developed and undeveloped recreation sites. Some areas could be closed to public use due to safety concerns. In general, access would become more difficult in affected areas of the forest. Fallen hemlock trees in streams and rivers would create physical barriers for activities such as boating, tubing and fishing. Access to and through some areas of remote backcountry and designated Wildernesses would also become increasingly difficult as hemlock trees fall across trails, streams and seldom used roads.

The task of manually removing these fallen trees would be laborious, dangerous and potentially cause further resource degradation in some locations. Wind events have the potential to quickly make even a recently maintained trail impassable again. Any restoration efforts would be evaluated and prioritized by forest managers.

Alternative 2 The suppression of HWA would be expanded across the CNF and include additional types of treatments. All areas of the CNF excluding designated Wilderness would be available for HWA suppression treatments. The intended effect would be less hemlock tree mortality and decline across the CNF and greater chances for a continued presence of hemlock trees in the future. The potential to maintain the existing scenic integrity of valued recreation settings in the CNF would be greater in Alternative 2 due to increased treatments and treatment options. However, as described for Alternative 1, HWA would continue to cause noticeable negative impacts to the CNF recreation and scenery resources. Alternative 3 Excluding the option to use Safari pesticide in Alternative 3 would potentially limit the effectiveness of treating large hemlock trees heavily infested with HWA. Overall, the loss of large hemlock trees in recreation areas as well as general forested areas would be expected to be higher in Alternative 3 in comparison to Alternative 2, but lower than Alternative 1.

84 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Cumulative Effects The cumulative effects of all three alternatives would vary only in their potential to sustain hemlocks in the affected landscape. Alternative 2 would have the highest potential for successfully maintaining hemlock populations as it would allow for the greatest number of treatment options followed by Alternative 3 and Alternative 1, No Action. In the past, the southern pine beetle (SPB) outbreak that occurred from 1999-2003 impacted the scenic integrity and recreation opportunities across CNF. The past and present impacts of the SPB outbreak would have a cumulative effect with the present and reasonably foreseeable future HWA infestations.

In all alternatives, the accumulation of dead and dying hemlock would compound the remaining impacts of the SPB. Forest Service managers would continue to prioritize the restoration of access to affected trails, roads, waterways and recreation areas. Wilderness Resources ______Affected Environment “Wilderness” is defined in the Wilderness Act of 1964 as “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.”

Cherokee National Forest (CNF) presently includes 66,389 acres of congressionally designated Wilderness including all or part of eleven Wilderness areas. An additional 20,537 acres were allocated to wilderness study areas through the Forest Plan revision process. These areas are managed to protect their wilderness attributes until Congress determines whether or not to include them in the National Wilderness Preservation System.

Alternative 1 Beetle releases and the application of imidacloprid would continue in conservation areas located within designated Wilderness as described in the 2005 Decision and SIR. A total of six active treatment sites across five Wilderness areas would be allowed: Citico Creek (1 site), Joyce Kilmer/Slickrock (1 site), Unaka Mountain (2 sites), Pond Mountain (1 site) and Big Laurel Branch (1 site). Alternatives 2 and 3 HWA treatments would be limited to six active sites as described in Alternative 1. The opportunities to suppress additional HWA populations outside of Wilderness boundaries

85 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest would be increased in Alternatives 2 and 3. Alternative 2 would provide the most treatment options. Cumulative Effects

The intended cumulative effect of treatments to suppress HWA inside and outside Wilderness on the Cherokee National Forest and surrounding public lands would continue to help prevent unacceptable damage to resources on adjacent lands or unnatural loss to the wilderness resource due to exotic pests. The cumulative effects of all three alternatives would vary only in their potential to sustain hemlock trees in affected Wilderness. Alternative 2 would have the highest potential for successfully maintaining hemlock populations because it would allow for the greatest number of treatment options across the landscape as a whole. Even with active treatments to suppress HWA within and outside designated Wildernesses, hemlock mortality would be expected in untreated areas over time unless the predator beetles prove to be highly effective in suppressing HWA. An accelerated invasion of non-native species would be anticipated in areas of hemlock mortality. This would further degrade the “naturalness” of the affected Wilderness resources and create a need for additional actions to remove or control the non-native invasive species in the reasonably foreseeable future.

The past and present impacts of the southern pine beetle (SPB) outbreak would have a cumulative effect with the present and reasonably foreseeable future HWA infestations. In terms of recovery efforts to restore access along trails, Forest managers would have to establish priorities and consider projects related to the SPB and HWA infestations across the forest. In general, restoring access in remote Wilderness areas with primitive tools would require more time and resources than trails located outside Wilderness resource boundaries.

The past and present ongoing human influences within Wilderness resources that tend to reduce the perception of “wildness” would continue such as trail use impacts, trail maintenance, bridge repair, dispersed campsites and the sights and sounds of humans. Outside influences such as noise, water and air pollution and exotic plants or organisms would also continue to affect the “wildness” and “naturalness” of Wilderness resources in CNF.

Climate Change ______Affected Environment Climate change can affect the resources in the project area and the proposed project can affect climate change through altering the carbon cycle. Climate models are continuing to be developed and refined, but the two principal models found to best simulate future climate changed conditions for the various regions across the country are the Hadley Centre model and the Canadian Climate Centre model (Climate Change Impacts on the United States 2001). Both models indicate warming in the southern region of the US. However, the models differ in that one predicts little change in precipitation until 2030

86 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest followed by much drier conditions over the next 70 years. The other predicts a slight decrease in precipitation during the next 30 years followed by increased precipitation. These changes could affect forest productivity, forest pest activity, vegetation types, major weather disturbances (droughts, hurricanes), and streamflow. These effects would likely be seen across the Forest, thoughin some sensitive areas may be affected sooner than others. not have any such sensitive areas

Environmental Consequences

Alternatives 1, 2, and 3 In general, genetic diversity provides resilience to a variety of environmental stressors (Moritz, 2002, Reed and Frankham, 2003, Reusch et al., 2005). Climate change affects biodiversity directly by altering the physical conditions to which many species are adapted. In some instances, changes in precipitation patterns may disrupt animal movements and influence recruitment and mortality rates (Inouye et al., 2000). Evidence is accumulating to indicate that species interactions and competitive responses under changing climates are complex and unexpected (Suttle, Thompsend, and Power, 2007). Although species with large geographic ranges have a wide range of physiological tolerance, species that are rare, threatened, endangered, narrowly distributed, and endemic, as well as those with limited dispersal ability, would be particularly at risk under climate change (Pounds et al., 2006) because they may not be able to adapt in situ or migrate rapidly enough to keep pace with changes in temperature (Hansen et al., 2001; Wilmking et al., 2004; Neilson et al., 2005b). A key predicted effect of climate change is the expansion of native species’ ranges into biogeographic areas in which they previously could not survive (Simberloff, 2000; Dale et al., 2001). This prediction is supported by the observed northward shift in the ranges of several species, both native and introduced, due to the reduction of cold temperature restrictions (Parmesan, 2006).

Maintenance of genetic diversity provides resilience to a variety of environmental stressors. Climate change affects biodiversity by altering the physical conditions to which many species are adapted. Range distribution for species varies.

Projected changes in temperature and precipitation suggest that southern ecosystems may shift dramatically. Depiction of the northern shift of the jet stream and the consequent drying of the Southeast (Fu et al., 2006) varies among future climate scenarios, with some showing significant drying with others show increased precipitation (Bachelet et al., 2001). Even under many of the somewhat wetter future scenarios, closed-canopy forests of the Southeast may revert or in some areas, be converted under temperature- induced drought stress (Bachelet et al., 2001; Scholze et al., 2006). Temperature induced droughts in Mountain ecosystems are expected to contribute to forest diebacks (Bugmann, Zieri, and Schumacher, 2005; Millar, Westfall, and Delaney, forthcoming).

The interactions of climate change with other stressors such as insects (Volney and Fleming, 2000; Logan, Regniere, and Powell, 2003), disease (Pounds et al., 2006) would challenge the management of ecosystem services and biodiversity conservation in NF ecosystems. Older forests can be strong carbon sinks (Stoy et al. 2006), and older trees

87 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest absorb more CO2 in an elevated CO2 atmosphere, but wood production of these trees show limited or only transient response to CO2 (Korner et al. 2005). Studies of elevated CO2 on trees have been done with young trees (which show a positive growth response), but the one study on mature trees showed no growth response (Korner, et al. 2005). This is consistent with model results found in an independent study (Kirschbaum 2005). The general findings from a number of recent syntheses using data from the three American and European FACE sites (King et al. 2004; Norby et al. 2005; McCarthy et al. 2006a; Palmroth et al. 2006) show that North American forests will absorb more CO2 and might retain more carbon as atmospheric CO2 increases. In this study, thinning removed carbon form the stand (in the form or removed logs) and also resulted in substantial, but temporary, reduction in ANNP (aboveground net primary production). The reduction of ANNPP by thinning lasted only 1 year, and its recover was likely due to changes in the foliar mass and leaf traits. Finally, the data portrayed that there is a transient impact of thinning on ANPP (aboveground net primary production), but that there is no long-term effect of thinning on aboveground carbon uptake in oak forests. Although this study focused on oak forests, the same principles and effects would apply to the hemlock forests.

The alternatives (1, 2, and 3) would alter the carbon cycle in that it affects the carbon stock in any one of the pools. Each alternative would remove biomass as a result of down hemlock trees. This would reduce the amount of carbon stored in the treated stands. A portion of the carbon removed would remain stored for a period of time in wood products.

The increase in down, dead wood would temporarily convert stands from a carbon sink that removes more carbon from the atmosphere than it emits, to a carbon source that emits more carbon through respiration than it absorbs. These stands would remain a source of carbon to the atmosphere until carbon uptake by new trees and other vegetation exceeds the emissions from decomposing dead organic material. The stands would likely remain a carbon source for several years, and perhaps for more than a decade, depending on the amount of dead biomass left on site, the length of time before new trees become reestablished, and their rate of growth once reestablished. As the stands continue to develop, the strength of the carbon sink would increase until peaking at an intermediate age and then gradually decline but remain positive. Similarly, once new trees are established, carbon stocks would accumulate rapidly for several decades. The rate of accumulation would slow as the stands age. Carbon stocks would continue to accumulate, although at a declining rate, until impacted by future disturbances.

Recent scientific literature confirms this general pattern of changes in net ecosystem productivity (NEP)1 and carbon stocks over the period of forest stand development. Most

1 Net ecosystem productivity, or NEP, is defined as gross primary productivity (GPP) minus ecosystem respiration (ER) (Chapin et al. 2006). It reflects the balance between (1) absorbing CO2 from the atmosphere through photosynthesis (GPP) and (2) the release of carbon into the atmosphere through respiration by live plants, decomposition of dead organic matter, and burning of biomass (ER). When NEP is positive, carbon accumulates in biomass. Ecosystems with a positive NEP are referred to as a carbon sink. When NEP is negative, ecosystems emit more carbon than they absorb. Ecosystem with a negative NEP are referred to as a carbon source.

88 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest mature and old stands remained a net sink of carbon. Pregitzer and Euskirchen (2004) synthesized results from 120 separate studies of carbon stocks and carbon fluxes for boreal, temperate, and tropical biomes. They found that in temperate forests NEP is lowest, and most variable, in young stands (0-30 years), highest in stands 31-70 years, and declines thereafter as stands age. These studies also reveal a general pattern of total carbon stocks declining after disturbance and then increasing, rapidly during intermediate years and then at a declining rate, over time until another significant disturbance (timber harvest or tree mortality resulting from drought, fire, insects, disease or other causes) kills large numbers of trees and again converts the stands to a carbon source where carbon emissions from decay of dead biomass exceeds that amount of carbon removed from the atmosphere by photosynthesis within the stand.

The impacts of the action alternatives on global carbon sequestration and atmospheric concentrations of CO2 are miniscule. However, the forests of the United States significantly reduce atmospheric concentrations of CO2 resulting from fossil fuel emissions. The forest and wood products of the United States currently sequester approximately 200 teragrams2 of carbon per year (Heath and Smith, 2004). This rate of carbon sequestration offsets approximately 10% of CO2 emissions from burning fossil fuels (Birdsey et al., 2006). U.S. Forests currently contain 66,600 teragrams of carbon. The short-term reduction in carbon stocks and sequestration rates resulting from the proposed project are imperceptibly small on global and national scales, as are the potential long-term benefits in terms of carbon storage.

The currently large carbon sink in US forests is a result of past land use changes, including the re-growth of forests on large areas of the eastern U.S. harvest in the 19th century, and 20th century fire suppression in the western U.S. (Birdsey et al. 2006). The continuation of this large carbon sink is uncertain because some of the processes promoting the current sink are likely to decline and projected increases in disturbance rates such as fire and large-scale insect mortaily may release a significant fraction of existing carbon stocks (Pacala et al. 2008; Canadell et al. 2007). Management actions – such as those proposed – that improve the resilience of forest to climate-induced increases in frequency, and utilize harvested trees for long-lived forest products and renewable energy sources may help sustain the current strength of the carbon sink in US forests (Birdsey et al. 2007).

Cumulative Effects

For all alternatives, there is confidence that temperatures are changing at a global scale, yet it is difficult to predict the effect of climate change at local and regional scales because the relationship between climate change and the proposed project areas are at a minute scale. Thus the contribution of the proposed actions to the carbon cycle is extremely small. When looked at the implementation collectively, the risk and rate of additional carbon release through regeneration is minimal for the reasonably foreseeable future in comparison to extinction of species and the complete loss of a carbon source.

2 200 teragrams, or Tg, equals 196,841,306 US tons.

89 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Road Management ______No road construction, road reconstruction, changes to current road management, or road obliteration is proposed as part of this project. No roads analysis is needed. Soils/Groundwater ______Scope of the analysis This assessment of direct, indirect, and cumulative effects is limited to the proposed use of imidacloprid and dinotefuran and the possibility of ground and surface water contamination.

In a review of available information, there were sufficient references to the possibility of groundwater contamination to warrant additional analysis of the characteristics of imidacloprid and dintotefuran and the susceptibility and vulnerability of groundwater in the project area to possible contamination.

Affected Environment Eastern hemlock occurs on a variety of sites across the landscape that are generally characterized as being moist to very moist but with good drainage. Eastern hemlock is most common along streams and sites that are generally protected from fire. With the exclusion of fire over the last 50 years eastern hemlock can be found away from drainages on slopes where soils are well drained and dry during drought conditions. In the Southern Appalachians the most frequent occurrences are at elevations from 2,000 to 5,000 feet.

Carolina hemlock typically occurs in primarily in small patches in rocky bluff habitats, but can be found on valley bottoms, gorge slopes, or other protected landforms. Soils would typically be shallow and well drained.

Soil erosion and sedimentation are the primary causes of soil loss and are the result of factors including soil type, rainfall, as well as vegetation cover and management practices. While there are many ways to minimize erosion, vegetative cover is the most effective over the long term. When vegetation is removed, the rate of soil erosion is greatly accelerated. Environmental Consequence Direct and Indirect Effects (Alternatives 1, 2 & 3) None of the alternatives propose treatment activities that would increase or decrease soil productivity through soil erosion or soil compaction. While the loss of hemlocks as vegetative cover could potentially lead to increased erosion, it is likely that as hemlocks are lost from the overstory, understory vegetation will increase, thereby providing continuous vegetative cover and relatively minimal erosion.

90 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

All alternatives include the application of the systemic insecticide imidacloprid, injected into the soil at the base of the trees. The portion of the chemical that is not taken up by the tree could spread downward in the soil a few feet from the injection sites and bind with soil particles and would not be expected to reach ground water (Burkingstock et al. 1997). Its biodegradation rate in soil has been characterized as moderately slow, with about 50% of the applied residue dissipating in a range of 48 – 190 days (Felsot 2001). Soil-inhabiting invertebrates that come into contact with the chemical while it is still active in the soil would likely be impacted, but the properties of the soil itself would not change. Invertebrates would be expected to recolonize the soil near the base of the tree when the chemical was no longer active. The loss of invertebrates in the soil at the bases of the treated hemlocks would be localized and temporary. All alternatives propose the use of predatory beetles. The beetles are applied directly to the branches of the trees and any contact with the soil is incidental. Beetle treatments would not adversely affect soil productivity through soil erosion or soil compaction. Specifically, for Alternatives 2 and 3, the fungus Lecanicillium muscarium would be aerially applied to hemlock trees and therefore would not result in erosion or soil compaction. Additionally, very little fungus would ultimately reach the soil surface so there is minimal risk associated with effects to soil composition as a result of fungus application. The majority of the areas likely to be treated in any of the alternatives are not susceptible to shallow groundwater contamination due to one or more of the following: distance from surface water; water table depths in excess of 18 inches; soils that are loamy, silty, or clayey, and/or have relatively high percentages of organic material. Individual sites would be retreated at an every two year or longer interval until biological control are effective. This two year retreatment cycle is sufficiently long to allow for two or more half-lives of degradation of the applied chemical (Gervais, 2010; EPA, 2004). Each chemical has a upper limit of active ingredient per acre that can be applied per year, thereby limiting the likelihood of excess application at a site. Susceptible sites would either 1) not be treated during the time period when the local water table is high (see attached map unit descriptions) or 2) would be treated using the trunk spray technique. Based on the above information, there should be no direst or indirect effects to groundwater from the implementation of any of the alternatives if they are implemented with the pesticide label requirements, Forest Plan requirements, and design features. Cumulative Effects (Alternatives 1, 2, & 3) Since none of alternatives considered would impact soil productivity through erosion or compaction, this project would not contribute to any cumulative impacts. In regards to the soil injection of imidacloprid and dinotefuran, some additional use is ongoing on both public and private lands as infestation of HWA progresses. This additional use is expected to be very limited, due to cost and logistics of treatment. On these other lands, loss of invertebrates in the soil at the bases of the treated hemlocks would be localized and temporary. There would be no cumulative effects to the loss of soil invertebrates.

91 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Based on the above information, there should be no cumulative effects to groundwater from the implementation of any of the alternatives if they are implemented with the pesticide label requirements, Forest Plan requirements, and design features.

Heritage Resources______This project does not include activities related to timber sales, road construction, recreation development or land exchanges as listed in the Programmatic Agreement between the U.S. Department of Agriculture, Forest Service and the Tennessee State Historic Preservation Office (SHPO) dated June, 2005. The project would have no effect, adverse or beneficial to a heritage resource and no Section 106 compliance documentation is required.

Human Health and Safety ______Environmental Consequences Alternative 1 – No Action The treatment of hemlocks with beetles and imidacloprid, particularly in recreation areas, would reduce the threat from dying hemlocks injuring forest users. Beetles would primarily be released onto hemlock branches that can be reached from the ground. Should any tree climbing be necessary, qualified climbers would be employed. Soil injection and stem injection of imidacloprid would be performed by a combination of USDA Forest Service personnel and private contractors. All working crews would be under the supervision of certified pesticide applicators and/or commercially licensed personnel as appropriate and would be required to follow precautionary procedures proscribed by the manufacturer. The potential safety issues include: accidental spills of the product and accidental contamination of skin or clothing of the applicators. A spill response plan would be developed as part of implementation of this project. The potential for the imidacloprid used in this project to impact human health under any circumstances is extremely minimal as its toxicity to mammals in general including humans is very low. No imidacloprid is expected to reach ground water since no soil injection would occur in close proximity to water or in highly permeable soils. Also, the method of application deposits the agent within a small area approximately 10 inches beneath the surface of the soil. Imidacloprid is known to bind with the soil so that whatever is not taken up by the roots of the tree is not expected to move away from the application site. The U.S. Environmental Protection Agency Office of Pesticide Programs has evaluated imidacloprid and determined it to be non-cancer causing (US EPA, Science and Information Branch).

92 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Alternative 2 The risks associated with the application of imidacloprid and beetle release would be similar to those described for Alternative 1. SafariTM is a common trade name for the chemical dinotefuran and is approved for use in Tennessee for the control of hemlock woolly adelgid. Similar to imidacloprid, it is within the class of neonicotinoid pesticides. The human health and safety effects associated with the use of dinotefuran would be similar to those described for imidacloprid. No dinotefuran is expected to reach ground water since no soil injection would occur in close proximity to water or in highly permeable soils. The fungus Lecanicillium muscarium, currently being proposed for treatment is commercially listed as Mycotal and is registered in seven European Union countries, Turkey, and Japan. A Tier II evaluation for European registration, that included human health, environmental and ecotoxicological analysis was conducted for Mycotal’s European registration, concluded that Mycotal is not considered harmful to the environment. The approved Material Safety Data Sheet (MSDS) for Mycotal lists no human health hazards associated with application. Alternative 3 Potential effects to human health and safety would be similar to those described in Alternative 2 with the exception that there would no use of dinotefuran and therefore no risks associated with application of the chemical. Cumulative Effects All alternatives would still result in some level of hemlock mortality on the forest as not all trees will be treated. The level of mortality has the potential to be the greatest with Alternative 1 and less with Alternatives 3 and 2. Impacts of hemlock mortality and the threat from these trees being broken off or uprooted and falling on humans using the forests would add to the level of risk that always exists in the Forests for a tree to potentially fall on a person. This rarely occurs on the Forests.

Soil injection of imidacloprid for the treatment of hemlock woolly adelgid has been occurring on the forest since 2005. Individual trees are treated on a rotation of no less than two years. When proper application methods are used there are no cumulative risks to applicators associated with repeated exposure to approved pesticides.

Economics ______Scope of Analysis The scope of the analysis is the proposed hemlock areas. The time frame used in the analysis is from when the infestation was detected in early 2000s up to 10 years in the future. This time frame was chosen to disclose the potential increase and spread of the hemlock wooly adelgid population and for associated impacts to occur.

93 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Affected Environment Several local industries and sawmills utilize timber from the Cherokee National Forest. Forest workers and contractors contribute to local economies. Many local residents heat their homes with firewood, while others supplement their income by gathering other forest products. Hemlock is a minor component in the local timber industry. Hemlock has an important but less tangible economic value in terms of its contribution to visual quality and visitor experience. The quality of the visitor experience drives the local tourism economy.

Environmental Consequences Direct, Indirect, and Cumulative Effects

Alternative 1(No Action), 2 (Proposed Action), and 3 (No Additional Chemicals)

All alternatives may provide additional economic benefit to local employment and revenue in this area of upper east Tennessee. Alternative 2 would provide the greatest economic benefit followed by Alternative 3 and then Alternative 1. There are no direct revenues or associated costs with any alternative to the Forest. The impacts would be to the local community. With the continuing spread of the adelgid, the trees would eventually die without intervention. However, with treatment for the adelgid, the forest community could be saved which would benefit the recreationist and provide continual scenery/visual opportunities. This would in turn bring revenues into the local community. Collectively over time, the impacts from the social perspective (recreational, visual, and local industries ) would be minimal.

94 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

CONSULTATION AND COORDINATION

The Forest Service consulted the following individuals, Federal, State, and local agencies, tribes and non-Forest Service persons during the development of this environmental assessment: ID TEAM MEMBERS: Bob Lewis: Forester, Team Leader Doug Byerly: Recreation Program Manager/Landscape Architect Mark Pistrang: Forest Botanist/Ecologist Jim Herrig: Forest Aquatic Biologist Mary Miller: Forest Wildlife Biologist Katherine Foster: District Ranger, Hydrologist Consultant

FEDERAL, STATE, AND LOCAL AGENCIES: US Fish and Wildlife Service, Cookeville Field Office: for concurrence on effects on Threatened and Endangered species. US Forest Service, Forest Health Protection, Asheville Field Office: for technical guidance on detection, suppression methods, and monitoring. US Army Corps of Engineers Tennessee Valley Authority

TRIBES: Kialegee Tribal Town, Wetumka, OK Muscogee (Creek) Nation, Okmulge, OK United Keetoowah Band of Cherokee Indians, Tahlequah, OK Cherokee Nation of Oklahoma, Tahlequah, OK Eastern Band of Cherokee Indians, Cherokee, NC Alabama-Quassarte Tribal Town, Wetumka, OK

95 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

OTHERS:

Cletus Brown Caroline Bickel Ted & Denna Biderman

James E. Brown Matthew H. Brown Michael E. Callaway

Clifford E. Carden Jr. Max R. Carrol John P. Collins, Jr.

Charles M. Corn Jack Huck Tommie J. Davis

Philip Dooley Mary L. Mayfield Elliot William G. Epperson

Jerry Estes Robert F. Evans Beth Cate Evans

J. Max Everhart Kelly W. Feehrer S. Wayne Feehrer

Karen P. Fisher Troy D. Gilson Karen Kinney Griggs

Rufus Guthrie David P. Hairrell Burton A. Hall

Jerry Hamby Ken & Elizabeth Higgins John M. Higgins

F. Harvey Howalt Christopher Stuart Jenkins William & Jaqueline Johnson

James M. Johnston III Ann Leathers Dyal Donald H. Ledford

George H. Lessing Amy Card Lillios Benjamin Longley

John Nelson or Lawrence F. McDaris, Jr. Edna M. Melton Lori Emerson Nelson

Ida R. Minor Jimmie R. Neeley Phil Newman

Thomas P. Newman John & Georgia O’Brien, Jr. Neal Officer

J.H. Patton, IV William A. Petit Cora N. Preston

Farrell Preston Forrest & Kathleen Preston Bryan Preston

L. C. Rhodes Jr. Mike Rice Robert Robbins

Lloyd D. Rogers Clyde Rogers, Jr. Benton A. Sellers

Bobby R. Shaver Mark Souther Clarke Stamper

Janis W. Strunk David L. Turpin Douglas P. Swayne

96 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

William M. Torbett John Thomas & Lynn Turpin Susan Lackey

Chris Zahnd Ray Payne David Elliot

Matthew Higgins Steve Malloy Nicholas Lillios

97 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

LITERATURE CITED AND REFERENCES

Birdsey, R.; Jenkins, J.; Johnston, M.; Huber-Sannwald, E.; 2007. North American Forests. In King, A.W.; Dilling, L.; Zimmerman, G.P.; Fairman, d.M.; Houghton, R.A.; Marland, G.; Rose, A.Z.; Wilbanks, T.J.; eds. The First State of the Carbon Cycle Report (SOCCR): The North American carbon budget and implications for the global carbon cycle, a report by the US Climate Change Science Program and the Subcommittee on Global Change Research, National Oceanic and Atmospheric Administration, Asheville, NC; National Climatic Data Center: 117- 126.

Birdsey , R.; Pregitzer, K.; Lucier, A. 2006. Forest carbon management in the United States: 1600-2100. Journal of Environmental Quality 35: 1461-1469.

Blouch, R. 1984. Northern Great Lakes states and Ontario forests. In: L. Halls, ed. White-tailed deer: ecology and management. Harrisburg, PA: Stackpole Books and Wildlife Management Institute: 870 pp.

Brodo, I.M., Sharnoff, S.D., and S. Sharnoff. 2001. Lichens of North America.Yale University Press. New Haven, CT.

Caldwell, R.S. 2005. Land Snails of the Southern Appalachian Workshop: Spring 2005. Cumberland Mountain Research Center, Lincoln Memorial University. Harrogate, TN.

Callahan, E. V., R. D. Drobney, ad R. L. Clawson. 1997. Selection of summer roosting sites by Indiana bats (MYOTIS SODALIS) in Missouri. Journal of Mammalogy 78:818-825.

Canadell, J.G., D.E. Pataki, R. Gifford, R.A. Houghton, Y. Luo, M.R. Raupach, P. Smith, W. Steffen. 2007. Saturation of the Terrestrial Carbon Sink. In Candadell JG, Pataki D, Pitelka L (eds) Terrestrial Ecosystems in a Changing World. The IGBP Series, Springer-Verlag, Berlin Heidelberg, pp. 59-78.

Cash, B. Selected habitat and species characteristics of the Tellico salamander (Plethodon aureolus) in the Cherokee National Forest, TN final report, Dec 2010. Maryville College.

Chapin, F.S.; Woodwell, G.M.; Randerson, J.T.; Rastetter, E.B.; Lovett, G.M.; Baldocchi, D.D.; Clark, D.A.; Harmon, M.E.; Schimel, D.S.; Valentini, R.; Wirth, C.; Aber, J.D.; Cole, J.J.; Goulden, M.L.; Harden, J.W.; Heimann, M.; Howarth, R.W.; Matson, P.A.; McGuire, A.D.; Melillo, J.M.; Mooney, H.A.; Neff, J.C.; Houghton, R.A.; Pace, M.L.; Ryan, M.G.; Running, S.W.; Sala, O.E.;

98 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Schlesinger, W.H.; Schulze, E.-D. 2006. Reconciling Carbon-cycle Concepts, Terminology, and Methods. Ecosystems 9: 1041-150.

Cheah, C.A.S. –J., M.S. McClure. 2002. Pseudoscymnus tsugae in Connecticutt forests: the first five years. In: B. Onken, R. Reardon, J. Lashomb, eds., Proceedings, Hemlock woolly adelgid in the Eastern United States symposium; 2002 February 5-7; East Brunswick, NJ. New Brunswick, NJ: Rutgers University; 150-165. Cherokee National Forest TES List. 2001. Cherokee National Forest, Cleveland, TN.

Choate, J.R., J.K. Jones, Jr, and C. Jones. 1994. Handbook of Mammals of the South-central States. Louisiana State University Press. Baton Rouge, LA.

Churchel, Melissa; Hanula, James; Berisford, C. Wayne; Vose, James; Dalusky, Mark. 2011. Impact of Imidacloprid for control of Hemlock Woolly Adelgid on nearby aquatic macroinvertebrate assemblages. South. J. Appl. For. 35(1)26-32. Available at the time of this report at http://www.srs.fs.usda.gov/pubs/ja/2011/ja_2011_churchel_001.pdf.

Cleavitt, N.L., A.K. Eschtruth, J.J. Battles, and T.J. Fahey. 2008. Bryophyte response to eastern hemlock decline caused by hemlock woolly adelgid infestation. Journal of Torrey Botanical Society 135(1): 12-25.

Costa, S., Felton, K., and Reardon, R. 2010. Aerial application of the insect-killing fungus Lecanicillium muscarium in a microfactory formulation for hemlock woolly adelgid suppression.

Costa, Scott. 2008. Current taxonomic status of the insect-killing fungus, Lecanicillium muscarium (Verticillium lecanii), contained in Mycotal and assessment of its potential to negatively impact plants. University of Vermont, White paper.

Cowles, Richard. 2010. The facts about systemic insecticides and their impact on the environment and bee pollinators. Clippings, Spring/Summer 2010.

Cowles, Richard. Undated. Soil application of imidacloprid to control hemlock woolly adelgid: best management practices. Available at the time of this report at http://na.fs.fed.us/fhp/hwa/control/soil_app_control_hwa.pdf.

Crum, H.A. and L.E. Anderson. 1981. Mosses of Eastern North America, Vols I & II. Columbia University Press. New York, NY.

Dalusky, M. 2011. Using imidacloprid & dinotefuran for control of hemlock woolly adelgid. Available at the time of this report at http://www.savegeorgiashemlocks.org/Downloads/Volunteer_Facilitators_page/D alusky_-_Using_Imidacloprid_&_Dinotefuran_2011.pdf

99 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

DeGraaf, R. and d. Rudis. 1986. New England wildlife: habitat, natural history, and distribution. Gen. Tech. Rep. NE-108. Broomall, PA: USDA Forest Service, Northeastern Forest Experiment Station. 491 pp.

DeGraaf, R., M. Yamasaki, W. Leak, and J. Lanier. 1992. New England wildlife: management of forested habitats. Gen. Tech. Rep. NE-144. Radnor, PA: USDA Forest Service, Northeastern Forest Experiment Station. 271 pp.

Edminster, F. 1947. The ruffed grouse. New York, NY: Macmillan Co. 385 pp.

Elowe, K. 1984. Home range, movements, and habitat preferences of black bears (Ursus americanus) in western Massachusetts. Amherst, MA: University of Massachusetts. MS thesis. 112 pp.

EPA Fact Sheet, 2004.

EPA. 2004. Pesticide Fact Sheet – Dinotefuran. Available at the time of this report at http://www.epa.gov/opprd001/factsheets/dunotefuran.pdf

Evans, R.A. 2002. An ecosystem unraveling? In: B. Onken, R. Reardon, J. Lashomb, eds., Proceedings, Hemlock woolly adelgid in the Eastern United States symposium; 2002 February 5-7; East Brunswick, NJ. New Brunswick, NJ: Rutgers University; pp. 23-33.

Evans, R., E. Johnson, J. Shreiner, A. Ambler, J. Battles, N. Cleavitt, T. Fahey, J. Sciascia, and E. Pehek. 1996. Potential impacts of hemlock woolly adelgid (Adelges tsugae) on eastern hemlock (Tsuga canadensis). In: S. Salom, T. Tigner, R. Reardon, eds. Proceedings of the first hemlock woolly adelgid review. Morgantown, WV: USDA Forest Service, Forest Health Technology Enterprise Team: 42-57.

Fossen, M., 2006. Environmental Fate of Imidacloprid, California Department of Pesticide Regulation. http://www.cdpr.ca.gov/docs/empm/pubs/fatememo/Imidclprdfate2.pdf>

Field Museum Collections Database System. 2002. Chicago, Illinois. Available at http://www.fieldmuseum.org/exhibits/online-exhib.htm.

Gervais, J. A.; Luukinen, B.; Buhl, K.; Stone, D. 2010. Imidacloprid Technical Fact Sheet; National Pesticide Information Center, Oregon State University Extension Services. Available at t he time of this report at http://npic.orst.edu/factsheets/imidacloprid.pdf.

100 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Griffin, S. 2010. Hemlock Wooly Adelgid: Control Options. Georgia Forestry Commission. Available at the time of this report at http://www.gfc.state.ga.us/forestmanagement/documents/HWAControlOptionsrev 062010.pdf.

Gouli, S., V. Gouli, H. B. Teillon, A. Kassa, M. Skinner, C. Frank, B. Parker, 2008. Controlling hemlock woolly adelgid with new forms of Mycoinsecticides New formulations of Beauveria bassiana and Lecanicillium muscarium were used for management of hemlock woolly adelgid (HWA), Adelges tsugae Annand. The formulations were prepared with whey, a waste product from the production of cheese, and vegetative oil, pp. 179-187.

Hamel, Paul B. 1992. The Land Manager’s Guide to Birds of the South. The Nature Conservancy, Chapel Hill, NC.

Heath, L.S. and J.E. Smith, 2004: Criterion 5, indicator 26: total forest ecosystem biomass and carbon pool, and if appropriate, by forest type, age class and successional change. In: Data Report: A Supplement to the National Report on Sustainable Forests— 2003 [Darr, D.R. (coord.)]. FS-766A, U.S. Department of Agriculture, Washington, DC, 14 pp. Available at http://www.fs.fed.us/research/sustain/contents.htm

Hicks, M.L. 1992. Guide To The Liverworts Of North Carolina. Duke University Press, Durham, NC.

Hicks, M.L. and J.L. Amoroso. 1996. Broyphyte status survey: MEGACEROS AENIGMATICUS Schuster. North Carolina Natural Heritage Program and Endangered Species Field Office, US Fish and Wildlife Service, Asheville, North Carolina. Revised 1997. 10 pp. and addenda

Howe, R. and M. Mossman. 1995. The significance of hemlock for breeding birds in the western Great Lakes region. In: Conference proceedings on hemlock ecology and management. September 27-28, 1995. Iron Mountain, MI: 125-139.

Hubricht, Leslie. 1985. The Distributions of the Native Land Mollusks of the Eastern United States. Field Museum of Natural History.

Huot, J, F. Potvin, and M. Belanger. 1984. Southeastern Canada. In: L. Halls, ed. White- tailed deer: ecology and management. Harrisburg, PA: Stackpole Books and Wildlife Management Institute: 293-304.

Jordan, J. and W. Sharp. 1967. Seeding and planting hemlock for ruffed grouse cover. Res. Pap. NE-83. Upper Darby, PA: USDA Forest Service. 17 pp.

101 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

Kerle, EA; Jenkins, JJ; Vogue, PA. 2007. Understanding pesticide persistence and mobility for groundwater and surface water protection. Oregon State University Extension Service.

Libby, G.W. Pers. Communication 2004. Eco-tech Inc. Frankfort, KY. Draft report. A survey for the federally endangered Indiana bat (Myotis sodalis) on the Hiwassee, Nolichucky, Tellico, and Watauga Ranger Districts of the Cherokee National Forest, Tennessee. Eco-Tech, Inc.

Linzey, D.W. 1995. Mammals of Great Smoky Mountains National Park. The McDonald and Woodward Publishing Company. Blacksburg, Virginia. 140 pp.

Martof, B.S., W.M. Palmer, J.R. Bailey, and J.R. Harrison III. 1980. Amphibians and Reptiles of the Carolinas and Virginia. The Univ. of North Carolina Press. Chapel Hill, NC. 264 pp.

Mattfeld, G. 1984. Eastern hardwood and spruce/fir forests. In: L. Halls, ed. White-tailed deer: ecology and management. Harrisburg, PA: Stackpole Books and Wildlife Management Institute: 305-330.

Mitchell, L.J. 2001. Sensitive Species, Terrestrial Animals, Cherokee National Forest. Cherokee National Forest, Cleveland, TN.

National Pesticide Information Center. 2011. Imadacloprid Technical Fact Sheet. National Pesticide Information Center.

NatureServe 2004. Explorer: An online encyclopedia of life [web application]. 2004. Version 1.6 . Arlington, Virginia, USA: NatureServe. Available: http://www.natureserve.org/explorer.

NatureServe. 2005. NatureServe Explorer: An online encyclopedia of life [web application]. Version 4.0. NatureServe, Arlington, Virginia, USA. Available: http://www.natureserve.org/explorer. (Accessed: June, 2005).

NatureServe. 2011. NatureServe Explorer: An online encyclopedia of life [web application]. Version 4.0. NatureServe, Arlington, Virginia, USA. Available: http://www.natureserve.org/explorer. (Accessed: July, 2011).

Nicholson, C.P. 1997. Atlas of the Breeding Birds of Tennessee. University of Tennessee Press. Knoxville, TN. 426 pp.

North Carolina Heritage & USFWS. 1997. “Bryophyte Status Survey Abstracts.” Compiled by J.L. Amoroso.

102 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

North Carolina Natural Heritage Program. 1999. Natural Heritage Program List of the Rare Plant Species of North Carolina. North Carolina Department of Environment and Natural Resources. Raleigh, NC.

Orwig , D.A. and D.R. Foster. 1998. Forest Response to the Introduced Hemlock Wooly Adelgid in Southern New England, USA. Jornal of the Torrey Botanical Society, Vol 125, Issue 1: 60-73.

Pacala, R.; Birdsey, R.A.; Bridgham, S.D.; Conant, R.T.; Davis, K.; Hales, B.; Houghton, R.A.; Jenkins, J.C.; Johnston, M.; Marland, G.; Paustian, K. 2007. The North American carbon budget past and present. In King, A.W.; Dilling, L.; Zimmerman, G.P.; Fairman, d.M.; Houghton, R.A.; Marland, G.; Rose, A.Z.; Wilbanks, T.J.; eds. The First State of the Carbon Cycle Report (SOCCR): The North American carbon budget and implications for the global carbon cycle, a report by the US Climate Change Science Program and the Subcommittee on Global Change Research, National Oceanic and Atmospheric Administration, Asheville, NC; national Climatic Data Center: 117-126.

Petranka, J.W. 1998. Salamanders of the United States and Canada. Smithsonian Institution Press. Washington, D.C. 587 pp.

Pfeiffer, Michael. 2010. Groundwater Ubiquity Score. Pesticide Training Resources. Available at the time of this report at www.ptrpest.com.

Powell, S., E. York, J. Scanlon, and T. Fuller. 1997a. Fisher maternal den sites in central New England. In: G. Prouix, H. Bryant, and P. Woodard, eds. Martes: Taxonomy, ecology, techniques and management. Edmonton, Alberta, Canada: Provincial Museum of Alberta: 265-278.

Powell, S., E. York, and T. Fuller. 1997b. Seasonal food habits of fishers in central New England. In: G. Prouix, H. Bryant, and P. Woodard, eds. Martes: Taxonomy, ecology, techniques and management. Edmonton, Alberta, Canada: Provincial Museum of Alberta: 279-305.

Pregitzer, K.S. and Euskirchen, E.S. 2004. Carbon cycling and storage in world forests: biome patterns related to forest age. Global Change Biology 10: 2052-2077.

Reay, R.S.; Blodgett, D.W.; Burns, B.S.; Weber, S.J.; Frey, T. 1990. Management guide for deer wintering areas in Vermont. Vermont Fish and Wildlife Department. 35 pp.

Reid, W. R., 2003. Isolation and Characterization of Entomopathogenic Fungi Assaciated with the Hemlock Wooly Adelgid, Adelges tsugae Annand. M. S. thesis, Univ. of Vermont, p. 99pp Rhea, Rusty 2004. USDA Forest Service. Forest Health. Personal communication.

103 2011 Supppression of Hemlock Woolly Adelgid Infestations Environmental Assessment Cherokee Nationl Forest

SAMAB. 1996. The Southern Appalachian Assessment Terrestrial Technical Report. Report 5 of 5. Atlanta: U.S. Department of Agriculture, Forest Service, Southern region.

Tennessee Natural Heritage. 2004. Rare Plant List. Division of Natural Heritage, Tennessee Department of Environment and Conservation. Nashville, TN.

U.S. Environmental Protection Agency, Inventory Of U.S. Greenhouse Gas Emissions And Sinks: 1990 – 2007, March 2008 Public Review Draft.

U.S. Global Change Research Program, National Assessment Synthesis Team. 2001. Climate change impacts on the United States – the potential consequences of climate variability and change. United Kingdom: Cambridge University Press, 612 pp.

USDA Forest Service. 2004a. Final Environmental Impact Statement for the Revised Land and Resource Management Plan, Cherokee National Forest. Management Bulletin R8-MB 114B. Atlanta GA.

USDA Forest Service. 2004b. Appendices for the Final Environmental Impact Statement for the Revised Land and Resource Management Plan, Cherokee National Forest. Management Bulletin R8-MB 114F. Atlanta GA.

USDA Forest Service. 2005. Environmental Assessment: Conservation of Native Hemlock by Suppression of hemlock Woolly Adelgid Infestations. Cherokee National Forest.

USDA-Forest Service, 2007. Technical Guide to Managing Ground Water Resources. FS-881.

USDA Forest Service 2009. 2008 Monitoring and Evaluation Report for the Cherokee National Forest. Cherokee National Forest, Cleveland, TN.

USDA Forest Service 2010a. Aquatic database for the Cherokee National Forest. Cherokee National Forest, Cleveland, TN.

USDA Forest Service 2010b. Environmental Assessment for 2010 Suppression of Hemlock Woolly Adelgid Infestations. Nantahala/Pisgah National Forests, Asheville, NC 28801.

USDI Fish and Wildlife Service. 2011. FWS #2011-CPA-0802. Biological Evaluation for the Proposal for Expanding Treatment Options for the Suppression of Hemlock Woolly Adelgid (HWA) Infestations on the Cherokee National Forest.

104 Environmental Assessment 2011 Suppression of Hemlock Woolly Adelgid Infestations Cherokee National Forest

Valent USG Corporation. 2009. Safari TN SLN. Available at the time of this report at http://www.valent.com/Data/Labels/TN- 080013%20Safari%20Control%20of%20HWA%20on%20Hemlocks.pdf

Valent USA Corporation. 2011. Material safety data sheet. Safari 20 SG Insecticide. Valent USA Corporation, P.O. Box 8025, Walnut Creek, Ca 94596.

Weakley, A.S. 2011. Flora of the Southern and Mid-Atlantic States. Working Draft of May 15, 2011. University of North Carolina Herbarium and North Carolina Botanical Garden, Chapel Hill, NC.

Wofford, B.E. 1989. Guide to the Vascular Plants of the Blue Ridge. University of Georgia Press. Athens, GA. Yorks, Thad E., Donald J. Leopold, and Dudley J. Raynal. (1999). Vascular Plant Propagule Banks of Six Eastern Hemlock Stands and Potential Response to the Hemlock Woolly Adelgid in the Catskill Mountains of New York. From Proceedings: Symposium on sustainable Management of hemlock ecosystems in Eastern North America. June 22-24, 1999 Durham, New Hampshire.

105