1 Biological Species Report for the Bicknell's Thrush (Catharus

Biological Species Report for the Bicknell’s Thrush (Catharus bicknelli) Version 1.4a

Bicknell’s thrush (Photo credit: Alan Schmierer)

August 2017 U.S. Fish and Wildlife Service Northeast Region Hadley, MA

This document was prepared by Anthony Tur (USFWS-New England Field Office (FO)/Northeast Regional Office (RO)), Krishna Gifford (USFWS-Northeast RO), and Beth Forbus (USFWS-Headquarters (HQ)).

We greatly appreciate the assistance of Dr. Randy Dettmers (USFWS-Northeast RO), Dr. John Lloyd (Vermont Center for Ecostudies), Dr. Jason Hill (Vermont Center for Ecostudies), Chris Rimmer (Vermont Center for Ecostudies), and Nancy Green (USFWS-HQ) who provided helpful information or review of sections of the working draft document.

We would like to thank the following USFWS staff for their review and comments on version 1.0 of the document: Dr. Randy Dettmers (Northeast RO), Anna Harris (Maine FO), Dr. Cherry Keller (Chesapeake Bay FO), Dr. Mark McCollough (Maine FO), Martin Miller (Northeast RO), Sarah Nystrom (Virginia FO), Nicole Ranalli (Pennsylvania FO), Nicole Rankin (Southeast RO), and David Simmons (New England FO).

We also would like to thank the following State or Territory natural resources staff, as well as our peer reviewers, for their assistance in reviewing and providing constructive comments on version 1.1 of the document: John W. Ozard (New York Department of Environmental Conservation, Bureau of Wildlife-Avian Diversity Unit), John Kanter and Dr. Pam Hunt (New Hampshire Fish and Game, Nongame and Endangered Wildlife), Ramon Luis Rivera (Puerto Rico Department of Natural and Environmental Resources), Dr. Toni Morelli (U.S. Geological Survey-Northeast Climate Science Center), Alyssa Rosemartin (USA National Phenology Network, University of Arizona), Dr. Yves Aubry and Dr. Junior Trembley (Environment and Climate Change Canada-Science and Technology Branch), and Becky Whittam (Environment and Climate Change Canada-Canadian Wildlife Service).

Suggested reference: U.S. Fish and Wildlife Service (USFWS). 2017. Biological Species Report for the Bicknell’s thrush (Catharus bicknelli). Version 1.4a. August 2017. Hadley, MA.

Table of Contents EXECUTIVE SUMMARY ...... 6 LIFE HISTORY AND BIOLOGY ...... 10 Species Description and Taxonomy ...... 10 General Life Cycle ...... 11 Feeding ...... 12 Breeding Behavior and Habitat ...... 12 Fall Migration Behavior and Habitat ...... 14 Wintering Behavior and Habitat ...... 14 Spring Migration Behavior and Habitat ...... 16 SURVEY DATA AND POPULATION TRENDS ...... 17 Survey Efforts ...... 17 Population Estimates ...... 19 Abundance Trends ...... 20 Summary of Survey Data and Trends ...... 22 FACTORS INFLUENCING VIABILITY ...... 23 Existing Factors Influencing the Bicknell’s Thrush ...... 23 Breeding Range ...... 23 Summary of Existing Factors Influencing the Bicknell’s Thrush in its Breeding Range ..... 34 Wintering Range ...... 34 Migration Range ...... 36 Summary of Existing Factors Influencing the Bicknell’s Thrush in its Wintering and Migration Ranges ...... 37 Behavioral Disruptions and Mortality ...... 37 Summary of Current Factors Influencing the Bicknell’s Thrush...... 40 Future Factors Influencing the Bicknell’s Thrush ...... 42 Anticipated Changing Climate Parameters in the 21st Century ...... 42 Projected Future Effects of Climate Change to Bicknell’s Thrush Breeding Habitat ...... 43 Summary of Projected Future Effects of Climate Change to Bicknell’s Thrush Breeding Habitat ...... 49 Increased Interspecific Competition Due to Climate Change...... 50 Climate Induced Increase in Predation by Red Squirrels ...... 51 Anticipated Effects of Climate Change to Wintering Habitat ...... 52 Summary of Future Factors Influencing the Bicknell’s Thrush ...... 53 3

FUTURE SCENARIO CONSIDERATIONS ...... 54 Non-Climate Stressor Scenario ...... 56 Optimistic Scenario ...... 56 Low Warming Scenario ...... 57 High Warming Scenario ...... 58 Summary of Future Scenario Considerations ...... 59 REFERENCES CITED ...... 61

LIST OF TABLE AND FIGURES

Figure 1. Photo of Bicknell’s thrush...... 10

Figure 2. Range map for the Bicknell’s thrush...... 11

Figure 3. Example of Bicknell’s thrush habitat from Mt. Mansfield, VT...... 13

Figure 4. Example of potential Bicknell’s thrush habitat in the Dominican...... 15

Figure 5. Some locations with previously observed or declining numbers of Bicknell’s thrush ...... 20 Figure 6. Estimated acres of total timber versus acres of timber stand improvement treatment in Maine from 2000 to 2015...... 28

Figure 7. Estimated annual removal or harvest of growing stock trees in New Hampshire from 2006 to 2015...... 29

Figure 8. Total number of cords harvested for all tree species versus spruce-fir in Vermont from 1997 to 2014...... 30

Figure 9. Millions of green tons of softwood harvested for pulpwood and chips in New York from 1999 to 2014...... 31

Figure 10. Estimated total acres harvested vs harvested using commercial thinning in five Canadian provinces from 1990 to 2014...... 32

Figure 11. Observed and projected trends in global CO2 emissions under four RCP scenarios (figure from Sanford et al. 2014, p 165)...... 43

Table 1. Specific high-elevation landbird surveys in the northeastern United States and eastern Canada since 1991 (table adapted from Hart and Lambert 2010, p. 7)...... 18

Table 2. Extent of tree cover on four Caribbean islands occupied by wintering Bicknell’s thrushes in 2000 and the extent of decline for the period 2001 to 2014 (Hansen et al. 2017, entire)...... 35

Table 3. Projected change in global mean surface temperature for the mid- and late 21st century, relative to the 1986 to 2005 period (modified from IPCC 2014, p. 60)...... 42

Table 4. Summary cross walk of regional projections in average annual temperature for the northeastern United States...... 44

Table 5. Summary of Future Projection Scenarios...... 55

EXECUTIVE SUMMARY The Bicknell’s thrush (Catharus bicknelli), the smallest of North American Catharus thrushes, is a migratory songbird. The species breeds during the summer in areas of the northeastern United States and southeastern Canada, migrates via coastal routes in the spring and overland in the fall, and winters in the Greater Antilles, with the majority of birds on the island of Hispaniola, in Haiti and the Dominican Republic. The species can also be found on the islands of Cuba, Jamaica, and Puerto Rico. The Bicknell’s thrush feeds predominantly on insects on its breeding grounds, but during migration and on its wintering grounds, the species may shift its diet to include several varieties of small fruits.

Breeding habitat for the Bicknell’s thrush consists of dense tangles of both living and dead “stunted” trees that are predominately balsam fir, with lesser amounts of red spruce and white birch. Except in the case of the Canadian provinces, where the species has been found at lower elevations along the coast and in high elevation regenerating industrial forests, the species breeds mostly in stunted high elevation or montane spruce-fir forests located close to, but below, timberline (i.e., at elevations above 700 m (2,300 ft)). Although the Bicknell’s thrush exhibits some flexibility in its choice of elevation for breeding, the species demonstrates a strong preference for a specific vegetation structure.

In the Dominican Republic, where the majority of the best available information on wintering is derived, the Bicknell’s thrush can be found from sea level to 2,200 m (7,200 ft). Most are found in mesic to wet broadleaf montane forests higher than 1,000 m (3,300 ft) elevation (i.e., cloud forest), but some can be found in dry pine-dominated forests at lower elevations. Similar to the habitat structure selected during the breeding season, the species prefers wintering in dense thicket vegetation. During migration, the Bicknell’s thrush appears to be a habitat generalist and can be found in dense woodlots of variable tree species composition, along well-vegetated beaches, orchards and gardens.

Localized and regional surveys have been conducted across the species’ range with widely different levels of geographic, temporal, and methodological consistency, due to breeding and wintering habitat accessibility (e.g., remote terrain of breeding area and lack of access to Cuba). These differences present challenges to deriving precise population distributions, estimates, and trends. However, the best available data indicate that approximately 66 and 33 percent of the breeding population occurs in the United States and Canada, respectively (note, due to rounding, the estimates do not equal 100). The distribution of the United States breeding population is approximately 37 percent in New Hampshire, 29 percent in New York, 26 percent in Maine, and 7 percent in Vermont; there is no longer a breeding population in Massachusetts. There is sparse information readily available about the species’ wintering population. The best available global population estimate of Bicknell’s thrush is approximately 97,358 to 139,477 individuals. Although the methods vary, the best available data based on breeding population estimates from point counts, habitat modeling, other abundance data, and one dispersal study suggests the species’ abundance has been undergoing a long-term declining trend.

Due to the lack of specific data regarding survival rates by life stage or fecundity rates, we evaluated existing stressor related data and qualitatively assessed the individual and cumulative 6

effects of those stressors on individual Bicknell’s thrush, aggregates of Bicknell’s thrush in the breeding or wintering grounds, and at the species level. We reviewed the best available information from the published literature, academic reports, Bicknell’s thrush experts, and independent peer reviewers with expertise with the species or relevant stressors. From this assessment, we conclude that the habitat loss in the wintering range has most likely been a significant driver of the species’ current viability, with the additive effects associated with low productivity in some years due to nest predation from red squirrels also contributing to annual variation in the abundance. For example, loss of wintering habitat in the Caribbean due to forest conversion has been extensive and is ongoing, with no indication that it is likely to be abated. Contributing factors to the Bicknell’s thrush’s viability have included some forestry practices such as precommercial thinning and clearcutting in the Canadian portion of the species’ breeding range, which may result in the loss and fragmentation of important Bicknell’s thrush breeding habitat in the short term. However, the regeneration of young dense stands of conifers that follows can provide breeding habitat for the species for approximately 5 to 12 years post- clearcutting. The development of ski areas, wind turbines, and telecommunication facilities and their associated infrastructure (i.e., roads and transmission lines), primarily in the United States, has also resulted in the loss and fragmentation of habitat for the Bicknell’s thrush, but these activities have affected a relatively small proportion of the available breeding habitat and the species does show some ability to adapt and persist in the vicinity of ski slopes and wind turbines. We have no information regarding factors that may be influencing or have influenced the species during migration.

In assessing the Bicknell’s thrush’s viability, we considered what factors may be reliably predicted to influence the species into the future and then focused our assessment of future conditions solely on those stressors likely affecting Bicknell’s thrush at the species level. For the breeding range, this includes changes in habitat suitability and red squirrel predation. On the wintering grounds, this includes direct habitat loss. The best available information suggests that as a result of climate change, the spruce-fir habitat that support breeding Bicknell’s thrush may be substantially reduced, with the potential to be nearly eliminated, from the species’ current range in the northeastern United States and may decline in Canada by the end of this century, depending on the amount of green-house gases emitted to the atmosphere, habitat type (i.e., low vs. high elevation) and forest harvest management strategies. The effect of climate change may also result in an increase in competition between Bicknell’s and Swainson’s thrushes, at the expense of the Bicknell’s thrush, and an increase in predation from red squirrels. On the wintering grounds, the consequences of climate change will likely include a drying of the Caribbean region and an associated decline in the wet montane habitats where most Bicknell’s thrushes are found. It is also likely that socioeconomic pressures, especially in the Dominican Republic, may result in further losses of the species’ preferred habitat, as forests are converted to other land uses. We acknowledge that there is uncertainty with how quickly and to what extent suitable Bicknell’s thrush habitat will change and further uncertainty with how the Bicknell’s thrush may respond to any shifting or outright loss of suitable habitat. But, the best available data suggest that the net effect of climate change on the Bicknell’s thrush is expected to be negative. For a full discussion of what factors we assessed but deemed not to affect the Bicknell’s thrush at the species level (e.g., acid deposition, disturbance by recreationists), see the Existing Factors Influencing the Bicknell’s Thrush section of the report.

The viability of Bicknell’s thrush depends on maintaining breeding and wintering habitat that is capable of supporting multiple resilient populations over time. Given the uncertainty over the expected projections from the multiple best available climate models, as well as the potential climate policy mitigations that may occur in the future, we attempted to forecast the range of what habitat availability for Bicknell’s thrush could potentially look like over the next 50 to 83 years (i.e., to the end of the century). We’ve chosen four potential scenarios, each one using a different climate forecast, and made assumptions about effects to the species’ habitat based on information published in the scientific literature and from Bicknell’s thrush, avian ecology, phenology, and climate experts. These scenarios do not include all possible futures, but rather include four potential scenarios that represent examples from the continuous spectrum of possible futures. In addition to a Non-Climate Stressor scenario that projects current stressor levels in isolation from climate change, we assessed three additional scenarios that align with the IPCC models: the annual regional surface temperature average in the Northeast will increase by less than 2 °C (3.6 °F) above pre-industrial levels under the RCP 2.6 scenario (Optimistic), by 3 °C to 5 °C (5.4 °F to 9 °F) under the B1 scenario (Low Warming), and 5.3 °C to 6 °C (9.5 °F to 10.8 °F) under the A1F1 scenario (High Warming).

In all but the Non-Climate Stressor scenario, some amount of Bicknell’s thrush habitat loss in the breeding range is expected (see table 5). The amount and distribution of breeding habitat that may remain through the end of the century varies depending on climate change projection, location (United State vs. Canada), and potential for forest harvest strategies. In Canada, potential habitat loss may be partially mitigated by management strategies or elevation in the Optimistic and Low Warming Scenarios, but habitat loss is expected under the High Warming Scenario. In the United States, the eventual shift in tree species composition is projected to result in a decrease in spruce-fir habitat under the Optimistic Scenario; there is the potential for most habitat to be eliminated, although small isolated patches may persist in the highest elevations of New Hampshire and Maine, under the Low Warming Scenario; and the possibility that a complete elimination of habitat could occur, with a possibility that small isolated patches may persist in the highest elevations of New Hampshire and Maine, under the High Warming Scenario. The results of these projections are inclusive of potential conservation measures intended to address habitat-related stressors. Any remaining breeding habitat will likely have limited suitability due to the presence of competing Swainson’s thrushes and nest-predating red squirrels.

In all scenarios, including the Non-Climate Stressor scenario, loss and degradation of the species’ habitat across the wintering range is expected to continue. The amount of wintering habitat that likely remains through the end of the century varies from approximately 18 percent (Optimistic) to potentially zero (Low and High Warming). The results of these projections are inclusive of potential conservation measures intended to address habitat-related stressors.

declining pattern is consistent among all scenarios, and it is reasonable to assume, in the absence of evidence to the contrary, that the species’ response may not keep pace with the projected changes in habitat within the predicted timeframes.

LIFE HISTORY AND BIOLOGY The species’ life history and biology information below is summarized from numerous sources, as specifically cited. For more detailed information, please refer to those sources, particularly the Bicknell’s thrush account from the Birds of North America (Townsend et al. 2015, entire; https://birdsna.org/Species-Account/bna/species/bicthr/introduction; last accessed April 3, 2017).

Species Description and Taxonomy

The Bicknell’s thrush (Catharus bicknelli) is the smallest of North American Catharus thrushes in the family Turdidae, which includes all birds related to the robins (Townsend et al. 2015, unnumbered). Dorsal (back) coloration ranges from olive-brown to brown, while the belly is generally white with a light buffy wash and darker spots (see figure 1).

Figure 1. Photo of Bicknell’s thrush (Photo credit: Alan Schmierer)

Due to similar morphometric characteristics, positively identifying a Bicknell’s thrush from other North American Catharus requires close scrutiny. Although trained biologists can tell similar species apart, the gray-cheeked thrush (C. minimus) (Wallace 1939, p. 217; Townsend et al. 2015, unnumbered) presents the greatest challenge. The two species so closely resemble each other in appearance that the Bicknell’s thrush was considered a subspecies of the gray-cheeked thrush until 1993. However, careful evaluation of morphology, range, song, behavior, habitat, and genetics revealed significant differences that led Ouellet (1993, p. 568) to recommend that the Bicknell’s thrush be elevated to a full species. Analysis of mitochondrial DNA (mtDNA) further supported recognition of the two species as distinct entities (Outlaw et al. 2003, p. 305). The American Ornithologist Union (1995, p. 824), the taxonomic authority for bird species found in North and Central America, recognizes the Bicknell’s thrush as a full species. There is no information to suggest that there is scientific disagreement about the species’ taxonomy. Therefore, the USFWS considers the Bicknell’s thrush to be a valid taxonomic species.

General Life Cycle

The Bicknell’s thrush breeds during the summer (May to August) in areas of the northeastern United States and southeastern Canada. Individuals start migrating in late September or early October by following a coastal route south to Virginia where most birds depart, flying across the ocean to the Bahamas and Cuba, before finally arriving in the Greater Antilles (i.e., the grouping of larger islands in the Caribbean, including but not limited to the Bicknell’s thrush’s wintering areas in Cuba, Haiti, Dominican Republic, Jamaica, and Puerto Rico) sometime during mid- October through early November. Wintering occurs in the Greater Antilles (October to March), and migration occurs back overland through the Southeast United States in spring (April to May) to reach its breeding grounds. See figure 2.

Figure 2. General range map for the Bicknell’s thrush from Townsend et al. 2015, unnumbered.

Bicknell’s thrush live to approximately 11 years (Townsend et al. 2015, unnumbered). Adult Bicknell’s thrushes begin breeding in their second season. Following their first breeding attempt, adult birds may relocate to a new territory; however, by their third season adult birds typically return to the same breeding area throughout the remainder of their lives (Townsend et al. 2015, unnumbered).

Survival rates for second year birds are estimated to be 0.34 ± 0.06 in the Gaspe region of Quebec and 0.52 ± 0.03 in Vermont (Townsend et al. 2015, unnumbered). Following their second year of life, the survival rate of adults increases to 0.48 ± 0.07 for birds in the Gaspe region and 0.66 ± 0.02 for Vermont birds (Townsend et al. 2015, unnumbered). Significant variation in annual survival rates from year to year has been documented and may be related to annual fluctuations in weather conditions, stochastic events (e.g., mortality due to extreme weather events during migration), cycles in predator abundance, and variations in forage abundance (Townsend et al. 2015, unnumbered).

Feeding

On its breeding grounds, the Bicknell’s thrush feeds predominantly on insects, but during migration and on its wintering grounds, the species may shift its diet to include several varieties of small fruits (Beal 1915 in Wallace 1939, p. 295; Rimmer et al. 2001, pp. 9–10; Townsend et al. 2010, p. 517). Bicknell’s thrush forages for food among trees, feeding among the branches or hawking (pursuit in flight); however, most foraging activity takes place on or near the ground through litter pecking or gleaning (Wallace 1939, p. 295; Sabo 1980, p. 251; Rimmer et al. 2001, pp. 9–10).

Breeding Behavior and Habitat

Males begin arriving on the breeding grounds in May and commence singing to establish territories and attract a mate; breeding occurs in June (Wallace 1939, p. 311; Rimmer et al. 2001, p. 12). Both males and females will mate with multiple partners, resulting in clutches from multiple paternities within the same nest (Goetz et al. 2003, p. 1044–1053). Nest building and egg incubation are the sole responsibility of the female, but both males and females feed the chicks (Wallace 1939, pp. 323–325; Rimmer et al. 2001, pp. 15–17). The average clutch size is 3 to 4 eggs (Townsend et al. 2015, unnumbered). Although some birds may return to their natal nesting areas, there is a high incidence of natal dispersal, which complicates the development of survival estimates for juvenile birds (Townsend et al. 2015, unnumbered; Studds et al. 2012, entire). Fledging occurs at 9 to 14 days, at which time the young either stay in the vicinity of the nest or disperse to other areas, including down slope to hardwood-dominated habitats (Rimmer et al. 2001, p. 18). The sex ratio of Bicknell’s thrush can vary from 1 male (M):1.5 females (F) to 2 M: 1 F (Rimmer et al. 2001, p. 13; Townsend et al. 2009b, pp. 92–93).

The Bicknell’s thrush’s breeding range extends from the northern Saint Lawrence area of Quebec and the Canadian Maritime Provinces south through New England and New York to that State’s Catskill Mountains (Wallace 1939, pp. 258–259; Ouellet 1993, pp. 563–564; Rimmer et al. 2001, p. 1). Breeding habitat for the Bicknell’s thrush consists of dense tangles of both living and dead “stunted” trees that are predominately balsam fir (Abies balsamea) with lesser amounts of red spruce (Picea rubens) and white birch (Betula papyrifera var. cordifolia) (Wallace 1939, p. 285; Ouellet 1993, p. 561; Rimmer et al. 2001, p. 7; McKinnon et al. 2014, p. 2). See figure 3. Depending upon location, white spruce (P. glauca) or an occasional black spruce (P. mariana) can also provide nest sites, as can pin cherry (Prunus pennsylvanica), mountain ash

(Sorbus americanus), shadbush (Amelanchier spp.), and other deciduous species (Wallace 1939, pp. 285–286; Sabo 1980, p. 242; Ouellet 1993, p. 561; Rimmer et al. 2001, p. 7).

Except in the case of the Canadian provinces where the species has been found at lower elevations along the coast and in high elevation regenerating industrial forests, the species breeds mostly in stunted high elevation or montane spruce-fir forests located close to, but below, timberline (i.e., at elevations above 700 m (2,300 ft)) (Wallace 1939, pp. 248 and 286; Ouellet 1993, pp. 560, 561; Atwood et al. 1996, p. 652; Nixon et al. 2001, p. 38; Rimmer et al. 2001, p. 7; Glennon and Seewagen 2016, p. 134; Aubry et al. 2016, p. 304). The coastal areas of New Brunswick and Nova Scotia have cooler temperatures and higher precipitation levels that maintain dense spruce-fir stand similar to those found at higher elevations (Whittam in litt. 2017). In the industrial forests the Bicknell’s thrush still generally selects for the highest elevations available, which are lower in elevation than those areas available in the non-industrial forests in highland habitats of the northeastern U.S., but are the highest available in the Maritime provinces (Whittam in litt. 2017).

Figure 3. Example of potential Bicknell’s thrush habitat on Mt. Mansfield, VT (Photo credit: K. Gifford USFWS).

The spruce-fir forests that this species prefers for breeding is typical of chronically disturbed areas associated with altered growing conditions resulting from natural processes, as well as some human activities (e.g., creation of ski trails and some forest harvest practices). Natural disturbances include ‘terrific’ winds, which can exceed 45 meters per second (mps) (100 miles per hour (mph)) at some elevations, fire, insect outbreaks, and heavy snow and rime ice accumulation that occurs when supercooled water droplets undergo rapid freezing upon contact with a cold surface (Wallace 1939, p. 282; Rimmer et al. 2001, p 7). As a result of these conditions, trees are stunted. For example, the mean canopy height in areas where the Bicknell’s

thrush is found in the White Mountains of New Hampshire is approximately 4.8 m (15.7 ft) (Sabo 1980, p. 250).

Although the Bicknell’s thrush exhibits some flexibility in the elevation of its breeding habitats, the species demonstrates a strong preference for a specific vegetation structure. Breeding habitats, whether in montane habitats or in lower elevation areas, are characterized by dense vegetation, usually containing between two and four shrub or tree stems per square meter (m2) (Nixon et al. 2001, p. 39). Stems also tend to be relatively small in diameter with most falling below 2.5 centimeter (cm) (1 inch (in)) at diameter breast height (dbh), but ranging up to approximately 20 cm (7.87 in) at dbh (Nixon et al. 2001, p. 39). In industrial forests, these density and size conditions usually occur between 5 and 12 years after harvest or replanting (Nixon et al. 2001, p. 39) and remain for 10 to 20 years (Tremblay in litt. 2017).

In the northeastern United States, there is approximately 367,015 hectares (ha) (906,914 acres (ac)) of Bicknell’s thrush nesting habitat (Hill and Lloyd in litt. 2016). The last estimate of nesting habitat in Canada is 4,885,100 ha (12,000,000 ac) (COSEWIC 2009, p. 6). While these estimates provide context for the relative amount of suitable breeding habitat, we note that the species is not evenly distributed throughout this habitat. As a result of these differences in density, there are more birds nesting in the northeastern United States than in Canada (see Population Estimates section below).

Fall Migration Behavior and Habitat

Bicknell’s thrushes begin departing the breeding grounds during the last few days of September and by the end of the first week in October almost all birds will have departed on their fall migration (Wallace 1939, p. 259). Individuals move southward along the Atlantic coast where they will make frequent stops for extended periods to rest and feed, with most birds departing from Virginia and flying across the ocean to the Bahamas and Cuba, finally arriving in the Greater Antilles from mid-October through early November (Ouellet 1993, p. 564; Rimmer et al. 2001, pp. 6–7; McFarland et al. In prep., in Townsend et al. 2015, unnumbered ). The “leisurely” pace of fall migration is supported by collected data obtained from Wallace’s studies (1939, p. 259) and from other Bicknell’s thrush outfitted with geolocators (McFarland et al. In prep., in Townsend et al. 2015, unnumbered). The data demonstrate the mean duration of fall migration is 29 days, with individuals making temporary stops that range in duration from 6 to 33 days, at locations centered primarily on the western Caribbean, and to a lesser extent on the Atlantic Coast, from Long Island to Virginia (McFarland et al. In prep., in Townsend et al. 2015, unnumbered). During migration, the Bicknell’s thrush appears to be a habitat generalist and can be found in dense woodlots composed of variable tree species, along well vegetated beaches, orchards and gardens (Wallace 1939, p. 259; Wilson and Watts 1997, pp. 520–521).

Wintering Behavior and Habitat

Wintering occurs exclusively in the Greater Antilles, with the majority of Bicknell’s thrushes on the island of Hispaniola, in Haiti and the Dominican Republic; however, the species can also be found on the islands of Cuba, Jamaica, and Puerto Rico (Rimmer et al. 2001, pp. 3–4). In

Jamaica, the Bicknell’s thrush is considered “extremely rare” and observed in old growth forests (Strong in litt. 2016). The species’ information for Puerto Rico is scant (Rivera in litt. 2017). The Vermont Center for Ecostudies conducted surveys for Bicknell’s thrush in the winter of 2015 and 2016 but had difficulty locating the species; in 2015 they found three birds and in 2016 the survey team found seven in the high elevation protected areas of the central mountain range, but only after resorting to non-standardized survey methods (Rimmer 2016, entire). In the Dominican Republic, where the majority of wintering information about the species is derived, the Bicknell’s thrush can be found from sea level to 2,200 m (7,200 ft), although most occur in mesic to wet broadleaf montane forests above 1,000 m (3,300 ft) elevation (i.e., cloud forest) (Rimmer et al. 2001, p. 8). The Bicknell’s thrush can also be found in dry pine-dominated forests at lower elevations (Rimmer et al. 2001, p. 6). The species prefers wintering in dense thicket vegetation (Townsend et al. 2010, p. 520), similar to the habitat structure selected during the breeding season. Both males and females defend and maintain individual territories against other Bicknell’s thrushes (Townsend et al. 2010, p. 517).

Figure 4. Example of potential Bicknell’s thrush habitat in the Dominican Republic (Photo credit: Vermont Center for Ecostudies).

In the Dominican Republic, there is evidence that Bicknell’s thrush exhibit sexual segregation, with males predominating (70M:30F) in high-elevation cloud forest, but the sexes cohabitate in low to middle elevation (i.e., rain forest) sites in equal proportion (50M:50F) or with a female- bias (Townsend et al. 2012, p. 689). The body mass of males shows little variation among wintering habitats used; however, females occupying rainforest sites are significantly heavier than females occupying cloud forest habitats (Townsend et al. 2012, p. 687). This observation

may be attributed to increased thermoregulatory demands on smaller females than are imposed on larger males in the cooler cloud forest habitats (i.e., smaller birds need to expend more energy to stay warm in higher elevation), or increased antagonistic behavior on cloud forest females from territorial males than are experienced by rainforest females (Townsend et al. 2012, p. 688).

Food availability varies among sites, with mid-elevation sites containing more fruiting trees and less arthropod biomass than in cloud forest sites (Townsend et al. 2012, p. 686). Based on these behavioral and food availability observations, Townsend et al. (2012, p. 689) proposed that lower elevation rainforest habitats are considered to be “critically important to female survival on Hispaniola,” with cloud forests also being important to the species’ conservation.

Spring Migration Behavior and Habitat

In spring, the birds leave the Greater Antilles in late April through the first week in May (Rimmer et al. 2001, p. 5; McFarland et al. In prep., in Townsend et al. 2015, unnumbered). The spring migration period is shorter than the fall migration period, with most birds completing the journey in about 17 days (McFarland et al., In prep.). There is no evidence of spring stopovers from banding or geolocator data (Townsend et al. 2015, unnumbered). In contrast with the fall migration route, the spring route is largely overland through the Southeast United States, with birds passing northward through Florida, Georgia, and the Carolinas (Townsend et al. 2015, unnumbered). By the end of May, the birds can be found back in the mountains of New England and Canada (Wallace 1939, p. 259; Rimmer et al. 2001, p. 5). Males typically arrive sooner than the females (Rimmer et al. 2001, p. 5).

SURVEY DATA AND POPULATION TRENDS

Survey Efforts

Localized and regional surveys have been conducted across the Bicknell’s thrush’s range with widely different levels of geographic, temporal, and methodological consistency. It is possible that the species may currently occur in unknown areas, but, with the potential exception of Cuba, the distribution of Bicknell’s thrush on the wintering grounds and breeding grounds is well defined. Available survey data are presented below. Some general characterizations of the available data are noted here.

The patchy distribution of the Bicknell’s thrush’s remote breeding habitat presents logistical difficulties that have hindered implementation of comprehensive survey efforts for this species. Consequently, prior to 1991, there were no organized attempts to survey breeding Bicknell’s thrushes across their range in the northeastern United States and Canada (Hart and Lambert 2010, p. 6). Since 1991, several efforts produced models of the species’ distribution, occupancy, and habitat, as well as estimates of population size and trends at various spatial and temporal scales (Hart and Lambert 2010, p. 6; see table 1 below).

Studies from 1991 to 1999 yielded important scientific information that has helped researchers understand the biology and conservation status of the species, but there are uncertainties in using the survey data for long-term trends. For example, some of the challenges include: (a) the lack of probabilistic sampling methodology (i.e., random selection of survey sites) that limits statistical inferences; (b) variation in survey protocols and timing; (c) variation in data collected; and (d) redundant efforts in some locations (e.g., White Mountains in New Hampshire) but incomplete coverage in other areas (e.g., Quebec and Maine) (Hart and Lambert 2010, p. 7).

In 2000, the Mountain Birdwatch program coordinated systematic surveys across the species’ U.S. breeding range (Hart and Lambert 2010, p. 7). In 2010, the Mountain Birdwatch program (renamed to Mountain Birdwatch 2.0) was expanded to include sites in Canada. The sampling design was modified to collect data about abundance, site occupancy, distribution, and trends in several high-elevation breeding bird species, including the Bicknell’s thrush (Hart and Lambert 2010, p. 7).

Table 1. Specific high-elevation landbird surveys in the northeastern United States and eastern Canada since 1991 (table adapted from Hart and Lambert 2010, p. 7).

Program Lead Institution State/ Province Timeframe Technique Green Mountain Green Mountain Vermont 1991 – 2000 Point counts National Forest National Forest high-elevation bird monitoring Vermont Forest Vermont Institute Vermont, Maine 1991 – 2000 Point counts Bird Monitoring of Natural Science Program high- (VINS) elevation surveys Bicknell’s thrush VINS and Massachusetts, 1992 – 1994 Playback assisted distribution Manomet Center New York, presence-absence surveys for Conservation Vermont, New surveys Sciences Hampshire, and Maine White Mountain White Mountain New Hampshire 1993 – present Point counts National Forest National Forest high-elevation and Audubon bird monitoring Society of New Hampshire Bicknell’s thrush Canadian Wildlife Quebec 1998 – present Point counts with distribution Service playback surveys Mountain VINS, Vermont New York, 2000 – 2008 Point counts with Birdwatch Center for Vermont, New playback assisted Ecostudies, Hampshire, Maine presence-absence USFWS surveys Mountain Vermont Center New York, 2009 – present Point counts Birdwatch 2.0 for Ecostudies and Vermont, New numerous Hampshire, Maine; partnering entities Quebec, New Brunswick, Nova Scotia High-Elevation Bird Studies New Brunswick, 2002 – 2009 Point counts with Landbird Canada Nova Scotia time-of-detection Program information

Survey efforts and associated information for the Bicknell’s thrush at migratory and wintering locations is summarized as follows: ● There are no systematic survey efforts from the species’ migration or wintering grounds that are suitable for yielding information about trends in Bicknell’s thrush populations. ● The majority of the Bicknell’s thrush migration records are derived from sightings by professional and amateur birders reported in eBird (eBird.org) and a limited number of birds captured from various sites in the eastern United States (for an overview of migration records see Brooks and Sherony 2010, pp. 7–14). Migration data from birds

outfitted with geolocators are being analyzed (McFarland et al. In prep., in Townsend et al. 2015, unnumbered). ● Records from wintering areas are limited to sightings reported in eBird (eBird.org) and local bird checklists (e.g., Arendt et al. 2015, p. 12) and other published information about the species’ presence in localized areas (e.g., Oviedo et al. 2001, p. 134; Sosa et al. 2001, p. 136; Lloyd et al. 2016, pp. 7–8, 10). The only effort to systematically survey the species on its wintering grounds was through an effort to validate a winter habitat model using a single season of data from the island of Puerto Rico (Lloyd and Rimmer in. litt. 2016).

Population Estimates

Prior to the 2010 implementation of the Mountain Birdwatch survey protocol across the species’ breeding habitats in the United States and Canada, there were no organized systematic efforts to collect rangewide survey data. Lack of systematic data results in a high degree of variability among historical population estimates for the Bicknell’s thrush. Population estimates, based on survey point count data collected through the initial Mountain Bird Watch effort (pre-2010, United States only), the High-Elevation Landbird Program, and similar data collection efforts conducted in Quebec, were used to derive the global population of the Bicknell’s thrush in the 2009 assessment and status report prepared by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC 2009, pp. 23–24). The estimate was derived by calculating the number of birds likely to occur in the available breeding habitat, as predicted by a habitat model that extrapolated bird densities as calculated from 1,302 survey points collected in 2005 and 2008 (COSEWIC 2009, p. 24). Based on this methodology, the best available Bicknell’s thrush population size estimate in 2009 is 57,480 to 76,640 in the United States and 40,570 to 49,258 in Canada, for a maximum global population of between 98,050 and 125,898 (COSEWIC 2009, p. 24). In 2016, researchers revised the Bicknell’s thrush U.S. population estimate using updated Mountain Birdwatch data collected from 2011 to 2016 (Hill and Lloyd in litt. 2016). The updated breeding density data resulted in an estimated population of 71,000 (range 56,788 to 90,219) in the United States, which is consistent with the 2009 estimate provided by COSEWIC. There is no updated population estimate for Canada. Therefore, by combining the 2016 U.S. estimates with the 2009 Canadian estimates, the best available data that extrapolates bird densities from modeled habitat indicate that the current Bicknell’s thrush global population is approximately 97,358 to 139,477.

From the available estimates, the United States’ breeding range supports approximately 66 percent of the global population of the Bicknell’s thrush (COSEWIC 2009, p. 246; Hill and Lloyd in litt. 2016). Within the United States, approximately 26,449 birds (37 percent) breed in New Hampshire, 21,072 birds (29 percent) breed in New York, 18,802 birds (26 percent) breed in Maine, and 5,297 birds (7 percent) breed in Vermont (Hill and Lloyd in litt. 2016). There is no longer a breeding population in Massachusetts (see Population Trends below). The remainder of the global population of the Bicknell’s thrush, 40,570 to 49,258 birds (approximately 33 percent), breeds in Canada (COSEWIC 2009, p. 24).

Abundance Trends

Notwithstanding the above mentioned challenges associated with conducting population surveys within the Bicknell’s thrush breeding range, the best available data suggest the species’ abundance has been undergoing a long-term declining trend. Although the methods vary, the conclusions from each of the data types are similar; the Bicknell’s thrush appears to be declining.

For example, analysis of Canadian Breeding Bird Survey data, collected over the 40-year period from 1968 to 2008, showed a decline in Bicknell’s thrush of 9 percent per year (COSEWIC 2009, p. 25). Updated Breeding Bird Survey trends available from 1970 to 2015 show a decline of 5.3 percent, but with low reliability (Tremblay in litt. 2017). Additionally, 10 years of data (2002–2012) show the number of Bicknell’s thrushes reported on survey routes in New Brunswick declined by 11.5 annually, but the rate of decline slowed in the latter 5 years of the study (Campbell and Stewart 2012, pp. 7, 13). Consistent with these declines, the species’ distribution appears to have contracted. For example, breeding Bicknell’s thrushes from Seal and Mud Islands in Nova Scotia (COSEWIC 2009, p. 9) have disappeared, despite being relatively common at the time of Wallace’s writing (1939, p. 331) when at least a dozen nests were found on Seal Island (see figure 5). The bird is also absent from formerly occupied habitats on Cape Breton Island and Cape Forchu, Nova Scotia (COSEWIC 2009, p. 9; Rimmer et al. 2001, p. 4) (see figure 5).

Figure 5. Some locations with previously observed or declining numbers of Bicknell’s thrush.

The distribution and probability of observation, measured through the Second Atlas of Breeding Birds of the Maritime Provinces, have also declined across the Maritimes. Bicknell’s thrush can 20

no longer be detected in some parts of the region, and the probability of observing the species has decreased in both Nova Scotia and New Brunswick (Whittam et al. 2015, pp. 391). In New Brunswick, since the 1980s, the species has apparently become absent as a breeder from the southern half of the province, including from Grand Manan Island and the Rapidy Brook area (COSEWIC 2009, p. 9). In Quebec, the Bicknell’s thrush was not observed from 1999 to 2008 in the following locations: Montagne Noire; Monts Sir-Wilfrid, des Éboulements, Comi, and St- Pierre; at some previously occupied sites in the zec des Martres; Métis-sur-Mer; and on Bonaventure and Magdalen Islands (COSEWIC 2009, p. 9). At Mont Gosford, annual surveys conducted from 2001 to 2007 showed a clear decline in the number of stations occupied by the Bicknell’s thrush (Aubry, unpubl. data in Campbell et al. 2007, p. 7). It is not clear from the best available data exactly when or why the extirpations occurred.

As with Canada, local extirpations have been detected from former breeding habitats in the United States (Rimmer et al. 2001, p. 4). For example, in Massachusetts, the Bicknell’s thrush breeding population on Mount Greylock gradually declined in apparently suitable habitat from 10 pairs in 1950 to 0 pairs in 1973, and visits to nearby Saddle Ball Mountain during the period 1992 to 1995 failed to detect the species (Atwood et al. 1996, p. 657). Attwood et al. (1996, entire) thought that these local extirpations were attributed to increased human disturbance by visitors to the summit rather than habitat loss (Petersen and Meservey 2003, p. 427; Rimmer and McFarland 2013, p. 9). However, studies conducted in the White Mountains of New Hampshire do not support the notion that human disturbance causes extirpation, since there was no evidence that recreational hiking trails were having adverse effects on the abundance, detection probability, abundance stability, or recruitment of the Bicknell’s thrush (Deluca and King 2014, p. 498). In addition to the extirpations observed at the Massachusetts sites, Atwood et al.’s (1996, p. 657) 4-year survey also failed to detect the species where it had historically occurred in Vermont on Glebe and Molly Stark Mountains, as well as Mounts Aeolus and Ascutney; and in New Hampshire on Mounts Pemigewasset, Monadnock and Sunapee, and North Moat Mountain. As with Canadian sites, it is not clear from the best available data exactly when or why the extirpations occurred.

The observations documenting local declines and extirpations at sites across the species’ breeding range in the United States is supported by an analysis of combined datasets collected from multiple standard single-observer point count surveys conducted in association with multiple monitoring programs (Ralston et al. 2015, p. 273). Since comparison of data collected under different protocols can bias estimates of detection and population trends, the researchers used modern analytical methods to control for the effects of varying methodologies in field sampling (Ralston et al. 2015, p. 272). The results of the analysis indicate the Bicknell’s thrush has undergone a statistically significant decline since the early 1990s, as indicated by an overall trend estimate of 0.977 (value of 1.0 indicates stable populations) (Ralston et al. 2015, p. 272).

In addition to point count, habitat modeling, and other abundance data, one dispersal study based on stable-hydrogen isotope values from the species’ feathers suggests that the Bicknell’s thrush may be declining (Studds et al. 2012, p. 920). Researchers can use location-specific ratios of stable-hydrogen isotopes to determine if an individual bird has returned the following year to the same area where it was hatched (i.e., its natal area) or reveal when it has dispersed to another

location (Hobson and Wassenaar 1997, p. 142; Studds et al. 2012, p. 920). The Bicknell’s thrush, as with other migratory songbirds, retains the feathers grown as a nestling through the first year of its life, and then moults as an adult the following summer, after the completion of the first nesting season (Studds et al. 2012, p. 920). By analyzing the stable-hydrogen isotope values in feathers collected from adult birds during their first breeding attempt, insights into the dispersal behavior of individuals can be revealed (Studds et al. 2012, p. 920). Bicknell’s thrush feather samples were collected from 500 individuals at 25 montane forest breeding sites in the United States and Canada from 1996 to 2005 (Studds et al.2012, p. 921). Based on the stable- hydrogen isotope values, researchers assigned birds to six breeding regions of origin: the Catskill Mountains of New York, southern Vermont, the Adirondacks and northern New England, southern Quebec, New Brunswick, and the Gaspé Peninsula of Quebec (Studds et al. 2012, p. 922). Of the 192 Bicknell’s thrushes sampled during their first breeding season, 59 percent were classified as dispersing away from their natal breeding region and 41 percent were classified as faithful to their natal region (Studds et al. 2012, p. 923).

Analysis also revealed that the probability of natal dispersal has declined by 30 to 38 percent over the 10-year period from 1996 to 2005 throughout the breeding range (meaning more birds had previously dispersed, indicating existing territories were at capacity). These findings are consistent with many of the independently observed local population declines and extirpations discussed above (Studds et al. 2012, p. 920). This decline in dispersal suggests first year birds are finding ample territory vacancies within close proximity to their natal habitats, which may provide additional support for recent population declines (Studds et al. 2012, p. 920). As mentioned above in the Breeding Behavior and Habitat section, by their third breeding season, Bicknell’s thrush typically return to the same breeding area (Townsend et al. 2015, unnumbered).

Counter to the above findings, Lloyd et al. (2016, entire) conducted a 13 year study on resident and neotropical migrant birds at two remote, undisturbed locations within the montane cloud forest in the Dominican Republic’s Sierra de Bahoruco National Park. Mist net surveys were conducted at least once annually, with no surveys in 1999 due to hurricane damage on the island, during January to March. While there were temporal declines detected in resident species, the authors did not detect a temporal trend in abundance of Bicknell’s thrush. A total of 149 individual Bicknell’s thrush were captured over the course of their study.

Summary of Survey Data and Trends

The range of global population estimates for the Bicknell’s thrush extrapolated from habitat modelling in 2009 compared to 2016, indicate an overall stable trend. In contrast, the best available data, based on breeding population abundance, indicate a slight long-term declining trend in the Bicknell’s thrush breeding population from historical levels. There is sparse information readily available about the species’ wintering population.

FACTORS INFLUENCING VIABILITY Existing Factors Influencing the Bicknell’s Thrush

Breeding Range

Contemporary Impacts to Breeding Areas Resulting From the Effects of Climate Change

Relatively small scaled studies conducted in montane spruce-fir habitat suggest that climate change-induced habitat loss has already occurred within the range of the Bicknell’s thrush. The spruce-fir/deciduous ecotone is correlated with elevation areas that have a mean July temperature of approximately 17 oC (63 °F); consequently, montane spruce-fir forests are restricted to upper elevations within the northeast United States (Cogbill and White 1991, pp. 169 and 171). Analysis of forest plots in the Green Mountains of Vermont from 1964 to 2004 demonstrated a 19 percent increase in the dominance of northern hardwood species in the northern hardwood- boreal forest ecotone, at the expense of red spruce, balsam fir, and montane paper birch (Beckage et al. 2008, p. 4197), the preferred tree species components of Bicknell’s thrush breeding habitat. This tree species shift was corroborated by remotely sensed data from 1962 to 2005 that indicated a 92 m (302 ft) and 119 m (390 ft) upslope movement in the northern hardwood to boreal ecotone on Mount Abraham, which supports a breeding population of the Bicknell’s thrush (Rimmer et al. 2005a, p. 27), and Camels Hump, respectively. This change coincided with a 1.1 oC (1.98 °F) increase in annual temperature during the same period, and Beckage et al. (2008, p. 4201) propose that ongoing climate change impacts include the growth and recruitment of northern hardwoods at higher elevations. These authors also suggest that the increases in northern hardwood species was made possible by vacancies left by boreal forest species that had, possibly, succumbed to the effects of acid rain depositions, which has been identified as a source of mortality in red spruce (Beckage et al. 2008, p. 4201). In conclusion, the authors suggested “that high-elevation forests may be jeopardized by climate change…” (Beckage et al. 2008, p. 4197). Similar shifts in habitat have been documented at other Vermont (Friedland 1989, pp. 240–241) and New York sites (Cook 1985 and Johnston et al. 1988 in Friedland 1989, p. 242).

In contrast with these relatively localized studies of retraction in the montane spruce-fir ecotone, other investigations using remote sensed data collected over larger spatial scales provide conflicting information regarding ongoing elevational changes in the montane spruce-fir ecotone resulting from climate change. For example, evaluation of Landstat imagery for an area of the White Mountains in New Hampshire collected from 1987 to 2010 found an increase in conifer dominance in the sub-canopy of deciduous habitats towards the lower parts of mountain slopes along elevational gradients (Vogelmann et al. 2012, pp. 95, 99). A similar study spanning the northern portion of the Green Mountains in Vermont and all of the White Mountain range in New Hampshire found variability in the movement of the montane spruce-fir to hardwood ecotone from 1991 to 2010 among individual mountain slopes (Foster and D’Amato 2015, p. 4501). Foster and D’Amato (2015, p. 4504) found that the montane spruce-fir to hardwood forest ecotone has shifted downslope over the past 20 to 30 years, rather than upslope, as would be expected if climate change had been influencing tree species distribution. These observed departures from the expected pattern may be the result of: (a) recovery of red spruce from recent declines resulting from atmospheric deposition of pollutants, as discussed in the Atmospheric 23

Acid and Nitrogen Deposition in Breeding Areas section of this document, below; (b) disproportionate stress on high-elevation hardwoods resulting in reduced competition; and (c) recovery from stand alterations resulting from past cutting practices that selectively removed red spruce or other microhabitat components (e.g., soil) (Lenoir et al. 2010, entire; Foster and D’Amato 2015, pp. 4504–4505).

Summary of Contemporary Impacts to Breeding Areas Resulting From the Effects of Climate Change

Relatively small scaled studies conducted in the montane spruce-fir habitat of Vermont (Friedland 1989, pp. 240–241; Beckage et al. 2008, pp. 4197, 4201), New Hampshire (Rimmer et al. 2005a, p. 27), and New York (Cook 1985 and Johnston et al. 1988 in Friedland 1989, p. 242) suggest that climate change-induced habitat loss has already occurred within the range of the Bicknell’s thrush. However, based on the results of larger spatial studies (Vogelmann et al. 2012, pp. 95, 99; Foster and D’Amato 2015, pp. 4501, 4504–4505), we conclude that the extent to which the effects of climate change have already affected the montane breeding habitat of the Bicknell’s thrush is somewhat ambiguous, but appears to be negligible.

Historical and Current Recreational, Telecommunication, and Wind Energy Development

Prior development for recreation (i.e., ski areas), especially the cumulative effect of multiple ski areas, has directly resulted in the loss and fragmentation of Bicknell’s thrush breeding habitat. However, the cumulative effects of these development-driven habitat losses across the range of the Bicknell’s thrush is poorly known (Rimmer et al. 2001, p. 21; Bredin and Whittam 2009, pp. 12, 13; COSEWIC 2009, p. 32) since biological results are available only from localized studies.

In Vermont, 13 mountains that are greater than 915 m (3,000 ft) elevation are developed for recreational skiing, and many of these offer mountain bike activities on cleared ski slopes during the Bicknell’s thrush breeding season (Rimmer et al. 2001, p. 21). Similar year-round recreational uses may occur in New Hampshire and Maine, but less so in the Catskills and Adirondacks in New York (Rimmer et al. 2001, p. 21), and in Canada (COSEWIC 2009, p. 32). In the short term, construction of these recreational developments resulted in the loss of some amounts of Bicknell’s thrush habitat (Rimmer et al. 2001, p. 21). For example, approximately 4.8 ha (11.8 ac) of Bicknell’s thrush breeding habitat was removed, and an additional 1.8 ha (4.4 ac) isolated, during construction of an expansion to the Whiteface Mountain trail system in New York’s Adirondack Mountains (Rimmer et al. 2004, p. 8). This loss constituted up to 0.26 percent of the suitable habitat in the Adirondack Park’s Whiteface Mountain Habitat Unit that includes high-elevation songbird habitat on Whiteface Mountain, Little Whiteface Mountain, Esther Mountain, Lookout Mountain, and Baldwin Hill, and less than 0.001 percent of the total breeding habitat available in the northeastern United States (Rimmer et al. 2004, p. 10). Post construction monitoring of the site revealed a slight but statistically insignificant decline in Bicknell’s thrush abundance within the expansion area (Glennon and Seewagen 2016, p. 126).

The Stowe Mountain Resort on Mount Mansfield and the Stratton Mountain Resort in Vermont demonstrate that there are few differences in various Bicknell’s thrush population and

reproductive parameters (including nest predation, nest success, parental care, movement patterns, survivorship, and productivity) between habitat patches at the ski areas and habitat patches at natural forest sites on each of the resorts’ mountains (Rimmer et al. 2004, p. 2). However, radio telemetry data did reveal that adult thrushes avoid trail crossings wider than 50 m (164 ft), and trails 35 to 40 m (115 ft to 131 ft) in width exhibit some restrictions on Bicknell’s thrush movements (Rimmer et al. 2004, p. 2). Yet in a different study, Glennon and Karasin’s (2004, p. 1) investigations of existing ski trails and glades showed no statistical differences in abundance of the Bicknell’s thrush. These results suggest that, while the construction of ski areas resulted in the immediate loss of Bicknell’s thrush habitat, the birds may be able to adapt and use most of the remaining habitat without long-term consequences, provided consideration is taken to avoid fragmenting habitat to a degree that may inhibit the movement of adult Bicknell’s thrushes.

In addition to ski area development, infrastructure development for telecommunication and wind energy projects may cause loss and fragmentation of Bicknell’s thrush habitat. Wind and telecommunications structures are often placed on exposed high-elevation areas, which may include areas of suitable Bicknell’s thrush breeding habitat. In some instances, construction of these facilities can directly impact Bicknell’s thrush habitat (Rimmer et al. 2001, p. 21; McFarland et al. 2008, p. 1; COSEWIC 2009, p. 32). For example, Noble Environmental Power (in litt. 2008) calculated that its Granite Reliable wind power project, located on Owl’s Head Mountain and Mount Kelsey in New Hampshire, would result in the removal of approximately 23.5 ha (58 ac) of high-elevation spruce and spruce-fir forest, some of which is known to be occupied by the Bicknell’s thrush. In addition, several wind power projects have been located within Bicknell’s thrush habitat in Quebec and New Brunswick (COSEWIC 2009, p. 32; Tremblay in litt. 2017) and wind development is considered to be continuous, localized threat with a very large population effect in Canada’s proposed recovery plan for the species (Environment and Climate Change Canada 2016, pp. 6–7, 10).

As of February 2017, two wind power projects have been constructed within occupied Bicknell’s thrush breeding habitat in the United States. Additional wind power projects could be built in the future due to the high wind exposure characteristic of these sites (McFarland et al. 2008, entire; Parrish 2013, entire; Leppold in litt. 2016; Todd in litt. 2016; Buck in. litt. 2016). McFarland et al. (2008, p. 8) assessed the relationship of Bicknell’s thrush breeding habitat to available wind resources. The authors determined that nearly 94 percent of the potential Bicknell’s thrush habitat found in the Northeastern Highlands region of Vermont overlaps areas of wind power Class 4 (greater than 7 mps (15.7 mph)) or higher, which are considered good resources for generating wind power with large turbines. However, the area of overlap between Bicknell’s thrush habitat and Class 4 or higher wind areas represents 7 percent of the total available high-value wind resource area. This suggests that a large portion (93 percent) of the potentially suitable wind power terrain could be developed without directly impacting Bicknell’s thrush habitat (McFarland et al. 2008, p. 8). Visual comparison of modeled Bicknell’s thrush habitat with wind resource data from throughout the Bicknell’s thrush’s range appears similar.

Except for the aforementioned Granite Reliable wind power project, there is limited information available from existing or proposed wind turbine sites (McFarland et al. 2008, p. 5; Parrish 2013,

entire). To assess habitat utilization response by the Bicknell’s thrush to the Granite Reliable project, 30 radio transmitters were deployed during construction and operation to collect spatial data on individual birds (Parrish 2013, pp. 38–42). Significant changes in Bicknell’s thrush home range size due to the development and operation of the wind facility were not detected during the 2011 and 2012 breeding seasons (Parrish 2013, p. 43). Furthermore, the radio telemetry data revealed that birds used suitable habitat when it was present in areas immediately adjacent to turbine pads (Parrish 2013, p. 43). Some turbine avoidance, attributed to rotor noise, was detected and resulted in increased home range size for individuals with home ranges containing turbines (Parrish 2013, p. 43–44). These observations suggest the habitat near pads remains suitable during the operation of wind turbines, but may be less preferable than habitat farther away from turbines where noise levels are lower (Parrish 2013, p. 47).

Summary of Historical and Current Recreational, Telecommunication, and Wind Energy Development

Development of ski areas, wind turbines, telecommunication facilities and their associated infrastructure (i.e., roads) has resulted in the loss and fragmentation of habitat for the Bicknell’s thrush (IBTCG 2010, p. 12; Environment and Climate Change Canada 2016, pp. 6–7, 10). However, the Bicknell’s thrush does show some ability to adapt and persist in the vicinity of ski resorts (Rimmer et al. 2004, p. 1) and the species may respond similarly to the presence of wind turbines, although there is evidence of wind turbine avoidance that is attributed to rotor noise (McFarland et al. 2008, p. 56, Parrish 2013, p. 47).

Historical and Current Logging and Forest Fragmentation

As discussed above in the Breeding Behavior and Habitat section, Bicknell’s thrush prefer a densely structured spruce-fir dominated habitat. The dense structure is often maintained through natural disturbances (wind throws, fire, etc.), but some forestry practices can mimic these natural disturbance regimes. Campbell and Stewart (2012, pp. 13–14) note that “Natural mountain top disturbances are usually of smaller size than industrial forest ones, but take more time to regenerate. As a result, the landscape in natural mountainous regions is usually characterized by a mosaic of small patches in various stages of regeneration. Disturbances on the industrial landscape tend to be large scale and result in a landscape of large patches of even-aged regenerating forest. This could lead to a boom-or-bust scenario, where large amounts of regenerating habitat are initially available 10 to 15 years clearcutting, followed by a substantial drop in suitable habitat once large patches are thinned or mature.” Throughout the industrial (i.e., managed) highlands of Canada and northern Maine, the practice of clearcutting may affect the Bicknell’s thrush by temporarily removing forest habitat; however, the resulting regeneration of balsam fir and spruce in these areas is known to result in the creation of breeding habitat at a later time (Ouellet 1993, p. 566; Nixon et al. 2001, p. 34; Chisholm and Leonard 2008, p. 218; COSEWIC 2009, p. 31; IBTCG 2010, p. 11). McKinnon et al. (2014, pp. 259, 268) conducted a nest patch study in New Brunswick that compared nest selection in pre- and post- precommerically thinned habitat. The study results show that individual birds relocated their nests from one year to the next if the original site was no longer suitable due to clear-cutting or thinning activity, with the new nest occurring in adjacent habitat (McKinnon et al. 2014, p. 264).

From this study, the researcher suggests that small patches of dense balsam fir dominated habitat within areas that have been thinned may be sufficient to support returning breeding birds; however, because nest productivity was not evaluated, it is unclear whether these treated areas were potential sinks (McKinnon et al. 2014, p. 268).

Although the Bicknell’s thrush occupies second growth stands of 25 to 40 years old, optimal conditions within managed patches for the species occur in regenerating clear cuts ranging from a minimum of 5 to 12 years old (Nixon et al. 2001, p. 39; Connolly et al. 2002, p. 338; Chisholm and Leonard 2008, p. 222). Some forestry practices, specifically the use of precommercial thinning that reduces the stem densities of regenerating trees to maximize growth in remaining trees, results in the reduced abundance of the Bicknell’s thrush (Chisholm and Leonard 2008, p. 222). When precommercial thinning is conducted during the bird’s nesting season (Makepeace and Aubry, unpubl. data in COSEWIC 2009, p. 31), there is the potential for direct mortality to occur to Bicknell’s thrush eggs and nestlings.

Historical and Current Timber Harvest Trends

The best available timber harvest data are inconsistent across the Bicknell’s thrush’s breeding range in the United States and Canada. However, the trend in harvest level is largely consistent across the data; most show a relatively stable or declining trend in the overall timber, softwood, or spruce-fir specific levels harvested. One state, Vermont, shows a variable overall trend in timber harvest, but a declining and stabilized trend specific to spruce-fir harvest.

In Maine, available data suggest that all tree species harvested, statewide, declined from 2000 to 2015 and that estimated timber stand improvement treatments (e.g., pre-commercial thinning) in the four counties within Bicknell’s thrush breeding habitat (Franklin, Oxford, Piscataquis, and Somerset) remained at relatively low and stable levels during the same timeframe (Maine Forest Service (MFS) 2001, pp. 3, 6; MFS 2002 pp. 3, 6; MFS 2003 pp. 3, 6; MFS 2004 pp. 3, 6; MFS 2005 pp. 3, 6; MFS 2006 pp. 3, 6; MFS 2007 pp. 3, 6; MFS 2008 pp. 3, 6; MFS 2009 pp. 3, 6; MFS 2010 pp. 3, 6; MFS 2011 pp. 3, 6; MFS 2013 pp. 3, 6; MFS 2014a pp. 3, 6; MFS 2014b pp. 3, 6; MFS 2015 pp. 3, 6; MFS 2016, p. 3, 6). See figure 5. Harvest data specific to spruce-fir trees were not available.

Figure 6. Estimated acres of total timber versus acres of timber stand improvement treatment in Maine from 2000 to 2015.

In New Hampshire, available data suggest that all species harvest, statewide, declined from 2006 to 2008, increased from 2008 to 2011, and then remained relatively stable from 2012 to 2015 (Morin and Tansey 2008, p. 1; Morin et al. 2010b, p. 1; Morin 2011, p. 1; Morin and Nelson 2011, p. 1; Morin and Woodall 2012, p. 1; Morin and Lombard 2013, p. 1; Morin and Pugh 2014, p. 1; Morin and Riemann 2015, p. 1; Morin and Widmann 2016, p. 1). No comparable data was available for 2007 (Morin et al. 2010a, entire). See figure 6. The estimated harvest data were available only at the statewide scale, so these values likely overestimate the amount of potential harvest within the three counties (Carroll, Coos, and Grafton) known to include Bicknell’s thrush habitat. Growing stock includes the “trees in a stand that have potential to produce salable products now or in the future. Excludes cull trees.” (New Hampshire Timber Harvest Council 2011).

NH Est. annual removal/harvest of growing stock trees 160,000 140,000

120,000 100,000 80,000 60,000 1,000 ft3/yr 1,000 40,000 20,000 0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Year

Figure 7. Estimated annual removal or harvest of growing stock trees in New Hampshire from 2006 to 2015.

In Vermont, available data suggest that the overall harvest, statewide, was variable from 1997 to 2014, with periods of both decreasing and increasing harvest levels. However, the level of spruce-fir harvest decreased from 1997 to 2005, and remained stable at a lower level from 2005 to 2014 (Vermont Department of Forests, Parks & Recreation (VTDFPR) 1997, entire; VTDFPR 1998, entire; VTDFPR 1999, entire; VTDFPR 2000, entire; VTDFPR 2001, entire; VTDFPR 2002, entire; VTDFPR 2003, entire; VTDFPR 2004, entire; VTDFPR 2005, entire; VTDFPR 2006, entire; VTDFPR 2007, entire; VTDFPR 2008, entire; VTDFPR 2009, entire; VTDFPR 2010, entire; VTDFPR 2011, entire; VTDFPR 2013, entire; VTDFPR 2014, entire). Comparable data for 2012 were not obtained. See figure 7. The estimated harvest data were available only at the statewide scale, so these values likely overestimate the amount of potential harvest within the seven counties (Addison, Bennington, Essex, Lamoille, Orleans, Rutland, and Washington) known to include Bicknell’s thrush habitat.

VT Spruce/Fir vs. Total Harvest Data 1,400,000

1,200,000

1,000,000

800,000

600,000 Spruce/Fir Harvest # of cords of # All Species Harvest 400,000

200,000

0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Year

Figure 8. Total number of cords harvested for all tree species versus spruce-fir in Vermont from 1997 to 2014.

In New York, available data suggest that the level of timber harvested for softwood pulpwood and chips, mainly eastern hemlock (Tsuga canadensis), white pine (Pinus strobus), and spruce (Picea spp.) trees, was largely stable, with small levels of variability, from 1999 to 2014 (New York Department of Environmental Conservation (NYDEC) 2014, p. 3). See figure 8. However, Bicknell’s thrush habitat occurs within the Adirondack and Catskill Parks and is almost exclusively owned by the State and protected as “forever wild” by Article XIV of the New York State Constitution, which prohibits timber harvest (NYDEC, retrieved from http://www.dec.ny.gov/lands/4960.html, accessed December 28, 2016). Therefore, the historical harvest trend in New York does not affect the species.

New York Pulpwood and Chips Harvest (1999-2014) 3

2.5

1.5

1 Millions of Green Tons 0.5

0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Year

Figure 9. Millions of green tons of softwood harvested for pulpwood and chips in New York from 1999 to 2014. Chart reproduced from New York Department of Environmental Conservation 2014 “New York State Industrial Timber Harvest Production and Consumption Report-2014.”

In Canada, available data suggest that the total amount of timber harvest within the three provinces (New Brunswick, Nova Scotia, and Quebec) containing Bicknell’s thrush breeding habitat declined over a 24-year period. The harvest increased from 1992 to1997, mostly decreased from 1998 to 2008, and overall increased, but at a reduced level from the previous decade, from 2009 to 2014. During this same period, 1990 to 2014, the amount of this timber harvested using commercial thinning (trees harvested from the site are used for commercial purposes) practices stayed relatively consistent (Data retrieved from the Canadian National Forestry Database, http://nfdp.ccfm.org/silviculture/jurisdictional_e.php, last accessed November 10, 2016). See figure 9. Information on precommercial thinning was unavailable, as was specific information for just spruce-fir species.

Figure 10. Estimated total acres harvested vs harvested using commercial thinning in five Canadian provinces from 1990 to 2014.

Summary of Historical and Current Logging and Forest Fragmentation

Some forestry practices may result in the loss and fragmentation of important Bicknell’s thrush breeding habitat, particularly in the Canadian proportion of the species’ range. Regeneration of clearcuts may be beneficial by creating additional breeding habitat for the species after a lag period of a minimum of 5 to 12 years, but existing information does not assess whether these or adjacent areas become habitat sinks (IBTCG 2010, p. 12; McKinnon et al 2014, p. 268). In contrast, there appears to be strong evidence that precommercial thinning that occurs in occupied breeding habitat renders the area immediately unsuitable for nesting, thereby contributing to the 32

loss of habitat, although the species shows some adaptability if there are sufficient pockets of adjacent suitable habitat (McKinnon et al. 2014, p. 264).

Atmospheric Acid and Nitrogen Deposition in Breeding Areas

During the 20th century, global rates of reactive nitrogen (N) production more than doubled (Vitousek et al. 1997, p. 739), and sulfur dioxide (SO2) emissions increased fivefold (Smith et al. 2011, p. 1109). Deposition of these acids may have several implications for the Bicknell’s thrush and its habitat. First, deposition of these acidic ions is known to reduce soil calcium, which likely leads to calcium deficiencies that render red spruce needles vulnerable to freezing damage. This damage reduces a tree’s tolerance to low temperatures and increases the occurrence of winter injury and subsequent mortality (DeHayes et al. 1999, p. 798). Second, acidic deposition may also increase soil aluminum availability, which may limit the ability of red spruce to take up water and nutrients through their roots (Cumming and Brown 1994, p. 597).

Similar to the deposition of acid, deposition of nitrogen, a major plant nutrient, may affect Bicknell’s thrush habitat. In high elevation spruce-fir forests, nutrient cycling is naturally low due to slower decomposition and low biological nitrogen demand; however, high elevation areas receive greater amounts of atmospheric nitrogen than do low elevation areas (McNulty et al. 1991, p. 16). Several research studies have documented a shift in species vegetation that favors hardwood tree species when montane spruce-fir stands were exposed to naturally occurring and artificially manipulated levels of atmospheric nitrogen (McNulty et al. 1996, p. 109; McNulty et al. 2005, p. 290; Beckage et al. 2008, p. 4201). The resulting vegetation shift towards more hardwoods may be decreasing the quality of foraging or nesting areas for the Bicknell’s thrush (IBTCG 2010, p. 13).

The results of these empirical studies suggest that Bicknell’s thrush breeding habitat within the United States may have decreased as a result of atmospheric acid and nitrogen deposition, which has been identified as a source of historical declines in red spruce and balsam fir in montane habitats, and may have facilitated the establishment of hardwood species throughout the species’ breeding range. However, the Acid Rain Program (ARP), the NOx Budget Trading (NOx), and the Clean Air Interstate Rule (CAIR) cap and trade programs have resulted in significant reductions in emission and deposition of sulfur and nitrogen from approximately 1995 to 2014 (U.S. EPA 2014, entire). This reduction in acid deposition may partially explain recent increased growth rates of red spruce (Vogelmann et al. 2012, entire; Foster and D’Amato 2015, entire; Engel et al. 2016, pp. 89–90). Therefore, based on the observed significant reductions in emissions of sulfur and nitrogen acid compounds and recent expansion of montane spruce fir habitats, along with increased growth of red spruce, we conclude that acid deposition, by contributing to habitat loss and degradation was a stressor to the Bicknell’s thrush, but that it is no longer a stressor.

Summary of Existing Factors Influencing the Bicknell’s Thrush in its Breeding Range

The best available data indicate that cumulative habitat fragmentation and loss due to some logging practices, particularly precommercial thinning and clearcutting conducted during the breeding season, and the development of recreational, telecommunications, and wind energy projects have most influenced the species within its breeding range. While clearcutting can result in regeneration of balsam fir and spruce in these areas which may become suitable Bicknell’s thrush breeding habitat, there can be a minimum of 5 to 12 year lag period before the area becomes suitable. Lastly, the change in tree species composition resulting from the current effects of climate change and atmospheric acid and nitrogen deposition are not currently considered significant influences on the Bicknell’s thrush within its breeding range.

Wintering Range

Historical and Current Forest Conversion

Some incompatible forestry practices have caused the loss and fragmentation of not only Bicknell’s thrush breeding habitat, but also its wintering habitat. Logging, subsistence farming, and human-caused fires have resulted in significant forest loss across the species’ wintering range (Rimmer et al. 2001, p. 4; Rimmer et al. 2005b, p. 228; Townsend and Rimmer 2006, p. 454; COSEWIC 2009, p. 32; IBTCG In prep., p. 26). As discussed above in the Life History and Biology section, the Bicknell’s thrush winters exclusively in the Greater Antilles. Despite the species’ apparent flexibility in the selection of wintering habitat types and elevations, the species has likely been affected by the overall amount of lost forest habitat, including preferred montane forests, because the loss of suitable wintering habitat has been substantial (Butler 2006, entire). For example, Cuba was estimated to be 90 percent forested at the time of Columbus’s arrival in the late 15th century, but by the turn of the 20th century, the forests had been reduced to cover only 5 percent of the island (Butler 2006, entire).

Analysis of global satellite imagery made available through the National Aeronautics and Space Administration (NASA) Landsat program has allowed researchers to quantify forest cover changes (Hansen et al. 2013, entire) across the world. The analysis of these global data show that the tropical forest domain has experienced the greatest change in total forest loss and gain of the four climate domains (tropical, subtropical, temperate, and boreal), as well as the highest ratio of loss to gain (3.6 for greater than 50 percent canopy density), which indicates the prevalence of deforestation (Hansen et al. 2013, p. 850). Data used for this analysis are accessible and being updated at Global Forest Watch (GFW) (www.globalforestwatch.org, accessed February 1, 2017), an interactive online forest monitoring and alert system designed to provide information about forest landscapes. Analysis of the information available at GFW shows that the extent of tree cover across the wintering range of the Bicknell’s thrush has declined over the 15 year period from 2000 to 2014 (see table 2 below).

Table 2. Extent of tree cover on four Caribbean islands occupied by wintering Bicknell’s thrushes in 2000 and the extent of decline for the period 2001 to 2014 (Hansen et al. 2017, entire).

Country Extent Tree Cover Percent Tree Tree Cover Loss Percent (2000) Cover (2001-2014) Tree Cover (2000) (2014) Dominican 2,000,000 ha 47% 209,756 ha 42% Republic (4,942,107 ac) (518,318 ac) Haiti 657,000 ha 24% 28,297 ha 23% (1,623,482 ac) (69,923 ac) Jamaica 730,000 ha 67% 35,077 ha 64% (1,803,869 ac) (86677 ac) Cuba 4,000,000 ha 33% 161,563 ha 32% (9,884,215 ac) (399,230 ac)

Forest cover loss is also occurring in protected areas in the Dominican Republic (Batlle 2015a, entire). Forest cover losses documented in nine protected areas from 2000 to 2014 ranged from 89 square kilometers (km2)(34.4 square miles (mi2), or 12 percent, lost in José del Carmen Ramirez National Park, to 16 km2 (6.2 mi2), or 15 percent, lost in Las Caobas Natural Monument (Batlle 2015a). More specifically, loss of forest cover has affected some habitats that are known to support wintering Bicknell’s thrush. For example, 58 km2 (22.4 mi2) of forest cover was lost at the Sierra de Bahoruco National Park, or 5.33 percent, the majority of which occurred in moisture rich (cloud, wet, and semi-deciduous) forests located on southern slopes (Batlle 2015a, entire; Batlle 2015b, entire). While the Dominican government has established a number of areas (e.g., Sierra de Bahoruco National Park) to protect important forest habitat (Latta et al. 2003, p. 180), habitat loss due to illegal logging and slash-and-burn agriculture continues1 both there and in Haiti (Rimmer et al. 2005b, p. 1; Rimmer et al. 2005b, unnumbered page; Townsend and Rimmer 2006, p. 452; IBTCG 2010, p. 12; Timyan et al. 2012, entire; León et al. 2013, entire; Batlle 2015a, entire; Batlle 2015b, entire; Pasachnik et al. 2016, entire). Furthermore, subsistence farming, involving free-ranging cattle and the presence of feral pigs, severely damages forest understory growth at some wintering sites in Hispaniola and degrades the quality of Bicknell’s thrush wintering habitat (IBTCG 2010, p. 12).

Current Management Efforts in Wintering Areas

The Migratory Bird Conservation Act, the Neotropical Migratory Bird Conservation Act, and the identification of birds of management concern through the Birds of Conservation Concern apply to the Bicknell’s thrush. These regulatory and non-regulatory actions are intended to foster conservation of migratory birds. As such, the USFWS is working with the U.S. Department of State, Canada, and the Dominican Republic Ministry of Environment and Natural Resources (Ministry) to implement a Memorandum of Agreement (MOA) that would provide the structure for formal assistance in addressing the challenges facing the Dominican Republic’s ability to

1 Links to four videos documenting habitat loss in the Dominican Republic (last accessed January 27, 2017): https://www.youtube.com/watch?v=wZJYKeAQYPs; https://www.youtube.com/watch?v=Jh-m__qGaKg; https://www.youtube.com/watch?v=w_6u6hVbyh0; https://www.youtube.com/watch?v=sRGOL7vZZ0Y 35

conserve its natural resources. A Letter of Intent approved by the USFWS’ International Affairs program and the State Department was sent to Canada and the Ministry in early 2016. However, full development of the MOA may still take some time. When completed, it will be a non- regulatory agreement if and when it is implemented (USFWS in litt. 2016; Dettmers in litt. 2016; Gifford in litt 2016).

The Ministry is working on a management plan (Plan) for the Sierra de Bahoruco National Park using an “open standard” conservation process for developing the Plan. The open standard is a structured approach to ranking threats with results chains, conservation targets, and strategies to reduce the threats. Development of the Plan includes participation from surrounding stakeholders and, as a result, should account for some of the socioeconomic factors driving some of the threats to Bicknell’s thrush wintering habitat on the island. The Plan will also include areas for restoration potential. The habitat can regrow if it is not subject to the same deforestation that initially made it unsuitable (Gifford in litt. 2016). The Ministry had hoped to have a Draft Plan completed by the end of calendar year 2016 (Gifford in litt. 2016), but the plan is approximately one-third complete (Dettmers in litt. 2016). In addition, other parks are working on pilot projects to establish sustainable forestry practices adjacent to protected area boundaries. These projects are meant to relieve deforestation pressure on currently protected areas (Gifford in litt. 2016).

Summary of Wintering Range

Logging, subsistence farming, and human-caused fires have resulted in significant forest loss across the species’ wintering range. Analysis of the information available at Global Forest Watch shows that the extent of tree cover across the wintering range of the Bicknell’s thrush has declined over the 15 year period from 2000 to 2014. More specifically, loss of forest cover has affected some habitats that are known to support wintering Bicknell’s thrush. While the Dominican government has established a number of areas (e.g., Sierra de Bahoruco National Park) to protect important forest habitat, habitat loss due to illegal logging and slash-and-burn agriculture continues both there and in Haiti. Under the Migratory Bird Conservation Act and the Neotropical Migratory Conservation Act, the USFWS is working with the U.S. Department of State, Canada, and the Ministry to implement a MOA that would provide the structure for formal assistance in addressing the challenges facing the Dominican Republic’s ability to conserve its natural resources. However, the MOA is still in development. In addition, the Ministry is also working to develop a management plan for the Sierra de Bahoruco National Park and implement projects in other parks to establish sustainable forestry practices meant to protect the parks’ boundaries. In the absence of these fully developed and implemented conservation measures, the Bicknell’s thrush wintering areas continue to be affected by incompatible habitat uses.

Migration Range

We have no substantive data on potential or existing stressors to Bicknell’s thrush migration habitat. As explained above in the Fall Migration Behavior and Habitat section, the species

appears to be a habitat generalist during migration and no areas of frequent, consistent use have been identified.

Summary of Existing Factors Influencing the Bicknell’s Thrush in its Wintering and Migration Ranges

The best available data indicate that loss of habitat due to incompatible logging and subsistence farming is a significant driver influencing the species in its wintering range. We have no information regarding factors that may be influencing the species during migration.

Behavioral Disruptions and Mortality

Predation

Only a few predators of breeding or wintering adult Bicknell’s thrushes have been documented. There are no documented predation events on migrating Bicknell’s thrush.

Breeding Of eight depredation events on radio-tagged breeding adults in Vermont, seven were attributed to the sharp-shinned hawk (Accipiter striatus) and one to the long-tailed weasel (Mustela frenata) (Rimmer et al. 2001, pp. 13–14).

On the breeding grounds, predation of eggs and nestlings by the sharp-shinned hawk, American marten (Martes americana), long-tailed weasel, deer mouse (Peromyscus maniculatus), and blue jay (Cyanocitta cristata) has been documented; however, the red squirrel (Tamiasciurus hudsonicus) is the only predator known to have a significant effect on the demographic characteristics of the Bicknell’s thrush (Wallace 1949, p. 216; COSEWIC 2009, p. 19; IBTCG 2010, p. 6). Wallace (1949, p. 215) suggested that high mortality from red squirrel predation and low breeding rate contributed to the restricted distribution of the Bicknell’s thrush. He noted that 9 of 13 observed nests on Vermont’s Mount Mansfield failed due to predation, while only 2 of the remaining nests were successful at fledging all nestlings. While acknowledging the limitations of his small, 1-year sample size, Wallace (1949, p. 215) concluded that the Bicknell’s thrush population was either stable or declining because the production of 0.85 young fledged per pair constituted a rate at which adults would be unable to replace themselves during two seasons.

Since Wallace’s observations, additional evidence from the United States demonstrates a somewhat loose 2-year (biennial) cycle in nest survival rates on Stratton Mountain and Mount Mansfield, Vermont (Rimmer et al. 2001, p. 19). This Bicknell thrush biennial pattern is attributed to the biennial pattern of balsam fir cone crop production and red squirrel abundance. A fall season with abundant cone production is followed by a spring and summer with high numbers of red squirrels, and results in high nest predation rates and low productivity in the Bicknell’s thrush. In some years, there may be no Bicknell’s thrush young produced (COSEWIC 2009, p. 17). The second part of the biennial cycle is explained when years of abundant cone

production are followed by years when few cones are produced; accordingly, red squirrel numbers drop, along with nest predation rates, and Bicknell’s thrush nesting success can reach as high as 90 percent (Rimmer et al. 2001, p. 19). A similar correlation has not yet been observed in Nova Scotia or New Brunswick (Environment and Climate Change Canada 2016, p. 13).

Climate conditions are known to regulate cone production of northern conifers, and leads to mast seeding with near absolute synchrony among individual trees across extensive landscapes (Koenig and Knops 1998, p. 225–226; Messaoud et al. 2007, entire; Krebs et al. 2012; pp.116– 118). This intermittent phenomenon leads to a high degree of variability in annual cone production, with years of heavy cone production followed by years of relative scarcity. This pattern is strongly influenced by two climatic variables: the maximum temperature of the warmest month; and the sum of growing degree days greater than 5 oC (41 oF) in the year prior to cone production (Messaoud et al. 2007, p. 751). Therefore, the higher maximum temperatures of the warmest month and increased number of total growing degree days greater than 5 degrees resulting from climate change is anticipated to increase cone production. The cyclical seed crop production is known to drive red squirrel population dynamics in boreal habitats (Kemp and Keith 1970, entire; Boutin et al. 2006, pp. 1928–1930; Krebs et al. 2014, p. 74). Specifically, increased cone availability has been shown to advance red squirrel parturition dates, increase litter size, increase growth rate of offspring, and increase the number of second litters (Réale et al. 2003, p. 593; Boutin et al. 2006, pp. 1928–1929).

Wintering

On the wintering grounds in the Dominican Republic, of 53 radio-tagged individuals, 5 were depredated by introduced Norway (Rattus norvegicus) and black rats (Rattus sp.), presumably while the adult birds were sedentary on their nocturnal roosts (Townsend et al. 2009a, p. 565). Lloyd et al. (2016, p. 12) also documented “large populations” of Norway rats at their two study sites in Sierra de Bahoruco.

Summary of Predation

The best available data indicate that, at current levels, predation of adult Bicknell’s thrushes is not a significant source of mortality to the species, although it may influence winter roost site selection (Townsend et al. 2009a, p. 568). Red squirrel predation of eggs and nestlings has been substantial in some years, but is currently considered part of the normal fluctuation in a natural cycle that has persisted since the species was first studied (Wallace 1937, p. 215). However, existing nest predation by red squirrels may increase to a level affecting the species throughout its breeding range if climate change-induced warmer temperatures result in an increase in balsam fir cone production and subsequent red squirrel numbers (see Anticipated Future Changes below).

Mercury

Elevated levels of toxic mercury have been found in Bicknell’s thrush tissue. These findings may be cause for concern (IBTCG 2010, p. 13; IBTCG In prep.) because elevated levels of

mercury have been documented to cause negative physical, behavioral, and reproductive changes in other birds (Evers 2005, entire). Mercury in the northeastern United States and eastern Canada is derived from local, regional, and global emissions, with approximately 60 percent of mercury in this area being derived from sources located within the United States (Evers 2005, p. 5). The toxicity of mercury to a species is largely dependent upon whether it is converted to the bioavailable toxic form known as methylmercury, as well as the species’ trophic position (i.e., its level in the food chain). Trophic position influences mercury exposure through the processes of bioaccumulation (increase in the body over time) and biomagnification (increase in concentration from one trophic level to another) (Evers 2005, p. 6). Generally, a species that is higher in the food chain has a greater exposure to mercury if its food source contains mercury at the time of consumption.

Mercury deposition is highest on high mountain summits in comparison to other landscape positions (Miller et al. 2005, p. 63). Compounding this problem, evergreen foliage generally exhibits higher mercury concentrations than deciduous foliage at the same site due to the longer retention time of needles, as well as the mercury in needles, as compared to leaves on deciduous trees, which are typically shed annually (Miller et al. 2005, p. 62). Consequently, high-elevation montane insectivores like songbirds that consume insects that have fed on evergreens, contain relatively high levels of mercury when compared with other songbirds from low-elevation habitats. Of those montane insectivores, the Bicknell’s thrush has the highest concentrations of mercury, ranging from 0.08 to 0.38 micrograms/grams (ug/g) across 21 distinct breeding sites (Rimmer et al. 2005c, pp. 227, 232). Mercury concentrations in the blood and feathers of Bicknell’s thrushes from southern sections of the species’ breeding range were higher, which implies greater atmospheric deposition rates and increased bioavailability in southern portions of the breeding range (i.e., the United States versus Canada) (Rimmer et al. 2005c, p. 235). In addition, blood mercury concentrations in wintering birds were generally two to three times higher than in birds sampled on their breeding sites (Rimmer et al. 2005c, p. 230).

The specific pathway by which the Bicknell’s thrush consumes mercury, and the effects that burden has on the birds, are unknown (Rimmer et al. 2005c, p. 237; Evers 2005, p. 16). Although species-specific responses to mercury concentrations make direct comparisons unreliable, studies of aquatic birds (e.g., mallard ducks and common loons) indicate changes in behavior, reproduction, and body chemistry are possible (Evers 2005, p. 6; IBTCG 2010, p. 13). Additional research is needed to evaluate the consequences, if any, that mercury exposure may be having on the Bicknell’s thrush (IBTCG In prep.).

Acid Deposition

Acid deposition has the potential to indirectly and directly affect the Bicknell’s thrush. First, acid deposition indirectly affects Bicknell’s thrush by reducing calcium availability that has been shown to influence survival of red spruce, a primary component of the species’ breeding habitat. And second, acid deposition also can directly alter calcium availability for breeding songbirds, which may affect eggshell production (DeHayes et al. 1999, p. 798; IBTCG 2010, p. 13). Acid deposition leaches calcium from forest soils, including soils from many Bicknell’s thrush breeding sites (DeHayes et al. 1999, p. 798; Driscoll et al. 2001, p. 11). This reduction in the

availability of calcium has been linked to declines in some invertebrate prey items and reduced dietary calcium for songbirds through the bioaccumulation and biomagnification processes mentioned above (Mänd et al. 2000, p. 64; Hames et al. 2002, pp. 11238–11239). Although it has not been confirmed, calcium depletion and lower availability may affect egg formation and productivity in the Bicknell’s thrush (King et al. 2008, p. 2697). While these effects may occur to individual Bicknell’s thrush, we have no information to conclude that acid deposition is having a species level effect.

Collision and Disturbance by Stationary and Moving Structures

As described above, the exposed high-elevation areas where suitable Bicknell’s thrush breeding habitat occurs is characterized by high winds and may be targeted for future construction of wind power structures (McFarland et al. 2008, entire). Telecommunication towers are also placed in prominent areas, which may include exposed ridge tops. Construction of these facilities has the potential to remove habitat used by migratory songbirds, including the Bicknell’s thrush, and to increase direct mortality of individuals through collisions, especially with communication towers or the guy wires used to support them (Rimmer et al. 2001, p. 20; USFWS 2003).

Disturbance by Recreationists

Recreational use (hiking and biking) in Bicknell’s thrush habitat may be a stressor to individual Bicknell’s thrushes (IBTCG 2010, p. 12). The White Mountain National Forest in New Hampshire received approximately 1.7 million visits in 2005, and some of those visitors forayed into the high-elevation spruce-fir habitat occupied by the Bicknell’s thrush (USFS 2006, p. 6). By 2010, visitation to the White Mountain National Forest had increased to approximately 2.2 million (USFS 2012, p. 9) and, presumably, there was also a concomitant increase in visits to high-elevation spruce-fir habitats. Similarly, visitation to New York’s Essex County, which includes most of the Bicknell’s thrush’s habitat within the Adirondack Park, received a record 442,000 visitors in 2013 (Lake Placid Convention and Visitor’s Bureau 2014, p. 4). Research suggests that nesting Bicknell’s thrushes are able to tolerate high or moderate levels of human activity by becoming habituated to nearby disturbance, while females in undisturbed areas demonstrate greater sensitivity to disturbance (Rimmer et al. 2001, p. 21). Off-trail excursions by hikers into vegetation that may contain a Bicknell’s thrush is unlikely, given the thick habitat preferred by the species and the fact that these parks have “conveniently opened up many formerly inaccessible areas, and now only a freak ornithologist would think of leaving the trails for more than a few feet” (Wallace 1939, p. 285). Hiking trails occur in a very small portion of the available Bicknell’s thrush nesting habitat and therefore, it seems unlikely that recreational activities in Bicknell’s thrush breeding habitat is a significant stressor to the species.

Summary of Current Factors Influencing the Bicknell’s Thrush

Due to the lack of specific data regarding survival rates by life stage or fecundity rates, we evaluated existing stressor related data and qualitatively assessed the individual and cumulative effects of those stressors on individual Bicknell’s thrush, aggregates of Bicknell’s thrush in the breeding or wintering grounds, and at the species level. From this assessment, we conclude that

habitat loss in the wintering range has most likely been a significant driver of the species’ viability, with the additive effects associated with low productivity in some years due to nest predation from red squirrels also contributing to annual variation in the abundance of the Bicknell’s thrush. For example, loss of wintering habitat in the Caribbean due to forest conversion has been extensive and is ongoing, with no indication that it is likely to be abated, (Rimmer et al. 2001, p. 4; Rimmer et al. 2005b, p. 228; Townsend and Rimmer 2006, p. 454; COSEWIC 2009, p. 32; IBTCG In prep.; Butler 2006, entire; Latta et al. 2003, p. 180; IBTCG 2010, p. 12; Timyan et al. 2012, entire; León et al. 2013, entire; Pasachnik et al. 2016, entire). Contributing factors to the Bicknell’s thrush’s viability include some forestry practices such as precommercial thinning and clearcutting in the Canadian portion of the species’ range, which may result in the loss and fragmentation of important breeding habitat. However, the regeneration of young dense stands of conifers that follows can provide breeding habitat for the species for approximately 5 to 12 years post clearcutting (IBTCG 2010, p. 12; McKinnon et al 2014, pp. 264, 268). The development of ski areas, wind turbines, telecommunication facilities, and their associated infrastructure (i.e., roads and transmission lines) has also resulted in the loss and fragmentation of Bicknell’s thrush habitat (IBTCG 2010, p. 12), but these activities have affected a relatively small proportion of the available Bicknell’s thrush breeding habitat and associated individuals. The species does show some ability to adapt and persist in the vicinity of ski slopes and wind turbines (Rimmer et al. 2004, p. 1; McFarland et al. 2008, p. 56, Parrish 2013, p. 47).

The best available data indicate that the current level of predation of adult Bicknell’s thrushes is not a significant source of mortality to the species (Townsend et al. 2009a, p. 568). However, nest predation by red squirrels can significantly influence nesting success in some years.

Future Factors Influencing the Bicknell’s Thrush

To assess the Bicknell’s thrush’s viability, we must consider what factors may be reliably predicted to influence the species’ into the future.

This section will focus on the stressors likely to currently be having the greatest influence on the species’ viability and how those stressors change over time. For the breeding range, this includes changes in habitat suitability and red squirrel predation. For the wintering range, this includes direct habitat loss.

Anticipated Changing Climate Parameters in the 21st Century

One of the parameters influencing future habitat suitability on the Bicknell’s thrush breeding and wintering grounds is the likely environmental effects related to ongoing and projected changes in climate. The Fifth Assessment Report: Climate Change 2014 (hereafter referred to as AR5), prepared by the Intergovernmental Panel on Climate Change (IPCC) presents the best available science on global climate change (IPCC 2014). The AR5 provides evidence of ongoing changes in climatic patterns and makes projections of the extent of future climate changes and risks associated with impacts at the global scale (IPCC 2014, entire). The key drivers of emissions vary greatly over time and are dependent upon future human population size, economic activity, lifestyle, energy use, land use, technology, and climate policy (IPCC 2014, p. 8). These factors were used to construct Representative Concentration Pathways (RCPs), quantitative descriptions of concentrations of climate change pollutants in the atmosphere over time. Collectively, these scenarios attempt to take into account the full range of uncertainties related to population growth, economic development, implementation of new technologies and other factors that collectively influence atmospheric concentrations of greenhouse gases (van Vuuren et al. 2011, p. 6). The AR5 used the RCP emission scenarios (RCP 2.6, 4.5, 6.0, and 8.5 watts per square meter (m2), respectively) to predict the range of global mean temperature increase that could be expected (IPCC 2014, p.10). The global mean temperature under all emission scenarios is likely to increase within the range of 0.3 °C to 0. 7 °C (0.5 °F to 1.3 °F) in the next two decades (IPCC 2014, p.10), and to 4.8 oC (8.6 °F), by the end of century, depending upon which emission scenario is realized (IPCC 2014, p. 60, see also table 3).

Table 3. Projected change in global mean surface temperature for the mid- and late 21st century, relative to the 1986 to 2005 period (modified from IPCC 2014, p. 60).

In December 2015, the 196 parties to the United Nations Framework Convention on Climate Change (UNFCCC) met in Paris, France and agreed to balance anthropogenic greenhouse gas budgets within the period 2050 to 2100 (UNFCCC 2015, entire). The “Paris Agreement” is a voluntary and nonbinding agreement among the parties that discusses reducing greenhouse gas 42

emissions while taking actions to conserve and enhance sinks and reservoirs that will contribute to reducing atmospheric concentrations (UNFCCC 2015, pp. 5–6). The goal is to hold the increase in “the global average temperature to well below 2 oC (3.6 °F) above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5 oC (2.7 °F) above pre- industrial levels” to reduce risk and impacts associated with climate change (UNFCCC 2015, p. 3). This goal equates to following an emission trajectory similar to the RCP 2.6, the lowest emission scenario used by the IPCC (see table 3 and figure 9) (Sanford et al. 2014, p. 164; Sanderson et al. 2016, p. 7137; Williamson 2016, p. 153). However, the current global trajectory already far exceeds the desired RCP 2.6 (see figure 9), making this lower emission trajectory implausible (Sanderson et al. 2016, p. 7137). Adjusting greenhouse gas emissions to align with the RCP 2.6 scenario will require aggressive reductions and reliance upon unproven negative emission technologies for sequestering atmospheric carbon (Sanford et al. 2014, p. 164; Williamson 2016, p. 153; Sanderson et al. 2016, p. 7138).

Figure 11. Observed and projected trends in global CO2 emissions under four RCP scenarios (figure from Sanford et al. 2014, p 165).

Projected Future Effects of Climate Change to Bicknell’s Thrush Breeding Habitat

Although it is virtually certain the future will see warming temperatures, predictions indicate there will be regional variations, including ‘very high confidence’ that the Artic and continental 43

areas will warm at levels above the global averages (IPCC 2014, p. 60). Due to this and other variations in regional and local responses in climate, we evaluated “downscaled” projections for the northeastern United States with higher resolution information on the effect of global climate change within the breeding range of the Bicknell’s thrush.

Downscaled models using the IPCC’s Special Report on Emission Scenario (SRES) high (A1F1), mid-high (A2), and lower (B1) emissions scenarios were used to project future climate parameters for the northeastern United States (Hayhoe et al. 2007, p. 383; Kunkel et al. 2013, pp. 34–35; see table 4). Like all scenarios, these SRES projections are sensitive to changes in population, demographics, technology, international trade, other socio-economic factors, and policies that will influence the emission of greenhouse gases by end of century (i.e., by the year 2100) While there is uncertainty in calculating these projections, the results of these scenarios indicate the average annual regional surface temperature in the Northeast will likely increase by 2.9 oC (5.2 °F) under the B1 scenario, by 4.5 oC (8.1 °F) under the A2 scenario, and 5.3 oC (9.5 °F) under the A1F1 scenario by the end of the century (Hayhoe et al. 2007, p. 388).

The U.S. Department of Commerce’s National Oceanic and Atmospheric Administration, using more advanced climate models, predicted even greater increases in average annual regional temperature values for the Northeast United States (Kunkel et al. 2013, pp. 34–35). Their analysis predicted the mean temperature for the Northeast region is expected to increase 3.5 oC (6.3 °F) to 5.5 oC (9.9 °F) for B1 and 6.5 oC (11.7 °F) to 8.5 oC (17.8 °F) for A2 toward the end of the 21st century (2085) (Kunkel et al. 2013, p. 35).

In addition to temperature changes, examples of other climatic factors that may change under these scenarios include: (a) a decline in mean summertime rainfall across northern New England; (b) increased winter precipitation; (c) changes in daily rainfall rates; and other temperature variables, like the number of days with temperatures exceeding 35 oC (95 oF) and the length of the freeze-free season (Hayhoe et al. 2007, pp. 433–434; Kunkel et al. 2013, pp. 34–62; Bryan et al. 2015, entire). Statistical downscaling based on RCP scenarios yields similar results (Ning et al. 2015, entire).

Table 4. Summary cross walk of regional projections in average annual temperature for the northeastern United States.

Reference B1 A2 A1F1 Hayhoe et al. 2007 2.9 oC 4.5 oC 5.3 oC (5.2 oF) (8.1 oF) (9.5 oF) Kunkel et al. 2013 3.5 to 5.5 oC 6.5 to 8.5 oC Not evaluated (6.3 to 9.9 oF) (11.7 to 17.8 oF)

We have assessed the best available information pertaining to future projections in global and regional temperatures and conclude that regional temperatures are very likely to increase above the global average. Although the extent of the projected increase in temperature varies by emission scenario and time, the best available information suggests the regional increase will almost certainly exceed the 1.5 oC (2.7 °F) target set forth in the Paris Agreement (Karmalkar

and Bradley 2017, entire). The current observed increase in global mean temperature indicates that the projections of RCP 4.5 and the similar values provided by the SRES B1 scenario are most likely, meaning that the regional average temperature increase is predicted to exceed 2.9 oC (5.2 °F) by the end of the 21st century. Below we discuss what specific changes to Bicknell’s thrush habitat may be likely to occur as a result of the most likely temperature forecast.

Tree Species Composition

The projected climate change factors are anticipated to affect biological resources (Hayhoe et al. 2007, entire), including those that are relevant to the Bicknell’s thrush. However, the interaction between a species’ habitat and a species’ distribution is a complex ecological process. The best available models may not be able to fully incorporate these biological processes or those processes may not fully play out within the time frames of the climate change scenarios (i.e., by the end of this century). The results of the models presented below are efforts to project future tree or bird species distributions and indicate where the climate conditions these species currently occupy will be located (or the change in extent of those conditions across the region) in the future, but there can be increasing amounts of uncertainty with predicting where these species will actually occur at a particular time in the future. For example, underlying physical conditions (e.g., soil and bedrock) and microclimates, especially in high mountain areas, can mediate forecasted changes that are based on climate variables collected at lower elevations (Seidel et al 2009, entire; Dobrowski 2010, entire; Morelli et al. 2017, entire). There may also be an unknown time lag between changes in climate conditions and changes in the actual distribution of tree or bird species or even have initial shifts in species’ distribution that are counterintuitive to forecasted shifts (DeLuca and King 2016, unnumbered). In addition, there are aspects of how adaptable species will be to new conditions, if habitat management can mediate the effects of climate change, or how interspecific interactions might influence distribution that may not be fully addressed by these models. While acknowledging this uncertainty, the information below represents the best available data with respect to climate mediated effects to Bicknell’s thrush habitat and our interpretation of how the species may be affected.

For spruce-fir habitats within the range of the Bicknell’s thrush, the consequence of climate change is expected to manifest into a shift in tree species composition and a decline in spruce-fir habitat (Iverson and Prasad 1998, entire; Iverson et al. 1999, entire; Iverson et al. 2008, entire; Bourque 2010, entire; Bourque 2015, entire; Iverson et al. 2016, entire; Wang et al. 2016, entire; Périé and de Blois 2016, entire). In part, this decline is attributed to climatic shifts away from optimal growing conditions favored by balsam fir, which includes areas with a mean temperature of 2 oC (3.6 °F) to 4 oC (7.2 °F), a January average temperature ranging from -18 oC (-0.4 °F) to - 12 oC (10.4 °F), a July mean temperature ranging from 16 oC (60.8 °F) to 18 oC (64.4 °F), and a mean annual precipitation ranging from 760 to 1100 millimeters (mm) (29.9 to 43.3 in) (Bakuzis and Hansen 1965 In Burns and Honkala 1990, unnumbered). Similarly, red spruce grows best in cool, moist climates characterized by annual precipitation ranging from 910 to 1,320 mm (35.8 to 52.0 in), January maximum temperature ranging from -7 oC (19.4 °F) to -1 oC (30.2 °F), January minimum temperature ranging from -18 oC (-0.4 °F) to -13 oC (8.6 °F), and July maximum temperature ranging from 21 oC (69.8 °F) to 27 oC (80.6 °F) (Lull 1968 in Burns and Honkala 1990, unnumbered).

Researchers with the U.S. Department of Agriculture’s Forest Service have conducted several assessments to quantify the extent of climate change impacts to various tree species occurring in the Eastern United States (Iverson and Prasad 1998, entire; Iverson et al. 1999, entire; Iverson et al. 2008, entire; and Iverson et al. 2016, entire). These efforts included an analysis that modeled and mapped the response of 134 tree species currently occurring within the Eastern United States under the A1F1 and B1 emission scenarios (Iverson et al. 2008, p. 489). The results revealed red spruce and balsam fir will be among the most affected species in the Northeast, with the spruce- fir forest type likely to persist only under the most conservative climate scenario (e.g., currently not the most likely) (Iverson et al. 2008, pp. 489, 499–500). In 2015/2016, multi-model efforts to evaluate the effects of climate change scenarios projected large future reductions and the potential for near extirpation of spruce and fir, even under the B1 emission scenario (Iverson et al. 2016, pp. 12, 18; Wang et al. 2016, p. 12). Similarly, climate envelope models for red spruce and balsam fir project an almost complete loss of suitable habitat for these species from the northeastern United States by 2090 under the RCP 6 scenario, with those species possibly persisting in a few high elevation refugia (D’Amato 2016, p. 2).

The decline of spruce and fir within the current breeding distribution of the Bicknell’s thrush may result in declines in the amount of available suitable breeding habitat (Rodenhouse et al. 2008, entire; IBTCG 2010, p. 6; D’Amato 2016, entire). Rodenhouse et al. (2008, p. 525) suggest that because the occurrence of this habitat type is primarily controlled by climate-driven factors (e.g., temperature, number of days colder or wetter than normal), the projected warming trend has the potential to alter the abundance and suitability of the species’ breeding habitat, suggesting a change in the distribution and abundance of the Bicknell’s thrush would follow. To evaluate the consequences of climate change to Bicknell’s thrush habitat, Rodenhouse et al. (2008, p. 525) evaluated the potential impacts of a warming climate on modeled Bicknell’s thrush breeding habitat within the United States. Based on their results, regional warming of 1 °C (1.8 °F) is expected to reduce Bicknell’s thrush habitat by more than one half, while a 2 °C (3.6 °F) increase may result in the elimination of all breeding sites from the Catskill Mountains and most of Vermont. With a 3 °C (5.4 °F) increase, nearly all Bicknell’s thrushes will be eliminated from the northeastern United States, with only a few small patches (total of 75 ha) persisting through a 5 °C (9 °F) increase (Rodenhouse et al. 2008, p. 526). Since the B1 scenario is predicted to increase the annual regional surface temperature average in the Northeast by 2.9 °C (5.2 °F), it appears likely that breeding habitat within the United States for the Bicknell’s thrush will be nearly eliminated by the end of the century. Although D’Amato (2016, entire) does not quantify the extent of habitat loss under the RCP scenarios, however he projects the substantial loss of spruce fir habitat in the northeastern United States by 2090.

Some projections are similar for spruce and fir habitats within the Canadian portion of the Bicknell’s thrush’s breeding range. Currently, 86.5 percent of the Province of New Brunswick provides potential habitat for balsam fir (Bourque 2015, pp. 20–21). However, one study projects a decline in the extent of balsam fir to 17.5 percent to 2 percent of the province under the RCP 4.5 and 8.5 scenarios, respectively, by 2100 (Bourque 2015, p. 20–21). Changes in biophysical variables (e.g., temperature, precipitation, duration of the growing season) resulting from climate change are projected to result in similar declines in Nova Scotia, where potential

habitat for balsam fir is projected to decline from dominance in covering 99 percent of the province to only 7 percent by 2100 (Bourque et al. 2010, p. 24). Similar declines are projected for balsam fir habitats located in southern Quebec (Périé and de Blois 2016, entire). However, any actual reductions and changes in distribution may occur over longer timeframes or vary regionally due to the intricacies of ecological processes (including the potential for localized areas of overgrazing by moose (Environment and Climate Change Canada 2016, p. 21)), anthropogenic factors, and latitude (Boulanger 2016a, pp. 3, 11, 14).

More recently, Tremblay et al. (In prep. 2017) used a Landscape Disturbance and Succession (LANDIS-II) model that incorporated information from climate scenarios and habitat disturbances (i.e., fire, spruce budworm outbreaks, wind throw, and forest harvest methods) to project spatial and temporal changes to Bicknell’s thrush breeding habitat in Canada. The authors’ preliminary findings indicate that, under the RCP 2.6 emission scenario, using forest management strategies may help to maintain most Bicknell’s thrush suitable habitat at low elevations (i.e., less than 750 m (2,461 ft) in New Brunswick and less than 900 m (2,953 ft) in Quebec). Under the RCP 4.5 and RCP 8.5 scenarios, low elevation suitable habitat would likely decrease, but approximately 49 percent and 20 percent, respectively, of the suitable habitat would remain. Higher elevation breeding habitat (i.e., greater than 750 m (2,461 ft) in New Brunswick and greater than 900 m (2,953 ft) in Quebec), regardless of forest harvest strategies, may increase by approximately 43 percent under RCP 2.6 and by 22 percent under RCP 4.5 or decrease such that 87 percent remains under RCP 8.5 (Tremblay in litt. 2017; Tremblay et al. In. Prep. 2017). It is unclear why the LANDIS-II model predicted an increase in higher elevation habitat under two out of the three climate scenarios. The observed discrepancy between modeling efforts by Bourque (2015), Iverson et al. (2016), and Wang et al. (2016) and Tremblay et al. (In prep. 2017) indicates the inherent uncertainty in projecting the extent to which Bicknell’s thrush breeding habitat and the species itself may be influenced by the effects of climate change.

Forest Pests

In addition to the departure of climatic conditions away from those that promote optimal spruce and fir growth, other secondary effects resulting from climate change are expected to contribute to declines in the availability of Bicknell’s thrush breeding habitat. One of those anticipated changes is an increase in exposure to insect outbreaks (Candau and Fleming 2011, entire; Bouchard and Auger 2014, entire). The spruce budworm (Choristoneura fumiferana), the caterpillar of a native species of lepidopteran (i.e., moths), is the most problematic native pest of spruce and fir in the Northeast; this forest pest is capable of substantially modifying large areas of boreal forest (Fleming and Candau 1998, p. 236). The spruce budworm is a naturally outbreaking insect that can be extremely abundant for periods of 5 to 15 years, with populations reaching 108 fourth instar larvae per ha (greater than 40 million per ac), and can kill most trees in dense, mature balsam fir stands (Fleming and Candau 1998, pp. 236, 237; Gitay et al. 2001, p. 291). Historically, periods of abundant spruce budworm populations were followed by periods of 30 to 60 years when they were relatively rare (Wagner et al. 2016, p. 3). The forest management agencies in the northeastern United States and Canada annually monitor the level of spruce budworm infestation. The most recent reports indicate that the region may be

approaching the next significant outbreak (Natural Resources Canada 2016, entire; Maine Forest Service 2017, pp. 1, 5, 8; New Hampshire Department of Resources and Economic Development 2016, p. 3; Wagner et al. 2016, pp. 2–3, 12–13).

Budworm outbreaks frequently follow droughts or hot, dry summers that have been linked to increased egg production and disruptions in the synchrony of the budworm and several of its parasitoid predators, thereby resulting in increased population growth potential (Gitay et al. 2001, p. 291). Some predictive models suggest the distribution of spruce budworm outbreaks is likely to shift northward and towards higher elevations over the next 50 years in response to climate change (Régnière et al. 2012, p. 1580). This phenomenon is likely already occurring, as evidenced by observations of developing outbreaks in Quebec at locations farther north than during outbreaks observed in the 19th and 20th centuries (Régnière et al. 2012, p. 1580; Natural Resources Canada 2016, p. 1; Forest Fauna and Parks Quebec 2017). This shift in areas of infestation may result in a reduction in the prevalence of spruce budworm outbreaks in the warmer parts of the insect’s range, although there is some uncertainty with these predictions (Boulanger 2016b, entire). This includes those areas currently within the range of the Bicknell’s thrush, since increasing temperatures are expected to result in higher overwinter mortality to the insect (Régnière et al. 2012, p. 1581). Therefore, climate change is expected to affect spruce budworm populations by altering any of the relationships among host tree species, the budworm, and its natural enemies (Fleming and Candau 1998, p. 236), but the timing and extent to which this happens is uncertain (Boulanger 2016b, entire).

In the shorter term (next several decades), the Bicknell’s thrush may benefit from increased habitat disturbance resulting from increased spruce budworm infestations and associated high spruce and fir mortality, since this will likely lead to increased availability of suitable young and dense habitats (COSEWIC 2009, p. 10; Bredin and Whittam 2009, p. 13). Additionally, the species may benefit from increased forage opportunities since, as described above in the Life History and Biology section, the Bicknell’s thrush feeds upon lepidopteran larvae (Wallace 1939, p. 295), which may include the spruce budworm. However, in the long term, one potential consequence of climate change-induced intensification of spruce budworm outbreaks may be local extinctions of balsam fir (Fleming and Candau 1998, p. 246) and the associated loss of Bicknell’s thrush breeding habitat, while another consequence could be a northward shift in outbreaks but a decrease in the duration of outbreaks (Boulanger 2016b, entire).

Bicknell’s thrush habitat is also affected by the balsam woolly adelgid (Adelges piceae), an exotic pest of fir trees that was introduced from central Europe and is now known to occur throughout most of the northeastern United States (Iverson et al. 1999, p. 176; Ragenovich and Mitchell 2006, entire). In some localities in the United States, such as the southern Appalachians, the balsam wooly adelgid has been implicated in the elimination of firs, and the spread of this pest into previously uninfested areas has been documented (Ragenovich and Mitchell 2006, p. 1). Weather is an important factor in the survival of this insect because, in cold winters, only those adelgids below the snowline will survive temperatures of below -1oC (30oF) (Ragenovich and Mitchell 2006, p. 9). Furthermore, only the first instar can survive the winter. In montane spruce-fir habitats, the season may be too short for this insect to complete a second generation (Ragenovich and Mitchell 2006, p. 9), which may afford some protection to high-

elevation Bicknell’s thrush breeding habitat. There is the potential for future deleterious effects on Bicknell’s thrush breeding habitat quality from the woolly adelgid (Lambert et al. 2005, p. 7; IBTCG 2010, p. 14) as winter temperatures increase. However, the timing and extent of any potential loss is complicated by a complex set of ecological factors related to species interactions, climatic response to growth, including precipitation, growing season length, hydrology, temperature, pest dynamics, and fire regimes. In addition, like the spruce budworm, the loss of balsam fir will likely diminish the potential effect of wooly adelgids.

The hemlock looper (Lambdina fiscellaria) is another native insect to the United States and Canada that feeds on, and could result in declines of, hemlock, balsam fir, and white spruce (USDA Forest Service undated, entire; Natural Resources Canada 2015, entire). Outbreaks can occur, spread, and end quickly (Natural Resources Canada 2015, entire). Outbreaks of the hemlock looper occurred for the first time inland of Québec and killed mostly balsam firs and spruces (Tremblay in litt. 2017). We are unaware of any predictive models for the spread or duration of hemlock looper outbreaks, but assume that the species may follow similar patterns and have similar effects on Bicknell’s thrush habitat as discussed above under spruce budworm and woolly adelgid.

Fire Regimes

Weather and climatic changes are expected to alter fire regimes across many habitats, including the spruce-fir forests that provide breeding habitat for the Bicknell’s thrush (Parisien and Moritz 2009, entire; Boulanger et al. 2013, entire; Wang et al. 2015, entire; Kerr and DeGaetano 2016, entire). For example, an analysis of the projected changes in temperature, precipitation, wind speed and relative humidity under the SRES A2 scenario indicates unabated emissions will result in increased fire across most of the area covering the Canadian Provinces of New Brunswick, the southern portion of Quebec, and most of Ontario (Boulanger et al. 2013, entire). Similarly, the projected changes in climatic variables related to fire are expected to increase exposure across the northeastern and Great Lakes States (Kerr and DeGaetano 2016, entire). Attributes such as easily ignitable, thin, resinous bark and shallow roots make the balsam fir the least fire-resistant conifer in the northeastern United States (Uchytil 1991, entire). Balsam fir seed mortality can be high, which hinders reestablishment of this species in burned areas (Uchytil 1991, entire). Consequently, an increase in fire across the Bicknell’s thrush’s breeding range will likely contribute to declines in habitat.

Summary of Projected Future Effects of Climate Change to Bicknell’s Thrush Breeding Habitat

The projected climate change factors are anticipated to affect biological resources, including those that are relevant to the Bicknell’s thrush. However, the interaction between a species’ habitat and a species’ distribution is a complex ecological process. This process is influenced by factors such as soil type, air and soil temperature, precipitation, exposure, elevation, and natural and human mediated disturbance regimes. Depending on the emission scenario and habitat elevation, some models predict a near complete elimination of suitable Bicknell’s thrush breeding habitat by the end of the century while others suggest suitable breeding habitat may

increase or can be stabilized with forest harvest strategies. Therefore, it is unclear to what extent the Bicknell’s thrush may be affected by future climate change or the exact timing of when the projected changes may manifest in its habitat and then in the species itself.

Increased Interspecific Competition Due to Climate Change

The widespread Swainson’s thrush (Catharus ustulatus) occurs in many breeding habitat types occupied by the Bicknell’s thrush and has been identified as a congeneric competitor (Freeman and Montgomery 2016, entire). However, in montane habitats, the Bicknell’s and Swainson’s thrushes generally inhabit mutually exclusive elevation ranges, as evidenced by a study conducted at locations in the Adirondack and Green Mountains, that found Bicknell’s and Swainson’s thrushes inhabiting mutually exclusive ranges with little overlap and the Bicknell’s thrush in higher elevations (Able and Noon 1976, p. 289). Swainson’s thrushes were found cohabitating with Bicknell’s thrushes in lower elevation habitats among regenerating stands following commercial forestry operations in New Brunswick and Quebec (Nixon et al. 2001, p. 34; Aubry et al. 2016, p. 301). Observations of the two species demonstrate occasional antagonistic encounters on the breeding grounds, including chases and displacement of Bicknell’s thrushes by Swainson’s thrush from song-posts (Able and Noon 1976, p. 287; Rimmer et al. 2001, p. 13). Additional studies conducted in the Adirondack Mountains found interspecific aggression, with Swainson’s thrush dominating interactions (Freeman and Montgomery 2016 p. 174). The consequences of interspecific interactions on Bicknell’s thrushes have not been fully evaluated, but may negatively influence habitat use, productivity, foraging, and home range size (Freeman and Montgomery 2016, p. 175; Aubry et al. 2016, p. 307).

The Swainson’s thrush is known to use stands below the spruce-fir ecotone, where northern hardwoods are relatively common (Noon 1981). Consequently, while temperature may be an important factor in the distribution of these two thrush species (Holmes and Sawyer 1975 in Nixon et al. 2001, p. 38), invasion of high-elevation spruce-fir habitats by northern hardwoods (Beckage et al. 2008, p. 4197) may simultaneously enhance habitat characteristics for the Swainson’s thrush while degrading habitat quality for the Bicknell’s thrush. The Bicknell’s thrush is considered, based on metabolic studies (Holmes and Sawyer 1975 in Nixon et al. 2001, p. 38), to be better adapted to colder environments than is the Swainson’s thrush. Lambert et al. (2005, p. 7) theorized that a rise in summer temperatures could reduce separation between the two species by nullifying the Bicknell’s thrush’s greater tolerance for cold, thereby facilitating the establishment of the Swainson’s thrush at higher elevations. Later observations of increased occurrence of Swainson’s thrush in high-elevation habitats formerly occupied exclusively by the Bicknell’s thrush may support Lambert et al.’s theory (Aubry et al. 2016, p. 307), thereby suggesting that the observed warming trends have allowed the Swainson’s thrush to invade higher elevation habitats. In New Brunswick, in areas where Bicknell’s thrush observations declined by 11.5 percent annually, observations of Swainson’s thrush increased by 8.9 percent (Campbell and Stewart 2012, p. 7). Due to competition, increased co-occurrence with the Swainson’s thrush may result in reduced suitable habitat available to support the Bicknell’s thrush (Aubry et al. 2016, p. 307; Freeman and Montgomery 2016, p. 175). However, there is uncertainty as to when habitat may shift or the extent to which the Bicknell’s thrush or Swainson’s thrush may respond to the habitat shifts.

Climate Induced Increase in Predation by Red Squirrels

As mentioned above, the cyclical seed crop production is known to drive red squirrel population dynamics in boreal habitats (Kemp and Keith 1970, entire; Boutin et al. 2006, pp. 1928–1930; Krebs et al. 2014, p. 74). The relationship between cone production and climatic conditions has led some researchers to hypothesize an increase in masting by balsam fir as a consequence of the effects of climate change, which may, in turn, translate into higher rates of red squirrel predation and reduced productivity in the Bicknell’s thrush (Messaoud et al. 2007, p. 753; IBTCG In prep.). The relationship between changing climate and cascading increases in cone production and shifts in parturition dates for red squirrel populations has been demonstrated for southwest Yukon Territory in Canada (Réale et al. 2003, pp. 593–594); consequently, we conclude the projected future change in climate may increase red squirrel predation and reduce Bicknell’s thrush productivity because of the predictions of increased masting in balsam fir (Messaoud et al. 2007, entire).

Summary of the Effects of Climate Change to Breeding Habitat

The best available information indicates that there may be both shorter and longer term effects to Bicknell’s thrush breeding habitat, and thus to the species itself, as a result of climate change. The current observed increase in global mean temperature indicates that the regional average temperature is predicted to exceed 2.9 oC (5.2 °F) by the end of the 21st century. The projected climate change factors are anticipated to affect biological resources important to the Bicknell’s thrush. For example, warming temperatures may result in an eventual shift in tree species composition that favors hardwood species, and thus cause a decline in spruce-fir habitat within the species’ current breeding distribution, but the extent to which this happens, and when, will be influenced by both ecological and human mediated factors. Warming temperatures are anticipated to create favorable conditions in the shorter term for insect outbreaks of spruce budworm and woolly adelgid, and perhaps the hemlock looper, before potentially eliminating the pests’ preferred balsam fir habitat in the longer term, and create conditions favorable (i.e., increased temperature, decreased precipitation and relative humidity, and increased wind speed) for increasing fire frequency. In the shorter term (next several decades), the Bicknell’s thrush may benefit from increased habitat disturbance resulting from increased spruce budworm infestations and associated high spruce and fir mortality, since this will likely lead to increased availability of suitable young and dense habitats.

In addition to the eventual decline in habitat availability, climate change is anticipated to affect the quality of the remaining Bicknell’s thrush habitat through increased interspecific competition with Swainson’s thrush and red squirrel predation. As mentioned above, rising temperatures are anticipated to facilitate the establishment of northern hardwoods into high-elevation spruce-fir habitats. Warming of high elevation habitats and the potential concomitant shift in tree species composition may in turn enhance the availability of suitable habitat for the Swainson’s thrush, a competitor of the Bicknell’s thrush for space and food. Lastly, warming temperatures, in the shorter term, are likely to improve growing conditions and promote balsam fir cone production (i.e., increase in maximum temperature in the warmest month and increasing the number of growing degree days greater than 5 oC (41 oF) in the prior year). Increased cone production

results in an abundance of food for red squirrels, a known nest predator of the Bicknell’s thrush, thereby increasing red squirrel abundance, which is expected to decrease Bicknell’s thrush productivity. Thus, the Bicknell’s thrush may be affected by the effects of climate change in the short and long term due to reduced habitat availability and suitability, but to what extent and where those changes happen will likely vary depending on habitat type (i.e., low vs high elevation) and management strategies.

Anticipated Effects of Climate Change to Wintering Habitat

Downscaled climate projections have been generated for the islands in the Caribbean based on the A1F1, A2, A1B, B2, and B1 emission scenarios and the RCP 8.5 and RCP 4.5 (Nurse et al. 2014, p. 1628; Hayhoe 2013, entire; Khalyani et al. 2016, entire). At a regional scale, these projections show increases in mean and extreme temperatures along with an overall drying of the region, with results varying by scenario and model selection (Hayhoe 2013, p.35; Nurse et al. 2014, p. 1628; Khalyani et al. 2016). The finescale spatial variability associated with steep climate gradients and multiple land use practices complicates the assessment of effects associated with future scenarios across the Caribbean (Khalyani et al. 2016, p. 266). There is also a paucity of independent peer-reviewed scientific publications providing downscaled climate projections for tropical and subtropical islands (Nurse et al. 2014, p. 1628) and this complicates the ability to accurately assess future scenarios relevant to the Bicknell’s thrush with high confidence. Although limited, the best information available indicates the effects associated with climate change may be problematic for the Bicknell’s thrush. Results of an investigation using three IPCC emission scenarios (A2, A1B, and B1) project a shift among nine life zones in Puerto Rico, with a switching from humid to drier life zones through this century (Khalyani et al. 2016, p. 277). These changes in life zones show a shift in climatic conditions that result in the complete loss of subtropical rain, moist, and wet forests (Khalyani et al. 2016, p. 277). Since the wet, broadleaf forests provide the majority (82 percent) of Bicknell’s thrush wintering habitat in the Caribbean (McFarland et al. 2013, p. 3), the consequence of drying in the Caribbean and the associated loss of this habitat type could significantly reduce or eliminate the species’ wintering habitat. We have no information to evaluate whether the species could or would adapt to shifts in its habitat in a timeframe that matches or exceeds the projected changes.

We anticipate not only direct impacts but also indirect impacts of climatic change on Bicknell’s thrush wintering habitat. Specifically, small islands are reliant on important imports, such as food and fuel (e.g, wood and charcoal), because the islands have limited natural resources. However, if island nations become more financially challenged, their ability to import needed resources is expected to decline, and demand for domestic resources will likely increase (Nurse et al. 2014, p.1625; Briguglio et al. 2008, p. 5). The volatility of these economies is expected to result in increased pressure on existing island resources (Nurse et al. 2014, entire), which will likely include additional forest resources being converted to cropland and fuel. Climate change factors, such as sea level rise, will further decrease the amount of available land. Together, these climatic and socioeconomic factors will likely result in the further degradation and loss of Bicknell’s thrush wintering habitat.

Summary of Future Factors Influencing the Bicknell’s Thrush

The best available information suggests that, as a result of climate change, the spruce-fir habitat that support breeding Bicknell’s thrushes, may be substantially reduced, with the potential to be nearly eliminated, from the species’ current range in the northeastern United States and may decline in Canada by the end of this century, depending on the amount of amount of green-house gases emitted to the atmosphere, habitat type (i.e., low vs. high elevation) and forest harvest management strategies. The effect of climate change may also result in an increase in competition between the Bicknell’s and Swainson’s thrushes, at the expense of the Bicknell’s thrush, and an increase in predation from red squirrels. On the wintering grounds, the consequences of climate change will likely include a drying of the Caribbean region and an associated decline in the wet montane and cloud forest habitats where most Bicknell’s thrushes are found. It is also likely that socioeconomic pressures, especially in the Dominican Republic and Haiti, will result in further losses of the species’ preferred habitat, as forests are converted to other land uses. We acknowledge that there is uncertainty with how quickly and to what extent suitable Bicknell’s thrush habitat will change and further uncertainty with how the Bicknell’s thrush may respond to any shifting or outright loss of suitable habitat. But, the best available data suggest that the net effect of climate change, in addition to the projected loss of forested habitat in the wintering area, on the Bicknell’s thrush is expected to be negative in the long term.

FUTURE SCENARIO CONSIDERATIONS

The viability of the Bicknell’s thrush depends on maintaining suitable breeding and wintering habitat that is capable of supporting multiple resilient populations over time. Given the uncertainty over the expected projections from the multiple best available climate models, as well as the potential climate policy mitigations that may occur in the future, we attempted to forecast the range of what habitat availability for the Bicknell’s thrush could potentially look like over the next 50 to 83 years (i.e., to the end of the century (2100)). We’ve chosen four potential scenarios, three of which use a different climate projection, and made assumptions about effects to the species’ habitat based on information published in the scientific literature and from Bicknell’s thrush, avian ecology, phenology, and climate experts. These scenarios (see summary below and table 5) do not include all possible futures, but rather include several potential scenarios that represent examples from the continuous spectrum of possible futures. The Non- Climate Stressor scenario is used to evaluate the habitat loss, predation, and competition stressors separate from the effects of climate change. The Optimistic scenario is less plausible than the Low and High Warming scenarios due to the current rate of increase in global emissions and average surface temperatures.

1. Non-Climate Stressor Scenario— projects the current non-climate stressor levels in isolation from the potential effects of climate change discussed in the other scenarios.

2. Optimistic Scenario—the annual global mean surface temperature will increase by less than 2 oC (3.6 oF) above pre-industrial levels, the average regional surface temperature in the Northeast will increase by less than 2 °C (3.6 oF), and the average regional surface temperature in the Caribbean will increase by 2.9 oC (5.2 °F) under the RCP 2.6 scenario and applicable downscaled regional models.

3. Low Warming Scenario—the annual global mean surface temperature will increase by 1.8 oC (3.2 °F), the average regional surface temperature in the Northeast will increase by 3 oC to 5 oC (5.4 °F to 9 °F), and the average regional surface temperature in the Caribbean will increase by 5.3 oC (9.5 °F) under the SRES B1 and RCP 4.5 scenarios and applicable downscaled regional models.

4. High Warming Scenario—the annual global mean surface temperature will increase by 3.7 oC (6.6 °F), the average regional surface temperature in the Northeast will increase by 5.3 oC to 6 oC (9.5 °F to 10.8 °F), and the average regional surface temperature in the Caribbean will increase by 3.7 oC (6.7 °F) under the SRES A1F1 and RCP 8.5 scenarios and applicable downscaled regional models.

Table 5. Summary of Future Scenarios (full discussion of parameters on pages 42-60).

Scenario Timeframe Climate Future Model Conservation Measures Breeding Habitat Wintering Habitat Name 50-83 years (i.e., Not applicable: scenario addresses only non-Developing and implementing best Canada: No change in suitable habitat which continues to Deforestation in the Caribbean 2100, end of the climate influences. management practices (BMP’s) for support approx. 33% of breeding population. United States: continues at the current rate; the century) timber companies and establishing the No change in suitable habitat which continues to support Dominican Republic will continue to Non-Climate Memorandum of Agreement (MOA) approx. 66% of breeding population. Both: Periodicity of red lose its tree cover at the rate of Stressors between Canada, the U.S., and the squirrel predation and competition with Swainson's thrush approximately 5 percent every 15 Scenario Dominican Republic (DR) remain remains at current levels. years due to land use, economics, desired goals. and lack of enforcement of protected areas.

50-83 years (i.e., Most closely resembles the RCP 2.6 Significant curtailment of global Canada: In low elevation habitat (<900 m), forest harvest Approximately 18% of wintering 2100, end of the scenario with regional variation. Limit emissions plus implementation and strategies could help maintain most habitat such that approx. habitat remains. century) increase in global average surface effectiveness of negative emissions 82% remains. In higher elevations (>900 m), regardless of temperatures to below 2 °C (3.6 °F) above technologies (as assumed under RCP forest harvest strategies, habitat may increase by approx. Optimistic pre-industrial levels, but in the Caribbean, 2.6). Developing and implementing 43%. United States: An eventual shift in tree species Scenario the average surface temperature will BMP’s for timber companies; MOA composition projecting to result in a decrease in the extent of increase by 2.9 °C (5.2 °F). between Canada, the U.S., and the DR spruce and balsam fir. Both: The periodicity of red squirrel is finalized & implemented; logging predation and competition with Swainson's thrush likely within DR parks is eliminated. increases. 50-83 years (i.e., SRES B1 and RCP 4.5 with regional Some curtailment of global emissions, Canada: In low elevation habitat (<900 m), regardless of Projected complete loss of wintering 2100, end of the variations. Mean annual global surface as assumed under SRES B1 and RCP forest harvest strategies, suitable habitat could decrease, habitat. century) temperatures increase by 1.8 °C (3.2 °F), 4.5. Developing and implementing leaving approx. 49% remaining. In higher elevations (>900 the average regional surface temperature in BMP’s for timber companies; MOA m), regardless of forest harvest strategies, habitat may Low Warming the Northeast will increase by increase by between Canada, the U.S., and the DR increase by approx. 22%. United States: There is the potential Scenario 3 °C to 5 oC (5.4 °F to 9 °F), and the is established; logging within DR parks for most Bicknell’s thrush habitat to be eliminated, with small average regional surface temperature in the is curtailed to the extent practicable. isolated patches likely to persist in the highest elevations of Caribbean will increase by 5.3 °C (9.5 °F). NH and ME. Both: Periodicity of red squirrel predation and competition with Swainson's thrush likely increases in any remaining habitat. 50-83 years (i.e., SRES A1F1 and RCP 8.5 scenarios with Very limited curtailment of global Canada: In low elevation habitat (<900 m), regardless of Projected complete loss of wintering 2100, end of the regional variations. Mean annual global emissions, as assumed under SRES forest harvest strategies, suitable habitat could decrease, habitat. century) surface temperatures increase by 3.7 °C A1F1 and RCP 8.5. Developing and leaving approx. 20% remaining. In higher elevations (>900 (6.7 °F), the average regional surface implementing BMP’s for timber m), forest harvest strategies could help mediate anticipated High temperature in the Northeast will increase companies; MOA between Canada, the decreases in suitable habitat such that approx. 87% remains. Warming by increase by 5.3 °C to 6 °C (9.5 °F to U.S., and the DR is established; logging United States: Possibility of complete elimination of breeding Scenario 10.8 °F), and the average regional surface within DR parks is curtailed to the habitat, with the possibility of small isolated patches remaining temperature in the Caribbean will increase extent practicable. in the highest elevations of NH and ME. Both: Periodicity of by 3.7 °C (6.7 °F) red squirrel predation and competition with Swainson's thrush likely increases in any remaining habitat.

Non-Climate Stressor Scenario

In the non-climate stressor scenario, we describe the future projection of viability without an increase or decrease in the effects of current levels of non-climate related stressors, in isolation from the potential effects of climate change discussed in the other scenarios. While realizing the likelihood of this scenario occurring is exceptionally low, it helps explain how the current stressors on the ground are affecting the species and what those might look like in the future.

Climate Change: Not applicable.

Breeding habitat in Canada: Timber harvest in Canada has remained steady for the last 20 years and we assume this trend will continue over the next 50 years. Canada will continue to support approximately 33 percent of the species’ breeding population.

Breeding habitat in the United States: Land use activities, such as development of wind turbines and existing levels of recreational uses (i.e., ski slopes and hiking trails), and forestry activities that affect breeding habitat used by the Bicknell’s thrush will continue at their current average rate and habitat will not substantially decline. For example, Maine will continue to harvest an average of 4,098 ac (1,658 ha) using timber stand improvement treatments and 481,270 ac (194,763 ha) in total; New Hampshire will continue to harvest an average of 77,084 cubic feet (ft3)/year of growing stock trees; Vermont will continue to harvest an average of 98,237 cords of spruce-fir timber and 746,115 cords of total timber; and New York will continue to harvest an average of 2.1 million green tons of pulpwood and chip products. With this level of timber harvest, the United States will continue to support approximately 66 percent of the species’ breeding population, in roughly the same proportions as currently estimated: 37 percent in New York, 29 percent in New Hampshire, 26 percent in Maine, and 7 percent in Vermont.

Wintering habitat in the Caribbean: Deforestation in the Caribbean continues at the current rate and habitat declines due to land use, economics, and lack of enforcement of protected areas. We assume the Dominican Republic will continue to lose its tree cover at the rate of approximately 10 percent every 15 years.

Optimistic Scenario

In the optimistic scenario, we align our projections with the targets outlined in the Paris Agreement, which is within the range of the RCP 2.6 scenario and strives to limit the increase in global average temperatures to well below 2 °C (3.6 °F) above preindustrial levels while pursuing efforts to limit the temperature increase to 1.5 °C (2.7 °F) above preindustrial levels. However, some regional temperature projections are expected to increase above the Paris Agreement targets. As explained above, existing conditions have already surpassed the RCP 2.6 trajectory, meaning that our Optimistic Scenario is possible only if significant societal changes to cut emissions back to the projected track of the RCP 2.6 scenario occur globally.

This scenario also includes projected full implementation of conservation measures that address the species’ habitat-related stressors, including significant curtailment of global emissions plus 56

implementation and effectiveness of negative emissions technologies (as assumed under RCP 2.6); developing and implementing best management practices (BMPs) for U.S. and Canadian timber companies; finalizing and implementing the MOA between Canada, the United States, and the Dominican Republic (Dettmers in litt. 2016; Gifford in litt. 2016), leading to improved protection of occupied winter habitat (IBTCG 2010, p.1–2) in the form of eliminating illegal logging within the parks; and implementing winter habitat restoration projects (IBTCG 2010, p.1–2). This scenario also assumes a corresponding response of the Bicknell’s thrush to the potential shift or elimination of habitat, but we recognize that there is uncertainty as to when or to what extent the species may respond.

Breeding habitat in Canada: Under a climate scenario where the mean annual temperature increase is below 2 °C (3.6 °F) and timber harvest continues at the current rate, we expect that in low elevation habitat (<900 m), forest harvest strategies could help maintain most habitat such that approximately 82 percent remains. In higher elevations (>900 m), regardless of forest harvest strategies, habitat may increase by approximately 43 percent. There may be an elevated exposure to insect outbreaks and increased interspecific competition with the Swainson’s thrush (Rodenhouse et al. 2008, p. 526).

Breeding habitat in the United States: Under a climate scenario where the mean annual temperature increase is below 2 °C (3.6 °F), we expect an eventual shift in tree species composition, projecting to result in a decrease in the extent of spruce and balsam fir (Iverson et al. 2016, entire; D’Amato 2016, p. 2). There may be an elevated exposure to insect outbreaks and increased interspecific competition with the Swainson’s thrush (Rodenhouse et al. 2008, p. 526).

Wintering habitat: Under a climate scenario where the mean annual regional temperature in the Caribbean increases by 2.9 °F (5.2 °F), we expect the amount of Bicknell’s thrush suitable habitat is likely to decrease due to changing precipitation patterns becoming drier. A complete loss of all subtropical rain, moist, and wet forests as explained by Khalyeni et al. (2016, p. 277) would mean a loss of 82 percent of Bicknell’s thrush wintering habitat in the Caribbean (i.e., 18 percent remains).

Low Warming Scenario

In the low warming scenario, climate will change within the ranges projected by the SRES B1 and RCP 4.5 scenarios, which include a global mean surface temperature increase of 1.8 °C (3.2 °F), the average regional surface temperature in the Northeast will increase by 3 °C to 5 °C (5.4 °F to 9 °F), and the average regional surface temperature in the Caribbean will increase by 5.3 °C (9.5 °F).

This scenario also includes projected partial implementation of conservation measures that address the species’ habitat-related stressors, including some curtailment of global emissions, as assumed under SRES B1 and RCP 4.5; developing and implementing best management practices (BMPs) for U.S. and Canadian timber companies; establishing the MOA between Canada, the United States, and the Dominican Republic (Dettmers in litt. 2016; Gifford in litt.

2016), which provides a mechanism for improving protection of occupied winter habitat (IBTCG 2010, p.1–2) in the form of curtailing illegal logging within the parks; and planning to implement winter habitat restoration projects (IBTCG 2010, p.1–2). This scenario also assumes a corresponding response of the Bicknell’s thrush to the potential shift or elimination of habitat, but we recognize that there is uncertainty as to when or to what extent the species may respond.

Breeding habitat in Canada: Under the B1 scenario of a 1.8 °C (3.2 °F) mean annual temperature increase, we expect that in low elevation habitat (<900 m), regardless of forest harvest strategies, suitable habitat could decrease, leaving approximately 49 percent remaining. In higher elevations (>900 m), regardless of forest harvest strategies, habitat may increase by approximately 22 percent. There may be an elevated exposure to insect outbreaks and increased interspecific competition with the Swainson’s thrush (Rodenhouse et al. 2008, p. 526).

Breeding habitat in the United States: The scaled down projections for the Northeast suggest that with a 3 °C (5.4 °F) increase in mean annual regional temperatures, there is the potential for most Bicknell’s thrush habitat to be eliminated from the northeastern United States. A few small isolated patches of habitat are likely to persist in the highest elevations of New Hampshire and Maine through a 5 °C (9 °F) increase (Rodenhouse et al. 2008, p. 526). There may be an elevated exposure to insect outbreaks and increased interspecific competition with the Swainson’s thrush (Rodenhouse et al. 2008, p. 526) in any remaining suitable habitat.

Wintering habitat: With a Caribbean-specific climate scenario of 5.3 °C (9.5 °F) mean annual temperature increase, we expect a likely complete loss of wintering habitat by the year 2100 (Khalyani et al. 2016, p.277). Moist habitat types that provide winter habitat for Bicknell’s thrush will likely be replaced with drier forests.

High Warming Scenario

In the high warming scenario, climate will change within the ranges projected by the SRES A1F1 and RCP 8.5 scenarios with a global mean surface temperature increase of 3.7 °C (6.7 oF), the average regional surface temperature in the Northeast will increase by 5.3 °C to 6 °C (9.5 °F to 10.8 °F, and the average regional surface temperature in the Caribbean will increase by 3.8 °C (6.6 °F).

This scenario also includes projected partial implementation of conservation measures that address the species’ habitat-related stressors, including very limited curtailment of global emissions, as assumed under SRES A1F1 and RCP 8.5; developing and implementing best management practices (BMPs) for U.S. and Canadian timber companies; establishing the MOA between Canada, the United States, and the Dominican Republic (Dettmers in litt. 2016; Gifford in litt. 2016), which provides a mechanism for improving protection of occupied winter habitat (IBTCG 2010, p.1–2) in the form of curtailing illegal logging within the parks; and planning to implement winter habitat restoration projects (IBTCG 2010, p.1–2). This scenario also assumes a corresponding response of the Bicknell’s thrush to the potential shift or elimination of habitat, but we recognize that there is uncertainty as to when or to what extent the species may respond.

Breeding habitat in Canada: Under a RCP 8.5 scenario, in low elevation habitat (<900 m), regardless of forest harvest strategies, suitable habitat could decrease, leaving approximately 20 percent remaining. In higher elevations (>900 m), forest harvest strategies could help mediate anticipated decreases in suitable habitat such that approximately 87 percent remains. There may be an elevated exposure to insect outbreaks and increased interspecific competition with the Swainson’s thrush (Rodenhouse et al. 2008, p. 526).

Breeding habitat in the United States: A climate scenario where the mean regional annual surface temperature increases by 5.3 °C (9.5 °F) has the potential to result in elimination of the species’ breeding habitat within the United States (Rodenhouse et al. 2008, p. 526). It is possible that small isolated habitat patches could remain in the highest elevations of New Hampshire and Maine (Rodenhouse et al. 2008, p. 526). There may be an elevated exposure to insect outbreaks and increased interspecific competition with the Swainson’s thrush (Rodenhouse et al. 2008, p. 526) in any remaining suitable habitat.

Wintering habitat: With a Caribbean-specific climate scenario leading to a 3.7 °C (6.7 °F) increase in the mean annual surface temperature, we expect a likely complete loss of wintering habitat by the year 2100 (Khalyani et al. 2016, p.277). Moist habitat types that provide winter habitat for the Bicknell’s thrush will likely be replaced with drier forests.

Summary of Future Scenario Considerations

In all but the Non-Climate Stressor scenario, some amount of Bicknell’s thrush habitat loss in the breeding range is expected (see table 5). The amount and distribution of breeding habitat that may remain through the end of the century varies depending on climate change projection, location (United State vs. Canada), and potential for forest harvest strategies to maintain spruce- fir habitat. In Canada, potential habitat loss may be partially mitigated by management strategies or elevation in the Optimistic and Low Warming Scenarios, but habitat loss is expected under the High Warming Scenario. In the United States, the eventual shift in tree species composition is projected to result in a decrease in spruce-fir habitat under the Optimistic Scenario; there is the potential for most habitat to be eliminated, although small isolated patches may persist in the highest elevations of New Hampshire and Maine, under the Low Warming Scenario; and the possibility that a complete elimination of habitat could occur, with a possibility that small isolated patches may persist in the highest elevations of New Hampshire and Maine, under the High Warming Scenario. The results of these projections are inclusive of potential conservation measures intended to address habitat-related stressors. Any remaining breeding habitat, it is predicted, will likely have limited suitability due to the presence of competing Swainson’s thrushes and nest-predating red squirrels.

In all scenarios, including the Non-Climate Stressor scenario, loss and degradation of the species’ habitat across the wintering range is expected to continue. The amount of wintering habitat that likely remains through the end of the century varies from the current amount and distribution (Non-Climate Stressor) to approximately 18 percent (Optimistic) to potentially zero (Low and High Warming). The results of these projections are inclusive of potential conservation measures intended to address habitat-related stressors.

We recognize that the level of uncertainty about the likely effects of climate change increases the further into the future we attempt to project. We also recognize that we do not know how the Bicknell’s thrush will respond or if it has the potential to adapt to the potential habitat changes. However, while the amount of potential habitat affected varies with scenario and time, the declining pattern is consistent among all scenarios, and it is reasonable to assume, in the absence of evidence to the contrary, that the species’ response may not keep pace with the projected changes in habitat within the predicted timeframes.

REFERENCES CITED

Able, K. P. and B. R. Noon. 1976. Avian community structure along elevational gradients in the northeastern United States. Oecologia 26:275-294.

American Ornithologists’ Union. 1995. Fortieth supplement to the American Ornithologists’ Union check-list of North American birds. The Auk 112:819-830.

Arendt, W.J., J. Faaborg, M. Canals, and J. Bauer. 2015. Bird checklist, Guánica Biosphere Reserve, Puerto Rico. Research Note SRS-23. Asheville, NC: U.S. Department of Agriculture Forest Service, Southern Research Station. 23p.

Atwood, J.L., C.C. Rimmer, K.P. McFarland, S.H. Tsai and L.R. Nagy. 1996. Distribution of Bicknell’s thrush in New England and New York. Wilson Bulletin 108:650-661.

Aubry, Y., A. Desrochers, and G. Seutin. 2016. Regional patterns of habitat use by a threatened forest bird, the Bicknell’s thrush (Catharus bicknelli), in Quebec. Canadian Journal of Zoology. 94:301 – 309.

Batlle, J.R.M. 2015a. Las 9 áreas protegidas que mayor cantidad absoluta de cobertura boscosa han perdido entre 2000 y 2014. Retrieved from http://www.geografiafisica.org/2016/03/26/las-9-areas-protegidas-que-mayor-cantidad- absoluta-de-cobertura-boscosa-han-perdido-entre-2000-y-2014/, accessed February 2, 2017.

Batlle, J.R.M. 2015b. Cambios en la cobertura boscosa del parquet nacional Sierra de Bahoruco. Retrieved from http://www.geografiafisica.org/2016/03/25/cambios-en-la-cobertura- boscosa-del-parque-nacional-sierra-de-bahoruco/, accessed February 2, 2017.

Beckage, B., B. Osborne, D.G. Gavin, C. Pucko, T. Siccama, and T. Perkins. 2008. A rapid upward shift of a forest ecotone during 40 years of warming in the Green Mountains of Vermont. Proceedings of the National Academy of Sciences 105:4197- 4202.

Bouchard, M. and I. Auger. 2014. Influence of environmental factors and spatio-temporal covariates during the initial development of a spruce budworm outbreak. Landscape Ecology. 29: 111 – 126.

Boulanger, Y., S. Gauthier, D.R. Gray, H. Le Goff, P. Lefort, and J. Morissette. 2013. Fire regime zonation under current and future climate over eastern Canada. Ecological Applications. 23(4): 904 – 923.

Boulanger, Y., A.R. Taylor, D.T. Price, E. McGarrigle, W. Rammer, G. Sainte-Marie, A. Beaudoin, L. Guindon, and N. Mansuy. 2016a. Climate change impacts on forest landscapes along the Canadian southern boreal forest transition zone. Landscape Ecol. DOI 10.1007/s10980-016-0421-7. 17 pp. 61

Boulanger, Y. D.R. Gray, B.J. Cooke, L. De Grandpré. 2016b. Model-specification uncertainty in future forest pest outbreak. Global Change Biology (22) 1595–1607, doi: 10.1111/gcb.13142.

Bourque, C.P. 2015. Modelled potential tree species distribution for current and projected future climates for New Brunswick, Canada. Unpublished report. 40pp.

Bourque, C.P., Quazi K.H., and D.E. Swift. 2010. Modelled potential species distribution for current and projected future climates for the Acadian Forest Region of Nova Scotia, Canada. Report prepared for the Nova Scotia Department of Natural Resources, Truro, Nova Scotia. Retrieved from http://www.novascotia.ca/natr/forestry/reports/NS-Climate Change-Project.pdf, accessed December 7, 2016.

Boutin, S., Wauters, L.A., A.G. McAdam, M.M. Humphries, G. Tosi, and A.A. Dhondt. 2006. Anticipatory reproduction and population growth in seed predators. Science. 314(5807): 1928–1930.

Bredin, K., and Whittam, B. 2009. Conserving the Bicknell’s thrush: Stewardship and management practices for Nova Scotia’s high elevation forest. Bird Studies Canada. 23pp.

Bryan, A., A. Karmalkar, E. Coffel, L. Ning, R. Horton, E. Demaria, F. Fan, R. S. Bradley, and R. Palmer. 2015. Climate Change in the Northeast and Midwest United States in Staudinger, M.D., T.L. Morelli, and A.M. Bryan. 2015. Integrating Climate Change into Northeast and Midwest State Wildlife Action Plans. DOI Northeast Climate Science Center Report, Amherst, Massachusetts. Retrieved from http://www.cakex.org/sites/default/files/documents/Staudinger%20et%20al.%202015%20 Integrating%20Climate%20Change%20into%20NE%20and%20MW%20SWAPs_0.pdf, accessed May 19, 2017.

Briguglio, L.; G. Cordina; N. Faffugia; and S. Vella. 2008. Economic vulnerability and resilience: concepts and measurements. United Nations University. Research Paper No. 2008/55.

Brooks, E. and D. Sherony. 2010. Some notes on the fall migration of Bicknell’s and gray- cheeked thrushes in New York State and eastern United States. The Kingbird. 60(1): 7– 14.

Buck, J. 2016. Wind turbines in Bicknell’s thrush habitat in Vermont??? Email. Vermont Fish and Wildlife Department. 3p.

Burns, R.M. and B.H. Honkala. 1990. Silvics of North America: Volume 1, Conifers, Agriculture Handbook 654. Department of Agriculture, Forest Service, US Government Printing Office, Washington, DC, USA. Retrieved from

http://na.fs.fed.us/spfo/pubs/silvics_manual/Volume_1/vol1_Table_of_contents.htm, accessed December 8, 2016.

Butler, R. 2006. Caribbean Islands. Retrieved from http://rainforests.mongabay.com/20carib.htm, accessed December 6, 2016.

Candau, J-N and R.A. Fleming. 2011. Forecasting the response of spruce budworm defoliation to climate change in Ontario. Canadian Journal of Forest Research. 41: 1948 – 1960.

Campbell, G., B. Stewart. 2012. High Elevation Landbird Program. 10-year report. Bird Studies Canada, Sackville, NB, Canada. 22 pp.

Chisholm, S.E. and M.L. Leonard. 2008. Effect of forest fragmentation on a rare habitat specialist, the Bicknell’s thrush (Catharus bicknelli). Canadian Journal of Zoology 86: 217-223.

Cogbill, C.V. and P.S. White. 1991. The latitude-elevation relationship for spruce-fir forest and treeline along the Appalachian Mountain chain. Vegetatio 94:153-175.

Connolly, V., Gilles Seutin, JP. L. Savard, and G. Rompré. 2002. Habitat use by the Bicknell’s thrush in the Estrie Region, Quebec. The Wilson Bulletin 114(3): 333–341.

COSEWIC. 2009. COSEWIC assessment and status report on the Bicknell’s thrush Catharus bicknelli in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa. 44pp.

Cumming, J.R., and S.M. Brown. 1994. Effects of elevated nitrate and aluminum on the growth and nutrition of red spruce (Picea rubens) seedlings. Tree Physiology 14:589-599.

D'Amato, A.W. 2016. Modeling effects of climate change on spruce-fir ecosystems and associated priority bird populations. Final Project Report. Retrieved from https://www.sciencebase.gov/catalog/item/5707f150e4b06fa6ac664531, accessed May 19, 2017.

DeHayes, D.H., P.G. Schaberg, G.J. Hawley, and G.R. Strimbeck. 1999. Acid rain impacts on calcium nutrition and forest health. Bioscience 49:789-800

Deluca, W.V. and D.I. King. 2016. Montane birds shift downslope despite recent warming in the northern Appalachian Mountains. Journal of Ornithology. DOI: 10.1007/s10336-016- 1414-7.

Deluca, W.V. and D.I. King. 2014. Influence of hiking trails on montane birds. Journal of Wildlife Management. 78(3): 494–502.

Dettmers, R. 2016. Need BITH assistance on 3 items. Email. U.S. Fish and Wildlife Service. 2p.

Dobrowski, S.Z. 2011. A climatic basis for microrefugia: the influence of terrain on climate. Global Change Biology. 17(2): 1022-1035.

Driscoll, C.T., G.B. Lawrence, A.J. Bulger, T.J. Butler, C.S. Cronan, C. Eagar, K.F. Lambert, G.E. Likens, J.L. Stoddard, and K.C. Weathers. 2001. Acid Rain Revisited: Advances in scientific understanding since the passage of the 1970 and 1990 Clean Air Act Amendments. Hubbard Brook Research Foundation. Science Links Publication. Vol. 1, no.1.

Engel, B.J., P.G. Schaberg, G.J. Hawley, S.A. Rayback, J. Pontius, A.M. Kosiba, and E.K. Miller. 2016. Assessing relationships between red spruce radial growth and pollution critical load exceedance values. Forest Ecology and Management. 359: 83–91.

Evers, D.C. 2005. Mercury Connections: The extent and effects of mercury pollution in northeastern North America. BioDiversity Research Institute. Gorham, Maine.

Freeman, B.G. and G. Montgomery. 2016. Interspecific aggression by the Swainson’s thrush (Catharus ustulatus) may limit the distribution of the threatened Bicknell’s thrush (Catharus bicknelli) in the Adirondack Mountains. The Condor. 118: 169–178.

Fleming, R.A., and JN Candau. 1998. Influences of climatic change on some ecological processes of an insect outbreak system in Canada's boreal forests and the implications for biodiversity. Environmental Monitoring and Assessment. 49: 235–249.

Forest Fauna and Parks Quebec. 2017. http://mffp.gouv.qc.ca/forets/protection/TBE.jsp, translated to English via Google Translate, last accessed May 21, 2017.

Foster, J.R. and A.W. D’Amato. 2015. Montane forest ecotones moved downslope in northeastern USA in spite of warming between 1984 and 2011. Global Change Biology. 21:4497–4507.

Friedland, A.J. 1989. Recent changes in the montane spruce-fir forests of the northeastern United States. Environmental Monitoring and Assessment 12: 237-244.

Friedlingstein, P., R.M. Andrew, J. Rogelj, G.P. Peters, J.G. Canadell, R. Knutti, G. Luderer, M.R. Raupach, M. Schaeffer, D.P. van Vuuren, and C. Le Quéré. 2014. Persistent growth of CO2 emissions and implications for reaching climate targets. Nature Geoscience 7(10): 709–715.

Gifford, K. 2016. Bicknell’s thrush: summary of 7/9/16 conversation. Email. U.S. Fish and Wildlife Service. 2p.

Gitay, H., S. Brown, W. Easterling, and B. Jallow. 2001. Ecosystems and their goods and services. Pages 235–342 in J.J. McCarthy, O.F. Canziani, N.A. Leary, D.J. Dokken, and

K.S. White, eds. Climate change 2001: impacts, adaptation, and vulnerability. Contribution of Working Group II to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge and New York.

Glennon, M. and L. Karasin. 2004. Use of Whiteface Mountain by Bicknell's thrush and other montane forest bird species. 2004 End-of-season report submitted to: Olympic Regional Development Authority. 10pp. Available at www.wcsnorthamerica.org.

Glennon, M.J. and C.L. Seewagen. 2016. Conservation status and monitoring of Bicknell’s thrush in the Adirondacks and New England: A brief review. The Adirondack Journal of Environmental Studies. 2: 123–137.

Goetz, J.E., K.P. McFarland, C.C. Rimmer. 2003. Multiple Paternity and Multiple Male Feeders in Bicknell's Thrush (Catharus bicknelli). The Auk, Vol. 120, No. 4, pp. 1044– 1053.

Hames, R.S., K.V. Rosenberg, J.D. Lowe, S.E. Barker, and A.A. Dhondt. 2002. Adverse effects of acid rain on the distribution of Wood Thrush (Hylocichla mustelina) in North America. Proceedings of the National Academy of Sciences 99: 11235-11240.

Hansen, M. C., P. V. Potapov, R. Moore, M. Hancher, S. A. Turubanova, A. Tyukavina, D. Thau, S. V. Stehman, S. J. Goetz, T. R. Loveland, A. Kommareddy, A. Egorov, L. Chini, C. O. Justice, and J. R. G. Townshend. 2013. High-Resolution Global Maps of 21st- Century Forest Cover Change. Science 342 (15 November): 850–53.

Hansen, M. C., P. V. Potapov, R. Moore, M. Hancher, S. A. Turubanova, A. Tyukavina, D. Thau, S. V. Stehman, S. J. Goetz, T. R. Loveland, A. Kommareddy, A. Egorov, L. Chini, C. O. Justice, and J. R. G. Townshend. 2017. Hansen/UMD/Google/USGS/NASA Tree Cover and Tree cover Loss and Gain, Country Profiles. Retrieved from www.globalforestwatch.org, accessed February 1, 2017.

Hart, J.A. and J.D. Lambert. 2010. Mountain Birdwatch. Protocol and standard operating procedures for monitoring high-elevation landbirds in the northern Appalachian and Laurentian Regions. Version 2.0. Unpublished report.

Hayhoe, K., C.P. Wake, T.G. Huntington, L. Luo, M.D. Schwartz, J. Sheffield, E. Wood, B. Anderson, J. Bradbury, A. DeGaetano, T.J. Troy, and D. Wolfe. 2007. Past and future changes in climate and hydrological indicators in the US Northeast. Climate Dynamics 28:381-407.

Hayhoe, K. 2013. Quantifying key drivers of climate variability and change for Puerto Rico and the Caribbean.” Final Report to the Southeast Climate Science Center. Raleigh, NC, USA. 241 pp.

Hill, J.M. and J.D. Lloyd. 2016. Summary Findings: Breeding population size of Bicknell’s thrush. Vermont Center for Ecostudies. 7p.

Hobson, K.A. and L.I. Wassenaar. 1997. Linking breeding and wintering grounds of neotropical migrant songbirds using stable hydrogen isotopic analysis of feathers. Oecologia. 109(1): 142–148.

IBTCG. 2017. A Conservation Action Plan for Bicknell’s Thrush (Catharus bicknelli). J. A. Hart, C.C. Rimmer, R. Dettmers, R.M. Whittam, E.A. McKinnon, and K.P. McFarland, eds. International Bicknell’s Thrush Conservation Group. May 2017. Available at www.bicknellsthrush.org.

IBTCG. 2010. A Conservation Action Plan for Bicknell’s Thrush (Catharus bicknelli). J. A. Hart, C.C. Rimmer, R. Dettmers, R.M. Whittam, E.A. McKinnon, and K.P. McFarland, eds. International Bicknell’s Thrush Conservation Group. Available at www.bicknellsthrush.org.

IPCC. 2007. Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Pachauri, R.K and Reisinger, A., eds., IPCC, Geneva, Switzerland. Available at www.ipcc.ch.

IPCC. 2014. Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. [Core Writing Team Pachauri, R.K. and L.A. Meyer, (eds.)], IPCC, Geneva, Switzerland. 151 pp. Retrieved from https://www.ipcc.ch/report/ar5/syr/, accessed December 7, 2016.

Iverson, L.R. and A.M. Prasad. 1998. Predicting abundance of 80 tree species following climate change in the eastern United States. Ecological Monographs. 68(4): 465–485.

Iverson, L., A.M. Prasad, J. Betsy, and E.K. Sutherland. 1999. An atlas of current and potential future distributions of common trees of the eastern United States. General Technical Report. NE-265, Radnor, PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station. 245pp.

Iverson, L.R., A.M. Prasad, S.N. Matthews, and Matthew Peters. 2008. Estimating potential habitat for 134 eastern US tree species under six climate scenarios. Forest Ecology and Management 254: 390–406.

Iverson, L.R., F.R. Thompson III, S. Matthews, M. Peters, A. Prasad, W.D. Dijak, J. Fraser, W.J. Wang, B. Hanberry, H. He, M. Janowiak, P. Butler, L. Brandt, C. Swanston. 2016. Multi-model comparison on the effects of climate change on tree species in the eastern U.S.: results from an enhanced niche model and process-based ecosystem and landscape models. Landscape Ecology. Retrieved from http://dx.doi.org/10.1007/s10980-016- 04048, accessed December 7, 2016.

Karmalkar, A.V. and R.S. Bradley. 2017. Consequences of Global Warming of 1.5 ÊC and 2ÊC for Regional Temperature and Precipitation Changes in the Contiguous United States. PLoS ONE 12(1): e0168697. Retrieved from http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0168697, accessed May 19, 2017.

Kerr, G.H. and A.T. DeGaetano. 2016. Climate change effects on wildland fire risk in the Northeastern and Great Lakes States predicted by a downscaled multi-model ensemble. Theoretical and Applied Climatology. 1–15.

Kemp, G.A. and L.B. Keith. 1970. Dynamics and regulation of red squirrel (Tamiasciurus hudsonicus) populations. Ecology. 51(5): 763–779.

Khalyani, A.H., W.A. Gould, E. Harmsen, A. Terando, M. Quinones, and J.A. Collazo. 2016. Climate change implications for tropical islands: Interpolating and interpreting statistically downscaled GCM projections for management and planning. Journal of Applied Meteorology and Climatology. 55: 265–282. King, D., J.D. Lambert, J.P. Buonaccorsi, and L.S. Prout. 2008. Avian population trends in the vulnerable montane forests of the Northern Appalachians, USA. Biodiversity and Conservation 17:2691–2700.

Koenig, W.D. and J.M.H. Knops. 1998. Scale of mast-seeding and tree-ring growth. Nature 396: 225–226.

Krebs, C.J., J.M. LaMontagne, A.J. Kenney, and S. Boutin. 2012. Climatic determinants of white spruce cone crops in the boreal forest of southwestern Yukon. Botany. 90: 113– 119.

Krebs, C.J., R. Boostra, S. Boutin, A.R.E. Sinclair, J.N.M. Smith, B.S. Gilbert, K. Martin, M. O’Donoghue, and R. Turkington. 2014. Trophic dynamics of the boreal forests of the Kluane Region. Arctic. 67(1): 71–81.

Kunkel, K.E., L.E. Stevens, S.E. Stevens, L. Sun, E. Janssen, D. Wuebbles, J. Rennells, A. DeGaetano, and J.G. Dobson. 2013. Regional climate trends and scenarios for the U.S. National Climate Assessment: Part 1. Climate of the Northeast U.S. NOAA technical report NESDIS 142-1: 79 pp.

Lake Placid Convention and Visitor’s Bureau, Regional Office of Sustainable Tourism. 2014. Leisure Travel Study. 2013 Visitor profile and return on investment- conversion rate analysis for Essex County, New York. Retrieved from http://www.roostadk.com/wpcontent/uploads/2014/08/Final-Leisure-Visitor-Study- 2013.pdf, accessed December 9, 2016.

Lambert, J.D., K.P. McFarland, C.C. Rimmer, S.D. Faccio, and J.L. Atwood. 2005. A practical model of Bicknell’s thrush distribution in the northeastern United States. Wilson Bulletin. 117(1): 1–11.

Latta, S.C., C.C. Rimmer, K.P. McFarland. 2003. Winter bird communities in four habitats along an elevational gradient on Hispaniola. The Condor 105:179-197.

Lenoir, J., J-C Gégout, A. Guisan, V. Pascal, T. Wohlgemuth, N. E. Zimmermann, S. Dullinger, H. Pauli, W. Willner, and J-C Svenning. 2010. Going against the flow: potential mechanisms for unexpected downslope range shifts in a warming climate. Ecography. 33: 295–303.

León, Y.M., E. Garrido and E., J. Almonte. 2013. Monitoring and mapping broadleaf mountain forests of southern Sierra de Bahoruco, Dominican Republic. Grupo Jaragua, Santo Domingo, 36 pp. Retrieved from http://www.grupojaragua.org.do/documents/Informe_MonitoreoBicknell_BahorucoS_G 2012-2013_ENGLISH_ed.pdf, accessed December 7, 2016.

Leppold, A. 2016. BITH reference. Email. Maine Department of Inland Fisheries and Wildlife. 2 pp.

Lloyd, J. and C.C. Rimmer. 2016. Summary Findings: Winter distribution of Bicknell’s thrush, with a focus on Puerto Rico. Vermont Center for Ecostudies. 4p.

Lloyd, J. C.C. Rimmer, and K.P. McFarland. 2016. Assessing conservation status of resident and migrant birds on Hispaniola with mist-netting. PeerJ 3:e1541; DOI 10.7717/peerj.1541. 17 pp.

Maine Forest Service. 2001. Silvicultural activities report including annual report on clearcutting. 2000. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2002. Silvicultural activities report including annual report on clearcutting. 2001. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2003. Silvicultural activities report including annual report on clearcutting. 2002. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2004. Silvicultural activities report including annual report on clearcutting. 2003. Retrieved from

https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2005. Silvicultural activities report including annual report on clearcutting. 2004. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2006. Silvicultural activities report including annual report on clearcutting. 2005. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2007. Silvicultural activities report including annual report on clearcutting. 2006. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2008. Silvicultural activities report including annual report on clearcutting. 2007. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2009. Silvicultural activities report including annual report on clearcutting. 2008. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2010. Silvicultural activities report including annual report on clearcutting. 2009. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2011. Silvicultural activities report including annual report on clearcutting. 2010. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2013. Silvicultural activities report including annual report on clearcutting. 2011. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2014a. Silvicultural activities report including annual report on clearcutting. 2012. Retrieved from

https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2014b. Silvicultural activities report including annual report on clearcutting. 2013. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2015. Silvicultural activities report including annual report on clearcutting. 2014. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2016. Silvicultural activities report including annual report on clearcutting. 2015. Retrieved from https://www1.maine.gov/dacf/mfs/publications/annual_reports.html, accessed December 27, 2016.

Maine Forest Service. 2017. Spruce budworm (Choristoneura fumiferana) in Maine 2016. Retrieved from http://www.sprucebudwormmaine.org/wpcontent/uploads/2017/01/MFS_2016SpruceBud wormReport_1_4_2017.pdf, accessed January 29, 2017.

Mänd, R., V. Tilgar, and A. Leivits. 2000. Calcium, snails, and birds: a case study. Web Ecology 1:63–69.

McFarland, K.P., C.C. Rimmer, S.J.K. Frey, S.D. Faccio, and B.B. Collins. 2008. Demography, Ecology and Conservation of Bicknell's Thrush in Vermont, with a Special Focus on the Northeast Highlands. Vermont Center for Ecostudies, Norwich, VT. Technical Report 08-03. 67pp.

McFarland, K.P., C.C. Rimmer, J.E. Goetz, Y. Aubry, J.M. Wunderle Jr., A. Sutton, J.M. Townsend, A.L. Sosa, and A. Kirkconnell. 2013. A winter distribution model for Bicknell’s thrush (Catharus bicknelli), a conservation tool for a threatened migratory songbird. PLoS ONE 8(1): e53986. doi:10.1371/journal.pone.0053986

McFarland, K.P, Y. Aubry, C.C. Rimmer, B. Whittam, J. Townsend, B. Stewart, G. Campbell, R. Renfrew, P. Marra, S. Paradis, V. Afanaseyev and J.W. Fox. In Prep. Using light-level geolocation to illuminate migratory routes, stop-over sites and population connectivity for Bicknell’s Thrush (Catharus bicknelli).

McKinnon, E.A., H. Askanas, and A.W. Diamond. 2014. Nest-patch characteristics of Bicknell’s thrush in regenerating clearcuts, and implications for precommercial thinning. Northeast Naturalist. 21(2): 259–270.

McNulty, S.G., J.D. Aber, and R.D. Boone. 1991. Spatial changes in forest floor and foliar chemistry of spruce-fir forests across New England. Biogeochemistry 14:13–29.

McNulty, S.G., J.D. Aber, and S.D. Newman. 1996. Nitrogen saturation in a high elevation spruce–fir stand. Forest Ecology and Management 84:109–121.

McNulty, S.G, J. Boggs, J.D. Aber, L. Rustad, and A. Magill. 2005. Red spruce ecosystem level changes following 14 years of chronic N fertilization. Forest Ecology and Management 219: 279–291.

Messaoud, Y., Y. Bergeron, and H. Asselin. 2007. Reproductive potential of balsam fir (Abies balsamea), white spruce (Picea glauca), and black spruce (P. mariana) at the ecotone between mixed wood and coniferous forests in the boreal zone of western Quebec. American Journal of Botany 94:746–754.

Miller, E.K., A. Vanarsdale, G.J. Keeler, A. Chalmers, L. Poissant, N.C. Kamman, and R. Brulotte. 2005. Estimation and mapping of wet and dry mercury deposition across Northeastern North America. Ecotoxicology 14:53–70.

Morelli, T.L., C. Daly, S.Z. Dobrowski, D.M. Dulen, J.L. Ebersole, S.T. Jackson, J.D. Lundquist, C.I. Millar, S.P. Maher, W.B. Monahan, K.R. Nydick, K.T. Redmond, S.C. Sawyer, S. Stock, and S.R. Beissinger. 2016. Managing climate change refugia for climate adaptation. PLoS ONE 11(8): e0159909. Retrieved from http://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0159909&type=pri ntable, accessed May 19, 2019.

Morin, R.S. 2011. New Hampshire’s forest resources, 2009. Research Note NRS-84. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 4 pp.

Morin, R.S. and M. Tansey. 2008. New Hampshire’s forest resources, 2006. Research Note NRS-17. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 4 pp.

Morin, R.S., G.M. McCaskill, W. McWilliams, and M. Tansey. 2010a. New Hampshire’s forest resources, 2007. Research Note NRS-56. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 4 pp.

Morin, R.S., B.J. Butler, and M. Tansey. 2010b. New Hampshire’s forest resources, 2008. Research Note NRS-57. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 4 pp.

Morin, R.S, and M. Nelson. 2011. New Hampshire’s forest resources, 2010. Research Note NRS-109. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 4 pp.

Morin, R.S, and C.W. Woodall. 2012. New Hampshire’s forest resources, 2011. Research Note NRS-140. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 4 pp.

Morin, R.S. and K. Lombard. 2013. New Hampshire’s forest resources, 2012. Research Note NRS-172. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 4 pp.

Morin, R.S. and S.A. Pugh. 2014. New Hampshire’s forest resources, 2013. Research Update FS-29. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 4 pp.

Morin, R.S. and R. Riemann. 2015. New Hampshire’s forest resources, 2014. Research Update FS-55. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 4 pp.

Morin, R.S. and R.H. Widmann. 2016. New Hampshire’s forest resources, 2015. Research Update FS-71. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 4 pp.

Natural Resources Canada. 2015. Hemlock looper. Retrieved from https://tidcf.nrcan.gc.ca/en/insects/factsheet/8846, accessed May 21, 2017.

Natural Resources Canada. 2016. Spruce budworm. Retrieved from http://www.nrcan.gc.ca/forests/fire-insects-disturbances/top-insects/13383, accessed January 18, 2017.

New York Department of Environmental Conservation. 2014. New York State industrial timber harvest production and consumption report–2014. Retrieved from http://www.dec.ny.gov/lands/33295.html, accessed December 27, 2016.

New Hampshire Department of Resources and Economic Development. 2016. Division of Forests and Lands Forest Health Program. Annual newsletter. Retrieved from http://nhdfl.org/library/pdf/Forest%20Health/2016newsletter.pdf, accessed January 29, 2017.

New Hampshire Timber Harvest Council. 2011. Silvicultural Principles for New Hampshire Forest Types. Retrieved from https://extension.unh.edu/resources/files/Resource000235_Rep254.pdf, accessed May 22, 2017.

Ning, L., E.E. Riddle, and R.S. Bradley. 2015. Projected changes in climate extremes over the northeastern United States. Journal of Climate. 28: 3289–3310.

Nixon, E.A., Holmes, S., and A. Diamond. 2001. Bicknell’s thrushes (Catharus bicknelli) in New Brunswick clear cuts: their habitat associations and co-occurrence with Swainson’s thrushes (Catharus ustulatus). The Wilson Bulletin 113:33–40.

Noble Environmental Power. 2008. Granite Reliable Power high elevation mitigation plan. Noble Environmental Power. 4p.

Noon, B.R. 1981. The distribution of an avian guild along a temperate elevational gradient: The importance and expression of competition. Ecological Monographs 51:105–124.

Nurse, L.A., R.F. McLean, J. Agard, L.P. Briguglio, V. Duvat-Magnan, N. Pelesikoti, E. Tompkins, and A. Webb, 2014. Small islands. In: Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part B: Regional Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Barros, V.R., C.B. Field, D.J. Dokken, M.D. Mastrandrea, K.J. Mach, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, and L.L. White(eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 1613-1654.

Ouellet, H. 1993. Bicknell’s thrush: taxonomic position and distribution. The Wilson Bulletin 105:545-572.

Outlaw, D.C., G. Voelker, B. Mila, and D.J. Girman. 2003. Evolution of long-distance migration in and historical biogeography of Catharus thrushes: A molecular phylogenetic approach. The Auk 120(2): 299–310.

Oviedo, R., A. Llanes, Y. Aubry, A. Hernández, G. Rompe, and F. Shaffer. 2001. Elements of the composition and structure of vegetation in the habitat of Bicknell’s thrush in Cuba. El Pitirre. 14(3): 134.

Parisien, M. – A. and M.A. Moritz. 2009. Environmental controls on the distribution of wildfire at multiple spatial scales. Ecological Monographs. 79(1): 127–154.

Parrish, C.R. 2013. Impacts of wind development on the abundance and distribution of high elevation birds in northern New Hampshire, with a focus on Bicknell’s thrush (Catharus bicknelli). Master’s Thesis. Plymouth State University, Plymouth, NH.

Pasachnik, S.A., R.C. DeLeón, and Y.M. León. 2016. Protected only on paper? Three case studies from protected areas in the Dominican Republic. Caribbean Naturalist. 30: 1–19.

Périé, C. and S. de Blois. 2016. Dominant forest tree species are potentially vulnerable to climate change over large portions of their range even at high latitudes. PeerJ 4:e2218; DOI 10.7717/peerj.2218

Petersen, W.R. and W.R. Meservey. 2003. Massachusetts Breeding Bird Atlas. University of Massachusetts Press. 434pp.

Ragenovich, I.R., and R.G. Mitchell. 2006. Forest Insect and Disease Leaflet. No. 118: Balsam Woolly Adelgid. Available at http://www.na.fs.fed.us/pubs/fidls/bwa.pdf.

Ralston, J., D.I. King, W.V. DeLuca, G.J. Niemi, M.J. Glennon, J.C. Scarl, and J.D. Lambert. 2015. Analysis of combined data sets yields trend estimates for vulnerable spruce-fir birds in northern United States. Biological Conservation. 187: 270 – 278.

Réale, D., A.G. McAdam, S. Boutin, and D. Berteaux. 2003. Genetic and plastic responses of a northern mammal to climate change. Proceedings of the Royal Society of London. 270: 591–596.

Régnière J., R. St-Amant, P. Duval. 2012. Predicting insect distributions under climate change from physiological responses: spruce budworm as an example. Biological Invasions. 14: 1571–1586.

Rimmer, C.C., J.L. Atwood, K.P. McFarland, and L.R. Nagy. 1996. Population density, vocal behavior, and recommended survey methods for Bicknell’s thrush. The Wilson Bulletin 108: 639-649.

Rimmer, C. C., K. P. McFarland, W. G. Ellison, and J. E. Goetz. 2001. Bicknell’s thrush (Catharus bicknelli). In A. Poole and F. Gills, eds., The Birds of North America, No. 592. The Birds of North America, Inc., Philadelphia, PA. 28pp.

Rimmer, C.C., K.P. McFarland, J.D. Lambert, and R.B. Renfrew. 2004. Evaluating the Use of Vermont Ski Areas by Bicknell’s Thrush: Applications for Whiteface Mountain, New York. Vermont Institute of Natural Science. Woodstock, VT. 46 pp.

Rimmer, C.C., J.D. Lambert and K.P. McFarland. 2005a. Bicknell’s Thrush Conservation Strategy for the Green Mountain National Forest. Vermont Institute of Natural Sciences. Technical Report 05-5.

Rimmer, C.C., J.M. Townsend, A.K. Townsend, E.M. Fernández, and J. Almonte. 2005b. Avian diversity, abundance, and conservation status in the Macaya Biosphere Reserve of Haiti. Ornithologia Neotropical 16:219-230.

Rimmer, C.C., K.P. McFarland, D.C. Evers, E.K. Miller, Y. Aubry, D. Busby, and R.J. Taylor. 2005c. Mercury concentrations in Bicknell’s thrush and other insectivorous passerines in montane forests of northeastern United States and Canada. Ecotoxicology 14:223-240.

Rimmer, C.C. and K.P. McFarland. 2013. Bicknell’s thrush: A Twenty-year retrospective on the northeast’s most vulnerable songbird. Bird Observer. 41(1):9–16.

Rimmer, C.C. 2016. Puerto Rico Scorecard: Endemics 17, Bicknell’s Thrush 7. Retrieved from https://vtecostudies.org/blog/puerto-rico-scorecard-endemics-17-bicknells-thrush-7/, accessed May 21, 2017.

Rivera, R.L. 2017. State partner comments on the U.S. Fish and Wildlife Service’s Biological Species Report for the Bicknell’s Thrush (Catharus bicknelli), version 1.1 (April 2017). Electronic mail and attachment dated May 19, 2017.

Rodenhouse, N.L., S.N. Matthews, McFarland, K.P., Lambert, J.D., Iverson, L.R., Prasad, A., Sillett, T.S., and R.T. Holmes. 2008. Potential effects of climate change on birds of the Northeast. Mitigation and Adaptation Strategies for Global Change 13:517-540.

Rogelj, J., M. Meinshausen, and R. Knutti. 2012. Global warming under old and new scenarios using IPCC climate sensitivity range estimates. Nature Climate Change. 2: 248–253.

Rustad, L., J. Campbell, J.S. Dukes, T. Huntington, K.F. Lambert, J. Mohan, and N. Rodenhouse. 2012. Changing climate, changing forests: The impacts of climate change on forests of the northeastern United States and eastern Canada. U.S. Department of Agriculture, Forest Service, Northern Research Station. 50pp.

Sabo, S.R. 1980. Niche and habitat relations in subalpine bird communities of the White Mountains of New Hampshire. Ecological Monographs 50:241–259.

Sanderson, B.M., B.C. O’Neill, and C. Tebaldi. 2016. What would it take to achieve the Paris temperature targets? Geophysical Research Letters. 43: 7133–7142.

Sanford, T., P.C. Frumhoff, A. Luers, and J. Gulledge. 2014. The climate policy narrative for a dangerously warming world. Nature Climate Change. 4: 164–166.

Seidel, T.M., D.M. Weihrauch, K.D. Kimball, A.A.P. Pszenny, R. Soboleski, E. Crete, and G. Murray. 2009. Evidence of Climate Change Declines with Elevation Based on Temperature and Snow Records from 1930s to 2006 on Mount Washington, New Hampshire, USA. INSTAAR, University of Colorado. 41(3): 362-372.

Smith, S.J., J. van Aardenne, Z. Klimont, R.J. Andres, A. Volke, and S. Delgado Arias. 2011. Anthropogenic sulfur dioxide emissions: 1850–2005. Atmospheric Chemistry and Physics. 11: 1101–1116.

Sosa, A.L., Y. Aubry, Oviedo R., F. Shaffer, A.H. Marrero, and G. Rompré. 2001. Bicknell’s thrush (Catharus bicknelli): A winter resident in Cuba. El Pitirre. 14(3): 136.

Strong, Y. 2016. Re: Information on Bicknell’s Thrush (Catharus bicknelli). Letter from Yvette Strong, for Chief Executive Officer/Government Town Planner, National Environment & Planning Agency. Kingston, Jamaica. 10 May 2016.

Studds, C.E., K.P. McFarland, Y. Aubry, C.C. Rimmer, K.A. Hobson, P.P. Marra, and L.I. Wassenaar. 2012. Stable-hydrogen isotope measures of natal dispersal reflect observed population declines in a threatened migratory songbird. Journal of Conservation Biogeography. 18: 919–930.

Timyan, J.C., J.E. Goetz, and J.C. Martinez-Sánchez. 2012. Toward an improved conservation strategy for Bicknell’s thrush (Catharus bicknelli) in La Visite National Park, Haiti. Unpublished report for Canadian Wildlife Service and BirdLife International.

Todd, C. 2016. BITH. Email. Maine Department of Inland Fisheries and Wildlife. 3p.

Townsend, J. M., and Rimmer, C.C. 2006. Known natal and wintering sites of a Bicknell’s thrush. Journal Field Ornithology 77:452–454.

Townsend, J.M., C.C. Rimmer, J. Brocca, K.P. McFarland, and A.K. Townsend. 2009a. Predation of a wintering migratory songbird by introduced rats: can nocturnal roosting behavior serve as predator avoidance? Condor 111:565–569.

Townsend, J.M., C.C. Rimmer, and K.P. McFarland. 2009b. Investigating the limiting factors of a rare, vulnerable species: Bicknell’s Thrush. Pp. 91-95 in Rich, T.D., C. Arizmendi, D. Demarest and C. Thompson, eds. Tundra to Tropics: Connecting Birds, Habitats and People. Proceedings of the 4th International Partners in Flight Conference, McAllen, Texas. Partners in Flight.

Townsend, J.M., C.C. Rimmer, and K.P. McFarland. 2010. Winter territoriality and spatial behavior of Bicknell’s Thrush (Catharus bicknelli) at two ecologically distinct sites in the Dominican Republic. The Auk 127: 514–522.

Townsend, J.M., C.C. Rimmer, K.P. McFarland, and J.E. Goetz. 2012. Site-specific variation in food resources, sex ratios, and body condition of an overwintering migrant songbird. The Auk. 129(4): 683–690.

Townsend, J.M, K.P. McFarland, C.C. Rimmer, W.G. Ellison, and J.E. Goetz. 2015. Bicknell’s Thrush (Catharus bicknelli), The Birds of North America (P.G. Rodewald, Ed.). Ithaca: Cornell Lab of Ornithology; Retrieved from the Birds of North America: https://birdsna.org/Species-Account/bna/species/bicthr.

Tremblay, J. A., Y. Boulanger, D. Cyr, A.R. Taylor, and D.P. Price. 2017. In Prep. Using forest landscape models to support recovery strategies in a climate-change context: the Bicknell’s Thrush as a case study.

Tremblay, J.A. 2017. Peer review comments on the U.S. Fish and Wildlife Service’s Biological Species Report for the Bicknell’s Thrush (Catharus bicknelli), version 1.1 (April 2017). Electronic mail and attachment dated May 11, 2017.

Uchytil, R.J. 1991. Abies balsamea. In: Fire effects Information System, [Online}. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory (Producer). Retrieved from http://www.fs.fed.us/database/feis/plants/tree/abibal/all.html 12/, December 20, 2016.

United Nations Framework Convention on Climate Change. 2015. Adoption of the Paris Agreement. Report Number FCCC/CP/2015/L.9/Rev.1. Retrieved from http://unfccc.int/resource/docs/2015/cop21/eng/l09r01.pdf, accessed December 7, 2016.

United States Department of Agriculture Forest Service. Undated. Pest alert—hemlock looper. Retrieved from https://www.na.fs.fed.us/spfo/pubs/pest_al/hem-looper/hem-looper.htm, accessed May 21, 2017.

United States Environmental Protection Agency. 2014. 2014 Program Progress - Clean Air Interstate Rule, Acid Rain Program, and former NOx Budget Trading Program. Retrieved from https://www3.epa.gov/airmarkets/progress/reports/index.html, accessed December 6, 2016.

United States Fish and Wildlife Service (USFWS). 2003. USFWS interim guidance on avoiding and minimizing wildlife impacts from wind turbines. 55 pp. Available at: http://www.fws.gov/habitatconservation/wind.pdf.

United States Fish and Wildlife Service (USFWS). 2016. Draft Letter of Intent to Promote Conservation of Bicknell’s Thrush in the Dominican Republic, the United States of America, and Canada. 3 pp.

United States Forest Service. 2006. National visitor use monitoring results for the White Mountain National Forest. Unpublished report. 51 pp.

United States Forest Service. 2012. Visitor use report. White Mountain NF. Data collected FY 2010. Unpublished report. 59 pp. van Vuuren, D.P., J. Edmonds, M. Kainuma, K. Riahi, A. Thomson, K. Hibbard, G.C. Hurtt, T. Kram, V. Krey, J-F Lamarque, T. Masui, M. Meinshausen, N. Nakicenovic, S.J. Smith, and S.K. Rose. 2011. The representative concentration pathways: an overview. Climate Change. 109: 5–31.

Vermont Department of Forests, Parks and Recreation. 1997. Vermont Forest Resource Harvest Summary. 1997. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 1998. Vermont Forest Resource Harvest Summary. 1998. Retrieved from

http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 1999. Vermont Forest Resource Harvest Summary. 1999. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2000. Vermont Forest Resource Harvest Summary. 2000. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2001. Vermont Forest Resource Harvest Summary. 2001. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2002. Vermont Forest Resource Harvest Summary. 2002. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2003. Vermont Forest Resource Harvest Summary. 2003. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2004. Vermont Forest Resource Harvest Summary. 2004. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2005. Vermont Forest Resource Harvest Summary. 2005. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2006. Vermont Forest Resource Harvest Summary. 2005. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2007. Vermont Forest Resource Harvest Summary. 2005. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2008. Vermont Forest Resource Harvest Summary. 2005. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2009. Vermont Forest Resource Harvest Summary. 2005. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2010. Vermont Forest Resource Harvest Summary. 2005. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2011. Vermont Forest Resource Harvest Summary. 2005. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2013. Vermont Forest Resource Harvest Summary. 2005. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vermont Department of Forests, Parks and Recreation. 2014. Vermont Forest Resource Harvest Summary. 2005. Retrieved from http://fpr.vermont.gov/forest/forest_business/forest_statistics/harvest_reports, accessed December 27, 2016.

Vitousek, P.M., J.D. Aber, R.W. Howarth, G.E. Likens, P.A. Matson, D.W. Schindler, W.H. Schlesinger, and D.G. Tilman. 1997. Technical Report: Human alteration of the global nitrogen cycle: Sources and Consequences. Ecological Applications. 7(3): 737–750.

Vogelmann, J.E., G. Xian, C. Homer, and B. Tolk. 2012. Monitoring gradual ecosystem change using Landsat time series analyses: Case studies in selected forest and rangeland ecosystems. USGS Staff- Published Research. Paper 528. Retrieved from http://digitalcommons.unl.edu/usgsstaffpub/528.

Wagner, R.G., J. Bryant, B. Burgason, M. Doty, B.E. Roth, P. Strauch, D. Struble, and D. Denico. 2016. Coming Spruce Budworm Outbreak: Initial Risk Assessment and Preparation & Response Recommendations for Maine’s Forestry Community. Cooperative Forestry Research Unit, University of Maine, Orono. 77p.

Wallace, G. J. 1939. Bicknell’s thrush, its taxonomy, distribution, and life history. Proceedings of the Boston Society of Natural History. 41:211–402.

Wallace, G. L. 1949. Bicknell's thrush. Pp. 199–217 In Life histories of North American thrushes, kinglets, and their allies. A. C. Bent, Ed. U.S. National Museum Bulletin 196.

Wang, X., D.K. Thompson, G.A. Marshall, C. Tymstra, R. Carr, and M.D. Flannigan. 2015. Increasing frequency of extreme fire in Canada with climate change. 130(4): 573–586.

Wang, W.J., H.S. He, F.R. Thompson III, J.S. Fraser, and W.D. Dijak. 2016. Changes in forest biomass and tree species distribution under climate change in the northeastern United States. Landscape Ecology. 1–15.

Whittam, B. 2017. Peer review comments on the U.S. Fish and Wildlife Service’s Biological Species Report for the Bicknell’s Thrush (Catharus bicknelli), version 1.1 (April 2017). Electronic mail and attachment dated May 2, 2017.

Whittam, B. 2015. Bicknell’s Thrush, p. 390–391 in Stewart, R.L.M., K.A. Bredin, A. Couturier, A. G. Horn, D. Lepage, S. Makepeace, P.D. Taylor, M.A. Villard, and R.M. Whittam (eds). 2015. Second Atlas of Breeding Birds of the Maritime Provinces. Bird Studies Canada, Environment Canada, Natural History Society of Prince Edward Island, Nature New Brunswick, New Brunswick Department of Natural Resources, Nova Scotia Bird Society, Nova Scotia Department of Natural Resources, and Prince Edward Island Department of Agriculture and Forestry, Sackville, 528 + 28 pp.

Williamson, P. 2016. Scrutinize CO2 removal methods. Nature 530: 153–155.

Wilson, M.D. and B.D. Watts. 1997. Autumn migration of grey-cheeked and Bicknell’s thrushes at Kiptopeke, Virginia. Journal of Field Ornithology. 68(4):519–525.