0

2010 Progress Report to the Vermont Monitoring Cooperative



Part I. Demographic Monitoring of Montane Forest Birds on Mt. Mansfield
Part II. Population Connectivity of Bicknell’s Thrush: Insights from Geolocators
Part III. A Dynamic Occupancy Model of Bicknell’s Thrush Breeding Habitat
Part IV. Forest Bird Surveys on Mt. Mansfield and Lye Brook Wilderness




Submitted by:
Kent McFarland, Steven Faccio, and Christopher Rimmer
Vermont Center for Ecostudies
PO Box 420
Norwich, VT 05055
www.vtecostudies.org
Bicknell‟s Thrush with a backpack geolocator attached. ©KP McFarland

McFarland, Faccio, and Rimmer 2010 VMC Report
1




Part I. Demographic Monitoring of Montane Forest Birds on Mt. Mansfield

In 2010, we continued demographic monitoring of Bicknell‟s Thrush (Catharus bicknelli),
Swainson‟s Thrush (C. usutulatus), Blackpoll Warbler (Dendroica striata), Yellow-rumped
(Myrtle) Warbler (D. coronata coronata), and White-throated Sparrow (Zonotrichia albicollis),
completing our 19th consecutive field season on Mt. Mansfield.

Study Areas and Methods

We used mist-netting and banding to sample breeding populations of the five target species on an
established study plot on the Mt. Mansfield ridgeline between c. 1155-1190 m (3800-3900 ft)
elevation. We conducted banding sessions from the end of May into July during most years,
using 4-30 nylon mist nets (12 x 2.5-m and 6 x 2.5-m, 36-mm mesh) placed at sites that have
been used annually since 1992, primarily on the Amherst, Lakeview, and Long trails. Nets were
generally opened from late afternoon until dark and from pre-dawn until noon on the following
morning. Bicknell‟s Thrushes were captured both passively and through the use of vocal lures
(recorded conspecific vocalizations), while other species were passively captured. Each
individual was fitted with a uniquely-numbered U.S. Fish and Wildlife Service (USFWS) leg
band and in some cases a unique combination of 3 plastic colored leg bands. We recorded data
on age, sex, breeding condition, subcutaneous fat class, ectoparasites, flight feather wear, and net
site of capture. Standard morphometrics recorded include wing chord, tail length, weight, tarsal
length, culmen length, bill length from mid-nares, bill width, and bill depth. Several non-
destructive tissue samples were collected from mist-netted Bicknell‟s Thrushes for studies of
isotope markers and mercury burdens. During some years, we took a small blood sample (c. 50
µl) from the brachial vein. Each sample was stored in a heparinized capillary tube, refrigerated in
a vaccutainer in the field, and frozen within 12-48 hours. A tail feather (rectrix #5) on both sides
was clipped and stored in envelopes. Approximately 2 mm of claw tip from the middle toe of
both feet was collected using sharp dissection scissors and deposited in a small paper envelope.

We estimated and compared adult survivorship according to methods described in Lebreton et al.
(1992), Pradel et al. (1997) and Sillett and Holmes (2002), using the program MARK (White and
Burnham 1999). We defined a candidate model set that included a fully parameterized global
model and all reduced parameter models derived from the global model. We used mark-capture-
recapture data from three mountains along a north-south latitudinal gradient: Gaspé in Quebec,
Mt. Mansfield and Stratton Mountain in Vermont. We compared two habitats of interest at each
Vermont site: those with abrupt linear edges from ski trails and work roads, and natural areas
that contain only hiking trails. Because evidence suggested that individuals aged as second-year
often emigrate after their first breeding season, while older birds tend to have very high site
fidelity, we used transient in our modeling which compares Φ and t in year +1 as one group
(transients) and year 2+ as a second group. The model with the lowest Akaike Information
Criterium (AIC) was accepted as the most parsimonious model for our data. Model comparisons
within the candidate were made by deriving an index of plausibility using normalized Akaike

McFarland, Faccio, and Rimmer 2010 VMC Report
2

weights (Burnham and Anderson 2002). The ratio between weights of any two models indicated
the relative degree to which a particular model was better supported by the data.

Results and Discussion

In 2010 on Mt. Mansfield, we operated mist nets on 9 days between 3 June and 2 July,
accumulating 576 net-hours, with an average of 64 net-hours per day. We had 123 bird captures
and banded 70 new individuals of the five target species.

For our analysis of Bicknell‟s Thrush survivorship for data from 1993-2008, no models with year
effects for recapture probability (p) were supported. As expected, p was almost always higher for
males and on ski areas (Figure 1). The model with the lowest AICc in that candidate set was
{Phi(trans*site) p(site*sex*hab)}, in which survivorship was lower for transients than for
residents, lowest at Gaspé (the northernmost site), and very similar between Stratton and
Mansfield (Table 1 and 2). There was little statistical support for ski areas affecting Bicknell‟s
Thrush survivorship (Table 1). There was some support for annual variation in survivorship
among sites (Table 1). Little congruence among study sites in annual survivorship (Figure 2)
may have been caused either by local stochastic events or relatively high variance in annual
estimates. These estimates represent survivorship from one breeding season to the next; because
the species‟ annual cycle involves migration to and from the Caribbean, as well as a 6-month
overwintering period, a detailed understanding of population connectivity will be necessary to
fully elucidate demographic patterns and develop targeted conservation actions.

Acknowledgements
We are grateful to the Stowe Mountain Resort for allowing us access to the Mt. Mansfield toll
road and for overnight use of the ski patrol hut. We also thank Brendan Collins and Rosalind
Renfrew for their skilled and dedicated fieldwork, Scott Sillett for help with MARK analysis and
Yves Aubry of Canadian Wildlife Service for data from Canada study areas. Additional funding
for our work on Mt. Mansfield was provided by The Nature Conservancy, U.S. Forest Service
Office of International Programs, and the William P. Wharton Trust.

Literature Cited
Burnham, K. P. and D. R. Anderson. 2002. Model selection and multimodel inference: a
practical information-theoretic approach. Springer-Verlag Inc., New York, NY.

Lebreton, J-D., K. P. Burnham, J. Clobert, and D. R. Anderson. 1992. Modeling survival and
testing biological hypotheses using marked animals: a unified approach with case studies.
Ecological Monographs 62: 67-118.

Pradel, R., J. E. Hines, J-D Lebreton, and J. D. Nichols. 1997. Capture-recapture survival
models taking account of transients. Biometrics 53:60-72.

Sillett, T. S., and R. T. Holmes. 2002. Variation in survivorship of a migratory songbird
throughout its annual cycle. Journal of Animal Ecology 71:296-308.

White, G. C. and K. P. Burnham. 1999. Program MARK: survival estimation from populations
of marked animals. Bird Study 46 (Supplement): 120-138.

McFarland, Faccio, and Rimmer 2010 VMC Report
3


Table 1. Ten highest-ranked models for determining survivorship of Bicknell‟s Thrush adults.
There was little support for an effect of ski areas on survivorship, and only modest support for a
year effect.




Table 2. Survivorship results derived from the best fitting model {Phi(trans*site)
p(site*sex*hab)}.

McFarland, Faccio, and Rimmer 2010 VMC Report
4


Figure 1. Bicknell‟s Thrush recapture probabilities from the model p(site*sex*habitat).

McFarland, Faccio, and Rimmer 2010 VMC Report
5

Figure 2. Annual survivorship estimates for Bicknell‟s Thrush adults from the model
{Phi(year*trans*site) p(site*sex*hab)}. The data for period 2002-2003 were insufficient to allow
an estimate for Mt. Mansfield.

McFarland, Faccio, and Rimmer 2010 VMC Report
6

Part II. Population Connectivity of Bicknell’s Thrush: Insights from Geolocators.

Introduction

Natural selection acts on individual animals throughout the annual cycle, and events during each
phase of the annual cycle likely influence subsequent events. For migratory animals,
understanding these selection processes has been impossible because of our inability to follow
individuals year-round and determine where breeding populations winter, where winter
populations breed, as well as exact routes during migration. An understanding of these factors,
which could operate during breeding and/or non-breeding periods to limit and ultimately
determine bird abundance, is of urgent conservation concern. The most pressing need, and to
date the most seemingly intractable problem, has been to determine the movement patterns and
population connectivity of individuals between their breeding and wintering grounds. This is
critical to understand how limiting factors (e.g. habitat destruction, climate change, etc.) operate
in different parts of a species' annual cycle and to determine population size and local abundance.

Migratory connectivity is defined as the amount of population mixing between summer-breeding
and winter-non-breeding, as well as the stop-over and migratory pathways between them
(Webster et al. 2002). Although understanding space-use strategies and habitat preferences of
Nearctic-Neotropical migratory songbirds has grown substantially over the last 20 years,
knowledge of migratory connectivity remains poor (Remsen 2001, Marra and Webster 2005).
Information from bird banding has been limited by the scattered and irregular nature of banding
returns (Webster et al. 2002). Although stable isotope measures have provided latitudinal
gradients of habitat occupancy and a useful tool to assess migratory connectivity (Rubenstein et
al. 2002), this technique constitutes only a blunt instrument to assess migratory connectivity and
can sometimes fail to provide clear, meaningful information (Hobson et al. 2001). Many
migratory songbirds cross distances of over 5,000 km between breeding and wintering sites, yet
the exact timing, pathways followed and migratory connectivity remain speculative.

The critical importance of migratory connectivity to understanding the fundamental biology of
migratory birds has spurred a long history of research. Mark-recapture via bird-banding appears
to have been the first and most broadly-used technique to address this issue for songbirds.
However, despite banding millions of individual birds, there remains a relatively poor
understanding of migratory connectivity for most species because of extremely low recapture
rates (Webster et al. 2002). For example, only two Bicknell‟s Thrushes (Catharus bicknelli)
have been recaptured on their Hispaniolan wintering grounds despite the banding of over 3,000
individuals throughout the species‟ North American breeding range (Rimmer and McFarland
2001, Townsend and Rimmer 2006). The recovery rate is much higher than that for other
Nearctic-Neotropical songbird species, in which the numbers of individuals marked in one
portion of the migratory range and recaptured in another is exceedingly small.

One of the most studied and publicized conservations problems in the past 20 years has been the
population decline of Nearctic-Neotropical migratory songbirds (e.g., Robbins et al. 1989,
Askins et al. 1990). Between 1993 and 2003, a core breeding population of Bicknell‟s Thrush in
the White Mountain National Forest was estimated to number as few as 4,900 individuals (Hale
2006) and experienced annual declines of 7% per year along 40 survey routes (King et al. 2008,

McFarland, Faccio, and Rimmer 2010 VMC Report
7

Lambert et al. 2008). Seven years of data from New Brunswick and Nova Scotia (2002-2008)
indicate that Bicknell‟s Thrushes declined along established survey routes by as much as 20%
annually (Campbell et al. 2007, Whittam and Campbell unpubl. data), while annual surveys at
Mont Gosford, Quebec, from 2001-2007 showed a pronounced decline in the number of stations
occupied by Bicknell‟s Thrush (Aubry unpubl. data). In addition, climate change projections
derived from survey data indicate that suitable Bicknell‟s Thrush habitat may be lost from the
U.S. following increases in summer temperatures that are projected to occur this century
(Rodenhouse et al. 2008). Yet, incomplete information on fundamental aspects of the species‟
ecology and demographics throughout its annual cycle preclude a full understanding of the
limiting factors that underlie these striking declines.

Understanding the timing and extent of avian population limitation and regulation is complicated
in the case of migratory populations that spend different periods of their annual cycle in
ecologically disparate regions. The “seasonal interaction hypothesis” was first put forth by
Fretwell (1972), who argued that breeding density is determined by winter survival, which in
turn is related to events that occur during the breeding cycle. Recent studies of American
Redstarts (Setophaga ruticilla) support the seasonal interaction hypothesis (Marra et al. 1998,
Norris et al. 2003). In this species, winter habitat quality determined physical condition and
timing of spring migration departure, which influenced arrival date and physical condition on the
breeding grounds. The quality of each individual‟s winter habitat was determined by measuring
stable carbon isotope levels shortly after arrival on breeding territories. Subsequent monitoring
of redstart breeding demographics revealed a profound interaction between seasons. Robust tests
of the seasonal interaction hypothesis such as this require detailed knowledge of migratory
connectivity.

Bicknell‟s Thrush is among eastern North America‟s most rare, range-restricted breeding
passerines. Considered one of the Nearctic-Neotropical migrants at greatest risk of extinction and
thus of highest continental conservation concern (Pashley et al. 2000, Rimmer et al. 2001, Rich
et al. 2004, Wells 2007), Bicknell‟s Thrush is classified as globally “vulnerable” by the
International Union for the Conservation of Nature (IUCN). At both ends of its migratory range,
Bicknell‟s Thrush occupies a limited, highly fragmented distribution and faces multiple habitat
threats. These include climate change (Rodenhouse et al. 2008), acid ion deposition (Johnson et
al. 1992, Hames et al. 2002), mercury contamination (Rimmer et al. 2005), mountaintop
development (Rimmer et al. 2001, 2004), forestry operations (Leonard and Chisholm 2008,
Gardiner 2006), and winter habitat loss and degradation (Rimmer 2005; Rimmer et al. 2001,
2005). With respect to the latter, Marshall (2001) postulated that loss of breeding populations at
several sites in Nova Scotia was directly related to extensive deforestation of wintering habitat in
Haiti, implying strong population connectivity. Overall, documented population declines in
Canada and New Hampshire, combined with severe and ongoing habitat loss on Hispaniola, have
heightened concern about the conservation status of Bicknell‟s Thrush. This concern catalyzed
the 2007 formation of the International Bicknell‟s Thrush Conservation Group (IBTCG; see
www.bicknellsthrush.org). To date, the IBTCG has identified elucidation of migratory
connectivity as one of the species‟ most pressing conservation needs.

McFarland, Faccio, and Rimmer 2010 VMC Report
8

Breeding Distribution
Breeding from the northern Gulf of St. Lawrence and easternmost Nova Scotia southwest to the
Catskill Mountains of New York, Bicknell‟s Thrush is estimated to number fewer than 100,000
individuals across its naturally fragmented breeding range (Rimmer et al. 2001). This range has
been well documented (Wallace 1939, Ouellet 1993, Atwood et al. 1995, Rimmer et al. 2001),
and recent habitat modeling has provided further refinements (Figure 1; Lambert et al. 2005,
Aubry et al. 2009).

Numerous local Bicknell‟s Thrush extirpations have been documented. Historic breeding
populations disappeared on Mt. Greylock, Massachusetts (10 pairs in 1950s, 0 in 1973; Veit and
Petersen 1993); Magdalen Is., Quebec (Ouellet 1996, D. McNair pers. comm.); Seal and Mud Is.,
Nova Scotia (Wallace 1939, Erskine 1992, D. Busby pers. comm.); Cape Forchu, sw. Nova
Scotia (J. Marshall pers. comm.); Fundy National Park, New Brunswick (Christie 1993); and
Grand Manan I., New Brunswick (B. Dalzell pers. comm.). Further range contraction in the
Canadian Maritime provinces is suggested by mid-1990s surveys that showed fewer occupied
sites than during the 1986–1991 Breeding Bird Atlas (D. Busby pers. comm.) survey. The
species‟ core U.S. range, however, appears to have remained stable overall, as Bicknell‟s Thrush
was confirmed on 63 of 73 historic (pre-1992) breeding sites surveyed in 1992–1995 (Atwood et
al. 1995).

Non-breeding Distribution
The stationary portion of the species‟ non-breeding period is believed to be confined to the
Greater Antilles, mostly in mesic to wet broadleaf forest. Specimen and field-survey data
indicate the bulk of wintering Bicknell‟s Thrushes occur in the Dominican Republic (Wetmore
and Swales 1931, Ouellet 1993, Rimmer et al. 1997, 1999), where the species is widely
distributed and locally common from sea level to 2,220 m (Rimmer et al. 1999, 2001). There are
few records from Haiti where known populations are restricted to montane forest fragments in
the southwest (Massif de la Hotte) and east (Massif La Visite; Wetmore and Swales 1931;
Woods and Ottenwalder 1983, 1986; Rimmer et al. 2005, 2010). Bicknell‟s Thrush appears to be
uncommon and local in Jamaica, known only from the Blue Mtns. at 1,200-2,225 m elevation (R.
and A. Sutton unpubl.; Rimmer unpubl.). The species is a rare winter resident in e. and se. Puerto
Rico, in the Luquillo Mountains at 450–720 m elevation and Sierra de Cayey at 720 m (Arendt
1992, J. Wunderle unpubl.), and it has recently been reported from Vieques island off the east
coast (S. Colon, pers. comm.). Thrushes have been found in e. Cuba at 1,600–1,960 m in Sierra
Maestra (Rompré et al. 2000, Y. Aubry and G. Rompré pers. comm.). There are no confirmed
winter records elsewhere. The remote locations and rough terrain of occupied winter habitats
make it unlikely that we have yet fully documented all areas inhabited by Bicknell‟s Thrush.
However, recent modeling of Bicknell‟s Thrush winter grounds has proven to be a robust
predictor of its habitat in the Greater Antilles (VCE unpub. manuscript).

Migratory Routes
Analysis of the scant specimen and banding data, using wing-chord as identification criterion
(<94 mm = Bicknell‟s, >98 mm = Gray-cheeked), suggests an elliptical southern portion of
migratory route between the North American breeding grounds and Greater Antillean winter
range (Rimmer et al. 2001). Most southbound migrants may depart the East Coast from the Mid-
Atlantic States or Carolinas on an overwater flight to the Greater Antilles, as suggested by the

McFarland, Faccio, and Rimmer 2010 VMC Report
9

scarcity of fall records south of Virginia. Northward passage appears to be more concentrated
through the Southeast coast, as spring specimens from Florida, Georgia, both Carolinas, and
Virginia outnumber fall records nearly 2 to 1. The entire migration in both directions appears to
be concentrated east of the Appalachian Mountains. Stopover lengths during migration are not
well documented by banding stations, but a few transients appear to linger at stopover sites
during the fall. There is yet no evidence of spring stopovers.

No information exists on the timing of departure from the wintering grounds or the rate of spring
or fall migration. We suspect that birds depart wintering areas in late April or early May, as
individuals are still present in the Dominican Republic second week of April (VCE unpub. data)
and there are no verifiable U.S. records prior to May. Furthermore, given recent data from Wood
Thrushes (Stutchbury et al. 2009), we suspect the rate of migration to be about 200-250km/day.
The northward migratory route of Bicknell‟s Thrush and its spring stopover locations largely
remain a mystery.

In this study we used miniature daylight level data recorders (geolocators), recently developed by
the British Antarctic Survey (Afanasyev 2004), to track locations of individual Bicknell‟s
Thrushes during their annual cycle to help elucidate migratory connectivity between breeding,
stopover and winter sites.

Methods

Study Sites and Geolocators

During the 2009 breeding season, we captured adult male Bicknell‟s Thrushes in nylon mist nets
(12 x 2.5-m and 6 x 2.5-m, 36-mm mesh) at seven study areas across the entire breeding range,
using passive netting and playback of conspecific vocalizations to attract birds. Each individual
was fitted with a uniquely-numbered Canadian Wildlife Service / U.S. Fish and Wildlife Service
leg band and in some cases a unique combination of three plastic colored leg bands. We
attempted to recover and deploy additional geolocators during the 2010 breeding season at each
site. During February and March of 2010, we added a south-north component to this study by
attaching geolocators to overwintering thrushes at two ecologically distinct sites in the
Dominican Republic. One was at Pueblo Viejo in Sierra de Bahoruco (a strongly male-
dominated forest habitat), while the second was at Loma Guaconejo in Cordillera Septentrional
(a mid-elevation, moderately disturbed site with even sex ratios).

We deployed Mk10S light level geolocators developed and manufactured by the British
Antarctic Survey (BAS). Light sensors were mounted on a stalk ~15mm in length and at 20
degrees to horizontal to better clear plumage. These loggers take consistent readings of daylight
timing for up to two years. The recovered data are then interpreted to determine latitude and
longitude of the individual bird for every day the logger was attached and exposed to suitable
sunlight.

We attached geolocators to Bicknell‟s Thrushes using a leg-loop backpack harness (Rappole and
Tipton 1991) that has been deployed successfully on over 250 Bicknell‟s Thrushes with 1.2 g
radio transmitters, as well as on many other passerine species. During 2009 and winter 2010, we

McFarland, Faccio, and Rimmer 2010 VMC Report
10

used size 600 Kevlar thread for the harnesses, while in June and July of 2010 we switched to
3/16-inch Teflon ribbon (Stutchbury et al. 2009). Both methods resulted in geolocators being
attached without logistic problems of any kind or any detectable short-term effects on thrushes.
Several birds recaptured up to four weeks after attachment were in good condition and within
normal weight ranges. The total attachment weighed ~1.4 g which corresponded to ~5% mean
male body mass (28.18 g; Rimmer et al. 2001).

Light level data were downloaded following recovery of geolocators, and data were examined
for clock drift. Very little or no clock drift was detected (i.e., drift <2 minutes), so we did not
apply linear drift corrections. Mk10S geolocators use a sensor to measure light intensity every
minute and record the maximum measurement in each 10-min interval. We used a light level
threshold of two to define light transition events because of the mountainous terrain and heavy
shading encountered by Bicknell‟s Thrush. We used a 2-week post deployment period as
calibration data for the breeding grounds. Additionally, we used a 1-week deployment of a fixed
geolocator placed in the forest understory on a Bicknell‟s Thrush territory at Pueblo Viejo as
calibration data for wintering grounds (mean sun elevation angle = -2.9). The annual cycle was
divided into four seasons: breeding (June – August), fall migration (Sept.- Oct.), winter (Nov. –
March), and spring migration (April – May). Exact seasonal transitions are not known for this
bird, but we believed this to be a conservative estimate of annual timing. We used mean breeding
sun elevation data for birds released on winter grounds and the above winter sun elevation
estimate for those released on the breeding grounds. Sun elevation estimates for migration
periods are problematic. We used the winter estimate of sun elevation because we believed that
Bicknell‟s Thrush likely inhabited heavily shaded forest understory during migration stopover,
while on the breeding grounds birds often call and sing from elevated perches during transition
events.

For each recovered geolocator, light transition events were visually assessed and assigned a
confidence level using program TransEdit (BAS). Non-linear or apparent shading events during
the transition periods were scored lower. Data for 15 days before and after spring and fall
equinox were excluded from latitude calculations.

Geolocator light levels are referenced to an internal clock/calendar and from these the Greenwich
Meridian Time (GMT) of local sunrise and sunset are estimated. The GMT midway between the
times of sunrise and sunset is local noon GMT and local midnight GMT, from which longitude
can be determined. The day length on a particular date determines the latitude. Longitude tends
to be much more precise than latitude. Location estimates were calculated and plotted with
BirdTracker software (BAS). Further analyses were completed in ArcGIS 10.

We derived a mean and 95% confidence interval for longitude using all winter locations and
examined the intersection with modeled Bicknell‟s Thrush winter habitat to derive an estimated
wintering location for each individual. We used only high confidence transition events when
attempting to derive locations using both latitude and longitude throughout the annual cycle.

McFarland, Faccio, and Rimmer 2010 VMC Report
11

Results and Discussion

During 2009, VCE and collaborators from the Smithsonian Migratory Bird Center (SMBC),
SUNY Environmental School of Forestry, Canadian Wildlife Service, Bird Studies Canada, and
University of New Brunswick deployed geolocators on 70 Bicknell‟s Thrushes: 36 in the U.S.
and 34 in Canada (Table 1). During the winter of 2009-10 we attached geolocators to 15
individuals at Pueblo Viejo and 13 at Loma Guaconejo (Table 1).

Table 1. Numbers of geolocators attached and recovered on Bicknell‟s Thrushes across the
migratory range, 2009 and 2010.

Site Deployed Recovered in
2010
Recaptures
missing 2009
geolocators
2009 2010
Plateau Mt. (NY) 5 0 0 –
Slide Mt. (NY) 17 0 1 4
Mt. Mansfield (VT) 14 22 2 4
Mt. Washington (NH) 0 4 – –
Mt. Jefferson (NH) 0 4 – –
Mt. Gosford (QC) 10 0 3 0
Massif-du-Sud (QC) 6 0 3 0
Gaspesie (QC) 9 0 2 0
New Brunswick Highlands 5 7 0 –
Cape Breton Highlands (NS) 4 7 2 1
Pueblo Viejo (DR)a 15 0 2 2
Loma Guaconejo (DR)a 13 0 2 0
Total 98 41 17 11
a geolocators attached in February and March of 2010, recovered in November

Overall return rates of geolocatored birds from 2009 were lower than expected, 30% versus the
long-term average of 65% for Vermont-banded adults. Recoveries of geolocators themselves
were low (n = 17). We discovered that the Kevlar thread used to secure the 2009 backpack
harnesses had abraded and broken – 2 of the 3 birds recovered with intact geolocators had at
least some broken strands of Kevlar fibers. However, in Quebec, where Dacron fly fishing line
was used for the harnesses, all 8 recovered birds still had fully intact geolocators.
During November of 2010, we returned to both study sites in the Dominican Republic for
geolocator recovery. At Pueblo Viejo, we recaptured 4 of the 15 birds to which we had attached
geolocators in the previous February or March; 2 of these retained their geolocators. At Loma
Guaconejo, both recovered thrushes still had their geolocators securely attached.
During the summer of 2010, in addition to recovering 2009 geolocators, we concentrated on
attachment of new devices. We used Teflon webbing to attach geolocators to 30 Bicknell‟s
Thrushes: 17 ASY male and 5 ASY female Bicknell‟s Thrushes on Mt. Mansfield, 4 ASY males

McFarland, Faccio, and Rimmer 2010 VMC Report
12

on Mt. Washington, and 4 ASY males on Mt. Jefferson. Additionally, 14 geolocators were
deployed by our cooperators in New Brunswick and Nova Scotia.
The rangewide total of 17 recovered geolocators during 2010, while falling short of expectations,
provides an invaluable store of data. However, analyses and interpretation of the retrieved data
have proven extremely challenging to date. The data showed extreme shading for much of the
year, indicating that individuals are often under heavy forest canopy cover during daily sun
transitions. After initial tests and examinations of data, we have been able to partially analyze
five geolocators thus far (Nova Scotia = 2, Mt. Mansfield = 2, and Slide Mountain, NY = 1).
We have been able to complete analysis using only high confidence transition events for one
male from Mt. Mansfield, Vermont. Geolocator data yielded few reliable locations due to heavy
shading events (Fig. 1). During fall migration between 19-26 October 2009, this bird stopped in
the eastern Bahamas for approximately six days. During northward spring migration it stopped
over in the western Bahamas or southeast Florida from 6-8 May 2010. On 15-16 May the bird
progressed northward to the coastal Carolinas. It arrived at its Mt. Mansfield breeding site on 24
May, and we recaptured it on 2 June. While on migration from 8-15 May from a presumed
position in southeast Florida to coastal South Carolina, this bird traveled 850 km (121 km/day).
It then traveled 1,360 km at a rate of 170 km/day from 16-24 May to arrive on the breeding
grounds. Stutchbury et al. (2009) reported that spring migrating Wood Thrushes traveled 242
km/day from winter to breeding sites (range of 5 birds: 159-279 km/day). This Bicknell‟s Thrush
migrated nearly twice as slowly, covering only 123 km/day from its apparent arrival in
southeastern Florida to its breeding site. The general migratory pathway of this individual helps
to confirm the pattern previously documented from migration banding records, nocturnal
migration calls, tower kills and other reliable sightings (Rimmer et al. 2001), indicating that fall
migrants tend to pass from coastal Virginia and North Carolina across open waters and through
the eastern Bahamas to the wintering grounds, while spring migration appears to be through
eastern Florida and northward an inland route to the breeding grounds.
We used winter longitude locations for all five birds to determine approximate wintering
locations (Table 2, Fig. 2). All four birds were predicted to be in the Dominican Republic based
on the mean longitude location. However, significant variation in longitude locations places
confidence intervals to Cuba and Puerto Rico. We hope to make significant improvements by
filtering heavily shaded dates.
Table 2. Longitude location estimates for geolocators on Bicknell‟s Thrush during the winter
period (November – March).
Geolocator Mean SD
Catskills, NY -69.93772 -1.03
Mt. Mansfield, VT -70.20977 -2.16
Nova Scotia (7533) -71.76242 -1.75
Nova Scotia (7618) -68.92296 -2.91

McFarland, Faccio, and Rimmer 2010 VMC Report
13

Future Work
During the 2011 breeding season, VCE staff and collaborators will undertake intensive efforts to
recover geolocators in Vermont, New Hampshire, Nova Scotia, and New Brunswick. With the
improved attachment method used in 2010, we expect to recover ~20 individuals.

We will continue to refine analytical methods to improve interpretation of data, and we are
optimistic about achieving more results from the four Dominican Republic recoveries given the
behavior of the bird during the breeding season at daily transition times. We will combine
geolocator data from both 2010 and 2011 for a comprehensive analysis, which we expect to
result in a peer-reviewed paper by the end of 2011.

Acknowledgments
This project involves many cooperators from the International Bicknell‟s Thrush Conservation
Group (www.bicknellsthrush.org) including: Bird Studies Canada, British Antarctic Survey,
Canadian Wildlife Service, Smithsonian Migratory Bird Center, SUNY Environmental School of
Forestry, and the University of New Brunswick. We are grateful for core financial support
provided by the Vermont Monitoring Cooperative, the USDA Forest Service Office of
International Programs, and The Nature Conservancy, under the terms of Federal Financial
Assistance Award of Domestic Grant Agreement No. #10-DG11132762-165. Additional funding
for our geolocator work was provided by the MacArthur Foundation through the Cornell
Laboratory of Ornithology, the U.S. Fish and Wildlife Service, and the William P. Wharton
Trust. We thank the Stowe Mountain Resort for allowing us access to the Mt. Mansfield toll road
and for overnight use of the ski patrol hut, and to Howie Weymss for access to the Mt.
Washington Auto Road. Logistical assistance in the Dominican Republic was provided by Jorge
Brocca of the Sociedad Ornitológica de la Hispaniola, and our work there was authorized by the
Subsecretaría de Áreas Protegidas y Biodiversidad. We thank Brendan Collins, Pat Johnson,
Jason Townsend, Steve Faccio, and Rosalind Renfrew for their dedicated field work in Vermont
and New York. Pat Johnson, Juan Klavins, Hodali Almonte, Jesus Almonte, Jorge Brocca,
Esteban Garrido, Robert Ortiz, Ben Rimmer, and others assisted with field work in the
Dominican Republic. Assistance with data analyses was provided by Patrick Johnson, and we
are especially grateful to Nora Diggs at SMBC for her diligent efforts to interpret data from
recovered geolocators, and to Brant Ryder at SMBC for analytical advice.

Literature Cited

Afanasyev, V. 2004.A miniature daylight level and activity data recorder for tracking animals
over long periods. Mem. Natl Inst. Polar Res., Spec. Issue, 58:227–233.
Arendt, W. J. 1992. Status of North American migrant landbirds in the Caribbean region: a
summary. Pages 143-171 in Ecology and conservation of Neotropical migrant landbirds.
(Hagan III, J. M. and D. W. Johnston, Eds.) Smithson. Inst. Press, Washington, D.C.
Askins R.A., J.F. Lynch, and R. Greenberg. 1990. Population declines in migratory birds in
eastern North America. Current Ornithology. 7:1–57.
Atwood, J.L., C.C. Rimmer, K.P. McFarland, S.H. Tsai, and L.R. Nagy. 1995. Distribution of
Bicknell's Thrush in New England and New York. Wilson Bull. 108: 650-661.

McFarland, Faccio, and Rimmer 2010 VMC Report
14

Bearhop, S., R.W. Furness, G.M. Hilton, S.C. Votier And S. Waldron. 2003. A forensic approach
to understanding diet and habitat use from stable isotope analysis of (avian) claw
material. Functional Ecology 17: 270-275.
Caccamise, D.F. and R.S. Hedin. 1985. An aerodynamic basis for selecting transmitter loads in
birds. Wilson Bull. 97:306-318.
Campbell, G., B. Whittam, and G. Robertson. 2007. High Elevation Landbird Program 5-year
report. Unpublished Report. Bird Studies Canada (Atlantic Region), Sackville, New
Brunswick. 16 pp.
Canadian Council On Animal Care. 2003. Guidelines on: the care and use of wildlife. Ottawa,
Ontario.
Christie, D. S. 1993. Survey of breeding birds in Fundy National Park, 1992. Can. Parks Serv.
Contract Rep. no. FNP92-004.
Cochran, W.W., G.G. Montgomery and R.R. Graber. 1967. Migratory flights of Hylocichla
thrushes in spring: a radiotelemetry study. Living Bird 6:213-225.
Erskine, A. J. 1992. Atlas of breeding birds of the Maritime Provinces. Nimbus Publ. and Nova
Scotia Mus. Halifax.
Fretwell, S. D. 1972. Populations in a seasonal environment. Princeton, New Jersey. Princeton
University Press. 1972. 217 p.
Gardiner, R. J. 2006. The long-term effects of pre-commercial thinning on Bicknell's thrushes
and four other high elevation songbirds in the New Brunswick highlands. Thesis (B. Sc.),
Mount Allison University.
Graber, R.R. and S.L. Wunderle. 1966. Telemetric observations of a robin (Turdus migratorius).
Auk 83:674-677.
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, Ralph S., Kenneth V. Rosenberg, James D. Lowe, Sara E. Barker, and André A.
Dhondt. 2002. Adverse effects of acid rain on the distribution of the Wood Thrush
Hylocichla mustelina in North America. PNAS 99:11235-11240
Hobson, K.A., K.P. Leonard, L.I. Wassenaar, C.C. Rimmer. 2001. Linking Breeding and
wintering grounds of Bicknell‟s Thrushes using stable isotope analyses of feathers. Auk
118:16-23.
Johnson, A. H., S. B. McLaughlin, M. B. Adams, E. R. Cook, D. H. DeHayes, et al. 1992.
Synthesis and conclusions from epidemiological and mechanistic studies of red spruce.
Pp. 387-411 in The ecology and decline of red spruce in trhe eastern United States (C.
eager and M. B. Adams, eds.). Univ. Press of New England, Hanover, NH.
King, D., J. Lambert, J. Buonaccorsi, and L. Prout. 2008. Avian population trends in the
vulnerable montane forests of the Northern Appalachians, USA. Biodiversity and
Conservation 17:2691-2700.
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.
Leonard, M. L., and S. E. Chisholm. 2008. Effect of forest management on a rare habitat
specialist, the Bicknells Thrush (Catharus bicknelli). Canadian Journal of Zoology
86:217-223.
Marra, P. P., K. A. Hobson, and R. T. Holmes. 1998. Linking Winter and Summer Events in a
Migratory Bird by Using Stable-Carbon Isotopes. Science 282:1884-1886.

McFarland, Faccio, and Rimmer 2010 VMC Report
15

Marra, P. P., Webster, M.S., Ed. 2005. The importance of understanding migratory connectivity
and seasonal interactions. Pages 199-209 In Birds of Two Worlds (R. Greenberg and P.P.
Marra eds.). Johns Hopkins University Press, Baltimore, London.
Marshall, J. T. 2001. The Gray-cheeked Thrush, Catharus minimus, and its New England
subspecies, Bicknell's Thrush, Catharus minimus bicknelli. Publ. Nuttall Ornithol. Club,
no. 28.
Norris, D. R., P. P. Marra, T. K. Kyser, T. W. Sherry, L. M. Ratcliffe, and R. Montgomerie.
2007. Continent-wide variation in feather colour of a migratory songbird in relation to
body condition and moulting locality. Biology Letters 3:16-19.
Ouellet, H. 1993. Bicknell's Thrush: taxonomic status and distribution. Wilson Bull. 105:545-
572.
Ouellet, H. 1996. Bicknell's Thrush. Pages 784-787 in The breeding birds of Quebec. (Gauthier,
J. and Y. Aubry, Eds.) Assoc. québecoise des groupes d'ornithologues, Prov. of Quebec
Soc. for the protection of birds, Can. Wildl. Serv., Environ. Canada, Québec Region,
Montréal.
Pashley, D. N., C. J. Beardmore, J. A. Fitzgerald, R. P. Ford, W. C. Hunter, M. S. Morrison, and
K. V. Rosenberg. 2000. Partners in Flight: conservation of the land birds of the United
States. American Bird Conservancy, The Plains, Virginia.
Phillips RA, Silk JRD, Croxall JP, Afanasyev V and Briggs DR. 2004. Accuracy of geolocation
estimates for flying seabirds. Mar Ecol Prog Ser 266: 265–272.
Phillips, R.A., P.C. Silk, S. Bearhop, R. McGill, V. Afanasyev, and I. J. Strange. 2007.
Movements, winter distribution and activity patterns of Falkland and brown skuas: insights
from loggers and isotopes. Marine Ecology Progress Series 345:281-291.
Rappole, J. H. and A.R. Tipton. 1991. New harness design for attachment of radio transmitters to
small passerines. Journal of Field Ornithology 62: 335-337.
Remsen, J. V. 2001. True winter range of the veery (Catharus fuscescens): lessons for
determining winter ranges of species that winter in the tropics. Auk 118:838-848.
Rich, T. D., C. J. Beardmore, H. Berlanga, P. J. Blancher, M. S. W. Bradstreet, G.S. Butcher,
D.W. Demarest, E. H. Dunn, W. C. Hunter, E. F. Iñigo-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.
Rimmer, C. C. and K.P. McFarland. 2001. Known breeding and wintering sites of a Bicknell's
Thrush. Wilson Bull.113: 234-236.
Rimmer, C. C., K. P. McFarland, and J. E. Goetz. 1997. Distribution, habitat use and
conservation of Bicknell's Thrush and other montane forest birds in the Dominican
Republic. Unpubl. rep. Vermont Inst. Nat. Sci. Woodstock.
Rimmer, C. C., K. P. McFarland, D. C. Evers, E. K. Miller, Y. Aubry, D. Busby, and R. J.
Taylor. 2005. Mercury Concentrations in Bicknell‟s Thrush and Other Insectivorous
Passerines in Montane Forests of Northeastern North America. Ecotoxicology 14:223-
240.
Rimmer, C.C., Jason M. Townsend, Andrea K. Townsend, Eladio M. Fernández, & Jesus
Almonte. 2005. Avian diversity, abundance, and conservation status in the Macaya
Biosphere Reserve of Haiti. Ornitologia Neotropical 16: 219–230.
Rimmer, C.C., K.P. McFarland, and J.E. Goetz. 1999. Distribution, habitat use, and conservation
status of Bicknell's Thrush in the Dominican Republic. El Pitirre 12: 114.

McFarland, Faccio, and Rimmer 2010 VMC Report
16

Rimmer, C.C., K.P. McFarland, J.D. Lambert, 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.
Rimmer, C.C., K.P. McFarland, W.G. Ellison, And J.E. Goetz. 2001. Bicknell‟s thrush (Catharus
bicknelli). In The Birds of North America, No. 592 (A. Poole and F. Gill, eds.).The Birds
of North America, Inc., Philadelphia, PA.
Robbins, C.S., J R Sauer, R S Greenberg, and S Droege. 1989. Population declines in North
American birds that migrate to the neotropics. PNAS 86:7658-7662
Rodenhouse, N., S. Matthews, K. McFarland, J. Lambert, L. Iverson, A. Prasad, T. Sillett, and R.
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
Cuba. Cotinga 13:66.
Rosenberg, K.V. and J.V. Wells. 1995. Importance of geographic area to Neotropical migrant
birds in the Northeast. Unpubl. report to U.S. Fish and Wildlife Service, Hadley, MA.
Rubenstein, D. R., C.P. Chamberlain, R.T. Holmes, M.P. Ayres, J.R. Waldbauer, G.R. Graves,
N.C. Tuross. 2002. Linking breeding and wintering ranges of a migratory songbird using
stable isotopes. Science 295:1062-1065.
Sheldon, L. D., Eunice H. Chin Sharon A. Gill Gregory Schmaltz Amy E. M. Newman Kiran K.
Soma. 2008. Effects of blood collection on wild birds: an update. Journal of Avian
Biology 39:369-378.
Stutchbury, B. J. M., S. A. Tarof, T. Done, E. Gow, P. M. Kramer, J. Tautin, J. W. Fox, and V.
Afanasyev. 2009. Tracking long-distance songbird migration by using geolocators. Science
323:896.
Takahashi, A., D. Ochi, Y. Watanuki, T. Deguchi, N. Oka, V. Afanasyev, J. W. Fox, and P. N.
Trathan. 2008. Post-breeding movement and activities of two Streaked Shearwaters in the
north-western Pacific. Ornithological Science 7:29-35.
Townsend, J. M., and C. C. Rimmer. 2006. Known natal and wintering sites of a Bicknell's
Thrush. Journal of Field Ornithology 77:452-454.
Veit, R. R. and W. R. Petersen. 1993. Birds of Massachusetts. Massachusetts Audubon Soc.
Lincoln.
Webster, M. S., P.P. Marra, S.M. Haig, S. Bensch, R.T. Holmes. . 2002. Links between worlds:
unraveling migratory connectivity. Trends in Ecology and Evolution 17:76-83.
Wetmore, A. and B. H. Swales. 1931. The birds of Haiti and the Dominican Republic. Bull. Mus.
Comp. Zool. 63:340-341.
Woods, C. A. and J. A. Ottenwalder. 1983. The montane avifauna of Haiti. Proc. Jean Delacour
Symposium. Intl. Found. Conserv. Birds.
Woods, C. A. and J. A. Ottenwalder. 1986. The birds of Parc National La Visite and Parc
National Pic Macaya, Haiti. Unpubl. rep. U.S. Agency for Int. Develop. USAID/Haiti,
Port au Prince.

McFarland, Faccio, and Rimmer 2010 VMC Report
17


Figure 1. Migration stop-over locations determined
from a male Bicknell‟s Thrush from Mt. Mansfield,
Vermont carrying a geolocator during its annual cycle
in 2009-2010. Circle indicates the mean location for
that period while the box represents the standard
deviation for the point.

McFarland, Faccio, and Rimmer 2010 VMC Report
18




Figure 2. Mean of winter longitude estimates from Bicknell‟s Thrush geolocators. Each bird could have been
located along any portion of the longitude and the associated error of that mean.

McFarland, Faccio, and Rimmer 2010 VMC Report
19





























Figure 3. Geolocator from Nova Scotia Bicknell‟s Thrush using the mean sun elevation from summer data (left)
compared to the winter elevation from a test geolocator in the Dominican Republic (left).

McFarland, Faccio, and Rimmer 2010 VMC Report
20

Part III. A Dynamic Occupancy Model of Bicknell’s Thrush Breeding Habitat

Introduction

In previous studies (Lambert et al. 2005, Hart et al. 2009, McFarland and Rimmer 2009), we
used the Quantreg library in R software (http://lib.stat.cmu.edu/R/CRAN) to estimate the 0.05
quantile regression (Cade and Noon 2003) of elevation as a linear function of latitude for
locations where Bicknell‟s Thrush (Catharus bicknelli) were found during targeted surveys. This
produced an elevation threshold above which 95% of the detections occurred. We converted the
linear threshold into an elevation mask, formed as a raster data set of 30 x 30 m cells (Lambert et
al. 2005) or 90 x 90 m cells (Hart et al. 2009, McFarland and Rimmer 2009). The elevation mask
was then placed over a digital elevation model (DEM) of the northeastern U.S. Summits,
ridgelines, and slopes emerged above the mask as a vast complex of high-elevation habitat units
predicted to be occupied by Bicknell‟s Thrush during the breeding season.

In this study we extended our model using detection-non detection (presence-absence) data
collected from 2001-2009 for Bicknell‟s Thrush across its U.S. breeding range through a
volunteer-based monitoring project, Mountain Birdwatch (Hart and Lambert 2007) and Forest
Bird Monitoring Program survey routes (see section IV).

Methods

We used detection-non detection (presence-absence) data collected from 2001-2009 across the
U.S. breeding range of Bicknell‟s Thrush through the volunteer-based Mountain Birdwatch
(MBW) project (Hart and Lambert 2007). We removed all sites with fewer than six years of data
for a final sample size of n=113. Each site consisted of a ~1 km transect of five points 200-250m
apart. Transect locations were placed through prioritization of high-elevation habitat units,
although trail locations and volunteer availability ultimately determined which sites were to be
surveyed and where at each site the transect was placed (Hart and Lambert 2007).

As many as four surveys were conducted at each site in each year, although each site was not
always surveyed all four times nor surveyed in all years. These missing observations did not
affect our analysis, as the parameters associated with a missed survey were simply not estimated
using that survey‟s data (MacKenzie et al. 2006). The four detection/non-detection surveys
conducted at each site included: 1) the evening before a morning point count; 2) a 10-minute
morning (between 4:30-6:30 h EST) point count at each of the five points along the transect; 3) a
one-minute playback followed by a two-minute listening period at each of the five points along
the transect immediately following the point count surveys; and 4) a follow-up playback survey
every 100 m along the transect within two weeks of the previous surveys and/or before 15 July.
For each survey at each site, a detection or non-detection was recorded based on whether a
Bicknell‟s Thrush was heard or seen at or between any of the five points along the transect.

A multi-season robust occupancy model framework was used, following MacKenzie et al. (2003)
to estimate the probability of extinction and colonization. The robust model examines the state
variable, occupancy, and how it changes over time at the site level. The four parameters
estimated in this model are probability of detection (p), probability of initial site occupancy (ψ),

McFarland, Faccio, and Rimmer 2010 VMC Report
21

probability of site colonization (γ), and probability of site extinction (ε). Analyses were
conducted using program Presence (Hines 2006). Detection probability (p) was modeled as a
function of survey number.

We determined Bicknell‟s Thrush breeding habitat across the northeastern U.S. using our
predictive model (Hart et al. 2009, McFarland and Rimmer 2009) and the National Land Cover
Database 2001 (http://www.mrlc.gov/; Homer et al. 2004). Raster cells that were defined as
conifer within areas identified as potentially containing Bicknell‟s Thrush were considered
Bicknell‟s Thrush breeding habitat.

We calculated the amount of habitat within a radius of 1, 5, 10, and 25 km around MBW sample
points using „isectpolyrst‟ command in Geospatial Modeling Environment
(http://www.spatialecology.com/gme/) in ArcGIS 9.3.1. We used „zonal statistics‟ (command
v.what.rast) in GRASS GIS 6.4 to assign a value for patch size, habitat within 1, 5, 10, and 25
km to each sample site. We standardized values for patch size, habitat within 1, 5, 10, and 25 km
to obtain z-scores [(x-mean)/standard deviation].

Because our dataset was relatively sparse due to missing surveys, we kept candidate models
comparatively simple. We selected 19 models a priori to compare the importance of patch size,
habitat within 1, 5, 10, and 25 km in explaining the variability of three population parameters: 1)
probability of initial site occupancy (ψ1); 2) local site colonization (γ); and 3) local site extinction
(ε). We used Program Presence 2.4 to run 19 dynamic occupancy models and rank them each by
their AIC score. We then obtained the beta (describing intercept and effect size of habitat within
1 km) and parameter estimates and standard error for each site based on the highest ranked
model. Maximum likelihood techniques were used to estimate parameters, where ψ1 refers to the
initial occupancy in the first period and ε and γ determine ψt in the following seasons
(MacKenzie et al. 2003). A total of 19 models were compared using AIC model selection
procedures (Burnham and Anderson 2002). Models with an AIC value within 4.0 of the
minimum AIC were considered plausible models (Burnham and Anderson 2002).

We used command r.neighbors in GRASS GIS 6.4 to calculate the amount of habitat within 1 km
for each raster cell in the Bicknell‟s Thrush breeding habitat model. We then used GRASS GIS
6.4 to calculate the probability of occupancy, colonization, and extinction for each cell within the
US distribution model based on the amount of habitat within 1 km of each cell and beta estimates
from the highest ranked model.

Results and Discussion

The amount of habitat within a 1-km radius was a prominent factor in determining whether a site
occupied by Bicknell‟s Thrush became locally extinct or a vacant site was colonized (Table 1).
The highest ranked model, ψ(hab1km), γ(hab1km), ε(hab1km), p(survey), was >6 AIC points
lower than the next model. The top model expressed both γ and ε as a function of the effect of
the amount of habitat within a 1-km radius. Initial occupancy (ψ1) was a function of the amount
of habitat within a 1-km radius in the top 10 models.

McFarland, Faccio, and Rimmer 2010 VMC Report
22

There were 192.7 ha of habitat within 1 km of the Mt Mansfield ridgeline survey route (MANS)
and 167.7 ha around the Ranch Brook survey route (RABR). Based on the highest ranked model,
the initial occupancy was 0.94 for MANS and 0.91 for RABR (Fig. 1). Extinction and
colonization probabilities were 0.01 and 0.43 for MANS and 0.02 and 0.39 for RABR (Figs. 2
and 3). Only the smallest and most isolated peaks within the U.S. breeding range of Bicknell‟s
Thrush had an extinction probability above 0.2 (Fig. 3).

The U.S. breeding range of Bicknell‟s Thrush contained 173,354.3 ha of habitat. Of this total,
Maine had 61,353.9 ha (35.4%), New Hampshire 64,696.2 ha (37.3%), Vermont 12,820.3 ha
(7.4%), and New York 34,471.8 ha (19.9%). Massachusetts, where the species has been
extirpated since the early 1970s (Rimmer et al. 2001), had just 12.06 ha (0.01%).

Bicknell‟s Thrush habitat occupancy involves a complexity of factors with intricate links
between landscape and local scale features of the habitat (Frey 2008). Simply preserving large
tracts of habitat may not be sufficient to ensure future persistence, but could minimize local
extinction risk. Because it is well known that Bicknell‟s Thrush is a natural disturbance, mid-
succession specialist (Rimmer et al. 2001), careful consideration of local habitat attributes such
as natural disturbance regimes or anthropogenic uses and management that closely mimic these
processes will be vital to the continued persistence of Bicknell‟s Thrush.

Acknowledgements
We thank Sarah Frey, Oregon State University - Forest Ecosystems and Society, for statistical
advice and data analysis help. We are grateful for core financial support provided by the
Vermont Monitoring Cooperative for Forest Bird Monitoring Program annual survey routes on
Mt. Mansfield. We gratefully acknowledge the hundreds of volunteers who participate in
Mountain Birdwatch. This dedicated group was recruited with assistance from the Adirondack
Mountain Club, the Appalachian Mountain Club, the Appalachian Trail Conservancy, Audubon
New York, Maine Audubon, the Maine Department of Inland Fisheries and Wildlife, and the
Wildlife Conservation Society. We are thankful for permission to conduct surveys on lands
owned and/or managed by: UVM, the Carthusian Monastery, Plum Creek Timber Company,
Inc., the Green Mountain Club, the Maine Department of Inland Fisheries and Wildlife, the
National Park Service, the New York State Department of Environmental Conservation, the U.S.
Forest Service, and the Vermont Agency of Natural Resources. Mountain Birdwatch is funded
by the U.S. Fish and Wildlife Service through a cooperative agreement administered by Assistant
Nongame Bird Coordinator and Mountain Birdwatcher, Randy Dettmers. We also receive
generous support from the National Park Service, Vermont Agency of Natural Resources, US
Forest Service, New York State Department of Environmental Conservation, and private donors.

Literature Cited

Burnham, K. P., and D. R. Anderson. 2002. Model selection and multimodel inference: A
practical information-theoretic approach. Springer-Verlag, Inc., New York, NY.
Cade, B. S., and B. R. Noon. 2003. A gentle introduction to quantile regression for ecologists.
Frontiers in Ecology and the Environment 1:412-420.
Frey, Sarah. 2008. Metapopulation dynamics and multi-scale habitat selection of a montane
forest songbird. M.S. Thesis, Univ. of Vermont. 80pp.

McFarland, Faccio, and Rimmer 2010 VMC Report
23

Hart, Julie A., and J. Daniel Lambert. 2007. Mountain Birdwatch 2006: Final Report to the US
Fish and Wildlife Service. Unpubl. report. Vermont Institute of Natural Science,
Quechee, VT. VINS Technical Report 07-03. 30 pp.
Hart, J.A., K.P. McFarland, and C.C. Rimmer. 2009. Production of a Bicknell‟s Thrush Global
Breeding Distribution Model Based on Known Occurrence and Potential Habitat.
Unpublished Report to Environment Canada. 12 pp.
Homer, C. C. Huang, L. Yang, B. Wylie and M. Coan. 2004. Development of a 2001 National
Landcover Database for the United States. Photogrammetric Engineering and Remote
Sensing, Vol. 70, No. 7, July 2004, pp. 829-840.
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.
Mackenzie, D., J. Nichols, J. Hines, M. Knutson, and A. Franklin. 2003. Estimating site
occpancy, colonization, and local extinction when a species is detected imperfectly.
Ecology 84:2200-2207.
Mackenzie, D., J. Nichols, J. Royle, K. Pollock, L. Bailey, and J. Hines. 2006. Occupancy
Estimation and Modeling: Inferring Patterns and Dynamics of Species Occurrence.
Elsevier Inc., Oxford, UK.
McFarland, K.P. and C. C. Rimmer. 2009. Analysis of Potential Bicknell‟s Thrush Breeding
Habitat in the Vicinity of Sisk Mountain, Maine Using a Predictive Distribution Model.
Unpublished Report to BRI. 13pp.
Rimmer, C. C., K. P. McFarland, W. G. Ellison, and J. E. Goetz. 2001. Bicknell‟s Thrush
(Catharus bicknelli). In The Birds of North America, No. 592 (A. Poole & F. Gill, eds.).
The Birds of North America, Inc., Philadelphia, PA.

McFarland, Faccio, and Rimmer 2010 VMC Report
24

Table 1. AIC model selection results for determining the effects of patch size, habitat within 1, 5, 10, and 25 km of
survey site on initial occupancy (ψ), colonization (γ), and local extinction (ε). Detection probability (p) was
modeled as a function of survey number for all models. Each model is ranked by its AIC score, which represents
how well the model fits the data. A lower delta AIC value is indicative of a better model. Only models within 4
AIC points of the top model were considered plausible. The probability that the model (of the models tested) would
best explain the data is indicated by AIC weight. Model likelihood is the ratio of each model‟s AIC weight to the
top model‟s AIC weight.

McFarland, Faccio, and Rimmer 2010 VMC Report
25


Figure 1. Estimated probability (+/- 1 SE) of initial occupancy (ψ) as a function of the amount of habitat (ha) within
a 1-km radius from the top ranked model: psi(hab1km),gamma(hab1km),eps(hab1km),p(survey).

Figure 2. Estimated probability (+/- 1 SE) of local site colonization (γ) as a function of the amount of habitat (ha)
within a 1-km radius from the top ranked model: psi(hab1km),gamma(hab1km),eps(hab1km),p(survey).

McFarland, Faccio, and Rimmer 2010 VMC Report
26


Figure 3. Estimated probability (+/- 1 SE) of local site extinction (ε) as a function of the amount of habitat (ha)
within a 1-km radius from the top ranked model: psi(hab1km),gamma(hab1km), eps(hab1km),p(survey).

McFarland, Faccio, and Rimmer 2010 VMC Report
27

Part IV. Forest Bird Surveys on Mt. Mansfield and Lye Brook Wilderness Area

In 2010, breeding bird surveys were continued at 3 permanent study sites on Mt. Mansfield, and
on a single site at the Lye Brook Wilderness Area (LBWA) of the Green Mountain National
Forest. The Mt. Mansfield ridgeline was surveyed for the 20th consecutive year, while the
Ranch Brook site was censused for the 15th year in 2010 (the 2004 survey was not completed
due to inclement weather on attempted survey dates). Our permanent study site at Underhill
State Park was surveyed for the 18th year in 2010 (the site was not surveyed in 2003 or 2005).
The LBWA was surveyed for the 11th consecutive year in 2010.

The Underhill State Park site consists of mature northern hardwoods ranging from 609 to 731 m
(2000 to 2400 ft) elevation, while the Mansfield ridgeline site, at 1158 m (3800 ft), consists of
montane fir-spruce. The Ranch Brook site ranges between 975 and 1097 m (3200 and 3600 ft),
and is dominated by a paper birch-fir canopy. The Lye Brook study site, located in Winhall, just
north of Little Mud Pond, is characterized by mature northern hardwoods at an elevation of 701
m (2300 ft).

These four study sites are part of VCE‟s long-term Forest Bird Monitoring Program (FBMP).
This program was initiated in 1989 with the primary goals of conducting habitat-specific
monitoring of forest interior breeding bird populations in Vermont and tracking long-term
changes (Faccio et al. 1998). As of 2008, VCE had established 39 monitoring sites in 9 different
forested habitats in Vermont, with additional montane sites in New York, New Hampshire,
Maine, and Massachusetts. A complementary, volunteer-based, long-term monitoring program,
called Mountain Birdwatch, was initiated in 2000 to collect census data on five common
montane forest bird species throughout the Northeast. Also, through a cooperative agreement
with the National Park Service, VCE is coordinating breeding bird monitoring at 9 National
Parks and Historic Sites in the Northeast. Initiated in 2006, annual surveys are conducted at 19
study sites in New Jersey, Connecticut, New York, Massachusetts, Vermont, New Hampshire,
and Maine.

Methods

In 2010, surveys were conducted by VCE staff biologists at the Mt. Mansfield Ridgeline and
Ranch Brook, and by volunteer observers at the Lye Brook and Underhill sites. Each study site
consisted of 5 point count stations. Survey methods consisted of unlimited distance point counts,
based on the approach described by Blondel et al. (1981) and used in Ontario (Welsh 1995). The
count procedure was as follows:

1. Counts began shortly after dawn on days where weather conditions were unlikely to
reduce count numbers (i.e., calm winds and very light or no rain). Censusing began
shortly (< 1 min) after arriving at a station.

2. Observers recorded all birds seen and heard during a 10-min sampling period, which was
divided into 3 time intervals: 3, 2, and 5 mins. Observers noted in which time interval
each bird was first encountered, and placed birds into one of 2 distance categories (within
or beyond 50 m). To reduce duplicate records, individual birds were mapped on

McFarland, Faccio, and Rimmer 2010 VMC Report
28

standardized field cards, and known or presumed movements were noted. Different
symbols were used to record the status of birds encountered (i.e., singing male, pair
observed, calling bird, etc.).

3. The number of surveys at each site was dependent on elevation; montane fir-spruce sites
were sampled once, while LBWA and Underhill were sampled twice during the breeding
season, the first during early June (ca. 2-12 June) and the second during late June (ca. 14-
30 June). Observers were encouraged to space their visits 7-10 days apart. For each site
visit, all stations were censused in a single morning and in the same sequence.

In summarizing data for analysis, the maximum count for each species was used as the station
estimate for each year. All birds seen or heard were each counted as 1 individual unless a family
group or active nest was encountered, in which case they were scored as a breeding pair, or 2
individuals. Population trends were calculated for the 8 most commonly encountered species at
each site using simple linear regression. For each species, the annual population trend was
calculated by plotting the maximum count against year, and then calculating the mean annual
rate of change of a linear trendline inserted through the points (e.g. Percent Annual Trend = slope
÷ y intercept x 100). Regression and correlation analyses were done using SYSTAT 10.2.

Results

A combined total of 55 avian species were detected during breeding bird surveys at three study
sites on Mt. Mansfield from 1991-2010. Species richness was similar at both montane forest
sites, with a total of 30 species detected at the Mansfield ridgeline and 31 at Ranch Brook.
Surveys at Ranch Brook continue to average a greater number of individuals and species per year
than the higher elevation and more exposed Mansfield ridgeline site (Tables 1 and 2). Surveys at
the mid-elevation, northern hardwood study sites at Underhill State Park and Lye Brook
Wilderness showed similar species composition, with Underhill averaging 17.83 species per year
compared to Lye Brook‟s 15.91 (Tables 3 and 4).

Mount Mansfield
On the Mt. Mansfield ridgeline plot in 2010, both species richness and numerical abundance
were well below average, with 55 individuals of 9 species detected, the lowest species count in
the survey‟s 20-year history (Table 1). Of the 8 most commonly recorded species, 4 were below
the 20-year average, and 4 were above. Five species exhibited decreasing population trends,
with one species, Blackpoll Warbler, showing a significant decline of 2.75% per year (r2 =
0.306; P = 0.011). Three species showed non-significant increasing trends. The number of
Bicknell‟s Thrush dropped from last year‟s count of 10 individuals to just 6, the lowest since
2004.

At the Ranch Brook study site in 2010, species richness was slightly below the 15-year average,
while numerical abundance rebounded from last year‟s record low count of just 37 individuals to
79 (Table 2). Among the 8 most abundant species, half were below the 15-year mean and half
were above. Overall, just 2 of these 8 species showed increasing trends, while 6 declined. Two
species declined significantly; White-throated Sparrow, which continued a downward trend at a
rate of 4.35% per year (r2 = 0.465; P = 0.005), and Yellow-bellied Flycatcher, which declined at

McFarland, Faccio, and Rimmer 2010 VMC Report
29

a rate of 2.91% per year (r2 = 0.359; P = 0.018). Counts of Bicknell‟s Thrush, Swainson‟s
Thrush, and Dark-eyed Junco all recovered from last year‟s record low numbers.

At Underhill State Park in 2010, total number of individuals rebounded from last year‟s record
low count of 39 individuals to 68 birds of 17 species (Table 3). Among the 8 most common
species at the site, half were above the 18-year mean, and half were below. Overall, 6 species
showed increasing population trends, including significant increases for Black-throated Blue
Warbler (6.27%; r2 = 0.289, P = 0.022) and Black-throated Green Warbler (5.44%; r2 = 0.459, P
= 0.002). After a single Canada Warbler was detected in 2008, the first in 5 years, none were
encountered in 2009 or 2010, continuing its declining trend at 5.30% per year (r2 = 0.714, P <
0.001). This was also the second consecutive year that no Winter Wrens were detected at
Underhill State Park.

Lye Brook Wilderness
At Lye Brook Wilderness, species richness and numerical abundance were both below the 11-
year average, with 61 individuals of 12 species detected (Table 4). Among the 8 most common
species, five were above the 10-year average, while three (all warblers) were below. Of these 8
species, five exhibited increasing population trends, while three showed declines. Among
significant trends, Ovenbird declined at a rate of 3.52% (r2 = 0.434; P = 0.028), and Hermit
Thrush increased at 12.90% per year (r2 = 0.519; P = 0.012). The maximum counts for two
species (Yellow-bellied Sapsucker and Hermit Thrush) were the highest in site‟s 11-year history.

Discussion

Although the linear regression trend for Blackpoll Warbler showed a significant decline for the
fourth consecutive year, the species appears to be rebounding from its record low count of 2
individuals on the Mt. Mansfield Ridgeline site in 2007. Interestingly, 2007 was also a low year
for Blackpolls at the Ranch Brook site, suggesting that low counts may represent an accurate
index to the population.

At the Ranch Brook site, White-throated Sparrow continued its declining trend for the fourth
consecutive year. However, the unusually high maximum count of 22 White-throats recorded in
1995 is largely responsible for driving the trend‟s statistical significance. The mean count at
Ranch Brook was 8.8 over the 15-year study period, and 7.9 without the 1995 outlier. So, while
the biological significance of the White-throated Sparrow decline observed at Ranch Brook
appears to be low, it bears continued scrutiny.

It‟s encouraging to note that two of the most commonly detected species at Underhill State Park
(Black-throated Green Warbler and Black-throated Blue Warbler) show significantly increasing
trends, suggesting that the conditions of their preferred breeding habitat has improved over the
last 10 years or so. Considering that Black-throated Green Warbler numbers increased after
1999 and the upward trend for Black-throated Blue began in 2004, it‟s possible that impacts from
the 1998 ice storm may have resulted in changes to forest structure and composition, benefitting
both species. Black-throated Blue Warbler prefers hardwood forests with a dense understory of
hobblebush (Viburnum lantanoides), while Black-throated Green prefer stands with a mix of
hardwood and coniferous trees. The ice storm resulted in broken limbs and main stems,

McFarland, Faccio, and Rimmer 2010 VMC Report
30

primarily of hardwood canopy trees, which left many small- to medium-sized canopy gaps
(Faccio 2003). Since the storm had a disproportionate effect on hardwoods over conifers (which
can better withstand ice accumulation due to their structural morphology), percent conifer cover
may have increased, favoring Black-throated Green Warbler. In addition, canopy gaps allowed
more sunlight to reach the forest floor, which, after a lag-time of 2-3 years, increased the density
of the understory layer, including hobblebush (Faccio 2003), possibly creating additional habitat
for Black-throated Blue Warbler.

The site-specific trend estimates presented for the Mt. Mansfield and Lye Brook sites must be
interpreted carefully as these data are from a limited geographic sample with low power. Year to
year changes in survey counts may simply reflect natural fluctuations, variable detection rates,
and/or a variety of dynamic factors, such as prey abundance, overwinter survival, and habitat
change. Continued data collection, their correlation with other VMC data, and comparison with
survey data from other ecologically similar sites will be necessary to elucidate meaningful
population trends of various species at these sites.

Acknowledgements
Many thanks to Bobbie Jean Booth and Zoe Richards for conducting bird surveys at the Lye
Brook Wilderness Area and Underhill State Park, respectively.


Literature Cited

Blondel, J., C. Ferry, and B. Frochot. 1981. Point counts with unlimited distance. Pp. 414-420,
In C. John Ralph and J. Michael Scott (Eds.). Estimating numbers of terrestrial birds.
Studies in Avian Biology 6: 630pp.

Faccio, S.D. 2003. Effects of ice storm-created gaps on forest breeding bird communities in
central Vermont. Forest Ecology and Management 186: 133-145.

Faccio, S.D., C.C. Rimmer, and K.P. McFarland. 1998. Results of the Vermont Forest Bird
Monitoring Program, 1989-1996. Northeastern Naturalist, 5(4): 293-312.

Welsh, D.A. 1995. An overview of the Forest Bird Monitoring Program in Ontario, Canada.
Pp. 93-97, In C.J. Ralph, J.R. Sauer, and S. Droege, (Eds.). Monitoring bird populations by
point counts. General Technical Report PSW-GTR-149. Pacific Southwest Research
Station, Forest Service, U.S. Dept. of Agriculture, Albany, CA. 181pp.

McFarland, Faccio, and Rimmer 2010 VMC Report
31

Table 1. Maximum counts of individual birds, and population trends from linear regression analysis for the 8 most common species
(bold type) at Mt. Mansfield Ridgeline, 1991-2010.

Common Name

‘91

‘92

‘93

‘94

‘95

‘96

‘97

‘98

‘99

‘00

‘01

‘02

‘03

‘04

‘05

‘06

‘07

‘08

‘09

‘10

Mean

SD

r2
Annual
Trend (%)
Red Squirrel 1 0.05 0.22
Sharp-shinned Hawk 1 0.05 0.22
Hairy Woodpecker 1 0.05 0.22
Northern Flicker 1 0.05 0.22
Yellow-bellied Flycatcher 1 1 2 3 1 1 1 1 2 1 1 2 1 3 1.05 0.94
Alder Flycatcher 1 0.05 0.22
Red-eyed Vireo 1 0.05 0.22
Blue Jay 1 1 1 0.15 0.37
Common Raven 1 1 1 1 1 1 1 2 1 0.50 0.61
Red-breasted Nuthatch 1 2 3 1 3 1 1 2 1 1 1 0.85 0.99
Winter Wren 10 9 7 4 5 2 4 10 8 4 4 7 3 7 8 12 7 5 6 8 6.50 2.63 0.003 0.41
Golden-crowned Kinglet 1 0.05 0.22
Ruby-crowned Kinglet 2 1 1 1 1 0.30 0.57
Bicknell's Thrush 6 15 11 8 10 11 9 9 8 7 9 9 6 5 8 11 12 7 10 6 8.85 2.43 0.072 -1.10
Swainson's Thrush 3 8 1 1 3 6 7 5 4 3 3 2 2 1 2 5 1 5 3 5 3.50 2.06 0.019 -1.20
Hermit Thrush 1 1 0.10 0.31
American Robin 1 4 1 2 2 2 2 1 1 3 3 2 6 3 1 3 4 3 2 4 2.50 1.32 0.153 5.51
Cedar Waxwing 1 4 9 1 0.75 2.15
Nashville Warbler 2 2 3 1 1 1 1 1 0.60 0.88
Magnolia Warbler 1 2 3 1 1 1 3 1 4 1 0.90 1.21
Yellow-rumped Warbler 9 11 8 9 8 12 10 13 11 9 11 14 10 13 9 9 7 12 12 8 10.25 1.97 0.006 0.23
Blackpoll Warbler 8 9 9 7 7 15 10 10 9 8 8 3 3 9 8 8 2 4 5 5 7.35 3.03 0.306 -2.75*
Ovenbird 1 1 0.10 0.31
Canada Warbler 1 0.05 0.22
Lincoln's Sparrow 2 1 0.15 0.49
White-throated Sparrow 6 14 14 12 14 13 20 14 19 14 18 11 13 11 10 14 14 12 10 12 13.25 3.19 0.010 -0.40
Dark-eyed Junco 3 9 6 2 5 5 9 8 7 2 7 6 5 7 4 5 4 6 6 6 5.60 1.98 0.000 -0.53
Purple Finch 2 4 1 2 3 2 2 1 4 2 3 4 4 2 1 2 2 4 3 2.40 1.19
White-winged Crossbill 8 1 1 0.50 1.79
Pine Siskin 1 1 2 1 11 5 1 1.10 2.61
Evening Grosbeak 2 0.10 0.45
Species Richness a 13 16 15 11 14 15 17 14 15 13 15 12 15 14 11 13 13 11 11 9 13.35 2.03
Number of Individuals a 54 94 69 49 71 78 94 76 78 56 80 61 61 63 56 62 62 60 61 55 66.25 12.14
a Does not include counts of Red Squirrel
* P = 0.011

McFarland, Faccio, and Rimmer 2010 VMC Report
32

Table 2. Maximum counts of individual birds, and population trends from linear regression analysis for the 8 most common species
(bold type) at Ranch Brook, 1995-2010. Note that a survey was not conducted in 2004.

Common Name

‘95

‘96

‘97

‘98

‘99

‘00

‘01

‘02

‘03

‘04

‘05

‘06

‘07

‘08

‘09

‘10

Mean

SD

r2
Annual
Trend (%)
Eastern Chipmunk 1 0.07 0.26
Red Squirrel 4 1 7 4 1.07 2.15
Sharp-shinned Hawk 1 1 0.13 0.35
Mourning Dove 1 1 0.13 0.35
Ruby-throated Hummingbird 1 0.07 0.26
Hairy Woodpecker 1 0.07 0.26
Pileated Woodpecker 2 0.13 0.52
Yellow-bellied Flycatcher 4 4 4 3 3 4 2 4 4 3 2 4 3 2 1 3.13 0.99 0.359 -2.91*
Blue-headed Vireo 1 0.07 0.26
Red-eyed Vireo 1 0.07 0.26
Blue Jay 1 1 1 4 0.47 1.06
Common Raven 4 3 4 4 2 1 1 1 1.33 1.63
Black-capped Chickadee 1 1 0.13 0.35
Red-breasted Nuthatch 7 2 6 2 2 4 5 1 5 2.27 2.49
Winter Wren 8 3 7 10 9 10 5 5 9 10 11 6 8 5 9 7.67 2.38 0.016 0.87
Golden-crowned Kinglet 1 3 1 3 2 1 2 1 0.93 1.10
Ruby-crowned Kinglet 3 3 3 1 1 1 1 0.87 1.19
Bicknell's Thrush 5 6 7 5 5 6 2 8 1 8 2 5 5 2 7 4.93 2.25 0.030 -1.41
Swainson's Thrush 6 15 9 5 3 4 8 11 10 8 5 9 7 3 7 7.33 3.24 0.046 -1.67
Hermit Thrush 1 3 0.27 0.80
American Robin 2 2 2 1 1 1 1 3 4 5 2 2 3 6 2.33 1.63
Cedar Waxwing 1 1 1 0.20 0.41
Nashville Warbler 1 3 2 1 3 3 4 3 2 3 2 1 4 2.13 1.30
Northern Parula 1 0.07 0.26
Magnolia Warbler 2 4 4 2 3 5 4 2 4 2 3 1 2 2 6 3.07 1.39
Black-throated Blue Warbler 1 0.07 0.26
Yellow-rumped Warbler 5 6 4 5 7 11 9 11 8 4 8 8 6 4 7 6.87 2.33 0.002 0.35
Blackpoll Warbler 9 9 15 8 3 8 7 8 8 8 10 4 6 6 7 7.73 2.74 0.165 -2.35
White-throated Sparrow 22 11 12 9 8 7 7 10 10 7 4 8 4 5 8 8.80 4.35 0.465 -4.35**
Dark-eyed Junco 9 5 3 2 5 2 5 4 4 7 5 1 4 1 5 4.20 2.21 0.088 -2.51
Purple Finch 2 1 4 4 2 4 4 6 2 1 5 2.53 1.92
White-winged Crossbill 8 2 1 6 1.13 2.47
Pine Siskin 12 1 7 1 1 1.47 3.42
Species Richness
a
19 13 18 17 16 17 18 12 15 17 15 16 16 14 15 15.87 1.92
Number of Individuals
a
107 71 88 65 67 75 69 82 82 74 61 62 59 37 79 71.87 15.68
a Does not include counts of Eastern Chipmunk or Red Squirrel
* P = 0.018; ** P = 0.005

McFarland, Faccio, and Rimmer 2010 VMC Report
33

Table 3. Maximum counts of individual birds, and population trends from linear regression analysis for the 8 most common species
(bold type) at Underhill State Park, 1991-2010. Note that surveys were not conducted in 2003 or 2005.

Common Name

‘91

‘92

‘93

‘94

‘95

‘96

‘97

‘98

‘99

‘00

‘01

‘02

‘03

‘04

‘05

‘06

‘07

‘08

‘09

‘10

Mean

SD

r2
Annual
Trend (%)
Eastern Chipmunk 3 5 1 1 0.56 1.34
Red Squirrel 1 3 1 1 1 1 0.44 0.78
Broad-winged Hawk 1 0.06 0.24
Mourning Dove 1 1 0.11 0.32
Yellow-bellied Sapsucker 2 1 1 1 1 1 3 2 2 3 2 1.06 1.06
Downy Woodpecker 1 1 1 1 0.22 0.43
Hairy Woodpecker 1 1 1 2 2 2 0.50 0.79
Northern Flicker 1 0.06 0.24
Pileated Woodpecker 2 1 1 1 0.28 0.57
Least Flycatcher 2 0.11 0.47
Eastern Phoebe 1 0.06 0.24
Blue-headed Vireo 1 2 1 1 1 1 2 1 1 3 3 0.94 1.00
Red-eyed Vireo 3 4 4 6 9 8 7 6 10 8 8 7 5 7 8 6 2 5 6.28 2.14 0.000 0.01
Blue Jay 2 1 1 2 2 1 1 2 1 1 1 1 1 0.94 0.73
American Crow 1 1 0.11 0.32
Common Raven 4 1 1 1 1 2 0.56 1.04
Black-capped Chickadee 1 1 2 3 3 3 1 1 2 1 3 2 1.28 1.18
Red-breasted Nuthatch 1 0.06 0.24
White-breasted Nuthatch 1 1 0.11 0.32
Brown Creeper 1 1 1 1 1 1 1 0.39 0.50
Winter Wren 6 2 1 5 3 4 6 4 4 3 3 3 4 2 1 2.83 1.92 0.129 -2.84
Golden-crowned Kinglet 1 1 0.11 0.32
Veery 1 1 1 0.17 0.38
Swainson's Thrush 1 2 4 3 1 4 2 2 1 2 1.22 1.40
Hermit Thrush 4 1 6 7 3 4 4 2 4 5 4 4 7 1 4 3 3.50 2.09 0.014 1.33
Wood Thrush 1 1 0.11 0.32
American Robin 1 3 3 3 4 2 1 2 1 2 1 3 1.44 1.34
Magnolia Warbler 1 1 1 0.17 0.38
Black-th. Blue Warbler 4 9 5 6 7 8 6 5 6 5 5 5 11 15 8 11 5 14 7.50 3.28 0.289 6.27*
Yellow-rumped Warbler 2 2 2 3 3 1 1 3 2 1 1 1 1 1.28 1.07
Black-th. Green Warbler 5 7 6 7 7 7 9 5 8 10 10 8 13 15 12 10 7 11 8.72 2.76 0.459 5.44**
Blackburnian Warbler 1 1 1 1 0.22 0.43
Blackpoll Warbler 1 2 0.17 0.51
Continued on next page

McFarland, Faccio, and Rimmer 2010 VMC Report

Underhill State Park, cont.
34


Common Name

‘91

‘92

‘93

‘94

‘95

‘96

‘97

‘98

‘99

‘00

‘01

‘02

‘03

‘04

‘05

‘06

‘07

‘08

‘09

‘10

Mean

SD

r2
Annual
Trend (%)
Black-and-White Warbler 3 2 2 4 2 3 2 1 3 4 2 1 2 3 1 1 2.00 1.19
American Redstart 4 1 1 0.33 0.97
Ovenbird 4 10 11 11 13 12 12 10 13 10 13 6 11 11 15 14 7 14 10.94 2.88 0.095 1.53
Mourning Warbler 1 1 0.11 0.32
Canada Warbler 3 4 4 6 2 4 4 2 2 3 2 2 1 2.17 1.79 0.714 -5.30***
Scarlet Tanager 1 1 1 0.17 0.38
White-throated Sparrow 2 2 1 1 1 1 1 1 1 0.61 0.70
Dark-eyed Junco 3 1 3 4 3 5 2 2 1 2 2 1 5 5 2 4 3 2.67 1.50 0.102 4.16
Rose-breasted Grosbeak 4 2 1 3 1 2 1 1 0.83 1.20
Purple Finch 1 1 1 1 1 1 0.33 0.49
White-winged Crossbill 2 0.11 0.47
Pine Siskin 1 1 0.11 0.32
American Goldfinch 1 1 0.11 0.32
Species Richness a 15 19 14 18 20 20 23 16 21 16 20 16 17 22 18 17 12 17 17.83 2.85
Number of Individuals a 35 66 43 62 77 69 77 54 67 53 70 48 60 81 73 58 39 68 61.11 13.48
a Does not include counts of Red Squirrel or Eastern Chipmunk
* P = 0.022
** P = 0.002
*** P < 0.001

McFarland, Faccio, and Rimmer 2010 VMC Report
35

Table 4. Maximum counts of individual birds, and population trends from linear regression analysis for the 8 most common
species (bold type) at Lye Brook Wilderness Area, 2000-2010.

Common Name

‘00

‘01

‘02

‘03

‘04

‘05

‘06

‘07

‘08

‘09

‘10

Mean

SD

r2
Annual
Trend (%)
Eastern Chipmunk 2 1 0.27 0.65
Red Squirrel 1 1 0.18 0.40
Ruffed Grouse 1 2 0.27 0.65
Mourning Dove 1 0.09 0.30
Yellow-Billed Cuckoo 1 0.09 0.30
Barred Owl 1 0.09 0.30
Chimney Swift 2 0.18 0.60
Ruby-throated Hummingbird 1 1 0.18 0.40
Yellow-bellied Sapsucker 5 6 2 2 2 5 8 11 3.73 3.61 0.227 83.80
Downy Woodpecker 1 1 0.18 0.40
Hairy Woodpecker 2 1 2 1 1 1 0.73 0.79
Unidentified Woodpecker 3 0.27 0.90
Northern Flicker 1 0.09 0.30
Pileated Woodpecker 1 3 1 4 1 1 2 1 2 1.45 1.21 0.001 0.65
Eastern Wood-Pewee 1 0.09 0.30
Yellow-bellied Flycatcher 1 0.09 0.30
Least Flycatcher 2 0.18 0.60
Great Crested Flycatcher 1 0.09 0.30
Blue-headed Vireo 1 4 1 1 1 0.73 1.19
Red-eyed Vireo 10 6 9 4 6 6 4 5 13 14 10 7.91 3.51 0.143 7.26
Blue Jay 3 1 1 2 1 3 1.00 1.18
Common Raven 1 1 0.18 0.40
Black-capped Chickadee 1 1 2 1 2 1 1 0.82 0.75
White-breasted Nuthatch 1 1 0.18 0.40
Brown Creeper 1 0.09 0.30
Winter Wren 7 1 3 1 2 1.27 2.15
Ruby-crowned Kinglet 1 0.09 0.30
Veery 1 0.09 0.30
Swainson's Thrush 2 1 3 2 2 1 1 2 1.27 1.01
Hermit Thrush 4 2 6 5 4 4 4 5 6 7 8 5.00 1.67 0.519 12.90*
American Robin 1 1 3 1 1 2 0.82 0.98
Cedar Waxwing 1 1 0.18 0.40
Northern Parula 3 1 0.36 0.92
Magnolia Warbler 1 3 0.36 0.92
Black-throated Blue Warbler 9 7 10 9 8 12 11 8 8 8 5 8.64 1.91 0.110 -1.95
Yellow-rumped Warbler 2 1 0 1 0.36 0.67 l
Black-throated Green Warbler 8 10 4 6 8 9 12 3 11 9 6 7.82 2.82 0.003 0.60
Blackburnian Warbler 5 0.45 1.51
American Redstart 2 1 3 1 4 1.00 1.41
Ovenbird 15 13 19 11 14 13 12 12 8 12 10 12.64 2.84 0.434 -3.52**
Canada Warbler 1 0.09 0.30
Scarlet Tanager 1 3 2 2 2 1 1 1.09 1.04
White-throated Sparrow 2 2 4 2 0.91 1.38
Dark-eyed Junco 2 3 1 1 1 4 1 2 1.36 1.29 0.141 -6.50
Rose-breasted Grosbeak 2 1 0.27 0.65
Species Richness a 28 15 17 17 16 17 11 11 18 13 12 15.91 4.76
Number of Individuals a 98 58 73 57 60 65 51 41 66 66 61 63.27 14.31
a Does not include counts of Red Squirrel or Eastern Chipmunk
* P = 0.012, ** P = 0.028