2008 NORTHEASTERN NATURALIST 15(4):607–618 Decline of a New Hampshire Bicknell’s Thrush Population, 1993–2003 J. Daniel Lambert1,2,*, David I. King3, John P. Buonaccorsi4, and Leighlan S. Prout5 Abstract - Catharus bicknelli (Bicknell’s Thrush) is a rare inhabitant of mountain forests in the northeastern United States and southeastern Canada. Conservation planners consider the species to be at risk, although evidence of population decline has thus far been localized or inconclusive. In order to assess the status of Bicknell’s Thrush in the White Mountains of New Hampshire, we conducted point-count surveys on 40 forested, high-elevation routes from 1993 to 2003. Non-linear regression on aggregate counts revealed a 7% annual decline over this period (P < 0.1). We discuss possible threats to Bicknell’s Thrush, including winter habitat loss, pollution of mountain ecosystems, climate change, and human intrusion during breeding. A range-wide monitoring program that incorporates new survey methods is needed to help identify limiting factors and reduce potential sources of error and bias. This study underscores the importance of efforts to monitor and conserve Bicknell’s Thrush. Introduction Catharus bicknelli (Ridgway) (Bicknell’s Thrush) is a rare, forest-dwelling passerine that nests at upper elevations in New York, Vermont, New Hampshire, and Maine (Atwood et al. 1996, Lambert et al. 2005), and in scattered highland and coastal areas of southeastern Canada (Nixon 1999, Ouellet 1993). Its global population may number fewer than 50,000 individuals; however, habitat and density data are too scarce to permit an accurate estimate (Rimmer et al. 2001, 2005a). A Bicknell’s Thrush abundance model for New Hampshire’s White Mountains, derived from point counts, habitat measurements, and satellite imagery, produced an estimate of 5000 birds (95% CI = 900–23,000; Hale 2006). The White Mountains lie at the core of the species’ breeding range and contain one-third of the potential Bicknell’s Thrush habitat in the United States (Lambert et al. 2005; J.D. Lambert, unpubl. data). Bicknell’s Thrush primarily inhabits young or chronically disturbed forests dominated by Abies balsamea (L.) P. Mill. (Balsam Fir), with variable amounts of Picea rubens Sarg. (Red Spruce) and Betula papyrifera var. cordifolia (Regel) Fern. (Paper Birch) (Atwood et al. 1996, Connolly et al. 2002). It also utilizes regenerating stands of mixed forest in northern New England (Vermont Center for Ecostudies [VCE], Norwich, VT, unpubl. data), Québec (Ouellet 1Vermont Institute of Natural Science, PO Box 1281, Quechee, VT 05059. 2Current address - American Bird Conservancy, c/o Vermont Center for Ecostudies, PO Box 420, Norwich, VT 05055. 3US Forest Service Northern Experiment Station, 203 Holdsworth Natural Resources Center, University of Massachusetts, Amherst, MA 01003. 4Department of Mathematics and Statistics, University of Massachusetts, Amherst, MA 01003. and 5White Mountain National Forest, 719 North Main Street, Laconia, NH 03246. *Corresponding author - dlambert@abcbirds.org. 608 Northeastern Naturalist Vol. 15, No. 4 1993), and Nova Scotia (Campbell and Whittam 2006). This alternative habitat, which follows disturbance by fire or logging, contains the majority of New Brunswick’s breeding population (Nixon et al. 2001). Most typical and alternative habitats are characterized by high vegetation density and canopies less than 6 m in height (Connolly et al. 2002, Nixon et al. 2001, Rimmer et al. 2001, Sabo 1980). Wintering birds concentrate in montane broadleaf forests of the Dominican Republic (Rimmer et al. 2001), with isolated populations in Haiti (Rimmer et al. 2005b) and eastern Cuba (Rompré et al. 2000). Its limited distribution, low numbers, and specialized habitat requirements make Bicknell’s Thrush vulnerable to extinction. New York, Vermont, New Hampshire, Maine, and Nova Scotia all list the songbird as a species of special concern, as does the Committee on the Status of Endangered Wildlife in Canada (COSEWIC 1999). It is extirpated from Massachusetts (Veit and Petersen 1993). The North American Bird Conservation Initiative identifies Bicknell’s Thrush among the highest priority landbirds in the Atlantic Northern Forest (Dettmers 2003), while Partners in Flight has placed the species on its continental watch list, citing multiple causes for concern (Rich et al. 2004). Bicknell’s Thrush has disappeared from several island and coastal locations in Canada during the 20th century (Nixon 1999), and from low-mountain sites in the United States (Atwood et al. 1996, Lambert et al. 2001). Although these extirpations suggest a decrease in overall numbers, there has been no largescale assessment of population trends for this species. The North American Breeding Bird Survey (BBS), which is conducted along roads, has recorded just 74 Bicknell’s Thrushes since 1966, nearly half of these in Nova Scotia (USGS Patuxent Wildlife Research Center 2006). Two routes that pass through mountain notches account for most of the US observations. To address the gap in coverage, in 1993 the US Forest Service initiated standardized surveys of mountain birds in the White Mountain National Forest. In this paper, we present trend analyses for the first decade of monitoring (1993–2003) to provide information on the status of Bicknell’s Thrush populations in montane spruce-fir forests in the White Mountains. Methods Field-site description Forty high-elevation routes were established systematically along footpaths in the White Mountain National Forest, located in north-central New Hampshire. Routes were selected to provide broad geographic coverage and a representative sample of montane spruce-fir forest (Fig. 1). A typical route started at the lower spruce-fir ecotone, continued up the path through a variety of age classes, and over the mountain until it passed back out of the focal habitat. Sampled elevations ranged from 740 m to 1470 m, with some stations falling on exposed ridgelines or in patches of mixed forest adjacent to conifer-dominated stands. Field survey From 1993 to 2003, trained observers conducted annual point counts between 5 and 28 June. Observers performed their first count at a fixed point 2008 J.D. Lambert, D.I. King, J.P. Buonaccorsi, and L.S. Prout 609 and used a rope to measure 250-m intervals between each of the subsequent points. We reviewed field records to verify that points were consistently placed and eliminated counts if evidence indicated an error in point placement. Surveys were completed between 0500 and 1100 EST. Observer turnover was high, with 30 observers participating for one year, 7 participating for two years, and another 2 observers participating for three and four years, respectively. For each route, start times were standardized and count dates fell within the same 10-day window each year. Observers recorded all individuals seen and heard during silent, 5-min counts. Surveys were not conducted in the rain or when wind speeds exceeded 40 km/h. To assess the effect of wind on our trend estimates, we ran analyses with and without counts conducted in winds of 31 to 40 km/h. Figure 1. Location of routes surveyed for Bicknell’s Thrush population trend analysis, 1993–2003. Shaded areas represent potential Bicknell’s Thrush habitat within and bordering the White Mountain National Forest (Lambert et al. 2005). Squares depict routes with pronounced decline (trend + 2 SE < 1.0). 610 Northeastern Naturalist Vol. 15, No. 4 Analysis Our primary analysis estimates an overall trend by fitting total yearly counts, aggregated over routes, as a function of time. This approach calls for a common set of routes, and points within routes, to be surveyed for each year. Originally, our data did not meet this requirement for consistency because several surveys were canceled or cut short due to rain or high winds. To resolve this issue, we used two subsets of data, the first providing maximum temporal coverage (17 routes in 9 years) and the second providing larger geographic coverage (39 routes in 7 years). The effective route length averaged 10.6 points (± 4.2 SD) in the first group (range = 1–18) and 11.8 points (± 3.2 SD) in the second group (range = 4–18). Once a common set of routes and points was designated, we assumed the count in year y (Cy) followed the model, E(Cy) = a * by, where the trend coeffi- cient b is the annual rate of change in the population size. We fit this model using weighted non-linear least squares (via PROC NLIN in SAS), assuming the variance of the count was proportional to the expected count. This method is simply Poisson regression, allowing for over- or under-dispersion, and is equivalent to using the estimating equation approach of Link and Sauer (1994). The estimated trend applies to the total area defined by the collection of routes entered into the analysis. This analysis also produces standard errors and confidence intervals for both the coefficients in the model and for the estimated abundances at given points in time. We investigated the adequacy of the model both through residual analyses and by fitting a nonparametric curve (using LOESS in SAS), and our model appeared to fit the data well (see Fig. 2). Figure 2. Fitted curve (solid), LOESS fit (dashed), and 95% confidence interval (dotted) for Bicknell’s Thrush population trends in the White Mountains on (A) 17 routes surveyed in 9 years (1993–2000, 2003) and (B) 39 routes surveyed in 7 years (1993, 1995–2000). 2008 J.D. Lambert, D.I. King, J.P. Buonaccorsi, and L.S. Prout 611 Since the data come from a time series, we also examined the residuals to check for serial correlation in the error term. Plots of the residual versus time and versus the previous residual showed that serial correlation was relatively weak in both analyses. Numerically, the correlation between adjacent residuals was 0.11 and 0.20 for the analyses with nine and seven years, respectively. For these reasons, we have used the non-linear regression analysis, assuming uncorrelated errors. We carefully considered whether to include route or observer effects in our models, and determined that their inclusion was not warranted. Although route characteristics influence bird counts, our treatment of the routes as fixed essentially builds these effects into the model. Thus, route effects do not need to be explicitly modeled in order to estimate the overall trend. The commonly employed route-regression technique, which estimates overall trend with a weighted average of route-specific trends (e.g., Geissler and Sauer 1990), could not be applied as it requires routes to be placed randomly. Even with random routes, there are additional questions as to how the weighted average of individual trends estimates a regional trend (although a mixed-model framework has begun to address this issue; see Link and Sauer 2002). However, information on routespecific trends can be useful for examining spatial patterns in population change. We therefore fit a trend for each route, using a common set of points sampled in each year. The number of years of data varied among routes, depending on sampling history. In some cases, the model fit was poor due to low counts, as indicated by the large confidence intervals on estimated trends (see Fig. 3). We were not able to include observer effects in our analyses because observer turnover was very high (95% of the observers participated in just one or two years), making it impossible to separate observer effects from year effects. Figure 3. Routelevel estimates of Bicknell’s Thrush population trends in the White Mountains (± 2 SE). Route numbers on xaxis correspond with locations mapped in Figure 1. 612 Northeastern Naturalist Vol. 15, No. 4 Results We detected 280 Bicknell’s Thrushes on the 17 routes that were sampled in each of the nine survey years (1993–2000, 2003). We recorded 624 individuals on the 39 routes that were surveyed in seven of the years (1993, 1995–2000). The non-linear regression on aggregate counts revealed negative trends for the subset of data that incorporated the most years (trend ± SE = 0.93 ± 03, 95% CI = 0.85–1.01, P = 0.06) and for the subset that incorporated the most routes (trend ± SE = 0.93 ± 03, 95% CI = 0.85–1.01, P = 0.09) (Fig. 2). These findings correspond with a rate of decline averaging 7% per year. Both data series were characterized by an abrupt decrease in the total count between 1997 and 1998. On the 17 routes surveyed in 9 years, Bicknell’s Thrush numbers were 64% lower in 2003 (total = 47) than in 1997 (total = 17). On the 39 routes monitored in 7 years, the count was halved between 1997 (total = 116) and 2000 (total = 58). Discarding counts that were conducted during periods of elevated wind (31–40 km/h) led to the elimination of certain routes in certain years and therefore limited which routes could be used in the aggregate analysis. In fact, a curve could not be fit using all years. We were able to fit a curve to data from 1993 and 1995–2000, after the windy counts were eliminated. The resulting trend estimate (0.95) was similar to the original value (0.93); however, it lacked statistical significance (P = 0.5). We note that the trend of 0.95 was estimated from only 12 routes, compared to the 39 routes included in the analysis of counts made under all wind conditions less than 40 km/h in 1993 and 1995–2000. Route-specific trends, estimated for 40 White Mountain routes (Fig. 3), were negative on 29 routes and positive on 11 routes. The most pronounced population changes were declines, which were detected throughout the study area (Fig. 1) and across a wide range of elevations. The two greatest declines occurred in small habitat patches on Terrace Mountain (route 39; trend = 0.60) and Mount Crawford (route 26; trend = 0.66). Routes located in extensive habitat of the Presidential Range (routes 28–32) and the Carter-Moriah Range (routes 34–36) showed the least evidence of population decline (Figs. 1 and 3). Discussion The decline of Bicknell’s Thrush in our survey area is consistent with a pattern of local extirpations. During the 20th century, Bicknell’s Thrushes have disappeared from several island and coastal locations in Canada (Nixon 1999) and from low-mountain sites in the United States (Atwood et al. 1996, Lambert et al. 2001). A peripheral population on Mount Greylock, MA fell from an estimate of 10 pairs in the 1950s to 0 in 1973 (Veit and Petersen 1993). Data from the BBS indicate that Bicknell’s Thrushes have also disappeared from Dixville Notch, NH and Black Brook Notch, ME after occupying these sites for a number of years (USGS Patuxent Wildlife Research Center 2006). Road-based BBS counts have produced few other records of this mountain-dwelling species in the United States. Even when US and Canadian BBS records are combined, observations of Bicknell’s 2008 J.D. Lambert, D.I. King, J.P. Buonaccorsi, and L.S. Prout 613 Thrush are too sparse to meet minimum requirements for trend estimation. While local extirpations may be symptomatic of a declining population, they also occur in stable populations as a result of changing habitat conditions. It is especially difficult to generalize site-based information for species, like Bicknell’s Thrush, that utilize ephemeral habitat patches within a forest type that is subject to disturbance and re-growth. Surveys at migration stopover sites corroborate the evidence of decline in the breeding areas. Wilson and Watts (1997) reported a decrease in autumn capture rates for Bicknell’s Thrush and Catharus minimus Lafresnaye (Graycheeked Thrush) between 1968 and 1995 in coastal Virginia. The two species were analyzed together because C. bicknelli was considered a subspecies of C. minimus until 1995 and because morphometric overlap is considerable. The authors’ analytical approach, rank correlation, detected a significant reduction in captures over time, but yielded no estimate of trend. Capture rates of Gray-cheeked/Bicknell’s Thrushes also declined in coastal Massachusetts between early years (1970–1985) and late years (1986–2001), dropping 66% and 44% during spring and fall migrations, respectively (Lloyd-Evans and Atwood 2004). Mountain Birdwatch, an effort to monitor high-elevation breeding birds at sites in New York, Vermont, New Hampshire, and Maine, measured a 9% annual decline (P = 0.07) in the relative abundance of Bicknell’s Thrush between 2001 and 2004 (n = 47 routes; Lambert 2005). This trend was similar to what we observed in the White Mountain population between 1993 and 2003, though the two studies had only one year in common (2003). In adjacent areas of Canada, the High Elevation Landbird Program detected significant reductions in Bicknell’s Thrush numbers between 2003 and 2007 (Campbell et al. 2007; B. Whittam, Bird Studies Canada, Sackville, NB, Canada, pers. comm.). Forestry operations in the New Brunswick survey area (e.g., precommercial thinning) may have influenced results from that province. If negative trends persist in the White Mountains, the region’s estimated population of 5000 birds (Hale 2006) could be endangered within a few decades. Based on the historic pattern of extirpation (Nixon 1999, Veit and Petersen 1993), low mountains with small habitat patches would be at greatest risk of losing the species. Mount Crawford (951 m) and Terrace Mountain (1114 m), sites of the most abrupt declines, match this profile. In contrast, clusters of stable or increasing trends were observed in large habitat blocks, including the vast montane spruce-fir of the Presidential Range. Short-term changes in Bicknell’s Thrush abundance may not accurately reflect long-term trends, as bird populations may fluctuate over time (King et al. 2006, Peterjohn et al. 1995). The trends we report might even reverse themselves. Mountain Birdwatch results from 2005–2007 support this possibility, showing an increase in Bicknell’s Thrush numbers on New York and northern New England survey routes (VCE, unpubl. data) following a period of decline over the four previous years (Lambert 2005). It is also possible that the population decrease observed in this study was offset by gains elsewhere in the White Mountains. However, we believe that this is unlikely for three reasons. First, none of the widely distributed routes 614 Northeastern Naturalist Vol. 15, No. 4 detected a significant increase during the survey interval. Second, most of the population change occurred during a three- to six-year period, a span too short for significant shifts in the location of suitable habitat. Finally, Bicknell’s Thrush exhibits high site fidelity and can live to be 10 or more years under favorable conditions (VCE, unpubl. data). Factors underlying the observed decline are unknown, but potential threats to Bicknell’s Thrush include: winter habitat loss, atmospheric pollution, climate change, and disturbance by hikers. The Dominican Republic, where most Bicknell’s Thrushes overwinter, has lost approximately 90% of its native forest (Stattersfield et al. 1998), including large areas of montane broadleaf habitat. Although winter habitat loss has been extensive, evidence of winter limitation is lacking. Studies are needed to quantify the extent and use of remaining winter habitat. Changes to breeding habitat may also influence Bicknell’s Thrush populations. For example, an increase in canopy height or loss of subcanopy structure could reduce availability of suitable nest sites. Such changes are incremental in our study area, when and where they occur. In general, White Mountain spruce-fir forests grow slowly, experience regular natural disturbance, and are not subject to adverse silvicultural treatments, such as precommercial thinning (Chisholm 2005). Atmospheric pollution has also been cited as a potential threat to Bicknell’s Thrush (Atwood et al. 1996). The White Mountains receive high levels of acid deposition (Ollinger et al. 1993), which leaches calcium from soils and foliage (DeHayes et al. 1999). In Europe, depletion of calcium-rich invertebrate prey from acidified forests has been implicated in egg-laying irregularities and reduced reproductive success in forest passerines (Graveland and vanderWal 1996, Graveland et al. 1994). In a study of Hylocichla mustelina Gmelin (Wood Thrush) in the eastern United States, Hames et al. (2002) found a negative relationship between acid rain and breeding probability, especially at upper elevations where soils are thin and poorly buffered. Birds that nest in acidified ecosystems face an elevated risk of mercury contamination because acidic environments promote the conversion of inorganic mercury (Hg) to toxic methylmercury (MeHg) (Miskimmin et al. 1992). The large amount of mercury deposited on northeastern mountains (Miller et al. 2005) compounds the threat to high-elevation insectivores, such as Bicknell’s Thrush. Rimmer et al. (2005c) found elevated MeHg levels, with highest concentrations in older males. Studies of aquatic birds have shown that mercury contamination can limit avian reproduction and survival (Chan et al. 2003). Research is needed to determine toxicity thresholds in Bicknell’s Thrush. Another class of pollutants, greenhouse gases, may further impair Bicknell’s Thrush habitat. A warming climate is expected to significantly reduce or eliminate Balsam Fir forests in the Northeast (Iverson and Prasad 2002). At upper elevations, Red Spruce and Balsam Fir may be replaced by Betula alleghaniensis Britt. (Yellow Birch), Fagus grandifolia Ehrh. (American Beech), and Tsuga canadensis (L.) Carr. (Eastern Hemlock) (Lee et al. 2005). Even a modest temperature increase could confine Bicknell’s Thrush to the region’s highest peaks (Rodenhouse et al. 2008). Recreational disturbance is an additional source of concern. Foot traffic 2008 J.D. Lambert, D.I. King, J.P. Buonaccorsi, and L.S. Prout 615 is steadily increasing in the White Mountains (US Forest Service 2005), with unknown effects on birds. Studies conducted elsewhere have shown that human intrusions can reduce singing activity (Gutzwiller et al. 1994), influence nest placement (Knight and Fitzner 1985, Miller et al. 1998), and limit both density (Mallord et al. 2007) and reproductive success (Murison et al. 2007) in passerines. Trails may also influence species composition and rates of nest predation in adjacent forests (Miller and Hobbs 2000, Miller et al. 1998). Although each of these areas requires focused research, improved monitoring could contribute to an understanding of limiting factors, while strengthening the inference of future trend results. We suggest the following measures be incorporated into a regionally coordinated survey design: 1) random route placement within a discrete, rangewide sampling frame; 2) quantitative evaluation of observer ability and/or greater continuity in observers from year to year; 3) point-count methods that account for variable detection rates; and 4) repeat sampling to estimate and track changes in occupancy. Validation of trail-based surveys is also needed to determine whether populations monitored along footpaths differ from those located away from trails. Finally, stewards of mountain habitat require information on environmental covariates that may influence Bicknell’s Thrush populations. Conservation and management efforts would benefit from a better understanding of the role of: habitat change, climatic change, mercury exposure, calcium availability, predation risk, hiker impacts, and changes in the availability of wintering habitat. Field measurements and GIS models of these variables should be integrated into the design of future monitoring. Bicknell’s Thrush has received high conservation priority ranks on the basis of its rarity, limited distribution, and reviews of existing and potential threats. The lack of meaningful trend information has hampered conservation status assessments for the species. The detection of negative trends in the White Mountain National Forest between 1993 and 2003 underscores the importance of efforts to conserve Bicknell’s Thrush and improve monitoring of high-elevation birds in the Northeast. Acknowledgments The study was funded by the White Mountain National Forest, with additional assistance from the Stone House Farm Fund of the Upper Valley Community Foundation and friends and trustees of the Vermont Institute of Natural Science. We gratefully acknowledge the Audubon Society of New Hampshire and US Forest Service personnel who conducted the surveys. Laura Deming coordinated the point counts for several years and provided essential information and support. Kent McFarland prepared Figure 1. We thank Yves Aubry, Jon Bart, Becky Whittam, and an anonymous reviewer, who provided helpful comments on earlier versions of this manuscript. Literature Cited Atwood, J.A., C.C. Rimmer, K.P. McFarland, S.H. Tsai, and L.N. Nagy. 1996. Distribution of Bicknell’s Thrush in New England and New York. Wilson Bulletin 108:650–661. Campbell, G., and B. Whittam. 2006. Bicknell’s Thrush habitat in Nova Scotia’s 616 Northeastern Naturalist Vol. 15, No. 4 industrial forest: Final report to the Nova Forest Alliance and Stora Enso Port Hawkesbury, Ltd. Unpublished report. Bird Studies Canada (Atlantic Region), Sackville, NB, Canada. 17 pp. Campbell, G., B. Whittam, and G. Robertson. 2007. High Elevation Landbird Program 5-year report. Unpublished report. Bird Studies Canada (Atlantic Region), Sackville, NB, Canada. 16 pp. Chan, H.M., A.M. Scheuhammer, A. Ferran, C. Loupelle, J. Holloway, and S. Weech.2003. Impacts of mercury on freshwater fish-eating wildlife and humans. Human and Ecological Risk Assessment 9:867–883. Chisholm, S.E. 2005. The impact of precommercial thinning on abundance of Bicknell’s Thrush and other high-elevation birds. M.Sc. Thesis. Dalhousie University, Halifax, NS, Canada. Committee on the Status of Endangered Wildlife in Canada (COSEWIC) 1999. COSEWIC assessment and status report on the Bicknell’s thrush Catharus bicknelli in Canada. Ottawa, ON, Canada. v + 43 pp. Connolly, V., G. Seutin, J.-P. L. Savard, and G. Rompré. 2002. Habitat use by Bicknell’s Thrush in the Estrie Region, Québec. Wilson Bulletin 114:333–341. DeHayes, D.H., P.G. Schaberg, G.J. Hawley, and G.R. Strimbeck 1999. Acid rain impacts calcium nutrition and forest health. BioScience 49:789–800. Dettmers, R. 2003. Draft: Blueprint for the design and delivery of bird conservation in the Atlantic Northern Forest. Unpublished report. US Fish and Wildlife Service Northeast Region, Hadley, MA. 352 pp. Geissler, P.H., and J.R. Sauer. 1990. Topics in route-regression analysis. Pp. 54–57, In J.R. Sauer and S. Droege (Eds.). Survey Designs and Statistical Methods for the Estimation of Avian Population Trends. US Fish and Wildlife Service, Washington, DC. 171 pp. Graveland, J., and R. van der Wal. 1996. Decline in snail abundance due to soil acidification causes eggshell defects in forest passerines. Oecologia 105:351–360. Graveland, J., R. van der Wal, J.H. van Balen, and A.J. van Noordwijk. 1994. Poor reproduction in forest passerines from decline of snail abundance on acidified soils. Nature 368:446–448. Gutzwiller, K.J., R.T. Wiedenmann, K.L. Clements, and S.H. Anderson. 1994. Effects of human intrusion on song occurrence and singing consistency in subalpine birds. Auk 111:28–37. Hale, S.R. 2006. Using satellite imagery to model distribution and abundance of Bicknell’s Thrush (Catharus bicknelli) in New Hampshire’s White Mountains. Auk 123:1038–1051. 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 the Wood Thrush (Hylocichla mustelina) in North America. Proceedings of the National Academy of Sciences of the United States of America 99:11235–11240. Iverson, L.R., and A.M. Prasad. 2002. Potential redistribution of tree species habitat under five climate change scenarios in the eastern US. Forest Ecology and Management 155:205–222. King, D.I., J.H. Rappole, and J. Buonaccorsi. 2006. Non-linear route-regression of breeding bird census data: Long-term population trends of forest-dwelling Neotropical-nearctic migrant birds. Bird Populations 7:1–9. Knight, R.L., and R.E. Fitzner. 1985. Human disturbance and nest-site placement in Black-billed Magpies. Journal of Field Ornithology 56:153–157. Lambert, J.D. 2005. Mountain Birdwatch 2004: Final report to the US Fish and Wildlife Service. VINS Technical Report 05-1. Vermont Institute of Natural Science, Woodstock, VT. 15 pp. Lambert, J.D., K.P. McFarland, C.C. Rimmer, and S.D. Faccio. 2001. Mountain 2008 J.D. Lambert, D.I. King, J.P. Buonaccorsi, and L.S. Prout 617 Birdwatch 2000: Final report to the US Fish and Wildlife Service. VINS Technical Report 01-1. Vermont Institute of Natural Science, Woodstock, VT. 12 pp. 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 117:1–11. Lee, T.D., J.P. Barrett, and B. Hartman. 2005. Elevation, substrate, and the potential for climate-induced tree migration in the White Mountains, New Hampshire, USA. Forest Ecology and Management 212:75–91. Link, W.A., and J.R. Sauer. 1994. Estimating equations estimates of trends. Bird Populations 2:23–32. Link W.A., and J.R. Sauer. 2002. A hierarchical analysis of population change with application to Cerulean Warblers. Ecology 83:2832–2840. Lloyd-Evans, T.L., and J.L. Atwood. 2004. 32 years of changes in passerine numbers during spring and fall migrations in coastal Massachusetts. Wilson Bulletin 116:1–16. Mallord, J.W., P.M. Dolman, A.F. Brown, and W.J. Sutherland. 2007. Linking recreational disturbance to population size in a ground-nesting passerine. Journal of Applied Ecology 44:185–195. Miller, E.K., A. Vanarsdale, J.G. Keeler, A. Chalmers, L. Poissant, N. Kamman, and R. Brulotte. 2005. Estimation and mapping of wet and dry mercury deposition across northeastern North America. Ecotoxicology 14:53–70. Miller, J.R., and N.T. Hobbs. 2000. Recreational trails, human activity, and nest predation in lowland riparian areas. Landscape and Urban Planning 50:227–236. Miller, S.G., R.L. Knight, and C.K. Miller. 1998. Influence of recreational trails on breeding bird communities. Ecological Applications 8:162–169. Miskimmin, B.M., J.W.M. Rudd, and C.A. Kelly. 1992. Influence of dissolved organic carbon, pH, and bacterial respiration rates on mercury methylation and demethylation in lake water. Canadian Journal of Fisheries and Aquatic Science 49:17–22. Murison, G., J.M. Bullock, J. Underhill-Day, R. Langston, A.F. Brown, and W.J. Sutherland. 2007. Habitat type determines the effects of disturbance on the breeding productivity of the Dartford Warbler Sylvia undata. Ibis 149:16–26. Nixon, E. 1999. COSEWIC status report on the Bicknell’s Thrush Catharus bicknelli in Canada. Pp. 1–43, In COSEWIC Assessment and Status Report on the Bicknell’s Thrush Catharus bicknelli in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa, ON, Canada. Nixon, E.A., S.B. Holmes, and A.W. Diamond. 2001. Bicknell’s Thrushes (Catharus bicknelli) in New Brunswick clear cuts: Their habitat associations and co-occurrence with Swainson’s Thrushes (Catharus ustulatus). Wilson Bulletin 113:33–40. Ouellet, H. 1993. Bicknell’s Thrush: Taxonomic status and distribution. Wilson Bulletin 105:545–572. Ollinger, S.V., J.D. Aber, G.M. Lovett, S.E. Millham, R.G. Lathrop, and J.M. Ellis. 1993. A spatial model of atmospheric deposition for the northeastern US. Ecological Applications 3:459–472. Peterjohn, B.G., J.R. Sauer, and C.S. Robbins. 1995. Population Trends from the North American Breeding Bird Survey. Pp. 3–39 In T.E. Martin and D.M. Finch (Eds.). Ecology and Management of Neotropical Migratory Birds. Oxford University Press, Oxford, UK. 506 pp. Rich, T.D., C.J. Beardmore, H. Berlanga, P.J. Blancher, M.S.W. Bradstreet, G.S. 618 Northeastern Naturalist Vol. 15, No. 4 Butcher, D.W. Demarest, E.H. Dunn, W.C. Hunter, E.E. Inigo-Elias, J.A. Kennedy, A.M. Martell, A.O. Panjabi, D.N. Pashley, K.V. Rosenberg, C.M. Rustay, J.S. Wendt, and T.C. Will. 2004. Partners in Flight North American Landbird Conservation Plan. Cornell Lab of Ornithology, Ithaca, NY. 84 pp. Rimmer, C.C., K.P. McFarland, W.G. Ellison, and J.E. Goetz. 2001. Bicknell’s Thrush (Catharus bicknelli). No. 592, In A. Poole and F. Gill (Eds.). The Birds of North America. Academy of Natural Sciences, Philadelphia, PA and American Ornithologists’ Union, Washington, DC. 28 pp. Rimmer, C.C., J.D. Lambert, and K.P. McFarland. 2005a. Bicknell’s Thrush (Catharus bicknelli) conservation strategy for the Green Mountain National Forest. VINS Technical Report 05-5. Vermont Institute of Natural Science, Woodstock, VT. 27 pp. 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. Ornitología 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 levels in Bicknell’s Thrush and other insectivorous passerines in montane forests of northeastern North America. Ecotoxicology 14:223–240. Rodenhouse, N.L., S.N. Matthews, K.P. McFarland, J.D. Lambert, L.R. Iverson, A. Prasad, T.S. Sillett, 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. Rompré, G., Y. Aubry, and A. Kirkconnell. 2000. Recent observations of threatened birds in eastern Cuba. Cotinga 13:66. Sabo, S.R. 1980. Niche and habitat relations in subalpine bird communities of the White Mountains of New Hampshire. Ecological Monographs 50:241–259. Stattersfield, A.J., M.J. Crosby, A.D. Long, and D.C. Wege. 1998. Endemic bird areas of the world: Priorities for biodiversity conservation. Bird Life Conservation Serial no. 7. US Forest Service. 2005. Socio-economic assessment to provide a context for the White Mountain National Forest Plan Revision. US Forest Service, Laconia, NH. US Geological Survey Patuxent Wildlife Research Center. 2006. North American Breeding Bird Survey current data 1966–2005. Available online at http://www. pwrc.usgs.gov/bbs/retrieval. Accessed February 1, 2006. Veit, R.R., and W.R. Petersen. 1993. Birds of Massachusetts. Massachusetts Audubon Society, Lincoln, MA. 514 pp. Wilson, M.D., and B.D. Watts. 1997. Autumn migration of Gray-cheeked and Bicknell’s Thrushes at Kiptopeke, Virginia. Journal of Field Ornithology 68:519– 525.