COSEWIC Assessment and Status Report on the Bicknell's Thrush Catharus bicknelli in Canada (1999)
Table of Contents
Assessment and Status Report
COSEWIC status reports are working documents used in assigning the status of wildlife species suspected of being at risk. This report may be cited as follows:
Please note: Persons wishing to cite data in the report should refer to the report (and cite the author(s)); persons wishing to cite the COSEWIC status will refer to the assessment (and cite COSEWIC). A production note will be provided if additional information on the status report history is required.
COSEWIC 1999. COSEWIC assessment and status report on the Bicknell’s thrush Catharus bicknelli in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa. v + 43 pp.
Nixon E. 1999. COSEWIC status report on the Bicknell’s thrush Catharus bicknelli in Canada, in COSEWIC assessment and status report on the Bicknell’s thrushCatharus bicknelli in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa. 1-43 pp.
For additional copies contact:
c/o Canadian Wildlife Service
Également disponible en français sous le titre Évaluation et Rapport de situation du COSEPAC sur la situation de la Grive de Bicknell au Canada.
Bicknell’s Thrush -- J. Crosby, The Birds of Canada, by W. Earl Godfrey, Canadian Museum of Nature, Ottawa, ON.
©Her Majesty the Queen in Right of Canada 2004
Catalogue No. CW69-14/454-2005E-PDF
COSEWIC Assessment Summary
Assessment Summary – April 1999
Common name : Bicknell’s Thrush
Scientific name : Catharus bicknelli
Status : Special Concern
Reason for designation : This habitat specialist of high-elevation dense balsam fir forest is distributed locally and is not numerous anywhere. The species is of special concern in Canada primarily because large-scale forestry practices and budworm infestation may lead to significant losses of preferred habitat.
Occurrence : Quebec, New Brunswick and Nova Scotia
Status history : Designated Special Concern in April 1999. Assessment based on a new status report.
COSEWIC Executive Summary
Although Bicknell's Thrush (Catharus bicknelli) records can be traced back to 1881, much remains to be learned about this species due to its historic classification as a subspecies of the Gray-cheeked Thrush (Catharus minimus), its preference for inaccessible, high elevation habitat, and its elusive behaviour. While the species breeds in suitable habitat in Quebec, the Maritimes, and the northeastern United States, the full extent of its wintering range is largely unknown, as are population numbers and trends, and life history traits.
With a Canadian population estimated at 2000-5000 pairs and a global population of 5000-15 000 pairs, Bicknell's Thrush is one of Canada's rarest songbirds. As such, it is recognized as a "highest priority species" for monitoring, research and conservation by the Canadian Wildlife Service (Dunn 1997); it is ranked the number one avian conservation priority in the northeastern United States (Rosenberg and Wells 1995); and a status of "vulnerable" has been suggested by BirdLife's Globally Threatened Species Programme using IUCN's "Red List" criteria.
While there appear to be no large-scale changes in the Canadian distribution of Bicknell's Thrush, the species has vanished from a number of headlands and islands along the Nova Scotian coast, as well as a few other historically known breeding areas. In the U.S., intensive distribution surveys have detected changes in the southern limit of the species' range, where it can no longer be found breeding in Massachusetts. While distribution data suggest relatively stable "core" sub-populations, several peripheral sub-populations appear to be decreasing.
Recent research indicated that the species is facing habitat modification, degradation, and loss across its range. Wide-scale forestry operations are evident throughout eastern Canada, and, while the species has been detected in regeneration following clear cutting, the value of such habitat remains unclear. Additional threats include loss of breeding habitat associated with spruce budworm infestations, acid rain, ski area developments, transmission and microwave tower construction and, on the wintering grounds, deforestation due to subsistence agriculture, charcoal production, and sugarcane operations.
Assessment of the conservation status of Bicknell’s Thrush need to consider its small population, fragmented distribution and apparently low reproductive ability, coupled with rangewide habitat loss and modification.
The Committee on the Status of Endangered Wildlife in Canada (COSEWIC) determines the national status of wild species, subspecies, varieties, and nationally significant populations that are considered to be at risk in Canada. Designations are made on all native species for the following taxonomic groups: mammals, birds, reptiles, amphibians, fish, lepidopterans, molluscs, vascular plants, lichens, and mosses.
COSEWIC comprises representatives from each provincial and territorial government wildlife agency, four federal agencies (Canadian Wildlife Service, Parks Canada Agency, Department of Fisheries and Oceans, and the Federal Biosystematic Partnership), three nonjurisdictional members and the co-chairs of the species specialist groups. The committee meets to consider status reports on candidate species.
Species: Any indigenous species, subspecies, variety, or geographically defined population of wild fauna and flora.
Extinct (X) : A species that no longer exists.
Extirpated (XT) : A species no longer existing in the wild in Canada, but occurring elsewhere.
Endangered (E) : A species facing imminent extirpation or extinction.
Threatened (T) : A species likely to become endangered if limiting factors are not reversed.
Special Concern (SC)* : A species of special concern because of characteristics that make it particularly sensitive to human activities or natural events.
Not at Risk (NAR)* : A species that has been evaluated and found to be not at risk.
Data Deficient (DD)*** : A species for which there is insufficient scientific information to support status designation.
* : Formerly described as “Vulnerable” from 1990 to 1999, or “Rare” prior to 1990.
** : Formerly described as “Not In Any Category”, or “No Designation Required.”
*** : Formerly described as “Indeterminate” from 1994 to 1999 or “ISIBD” (insufficient scientific information on which to base a designation) prior to 1994.
The Committee on the Status of Endangered Wildlife in Canada (COSEWIC) was created in 1977 as a result of a recommendation at the Federal-Provincial Wildlife Conference held in 1976. It arose from the need for a single, official, scientifically sound, national listing of wildlife species at risk. In 1978, COSEWIC designated its first species and produced its first list of Canadian species at risk. Species designated at meetings of the full committee are added to the list.
Environment Canada Environnement Canada
Canadian Wildlife Service Service canadien de la faune
The Canadian Wildlife Service, Environment Canada, provides full administrative and financial support to the COSEWIC Secretariat.
Breeding (Figure 1)
In Canada, Bicknell's Thrush breeds mainly in high elevation, dense and stunted spruce forests in Quebec, New Brunswick, and Nova Scotia. The species reaches its western limit in Quebec, where it breeds along the northern shore of the St. Lawrence, the Magdalen Islands and the Gaspé Peninsula (see Bishop in Wallace 1939, Todd 1963, Ouellet 1995). In New Brunswick, the species has been detected mainly in the northwestern and northcentral highlands (Squires 1976, Erskine 1992, Nixon 1996), while Nova Scotian populations are confined largely to the Cape Breton highlands and, in smaller numbers, to a few of offshore islands (Erskine 1992, D. Busby pers. comm.). The species has also been recorded at several other locations in the Maritime provinces (see below), however, given the sporadic and limited nature of these observations, such areas do not appear to support viable populations.
Historically, the species was present in a number of additional locations in Nova Scotia. It was noted in the first half of the century in considerable numbers on two islands off the southwest coast (Seal Island and Mud Island; Wallace 1939, Erskine 1992), and in lesser numbers on small islands off the south coast of Cape Breton Island (see Tufts 1986).
In the United States, Bicknell's Thrush is geographically isolated on mountains in northwestern and central Maine (limited, sporadic coastal distribution as well), northern and central New Hampshire, and the interior of Vermont (particularly in the Green Mountains). It also breeds in the Adirondack Mountains of New York, with the southern terminus of its range in the Catskills (Kibbe 1985, Petersen 1990, Ouellet 1993, Atwood et al. 1994). Populations were also noted in Massachusetts, but have not been detected there since the mid-1950s (Rimmer et al. 1993, Veit and Petersen 1993).
Wintering (Figure 2)
The majority of Bicknell's Thrush records have originated from the 76 000 km2island of Hispaniola, with records from Haiti (Morne Malanga) and the Dominican Republic, including Puerto Plata, Sanchez, Aguacate, Santo Domingo, north of Cabo Rojo, Duarte, Pedernales, and the Parque Nacional Sierra de Bahoruco (see Wallace 1939, Marshall and Clapp unpublished ms, Rimmer and McFarland 1995). The first direct biological link between the Hispaniolan and North American populations of Bicknell's Thrush was established in 1995 when researchers recaptured a bird in the Dominican Republic which had been banded in Vermont (Rimmer et al. 1997).
Figure 1. Current breeding range of Bicknell’s Thrush in North America.
Figure 2. Winter range of Bicknell’s Thrush.
Bicknell's Thrush has also been sighted in Cuba, Puerto Rico, and St. Croix (see Wallace 1939, Wallace 1949, A.O.U. 1957, Sladen 1988, Petersen 1990, Ouellet 1993). Additionally, it is likely that winter records of the Gray-Cheeked Thrush in Jamaica and Mona Island represent C.bicknelli (Arendt 1992). There are no reliable sightings or specimen records in Central or South America (Ouellet 1993).
While migration patterns are not fully delineated (Ouellet 1993), records indicate that the majority of the population migrates along the eastern coast and Atlantic coastal plain from southern Quebec to South Carolina (Wallace 1939, Ouellet 1993). Sightings and specimens have been collected from Quebec, New Brunswick, Nova Scotia and Sable Island, and on or near the coasts of Massachusetts, Connecticut, Rhode Island, New York (particularly New York Harbour and Long Island), New Jersey, Pennsylvania, Maryland, the District of Columbia, Virginia, North Carolina, South Carolina, Georgia, Florida, and Cay Sal in the Bahamas (Cory 1891, Wallace 1939, Bond 1956, Ouellet 1993).
There is a scarcity of records in the southern U.S., suggesting that a portion of the population begins to leave the North American coast anywhere from North Carolina southwards, with others continuing down into Florida (Wallace 1939, Wallace 1949).
Bicknell's Thrush may also migrate through the interior of the continent, although in smaller numbers (Ouellet 1993). It is listed as a rare migrant in southern Ontario (James 1991). Specimens have been collected in Illinois, and possibly a number of Great Lakes states, prompting Wallace (1939) to propose the existence of a southward Mississippi valley route. However, records collected in the last 50 years suggest that these are more likely vagrants.
Casual or accidental
Bicknell's Thrush is listed as accidental in Newfoundland (west coast in 1912; Peters and Burleigh 1951), Saskatchewan (Maple Creek), southern Ontario (Long Point, possibly Toronto and Hamilton), Ohio (Toledo), and possibly in Indiana (Vincennes), Illinois (Cook County) and Louisiana (New Orleans; Wallace 1939).
Bicknell's Thrush populations reach their northern distributional limit in Canada, breeding primarily in Quebec along the north shore of the Gulf of St. Lawrence, the Gaspé Peninsula, and the Magdalen Islands (Wallace 1939; Ouellet 1993, 1995), in the interior highlands of New Brunswick, and Cape Breton Island (Christie 1980, Tufts 1986, Erskine 1992).
Figure 3. Breeding range of Bicknell’s Thrush in Canada.
The species has also been detected at a few locations along the Bay of Fundy coasts of New Brunswick and Nova Scotia, including Fundy National Park (Christie 1980) and Point LePreau (Squires 1976), on Grand Manan (B. Dalzell pers. comm.) and Campobello Island (Norm Famous pers. comm.), the southwest coast of Nova Scotia (Todd 1963, Tufts 1986), and Prince Edward Island (Erskine 1992). However, population numbers in these locations and their sporadic nature suggest that these are not viable populations.
Bicknell's Thrush reaches the northern limit of its range in southern Quebec, where it is found breeding at altitudes of 175-1160 m along the north shore of the St. Lawrence River, on the Gaspé Peninsula, and in the eastern townships (Ouellet 1993, 1995). Records obtained during the 1984-1989 Quebec breeding bird atlas period (Ouellet 1995) were patchy and widely dispersed. The extreme boundaries observed at that time were, in the north, the mouth of the Moisie River; in the south, around Mt. Gosford on the Maine border; in the west, in the Reserve La Verendrye north of Ottawa; and in the east, the Magdalen Islands.
Historically, scattered records exist along the north shore of the St. Lawrence River as far east as the St-Marguerite River, Pointe Natashquan, the mouth of the Little Mecatina River (Todd 1963), La Tabatière (Gillet 1935), and into Labrador as far north as Cape Charles (Townsend 1917). Todd (1963) believed the distribution of Bicknell's Thrush extended to the eastern extremity of the Strait of Belle Isle. According to Ouellet (1993), historic observations of the species from Pointe Natashquan and the Little Mecatina River eastward may be extralimital records of birds straying beyond their regular breeding grounds during spring migration, or they may indicate a decrease in size in the current breeding range of Bicknell's Thrush in the province. During the Quebec atlas period, no confirmed nesting record was obtained beyond the Tadoussac area, with cases of probable breeding noted in the vicinities of Godbout and Moisie (Ouellet 1995).
Bicknell's Thrush has been sighted on the Magdalen Islands, although observations are sporadic. One specimen was collected 15 July 1887 on Grindstone Island (Wallace 1939), and two were collected in late June 1901 on Grosse Isle (Todd 1963). Breeding was also confirmed at the northern tip of the islands during the Quebec atlas period (Ouellet 1995). Censuses on the islands from May to October 1992 failed to disclose any evidence of Bicknell's Thrush, even in areas where it had been historically observed (A.J. Marshall pers. comm.). Thus, it appears that breeding on the islands is sporadic.
New Brunswick(Figure 4)
Bicknell's Thrush has been detected predominantly at elevations above 300-450 m (Erskine 1992, Nixon 1996). Historical and recent records suggest that the species also breeds at lowland coastal and island locations, but only in small numbers.
The majority of New Brunswick records are concentrated in the northwestern and northcentral highlands, more particularly in extreme northwest N.R near the Quebec border, and in and around Mt. Carleton Provincial Park, particularly in the counties of Restigouche, Northumberland and Victoria (Squires 1976, Erskine 1992, Nixon 1996, Busby and Holmes unpubl. data). Although seemingly suitable habitat is found in the Albert County highlands in the southeast, no evidence of Bicknell's Thrush was found there during the Maritimes atlas period (Erskine 1992).
Figure 4. Bicknell’s Thrush sightings in New Brunswickand Cape Breton, N.S., from surveys conducted from 1995-1998 by the Canadian Wildlife Service and the Canadian forest Service. (Courtesy of D. Busby, CWS-Atlantic Region.)
There have also been records of Bicknell's Thrush in low-elevation areas in southwestern New Brunswick, along the east coast (Point Escuminac), in cool areas along the Bay of Fundy coastline, and further south on the Grand Manan archipelago (Grand Manan Island and Campobello Island; Erskine 1992, N. Famous pers. comm.). The most eastward coastal records came from Fundy National Park, where a breeding population was discovered in 1979 (Christie 1980). Birds were not detected during a subsequent census in 1992, however, and there has been no further evidence of Bicknell's Thrush in the park (Christie 1993). During the Maritime atlas period, low-elevation breeding was also confirmed in an area 20 km west of Fundy National Park (Erskine 1992). The sporadic nature of such low-elevation sightings and the few nest records suggest that breeding in these areas is sporadic and involves small numbers of individuals (Erskine, unpublished ms).
Nova Scotia(Figure 4)
The majority of the provincial population is found on Cape Breton Island. According to Tufts (1986), Bicknell's Thrush may summer there in higher frequencies than elsewhere in N.S., with records, including evidence of confirmed breeding, particularly frequent on the highlands right up to the northern tip (see Tufts 1986, Erskine 1992). Tuft's observations were backed by a 1996-1998 Canadian Wildlife Service distribution survey that also detected the species at a number of low elevation sites in and around Cape Breton, including on St. Paul Island and Scatarie Island (D. Busby un pub 1. Data). Erskine (1992) located the species near sea level on southern Cape Breton Island and Main-a-Dieu Island off southeastern Cape Breton Island, while records also exist for Kidston Island off of Baddeck (Tufts 1986).
Bicknell's Thrush has a very scattered distribution on the Nova Scotian mainland. Historically, it was listed as a regular summer resident at Cape Forchu, Yarmouth County, on the Southwest coast (Allen 1916). However, Erskine (1992) was unable to confirm breeding there during the Maritimes atlas period. Observations included one probable mainland sighting of a breeding bird near Cape Chignecto on the northwest coast, and possible sightings in three other areas: coastal and interior (highland) sightings east of Cape Chignecto, and in southwestern N.S. in the vicinity of Cape Forchu. The species was not found in the highlands (200500 m elevation) counties of Kings, Annapolis and Lunenberg, nor in the smaller highlands of Pictou County.
There has been no recent report of Bicknell's Thrush on any of the hundreds of small islands along Nova Scotia's southwest and Bay of Fundy coastlines, although there are historical accounts of breeding populations on two of them. In 1904, Bicknell's Thrush was described as "very abundant" on Seal Island (see Bent in Wallace 1939), and present in lesser numbers on Mud Island (Wallace 1939). By 1938, the species was considered extirpated from Seal Island (Tufts 1986). However, it was apparently sighted there in 1954 (Erskine 1955) and again in 1983 (see I.A. McLaren in Tufts 1986). There have been no records since then, leading Erskine (unpublished ms) to attribute the sightings to a series of colonizations and extirpations following the species' original disappearance, eventually culminating in its final elimination. The species' presence on
Mud Island has been less thoroughly documented, but based on the lack of recent records, it is likely that this population met with a similar fate.
Prince Edward Island
Neither the Maritimes breeding bird atlas nor the P.E.I. bird checklist present evidence of breeding of Bicknell's Thrush in the province. However, the species was recorded twice during the Maritimes atlas period (Erskine 1992). Probable breeding was reported in two locations: at the extreme northwestern tip of the island, and in the vicinity of the Tyne Valley (A.I. Erskine pers. comm.). Given the infrequent and isolated nature of the records, as well as the scarcity of suitable habitat, breeding on the island appears sporadic at best and requires confirmation.
Although Bicknell's Thrush is recognized by the Canadian Wildlife Service as one of the highest priority species for monitoring, research and conservation (Dunn 1997), and a status of vulnerable has been recommended under Birdlife's Globally Threatened Species Programme using IUCN's "Red List" criterias (BirdLife International 1998), the species is afforded no species-specific protection in Canada. Protection is afforded by the Federal Migratory Birds Convention of 1916, which prohibits the hunting, possession, and/or sale of migratory birds.
In the United States, the species has been ranked the number one conservation priority among neotropical migrants in the Northeast for the past three years (Rosenberg and Wells 1995). This ranking offers no legal protection; however, evidence of a population decline would result in the species being placed "first on the list of neotropical migrant bird species in need of immediate conservation action." At the national level, Bicknell's Thrush is listed as a Category 2 species under the U.S. Federal Endangered Species Act. This offers no legal protection but indicate that formal endangered species listing may be warranted pending further research. According to Atwood et al. (1994), this listing may also alert state and federal agencies to the existence of Bicknell's Thrush and hopefully stimulate further research.
On the wintering grounds, where deforestation and land conversion continue to claim habitats, the species is offered no known form of protection.
Population Sizes and Trends
There is little data regarding the population size and trend of Bicknell's Thrush. Although records date back to 1881, studies have tended to be qualitative rather than
quantitative, and have collected no data by which to assess the magnitude of population changes (Rimmer et al. 1993).
While researchers stress the lack of data with which to arrive at accurate population size estimates, they consistently estimate the North American population at between 5000 to 15 000 pairs. The most modest estimate was provided by Marshall (pers. comm.), who estimated the Canadian population at only 200 pairs. However, data recently collected in Quebec (Ouellet 1993, Rompré et al. 1997, Y. Aubry pers. comm.), New Brunswick (Erskine 1992, S. Holmes pers. comm.), and Nova Scotia (Erskine 1992, D. Busby pers. comm.) indicate that this estimate is clearly too low.
Erskine (unpublished ms) also estimated the total Bicknell's Thrush population to be in the order of 5000 pairs (range of 1000-10 000 pairs), based on extrapolations from breeding bird atlas data and areas of potential habitat in Canada and the United States. He believes that approximately 60% of the population breed in New England and New York, with the remainder breeding in Canada, particularly in southern Quebec, northern New Brunswick, and Cape Breton Island, N.S. (Table 1).
*probably does not breed regularly
Rimmer (pers. comm.) believes that estimates of New England and New York populations have been too cautious, and estimates the U.S. population at 5000 - 8000 pairs. He suggests that the Canadian population makes up about 20 - 30% of the total North American population. McParland (pers. comm.) places the U.S. population at approximately 15000 pairs and, employing what he describes as "extremely crude" estimates of available habitat (areas above 915 m) and a density estimate of 0,5 bird per hectare, suggest that it could be as high as 25000 pairs. However, as mentioned by Atwood et al. (1996), only a fraction of habitat above 915 m is used by the species and, consequently, McParland's estimate is certainly too high.
Based on sightings and surveys conducted during the last decade, there appear to be a number of significant areas for the species in the province. Ouellet (1993, pers. comm.) concluded that the provincial population is largely centered in two areas: the Gaspé Peninsula, and an area north of Quebec City defined by Rivière Batiscan in the west, Rivière Saguenay in the east, and Lac Saint-Jean in the north. Extensive surveys in 1997 and 1998 have pinpointed a number of specific locations supporting Ouellet's observations, including ZEC-des-Martres, located north of Quebec City, and the Parc de conservation de la Gaspésie and the Réserve faunique des Chic-Chocs, on the Gaspé Peninsula. Other areas of concentration include Mont-Mégantic and Mont-Gosford, Estrie region, and possibly Mont-Vallin, north of the Rivière Saguenay (G. Rompré and Y. Aubry, pers. comm.).
Erskine (unpublished ms) estimated the Quebec population at about 99 pairs. However, due to the large degree of uncertainty involved in population extrapolation, he suggested that a truer approximation would be closer to 1000 pairs (Table 1), with confidence limits ranging from a few hundred to a few thousand pairs. This estimate is supported by R. Ouellet (pers. comm.).
Data collected in 1997 and 1998 by a CWS/McGill University research team suggest numbers in the order of a few thousands (G. Rompré and Y. Aubry pers. comm.). They reached an estimate of 750-1250 pairs in the Estrie region (G. Seutin and G. Rompré, pers. comm.), 1000-2000 in the Gaspésie, and 500-2000 on the north shore of the St. Lawrence River (G. Seutin and Y. Aubry,pers. comm.).. Consequently, the population in Quebec may range between 3000 and 5000 pairs.
Breeding densities have been measured in three locations in Quebec. On Mont-Gosford, Rompré and Connolly (pers. comm.) estimated density at approximately 10 pairs/l0 ha. On Mont-Mégantic, densities were estimated at 2.5 pairs/l0 ha (Erskine 1984) and 12-15 pairs/l0 ha (Rompré et al. 1997). Finally, a density of 4.9 males/l0 ha was calculated for an area near the Petit Rivière Cascapedia, on the Gaspé Peninsula (Erskine 1980).
Ouellet (1993, 1995) infered from recent sightings in dense second-growth mixed forests that Bicknell's Thrush might be increasing in numbers in Quebec. However, Seutin (1998) advised caution in assessing population trends based on sightings in regeneration habitat. Reproduction in this habitat has not been confirmed; thus, it may be occupied by non-breeders (e.g., sexually immature individuals or failed breeders), or be population "sinks" maintained by continuous immigration from nearby productive "source" areas. The CWS/McGill University distributional surveys also indicates that the species may be limited in this habitat to forests of a certain successional stage and species composition (Rompré et al. 1997, Seutin 1998). Thus, a large amount of regeneration forest does not directly translate into a large amount of potential Bicknell's Thrush habitat (Seutin 1998; see Habitat section).
Populations were largely concentrated in New Brunswick and Cape Breton Island, where they are listed as scarce but regular at high elevations, mainly above 300 m (Erskine 1992). Using atlas and other data, Erskine (1992) estimates the Maritimes population at less than 1000 pairs, divided amongst New Brunswick (600 pairs), Nova Scotia (300 pairs), and Prince Edward Island (10 pairs). He has since slightly revised this estimate, boosting the N.B. population up to 700 pairs (Erskine unpublished ms). Confidence limits, described as "little more than guesses", put the population at one-third to three times this estimate, or 300 to 3000 pairs.
Recent estimates for the New Brunswick population are quite consistent. Extrapolating from results of a 1995 CFS/CWS distribution survey, as well as further work in 1996-1998, D. Busby (pers. comm.) estimates the population at 400-500 pairs. S. Holmes (pers. comm.) places the population at 600 pairs, based on habitat availability and census work in 1996 and 1997, with a confidence limits of 400-800 pairs.
New Brunswick populations are concentrated in the extreme northwest of the province and the north-central highlands. Populations there are located mainly at high elevations on peaks, and not within the connecting valleys. Populations are also small. Just south of Mt. Carleton Provincial Park, in an area that may be the most densely populated part of the province, five of the most productive peaks are estimated to contain populations of 50-100 pairs each (S. Holmes pers. comm.). Caution must be used in extrapolating these estimates to surrounding habitat, however, as seemingly identical peaks within the vicinity were found to contain much fewer or even no birds.
The Nova Scotian population is centred on Cape Breton Island. Distribution surveys by CWS indicated that populations are concentrated mainly in the northwest, from Cape North to midway along the peninsula. D. Busby (pers. comm.) places the island population at 400-500 pairs, based on suitable habitat, with populations divided amongst the Cape Breton Highlands National Park and immediate area to the north (200 pairs), the harvested forest highlands (100 pairs), Cape North (50 pairs), Scaterie Island (25 pairs), and St. Paul Island (25 pairs).
Although recent observations of Bicknell's Thrush on Prince Edward Island (Erskine 1992) have set a new record for the province, it is probable that this population is insignificant. Erskine (unpublished ms) estimates the total population at less than 5 pairs, and rejects the idea of anything other than sporadic breeding. Not only is breeding unconfirmed, but available habitat is apparently insufficient to support a sustainable population. Nonetheless, it is possible that Bicknell's Thrush has bred sporadically on the island for years.
Northeastern United States
The American Northeast may support the largest proportion of the species' total breeding population. Estimates range from 60% (Erskine unpublished ms) to 70 or 80% (c. Rimmer pers. comm.) of the total population. In total numbers, Erskine (pers. comm.) estimates the U.S. population at 3000 - 6000 pairs. Rimmer's (pers. comm.) estimate falls between 5000 - 10 000 pairs, and McFarland (pers. comm.), while describing his estimate as "embarrassingly crude," suggests that numbers may range from 15 000 – 25 000 pairs.
Based on distribution studies, Bicknell's Thrush populations in New England and New York appear to be declining. Atwood et al.(1994) found the species absent at 10 of 71 (14%) of historic (pre-1992) sites. Marshall (pers. comm.) has also noticed changes in populations along an elevational gradient, with populations receding upwards toward optimal habitat at tree-line. On Camel's Hump, Vermont, populations which he found previously to extend from tree-line to 1036m in 1984 had retracted to 1219 m in 1994. He noted a similar trend on Mt. Mansfield, Vermont, with populations retracting from 884 to 1067 m, again over a period of 10 years. This decline in altitudinal span, if true, was apparently not accompanied by reduced population densities at high elevations. On Mt. Mansfield, census plots monitored since 1992 at 1150 m revealed a stable population (Rimmer and McFarland 1996).
The history of Bicknell's Thrush population on Mont Greylock, Vermont, illustrates the fate of small isolated mountain top populations. The species was first detected on that summit in 1888 by Faxon (1889). Maynard (see Wallace 1939) observed about twelve pairs in 1910. From 1934 to 1960, a small, consistent breeding population was estimated at between 6 to 11 pairs, dwindling gradually to 1 to 2 pairs in 1972, and disappearing the following year (see Veit and Petersen 1993). None have been recorded since. Although the reasons for their disappearance are unknown, increased human disturbance near the summit may have played a role (Veit and Petersen 1993).
Population densities are largely unknown in Vermont, where Bicknell's Thrush is historically considered to be a rare or uncommon summer resident (Kibbe 1985). Bicknell's Thrush is not found along the U.S. Fish and Wildlife Service's Breeding Bird Survey routes, nor is it recorded in Breeding Bird censuses for Vermont published in American Birds. However, using extrapolation Erskine (unpublished ms) estimates the total state population at about 500 pairs. More recently, McFarland (pers. comm.) suggested that it may exceed 1000 pairs.
The highest numbers and densities of Bicknell's Thrush in Vermont occur on Mt. Mansfield, where Torrey (1889) described calls during the day "so persistent as to become almost a nuisance." Recent estimates based on 7 years of fieldwork placed the population at about 250 pairs (Rimmer et al. 1996). Spot-mapping of vocalizing males, corrected for partial territories, suggested population densities varrying annually from 48 - 62 pairs/40 ha (Rimmer et al. 1996). Other peaks in the Northeast appear to have much lower densities: 6 pairs/40 ha on Mt. Equinox and 14.5 pairs/40 ha on Belvidere Mountain, Vermont, and 19.7 pairs/40 ha on Plateau Mountain, N.Y. (McFarland and Rimmer 1996).
Palmer (1949) listed Bicknell's Thrush as a fairly common summer resident in the altitudinous spruce forests of Maine. Rimmer (pers. comm.) estimated the total Maine population at >1,000 pairs, using data from distribution surveys conducted 1993-1995 (Rimmer and McFarland 1996). Marshall (pers. comm.) estimated the population at less than 550 pairs. Highest densities are probably found on Mt. Katahdin, with estimates running as high as 500 pairs (J. Marshall pers. comm.). Sporadic sightings have occurred in coastal Maine, but a recent intensive survey failed to locate any birds there. Thus, Bicknell's Thrush is likely a rare, sporadic and possibly only transient inhabitant of such areas (Rimmer and McFarland 1996).
Due to the high proportion of suitable breeding habitat in New Hampshire and New York, Bicknell's Thrush populations may reach their highest numbers there. According to Erskine (unpublished ms), upwards of two-third of the total Northeast population may occur in these states, with populations in each estimated at 1000 pairs. Estimates by Marshall (pers. comm.) are of about 2,200 pairs for New York State and 350 pairs in New Hampshire. At the liberal end of the range of estimates, McFarland (pers. comm.) suggested that populations within the two states could exceed 15 000 pairs.
In New Hampshire, the breeding population appears to be centered in the White Mountains. Mean population densities for Mt. Moosilauke ranged from 3.2 birds/l0 ha in a subalpine valley to 2.0 birds/l0 ha along the ridge (Sabo and Holmes 1983). On Mt. Osceola, densities were measured at 22 pairs/40 ha (Morse 1979).
Bicknell's Thrush reaches the extreme southern limit of its breeding range in the Catskill Mountains of New York, where it is known as a fairly common breeder (Peterson 1988). It is also described as a fairly common inhabitant of the Adirondack Mountains. Peterson (1988) suggested no changes in distribution or abundance in the state. Accordingly, Atwood et al. (1994) found the species on 33 of 44 peaks (75%) surveyed from 1992-1993, and on 16 of 16 peaks (100%) known to have been occupied prior to 1992.
On its wintering grounds, Bicknell's Thrush appears to concentrate on the island of Hispaniola (Rimmer et al. 1997). Distribution surveys conducted from 1994-1997 detected thrushes at only 7 of 13 (54%) historic sites in the Dominican Republic. Additionally, intensive demographic studies indicated a possible decline in density at two locations. In Pueblo Viejo, the species declined from 77 territories/40 ha in 1995, to 60 territories/40 ha in 1996, and 50 territories/40 ha in 1997; at Loma del Toro, Bicknell's Thrush declined from 40 territories/40 ha in 1995 to 18 territories/40 ha in 1996. While these decreases may be a result of increasing research activity (Rimmer et al. 1997), populations may also be suffering from deforestation. Critical habitat is said to be disappearing "at an alarming rate throughout the island [of Hispaniola]," including within national parks and reserves (Rimmer et al. 1997).
In the United States, where most Bicknell's Thrush research has been conducted until the 1990s, the species is a strict habitat specialist. Described as a "mist-and-height-Ioving" bird (Gillet 1935) of the "tree-line" ecotone (Petersen 1990), Bicknell's Thrush populations are confined mainly to altitudes of 914 m to the tree line on rocky peaks (Fig. 7), with some scattered pairs breeding down to 762 m (Wallace 1939, Rimmer et al. 1993). Atwood et al.(1994) detected Bicknell's Thrush on 79% of all high-elevation peaks (≥914 m) surveyed in New England and New York, on only 11 % of all mid-elevation peaks (762-914 m), and at no low-elevation (762 m) areas in northern Maine.
The species also occurs predominantly in traditional subalpine habitat in Quebec (Ouellet 1995, Rompré et al. 1997), in remnant patches of subalpine habitat in New Brunswick, and in structurally similar habitat in Cape Breton (Holmes and Nixon 1997, Nixon 1996). In Quebec, a 1997 distribution survey located no birds at elevations below 620 m, even though censuses began at 315 m (Rompré et al. 1997). The species' preference for high elevations was also demonstrated during the CWS/CFS surveys in New Brunswick and Cape Breton (Figure 5). The species is only occasionally found at elevations as low as 200 m in Quebec (Ouellet 1995), and 300 m in New Brunswick and Cape Breton (Erskine 1992).
The subalpine forests favoured by Bicknell's Thrush are characterized by a wet, cool, windy climate that increases in severity with elevation (Reiners and Lang 1979, Sabo and Holmes 1983). In Quebec, Rompré et al. (1997) found mistnetting of the species to be difficult due to frequent cold weather, strong winds and the presence of snow in late-June. On Mt. Mansfield, Vermont, average monthly temperatures in the summer barely exceeded 10°C (Wallace 1939). Sabo and Holmes (1983) found the average June temperature at 600 m versus 1000 m in New Hampshire to drop from 16°C to 12°C, the annual precipitation to increase from 130 mm to 150 mm, and the mean above-canopy wind speed to increase from 12 km/hr to 30 km/h.
Under these climatic conditions, the habitat is very dense and twig-rich, dominated by small, stunted, even-aged balsam fir and, to a lesser degree, spruce (Wallace 1939, Palmer 1949, Able and Noon 1976, Sabo 1980), often underlain by a dense flush of young (mainly fir) growth (McFarland pers. comm.). According to Wallace (1939), these conifer stands become increasingly dense and stunted as altitude progresses, until "at the tree limit they are so closely matted together that they are often virtually impenetrable. Indeed, their branches may be so closely interlaced that it is sometimes possible to walk over the tops of them without sinking through."
Mortality is high in these stands, with many of the conifers dead or dying (Wallace 1939). There is often an additional number of taller snags scattered throughout (C. Rimmer pers. comm.).
Bicknell's Thrush habitat in three areas in Quebec had a balsam fir composition (as measured by numbers of tree stems) of 88.5%, 75.1 % and 71.1 % (Rompré et al. 1997). White birch (Betula papyrifera) was also a regular component, interspersed with smaller numbers of mountain ash (Sorbus sp.), pin cherry (Prunus pensylvanica), mountain maple (Acer spicatum), and other hardwood species. Extensive mats of Sphagnum, Pleurozium schreberi, Cladina sp., Polytrichum sp. and other mosses and lichens are found beneath the trees (Wallace 1939, Dilger 1956, Sabo 1980, Nixon 1996, Rompré et al. 1997).
Average canopy height of Bicknell's Thrush breeding habitat tends to range from 3-7 m in New Brunswick (Nixon 1996) and Cape Breton (D. Busby unpubl. data), with mean heights in Quebec of 5.4 m (Parc de la Gaspésie), 7.5 m (ZEC des Martres) and 14.1 m (Mont-Mégantic; Rompré et al. 1997). Sabo (1980) calculated a mean canopy height of 4.8 m in Bicknell's Thrush habitat in New Hampshire. Stands also tend to be dominated by small-diameter trees. In Quebec, small diameter size trees (.5 cm; measured 20 cm above ground level) make up 89.3%, 85.3%, 55.9% of stems on the previously mentioned sites. In New Hampshire, mean diameter at breast height (dbh) was 7 cm (Sabo 1980).
Figure 5. Occurrence of Bicknell’s Thrush in New Brunswickand Cape Breton, N.S., by elevation (m), based on surveys conducted from 1995-1998 by the Canadian Wildlife Service and the Canadian Forest Service. (Courtesy of D. Busby, CWS – Atlantic Region.)
As stand characteristics within the subalpine zone are in part determined by cold and often severe weather patterns, similar habitat (extremely dense stands of stunted, windswept fir and spruce) also occurs at some coastal and island locations. Historically, the species was known to breed in such coastal habitat at a number of locations in Canada (see Distribution section), but it is now limited to Cape Breton lowlands and peripheral small islands.
Recent observations in southern Quebec (Ouellet 1993, 1995), northern New Brunswick (Erskine 1992, Nixon 1996, Holmes and Nixon 1997) and Cape Breton (D. Busby peTS. comm.) indicate that in Canada, Bicknell's Thrush also frequents second-growth or regeneration forests that follow large, stand-level disturbances such as clearcutting or fire. This habitat is referred to as "non-typical" or "non-traditional" for two reasons. First, the species is virtually absent from this habitat in the U.S., where the vast majority of research on Bicknell's Thrush has been conducted until the 1990s. Secondly, the species' presence in this habitat has only been recently detected. In fact, it is possible that the species has been present in such habitat for years, but was overlooked (Ouellet 1993).
Bicknell's Thrushes are most commonly found in regeneration forests in New Brunswick. Findings from a 1995 distribution survey in the northcentral and northwestern parts of the province showed that 91 % of all detected birds were located in regeneration, and only 9% in "typical" habitat (Nixon 1996). More than 84% of all birds located during the survey were found on sites estimated to have been clearcut within the last five to 20 years (Nixon 1996). On Cape Breton, Nova Scotia, the species is still found primarily in "typical" habitat, but approximately 25% of the population is estimated to occur in regeneration (D. Busby unpubl. data). In Quebec, Ouellet (1993, 1995) suggested that a major shift in habitat preference has occurred (Table 2). However, Rompré et al. (1997) and Seutin (1998) contended, based on intensive surveys in 1997 and 1998, that alternative habitat is used in smaller proportion in Quebec than elsewhere in eastern Canada.
As in "traditional" habitat, the species is generally present in alternative stands at high elevations. In New Brunswick and Cape Breton, most birds detected during the atlas period were found above 300 m (Erskine 1992). No birds detected during distribution and habitat surveys in 1995-1997 were found below 426 m (Nixon 1996, Holmes and Nixon 1997, unpubl. data) and the species was found to increase steadily in density with elevations (Holmes and Nixon unpubl. data).
There appears to be some variability in the tree species composition of regeneration forests in which Bicknell's Thrush is found. In Quebec and Cape Breton, Bicknell's Thrush has been located in regeneration dominated by balsam fir, with some spruce. Hardwood species are also present (Prunus sp., Betula sp., Amelanchier sp., Acer spicatum, Populus sp., Sorbus sp.), but in small quantities (Ouellet 1993, Rompré et al. 1997, D. Busby pers. comm.). In New Brunswick, stem counts on Bicknell's Thrush territories indicate that white birch is the dominant species. In 1996 and 1997, white birch comprised 48 and 44% of all stems, followed by balsam fir (30 and 25%) and pin cherry (15 and 22%; Holmes and Nixon 1997, unpubl. data).
Generally, Bicknell's Thrush tends to be located in regeneration of low average canopy heights. In Quebec, Ouellet (1995) found that occupied stands were generally less than 10 m in height, results consistent with stand heights measured in Cape Breton (3-7 m; D. Busby pers. comm.) and New Brunswick (2.1-5.9 m; Holmes and Nixon 1997). Of 38 sites sampled by Holmes and Nixon (1997), average mean canopy height for 34 sites (89%) was 4.3 m or less. As well, used regenaration stands are generally dense (Erskine 1992) and closely spaced, with average dbhs ranging from 4-15 cm (Ouellet 1993). Small-stemmed trees (diameter of less than 2.5 cm) made up 74% of all stems on Bicknell's Thrush territories in New Brunswick (Holmes and Nixon unpubl. data).
A high degree of vertical ground cover has also been noted in regenerating stands. In northern New Brunswick, this cover was provided by closely spaced, low-branching conifers, mat-like, shrubby growths of mountain maple, dogwood (Comus sp.), pin cherry and elderberry, and ground plants such as raspberry (Rubus sp.), bracken and wood fern (Pteridium acquilinum, Dryopteris sp.), sarsaparilla (Aralia nudicaulis), and high grasses (Nixon 1996). Other common plant species, many of which have been previously noted under "typical" habitat, include Canada mayflower, boreal lily, bunchberry, Lycopodium sp., blueberry and snowberry (Nixon 1996, D. Busby pers. comm.).
Recent research in Quebec clearly indicated that not all coniferous or mixed regeneration forests in altitude are suitable for Bicknell's Thrush (Rompré et al. 1997, Seutin 1998). The precise characteristics of occupied stands are still being researched, but necessary factors likely include a very high tree density within a few meters from the ground (Seutin and Connolly, unpubl. data). Stand height (Holmes and Nixon 1997, Seutin 1998) and tree species composition (Seutin 1998) may also be important factors.
Observations of Bicknell's Thrush in regenerating forests in Quebec and the Maritimes raises the possibility that the species is less likely to be limited by habitat in Canada than in the United States. Currently, mixed second-growth forests are abundant in Quebec and in northern New Brunswick (ENixon peTs. obs.). For a number of reasons, however, caution must be exercised in assuming that abundant second-growth habitat can replace traditional breeding habitat. Firstly, while there is no doubt that Bicknell's Thrush occurs in regeneration habitats during the breeding season, there is as yet no proof that they successfully breed there; this habitat may actually represent a sink for Bicknell's Thrush. Secondly, there is no indication of a "shift" to second-growth habitat in the U.S. Northeast. It is possible that extensive logging in eastern Canada, coupled with outbreaks of spruce budworm, may have seriously reduced the extent of "typical" habitat available to the species in Canada, forcing it to use lower quality habitat. Thirdly, the species appears to select for specific age and structural characteristics within this habitat (Holmes and Nixon 1997, Seutin 1998), so only a portion of available regeneration is likely suitable at any given time. Also, because of the transitional nature of these forests, the quantity of suitable habitat will vary over space and time, depending on harvest schedules and silvicultural practices.
In southern Quebec, large area of apparently high-quality habitat is found along the Gaspé Peninsula. High-elevation peaks occur throughout much of the peninsular area; they are especially concentrated in the northeast, diminishing to smaller, more isolated areas west of Rimouski. Suitable habitat is also found along the north shore of the St. Lawrence River. A band of high-elevation areas extends along the shore from Quebec City in the west to Godbout in the east, with scattered areas occurring in the interior. Extensive surveys in 1997 and 1998 revealed however that only a fraction of this habitat is currently used (Rompré et al. 1997, pers. comm.). Small patches of habitat are also found in extreme southern Quebec in the Estrie region. They are found on the few summits that reach more than 900m (i.e., Mont Mégantic, Mont Gosford and Montagne de Marbre).
In the Maritime provinces, the largest, most continuous highlands are found in the northcentral and northwestern regions of New Brunswick, where elevations range mainly from 200-500 m (Figure 6). This is where the majority of Bicknell's Thrush detections have been reported. A much smaller highland area (mainly 200-500 m) occurs in the southeast comer of the province (Albert County Highlands), although there is no evidence of a breeding population (A.I. Erskine pers. comm.).
High-elevation sites are limited on the Nova Scotian mainland. Few peaks above 300 m are found in the north (A.I. Erskine pers. comm.) and a small highland area is found in the southwest. However, there is no breeding evidence of Bicknell's Thrush in either of these areas. Bicknell's Thrush is found on Cape Breton Island, however, where a large block of suitable, high-elevation habitat extends along the northern peninsula.
In Prince Edward Island, some suitable habitat is found along the northern tip, in the form of wind-exposed, stunted conifer forests. Much of the island, however, is of a relatively low elevation, with habitat unsuitable to Bicknell's Thrush.
As in Quebec, there are a number of areas in the Maritimes containing apparently suitable habitat which are currently uninhabited by Bicknell's Thrush (e.g., Albert County Highlands in N.B, southwest N.S., P.E.I.). It is possible that habitat size (A.I. Erskine pers. comm.) or degree of isolation from larger areas of suitable habitat (McFarland pers. comm.), not elevation, are the limiting factors. Habitat patches may simply be too isolated or too small to support other than sporadic breeding. Although there is no data on the impact of habitat fragmentation or isolation on the species, Atwood et al. (1994) found that Bicknell's Thrush occurred less frequently on high-elevation (≥ 914 m) New England and New York mountain masses with < 5 km2 of estimated land area than they did on larger mountains with similar habitat. Despite these findings, however, the species was still detected on 83% (n=76) of the surveyed mountains with < 5 km2 of land area at ³914 m elevation.
Figure 6. Elevation contour map of the Maritime provinces(from Erskine 1992).
Figure 7. Approximate distribution of Bicknell’s Thrush habitat in the United States(from Atwood et al. 1994).
Northeastern United States
In Maine, suitable habitat occurs on a number of scattered, isolated mountain peaks, found mainly in the southwest part of the state close to the New Hampshire border (Allen 1898, Wallace 1939) and continuing northward toward Baxter State Park (Atwood et al. 1994). There are also some coastal areas where harsh weather conditions have produced forests of a structure similar to that found at high elevations (Atwood et al. 1994). Scattered peaks are also present in northern New Hampshire, with most suitable habitat concentrated in the White Mountains in the centre of the state (Faxon and Allen 1888, Wallace 1939, Atwood et al. 1994). In Vermont, breeding habitat is restricted to the Green Mountains (Davenport 1903, Wallace 1939), a narrow band of peaks running down the centre of the state from Canada to Massachusetts, and to the Taconic Mountains in the southwest (Atwood et al. 1994). In New York, good breeding habitat is found in the Adirondack Mountains in the north and the Catskill Mountains in the south (Wallace 1939, Palmer 1949, Peterson 1988, Marshall and Clapp unpublished ms). Suitable habitat appears to be extremely isolated, and is relegated to a single peak in the northwest corner of the Massachusetts (Wallace 1939, Atwood et al. 1994).
Trend in Quantity and Quality of Habitat
In Canada, Bicknell's Thrush habitat has been substantially altered in Quebec and the Maritimes over the last century by major infestations of spruce budworm (Choristoneura fumiferana). Infestations have also occurred in the eastern U.S., mostly in Maine, although to a lesser degree (Hardy et al. 1986). Spruce budworm outbreaks occur cyclically in spruce-fir forests, where budworm feed voraciously on balsam fir and white, red, and black spruce (Kettela 1983). Damage to traditional breeding habitat was especially severe during the last infestation. During its peak year (1975), the outbreak covered between 57 million ha (Kettela 1983) and 72 million ha (Hardy et al. 1986) of forested land in eastern Canada and the U.S. (Figure 8). By the following year, significant volumes of dead and dying trees covered 3 million ha of land (Kettela 1983). Outbreaks and spruce-fir mortality were especially severe on Cape Breton Island, an area of prime breeding habitat for Bicknell's Thrush, and in Quebec, including the Gaspé Peninsula, where, by 1981, 10.2 million ha of land contained significant volumes of dead and dying trees.
Budworm populations were at their peak in 1975, and have since declined. However, budworm is still considered to be the most devastating forest insect pest in North America whether measured by tree mortality, hectares infested, or frequency and duration of outbreaks (Hardy et al. 1986). It has also been noted that budworm infestations have increased in frequency, duration and severity since the start of the
Figure 8. Areas of defoliation by the spruce budworm in eastern North America, 1975 (from Hardy et al. 1986).
20th century (Bailey 1924, Blais 1983, Hardy et al. 1986), with the last infestation also the most wide-spread (Kettela 1983). Nevertheless, Erskine (1992) concluded that Bicknell's Thrush populations in the Maritimes have not been obviously affected by spruce budworm damage to forests. In Quebec, however, R. Ouellet (pers. comm.) links the presence of Bicknell's Thrush in young, second-growth forests to previous budworm damage: the destruction of large expanses of spruce-fir habitat during the latest budworm infestation may have forced the species into young, regenerating forests. While the impacts of logging on Bicknell's Thrush are unknown at this time, large-scale forestry operations are occurring throughout important breeding areas, including the Gaspé Peninsula and in areas north of the St. Lawrence, throughout the northcentral and northwestern highlands of New Brunswick, and in Cape Breton. In New Brunswick, the distribution of high elevation habitat is virtually contained within timber license land. Clear cutting is in evidence throughout, with some peaks virtually denuded (E. Nixon pers. obs.).
The presence of Bicknell's Thrush in regeneration stands may be an indication of a greater adaptability or wider habitat niche than previously believed, or of a switching of habitat (Ouellet 1993). However, based on current information, it appears that Bicknell's Thrush may be selecting for very specific habitat features within managed forest. In Quebec, a 1997 survey indicated that birds had vacated sites which had been recently thinned during logging operations (Rompre et al. 1997, Seutin 1998). As thinning operations are a common forestry practice throughout eastern Canada, such operations undoutably reduce the supply of suitable habitat for the species.
Another potential threat to Bicknell's Thrush is the fact that the distribution of suitable regenerating habitat will certainly vary over space and time. In New Brunswick, where the species is found almost exclusively in regeneration stands, B. Donovan, a forestry management planner with a pulp and paper company operating in the highlands, predicts that the uneven age distribution of industrial forests will lead to a "crash" in young dense regeneration within the next few decades. This has already happened in parts of Quebec (e.g., sectors of Parc des Laurentides). Thus, while young regeneration is currently plentiful, its transitional nature may lead to the need for specific forest management practices in the future if there is to be sufficient habitat for Bicknell's Thrush, particularly in New Brunswick.
In the United States, where habitat is restricted to higher elevations, logging has not contributed significantly to habitat degradation. Due to lack of accessibility, it is unlikely that forestry operations will ever prove feasible in these areas. However, these high elevation habitats face different threats. In New England (except, perhaps, in Maine) and New York, a steady decline in habitat quality may have been occurring over the last 30-40 years (Marshall in Erskine unpublished ms). Research suggests that forest decline is due largely to the direct and indirect effects of atmospheric deposition (Vogelmann 1982, Rimmer and McFarland 1994). While the effects of atmospheric deposition, particularly in the form of acid precipitation, are most evident in Bicknell's Thrush habitat in the northeastern U.S., its impacts may potentially be felt throughout the entire breeding range of Bicknell's Thrush (Vogelmann 1982).
High-elevation forests (>800 m) are particularly vulnerable to the effects of atmospheric deposition due to the high levels of acidity transmitted through cloud moisture, the major source of precipitation within these areas (Vogelmann et al. 1968). These forests also receive higher levels of rain and fog than low-lying areas. In Vermont's Green Mountains, forests at 1067-1372m receive twice the annual precipitation of a low-elevation (122m) forest only 50 km away, more than doubling the acid load (Vogelmann 1982). Rain in the northeastern states is now calculated to be at least 30-40 times more acidic than that of pre-industrial times (Vogelmann 1982). Montane areas also receive more frequent exposure to other air pollutants, including lead, zinc, copper, vanadium and cadmium, all heavy metals known to be toxic to plants.
Severe declines in red spruce, an integral component of the species' breeding habitat in the Northeastern North America., have been documented since the 1960s, with dead and crown-damaged trees common in the Green Mountains of Vermont, the White Mountains of New Hampshire, New York's Adirondack Mountains, Quebec's Laurentians, and the Appalachians as far south as West Virginia (Siccama 1974, Siccama et al. 1982, Vogelmann 1982, Foster and Reiners 1983, Battles et al. 1992). Increasing mortality rates are evident in individuals of all size classes, and have been exacerbated by a lack of reproduction (Siccama et at. 1982). On one peak in the Green Mountains, a mortality rate of nearly 50% was recorded among spruces between 1965-1982 (Vogelmann 1982), with mortality especially severe among high-elevation spruces on windward slopes. Seedling production also declined by 50%, with little spruce regeneration evident. In New Hampshire's White Mountains, a forest sampled near the Hubbard Brook Experimental Forests showed a decline in saplings from 117 stems/ha in 1965 to 33 stems/ha in 1977 (Siccama et al. 1982). Larger trees were also affected, declining from 45 to 40 stems/ha.
High mortality rates are also evident in balsam fir (Miller-Weeks and Smoronk 1993), another essential component of Bicknell's Thrush breeding habitat. On Camel's Hump in Vermont, Siccama et al. (1982) found that balsam fir, the dominant species there, was not filling in the "growing space" released by the death of the spruces. Not only were relatively high numbers of fir dead or dying, but there was also an over-all decrease in growth.
The composition and structure of subalpine forests may be further altered by global warming, which is predicted to have a particularly strong impact upon ecosystems of cooler climates or high altitudes. In temperate breeding areas, predicted temperature increases range from 3°C in southern North America up to 9°C in the northern latitudes (Davis and Botkin 1985). Although global warming is an as-yet-unrealized threat, even minor changes in temperature and soil moisture could potentially alter montane forests dependent upon a cooler climate. Using computer simulations, a mean annual temperature increase of as little as 2°C could result in red spruce being replaced by deciduous tree species (Davis and Botkin 1985).
Tree-line communities may be particularly vulnerable to climactic change. Unlike other communities, which are predicted to experience a delayed response to changes, tree-line habitats are expected to undergo an extremely rapid response (Davis and Botkin 1985). Historically, a warming trend of only 0.6°C in 1880-1950 resulted in an upward movement in the alpine treeline in Lapland, and an expansion in the distribution of broad-leaved trees in southern Finland (Erkamo 1952, Davis and Botkin 1985). In the mid- to high-latitudes, warmer temperatures of 23°C during the last interglacial (about 120,000-80,000 years ago) resulted in broad expanses of deciduous forests extending north into areas currently occupied by mixed conifer-deciduous forests (Hohn 1980, Warrick et al. 1986). During a period of lesser warming (about 1°C, 1,200800 years ago), Canada's boreal forests extended well north of the current timberline (Lamb 1977).
Additional threats to habitat include intensive use by outdoor recreationalists such as hikers and bikers, ski resort developments, and the construction of transmission and microwave towers (Rimmer et al. 1993). Wind-power development may become a problem, if use of this technology becomes more widespread in the future (C. Rimmer pers. comm.).
Deforestation and land conversion continue to rapidly reduce the amount of broad-leaf, montane forests available to Bicknell's Thrush in the Dominican Republic and Haiti (Rimmer et al. 1998). Deforestation in Haiti is calculated to occur at an annual rate of 3.7%, with only 2-5% of the landscape still forested (Rappole 1995). In the Dominican Republic, 1980s forest inventories indicate that 14% of the native broad-leafed forest remained (Ottenwalder 1989, Anonymous 1995), with a deforestation rate of 0.6% (Rappole 1995).
Highlands are particularly vulnerable to conversion for agricultural use (Terborgh 1977, 1980). Terborgh (1980) reported that elevations between 1O00-2000 m in the Dominican Republic are particularly denuded. This is a concern for Bicknell's Thrush which was found at elevations ranging from 1676 m to 1829 m during recent field investigations in the Dominican Republic (Rimmer and McFarland 1995).
The larger availability of breeding versus wintering habitat means that many neotropical migrants occur in higher densities on the wintering grounds (Terborgh 1980). This concentration can severely amplify the effects of habitat loss, with Terborgh (1980) suggesting that the conversion of Iha of forest in the Neotropics may equal the loss of 5-8ha on the breeding grounds. He concluded that species such as Bicknell's Thrush, which concentrate in, or use primary forests in the Neotropical highlands, will be the first to experience heavy pressure due to habitat loss, and suggested the serious prospect of insufficient habitat for many neotropical migrants by the end of the century. These conclusions were supported by Rimmer et al. (1998). They predicted the loss of the eastern section of forest within the Dominican Republic within 2-3 years, and suggested that increasing fragmentation may affect the ability of forest fragments to support forest songbirds even sooner.
In Canada, small portions of the confirmed and potential breeding range of Bicknell's Thrush fall within the boundaries of provincial and federal parks, including the Cape Breton Highlands National Park on Cape Breton Island, Mount Carleton Provincial Park and Fundy National Park in New Brunswick, Forillon National Park, Murdochville Réserve, Parc de conservation de la Gaspésie, and Reserve Matane on the Gaspé Peninsula, and, in southern Quebec, a number of scattered parks and reserves extending from Mont Mégantic in Estrie to Parc des Grands-lardins in Charlevoix. Logging and/or mining activities are permitted in some of these parks and/or reserves, however.
In New England and New York, Bicknell's Thrush breeding habitat falls mainly within national and state forests, including Maine's Baxter State Park, Vermont's Green Mountain National Forest, New Hampshire's White Mountain National Forest (which also extends across the border into Maine), and New York's Catskill Forest Park. Additionally, in Vermont, highaltitude areas (> 762 m) are somehow protected under Act 250, a state land-use law requiring development plans on lands above 762m to undergo a rigorous review process (Kibbe 1985, C. Rimmer pers. comm.). In New York, the Adirondacks are located on state-protected land, and camping throughout the state is prohibited on peaks above 1219m (Peterson 1988).
While 10% of the Dominican Republic and 0.3% of the Haitian landscape are under some form of conservation protection (Rappole 1995, SEA-DED 1990), activities such as timber extraction and agricultural clearing have continued largely unabated. In the national park of Sierra de Neiba, 1995 observations estimated that deforestation had occurred on approximately 30-40% of the land; by 1997, two years after the park boundaries had been established, this had increased to 70-80% of the land (Rimmer et al. 1998). Park areas below 1600 m were found to be completely deforested. (c. Rimmerpers. comm.).
Taxonomy and Species Description
Full species status to Bicknell's Thrush was not conferred until 1995 (AOU 1995). Prior to this, it was listed as a southern subspecies of the more widely distributed Gray-cheeked Thrush, Catharus minimus (Wallace 1939, 1949, AOU 1957).
Bicknell's Thrush was first sighted in 1881 in the Catskill Mountains of New York (Bicknell 1882, Ridgway 1882). It was considered one of three populations of Gray-cheeked Thrush recognized on the basis of phenotysic differences and geographic disjunction: Catharus minimus aliciae, distributed throughout the boreal forest, from northeastern Siberia to Labrador; C. m. minimus, found only in Newfoundland; and C. m. bicknelli (Ridgway 1882, Wallace 1939, 1949, Godfrey 1986, Ouellet 1993).
Recently, however, strong evidence was put forward to support the recognition of Bicknell's Thrush as a distinct species (Seutin 1991, Ouellet 1993). It is genetically highly divergent from C. minimus, the two forms occupy completely different breeding and wintering ranges and they have different songs. Bicknell's Thrush is unresponsive to playbacks of both minimus or aliciae songs played on its territories during the breeding season, indicating reproductive isolation (Ouellet 1991, 1993). Accordingly, there are no known occurrences of intergradation or hybridization between bicknelli, aliciae or minimus (Ouellet 1993).
Bicknell's Thrush is smaller in size than C. minimus, with the mean wing length a full 10 mm shorter (Wallace 1939, Ouellet 1993). There is an overlap in wing size between the largest bicknelli male and the smallest minimus female which, due to a lack of intersexual plumage differences, can occasionally hinder identification (Wallace 1939). The over-all coloration of Bicknell's Thrush is generally browner in tone than that of aliciae and minimus (Ouellet 1993), although bicknelli is characterized by a bi-phasal colour condition dictated geographically. In Quebec and the Maritimes, the species is distinguished by a greyish or olive phase, and by a brownish colour phase in southern New England and New York. Taking into account this colour variation, Bicknell's Thrush is described as having a buffy throat, upperparts of a rich brown colour, a contrasting, uniquely pigmented, chestnut-coloured tail, and a slender bill with a yellow mandible (Ridgway 1882, Wallace 1939, Ouellet 1993).
Nesting biology information on this species is scarce and, where available, based largely on small sample sizes. Tufts (see Thayer 1907), describes the nest as "the most difficult to find of any small bird occurring in equal numbers on a given area. Rimmer and McFarland (1996) were only able to locate a total of 14 active or recently active nests during systematic nest searches from 1992-1995. Wallace's (1939) study, the most substantial study to date on Bicknell's Thrush nesting biology, was based on only 13 nests.
Bicknell's Thrush is an above-ground nester, building bulky, well-constructed nests in small or medium-sized spruce and fir (Thayer 1907, Wallace 1939, 1949, McFarland and Rimmer 1996). Nests have been found less commonly in alder and birch (Wallace 1949, Maritimes Nest Record Survey (MNRS) cards) and, rarely, in striped maple (Acer pensylvanicum; Aubry pers. comm.). Nest trees are small, ranging from 1.8-7.2 m in height, with dbhs of 1-19.1 cm (n=18; McFarland and Rimmer 1996). Nests are built relatively close to the ground (1-4.3 m above ground; McFarland and Rimmer 1996), although Tufts (see Macoun and Macoun 1909) reported finding two nests in Nova Scotia at heights of 4.5 m and 7.6 m above ground.
Nest construction begins shortly after the birds arrive on the breeding grounds, commencing in the first week of June and occurring throughout the first half of the month. Most first-attempt nests appear to be complete by mid-June. In Vermont, birds were observed constructing nest foundations as early as June 4 and June 6; and work on one partially constructed nest was assumed to have commenced even earlier. Completed nests have been noted as early as June 8.
Egg-laying dates observed by Wallace (1939) in Vermont ranged from June 9-25. In New Brunswick, nests with eggs have been found mainly between June 14-29 (MNRS cards). On Seal Island, nests with eggs were observed on June 3 (see Wallace 1939) and June 13-14 (see Tufts 1986). Other egg dates include June 12-14 in Nova Scotia (MNRS cards), June 8 to July 21 in Quebec (Y. Aubry pers. comm.) and June 18 in Massachusetts (see Veit and Petersen 1993). Nests with young have been recorded as early as June 19 and June 23 (Tufts 1986). There is no confirmed evidence of double brooding, but Wallace (1949) suggests that eggs laid after mid-June may be due to previous, failed nesting attempts.
Clutches range from 3-4 greenish-blue eggs, lightly spotted with brown. Eggs are incubated by the female, with incubation commencing with the laying of the third egg. (Wallace 1939).
Brooding may be carried out solely by the female and, perhaps because of the extreme climate in which the species nests, occurs more frequently and intensely than in other passerines (Wallace 1939). During the first six to seven days following hatch, from one-third to more than one-half of all daylight hours are spent brooding the young (Wallace 1939).
Nesting success in general appears to be quite low. Nine of 13 nests monitored by Wallace (1939) suffered total mortality. Fledging was only partially successful in two and completely successful in the remaining two. A total of 11 young fledged from 45 eggs, for a nest success rate of less than 25%. Egg and chick mortality was attributed to predation (7 nests), desertion (1 nest), and disease (1 nest). McFarland and Rimmer (1996) also found nesting. success to be low. Of 21 located nests, only 9 were successful. The remainder were depredated (4), abandoned (3), failed due to severe weather (2) or unknown circumstances (1), or were never occupied during observation and so were of unknown status (2). A total of 8 of 18 active nests fledged young, or 44.4%. The Mayfield predicted nest success was calculated at 26%.
Both parents feed the chicks, whose diet consists largely of Lepidoptera larvae, ants, wasps and other Hymenoptera, beetles, grasshopper nymphs, flies, millers and moths. The summer diet of both chicks and adults consists almost entirely of animal matter, but they become partly frugivorous in the late summer and fall, supplementing their diet with berries such as cornels, deadly nightshade, barberry, spice-bush, wild grape, woodbine, and poison ivy (Brewster 1906, Wallace 1949).
Fledging generally occurs at 10 to 13 days, with an average of 11 days (Wallace 1939; Ouellet 1995). Chicks are able to fly at 11 to 13 days-of-age (Wallace 1949), but remain with, and are fed by, their parents for a number of days following fledging.
Preliminary survivorship studies by Rimmer and McFarland (1995) report a combined female/male average adult survival of 36.3-53.6% (this calculation does not include juvenile mortality).
Bicknell's Thrush arrives relatively late on its spring breeding grounds. In Canada, arrival dates occur generally at the end of May and the beginning of June. In southern Ontario, where Bicknell's Thrush is listed as a rare migrant (James 1991), it has been observed May 24, 1892, June 2, 1892, and June 4, 1927 (see Wallace 1939). In Charlevoix, Quebec, the species can be mistnetted in the last week of May and the first week of June (R. Ouellet, pers. comm.). In northern N.H., the species has been spotted at the top of mountain peaks on May 30 and June 1 (Nixon 1996), and it is likely some birds arrive earlier.
In the U.S., Bicknell's Thrush reaches the southern states from its wintering grounds in early May, with the earliest continental records being May 2 for South Carolina and May 3 for the northern coast of Florida (Wallace 1949). The species continues northward up the coast, reaching Washington, D.C. in about mid-May, and the southern New England states in the last two weeks of May. Of the specimens examined by Wallace (1939, 1949), 18 were collected between May 20-30, 5 prior to May 20 (one as early as May 15), and one as late as June 11.
Returning birds have been recorded on Mt. Moosilauke in New Hampshire from May 25-30 (Faxon in Allen 1902) and, historically, on Mt. Greylock, Massachusetts, from May 28-30 (Wallace 1949). Arrival dates, particularly on high peaks in the U.S., may be affected by slowly retreating snows. During a late spring on Mt. Mansfield in 1935, few thrushes were present before May 27. Their numbers peaked, however, by May 31 - their arrival coinciding with a spring thaw. In Vermont, Bicknell's Thrushes are described as one of the last of the thrushes to return, with migrants found at lower elevations well into early June (Kibbe 1985). Birds have been mistnetted on peaks as early as May 16 (K. McFarland pers. comm.) and May 17 (c. Rimmer pers. comm.), however, when patches of snow were still present on one peak.
Preparations for fall migration begin in early September, with the birds becoming more active, giving loud and frequent calls, singing, and engaging in flight songs (Wallace 1939, J. Marshall pers. comm., C. Rimmer pers. com.). Marshall (pers. comm.) heard them singing at Percé, Quebec from Sept. 17-19, apparently giving no indication of migratory movement. Along the north shore, R. Ouellet (pers. comm.) says that Bicknell's Thrush generally begin to pass through during the third week of September, with movement continuing into the second week of October. There are few migratory records in New Brunswick, although one migrant was recorded Sept. 27, 1978 (Christie 1980).
In the northeastern U.S., Wallace (1939) found Bicknell's Thrush to be common on Mt. Mansfield, Vermont, on September 10, with few to none having begun migration. They were still reported on their breeding grounds in New Hampshire on Sept. 15 (Allen 1902), and were heard calling on Mt. Abraham, Maine on Sept. 20 (Sweet 1906). Marshall (pers. comm.) found thrushes singing and still on territory on Mt. Katahdin, Maine as late as Sept. 1, 1994. In Massachusetts, the earliest migration record is Sept. 18 (Wallace 1939). According to records, migratory movement is most concentrated at the very end of September and the beginning of October (see Wallace 1939). The species is listed as a very uncommon migrant in Massachusetts (Veit and Petersen 1993), with most migrants seen from late May to early June, and from late September to mid-October.
Although Bent remarked on the fearless, "quite friendly" behaviour of Bicknell's Thrush and Torrey found them so abundant and noisy as to be "almost a nuisance" (see Wallace 1939), this is not the general concensus. Wallace (1939) describes the female as tame and fearless at the nest, but extremely elusive during the nesting period. "Ordinarily, only fleeting glimpses are caught of a brownish-olive form slipping quickly from view into concealment" (Wallace 1949). He describes both the male and female as generally "wary," "easily alarmed," and "suspicious." Brewster (see Wallace 1939) calls Bicknell's Thrush the shyest of the smaller birds during the breeding season, stating that "one may spend hours within their chosen haunts without getting a fair view of a single individual." And, while Howell "heard them all about" on Mt. Mansfield, "so shy were they" that he observed them less than a dozen times (see Wallace 1939). Wallace (1939) attributes the paucity of data on Bicknell's Thrush mainly to its behaviour, writing that "its shyness and seclusiveness, its habit of breeding in only the most inaccessible places, and its almost unbroken silence during most of the year have kept the taxonomic, distributional, and life-history facts concerning it in mystery so long that it has been correctly regarded as one of the least known of American passerine birds."
The most intense period of singing in the northeastern U.S. occurs from late May to mid June, during the peak of courtship and mating activities and prior to incubation (e.g., Rimmer et al. 1996, Rompré et al. 1997). Courtship and mating activities in eastern Canada likely commence a few days later. During this time, singing is heard sporadically throughout the day, sometimes beginning in the pre-dawn hours (before 3 am on Mt. Mansfield, and 3:30 to 4 am in Quebec and eastern Canada). It continues for about a half hour or more following sunset, terminating with darkness.
The species' daily singing cycle is extremely variable, and has played a large role in the paucity of population data regarding Bicknell's Thrush. Although peak singing periods generally occur at dawn and dusk, with the latter time often the more intense, R. Ouellet (pers. comm.) has found singing duration and frequency to be largely dependent upon weather conditions. Warm, humid, windless conditions appear to be optimal. Even then, however, singing is of short duration. On a good morning, singing may last for a period of 15-20 minutes (c. Rimmer pers. comm.), and perhaps up to 30 minutes (R. Ouelletpers. comm.), occurring mainly between one half hour before to two hours following sunrise in Quebec (Rompré et al. 1997) and New Brunswick (Nixon and Holmes unpubl. data). Singing may continue for a bit longer at dusk (c. Rimmer pers. comm.), with groups of birds occasionally singing for more than an hour at a time (Nixon pers. obs.).
Singing continues throughout the egg-laying period, but wanes rapidly in mid-June with the onset of incubation. It remains irregular throughout the hatching and brooding stages, but does not cease altogether. Interestingly enough, there are accounts of females singing while on the nest throughout the entire nesting cycle (Wallace 1939). This report is not confirmed, mainly because of difficulties in distinguishing males from females in the field, and opinions are mixed regarding its probability (A.J. Erskine pers. comm., J. Marshallpers. comm., R. Ouellet pers. comm.). If confirmed, it could have implications for past census data and for present census methods.
Sporadic singing in late June has been attributed mainly to birds that are renesting after unsuccessful first nesting attempts. Singing intensifies again for about a week in early July, once the young have fledged (Wallace 1939), but subsides to almost total silence through August, as the post-nuptial molt takes place. Infrequent singing is heard again in early September during pre-migratory activities. Marshall (pers. comm.) describes the species at this time as "very vocal," and found it a much more effective time at which to census populations. Rimmer (pers. comm.), however, maintains the first three weeks of June present the optimal censusing period. In 1992, Marshall (pers. comm.) found peak fall singing (and flight songs) on Mt. Mansfield to occur from Aug. 28-Sept. 7. The peak period may be a few days earlier in eastern Canada.
Bicknell's Thrush populations are presently facing a number of threats. In Canada, large amounts of habitat are contained within industrial forest lands, a cause for concern given the inconclusiveness of data regarding the effects of forestry operations on the species. In New Brunswick, virtually the entire range of high-elevation habitat is contained within industrial lands, much of which is in various stages of regeneration following clear cutting (Nixon pers. obs.). Logging activities are also widespread in Cape Breton, with an estimated 25% of the population thought to occur within these areas, and in Quebec.
The species' presence in regeneration opens up a number of issues connected to logging practices and their impacts upon habitat. Firstly, it is possible that regeneration habitat is suboptimal and act as a "sink" for the species (Seutin 1998). Secondly, in Quebec, the species was found to vacate regenerating forests which had undergone precommercial thinning (Rompré et al. 1997, Seutin 1998). Bicknell's Thrush may also occur at lower densities in thinned areas in New Brunswick, seeming to prefer higher stem densities within the 5-10cm size class (Holmes and Nixon unpubl. data). Thinning contribute to lowered foliage density and/or concealment at or in the vicinity of nests, a condition which has been shown to increase nest predation rates (Nolan 1978, Martin and Roper 1988).
Assuming that regeneration provides suitable breeding habitat for the species, there is still the problem of uneven distribution of regenerating stands over space and time discussed in the “Trend in Quantity and Quality of Habitat” section.
Atmospheric deposition, including acid precipitation, has also affected habitat throughout the breeding range (see Trend in Quantity and Quality of Critical Habitat section).
The high-elevation, cool habitats preferred by Bicknell's Thrush may also be particularly sensitive to global warming.
It is presently impossible to determine the impacts of habitat degradation on Bicknell's Thrush populations in the absence of long-term monitoring studies of both the species and its habitat (Rimmer and McFarland 1994). However, we do know that southern breeding grounds are far more restricted than those in Canada. As well, they occur in areas highly vulnerable to degradation by acid precipitation and atmospheric deposition, and appear more vulnerable to the effects of global warming. Thus, although a smaller proportion of the total Bicknell's Thrush population may breed in Canada rather than the northeastern U.S., Canadian habitat may take on greater importance in the future if current trends continue.
Predation & Disturbance
Although little is know of predation rates on Bicknell's Thrush, it is possible that it may account for a high proportion of nest losses. On Seal Island, off the coast of Nova Scotia, abundant numbers of ravens and crows, as well as breeding populations of feral cats and introduced squirrels, may have devastated the population and been the ultimate cause of the species' final disappearance from the island (Wallace 1939, see Dalzell in Erskine unpublished ms). No numbers are available to document predation rates on this site; however, on Mt. Mansfield, Vermont, predation of eggs and chicks accounted for 46% of all nest losses, or six of 13 nests (Wallace 1939). Chicks as old as 6-days were taken. As well, of five empty nests found later in the summer, at least three showed signs of predation.
In Vermont, red squirrels (Tamiasciurus hudsonicus) appear to be the primary nest predators (Rimmer and McFarland 1996), as well as blue jays (Cyanocitta cristata; Wallace 1939). Weasels (Mustela sp.), eastern chipmunks (Tamias striatus), raccoons (Procyon lotor), northern ravens (Corvus corax), sharp-shinned hawks (Accipiter striatus) and gray jays (Perisoreus canadensis) have also been identified as possible predators (Wallace 1939, Rimmer and McFarland 1994, 1996, G. Rompré pers. comm.). While Wallace (1939) notes that blue jays and crows seldom frequent altitudes greater than 914 m in Vermont, these predators may prove to be a larger problem in Canada, where. Bicknell's Thrush habitat is found at lower elevations.
Substantial egg and chick losses can be effected through other agents. Bicknell's Thrush appears to be sensitive to nest disturbances, particularly from humans (Wallace 1939, Rimmer and McFarland 1994, McFarland and Rimmer 1996). In a study of Bicknell's Thrush populations in Vermont, two of four active nests were abandoned at the egg stage within days of discovery (Rimmer and McFarland 1994). Eggs from one nest were estimated to be about 7 days-old. In Wallace's study (1939), desertion accounted for nearly 9% of all observed mortality, infertile eggs and defective embryos for more than 11 %, and disease for nearly 7%.
These high mortality rates may compound an already low reproductive or breeding capacity. As mentioned previously, single brooding is the norm, and clutches tend to be small. Wallace (1939) found that females average 3.46 eggs per nest, significantly below the average for most passerine - but not Catharus - species. Clutch size declines later in the season.
In eastern Canada, it is possible that some degree of habitat partitioning takes place due to competition. In southern Quebec and the Maritime provinces, Bicknell's Thrush populations generally occur alongside larger numbers of Swainson's Thrush (Catharus ustulatus), a more wide-spread habitat generalist. While Bicknell's Thrush is said to inhabit higher elevations (Noon 1981), there is considerable area of altitudinal overlap between the two species. During surveys in Quebec, Swainson's Thrush was detected at all point count stations where Bicknell's Thrush were found (Rompré et al. 1997). In New Brunswick, Swainson's Thrush was found to respond aggressively to playbacks of Bicknell's Thrush vocalizations on numerous occasions (Nixon and Holmes unpubl. data). The species was also present at all 59 of the 90 census stations where Bicknell's Thrush were detected, and often at much higher densities.
Although Bicknell's Thrush is said to have relatively few competitors in the high-altitude habitats of the northeastern United States, it is possible that Swainson's Thrush is playing a role in habitat selection and population levels here as well. Wallace (1939) concluded there were less than a dozen avian species common above 914 m, and Noon (1981) found little evidence of habitat overlap (as defined along elevational gradients) between Bicknell's Thrush and four other insectivorous thrushes on two Vermont peaks. McFarland (pers. comm.), however, has detected Swainson's Thrush at elevations greater than 1,000 m, and reports aggressive responses by the species to playbacks of Bicknell's Thrush calls and songs. Rimmer (pers. comm.) has found Swainson's Thrush to be virtually absent as breeders in krummholz, but has observed incidences of complete overlap between the two species on plots at the lower altitudinal range of Bicknell's Thrush, with Bicknell's Thrush invariably occurring at lower densities. Thus, while there is as yet no hard data on competition between these two species, inter-specific competition can not be ruled out.
Catharus thrushes are known for their "lurking" behaviour and secretive habits; Bicknell's Thrush, however, takes these attributes one step further. When Wallace (1939) began his famous study on Bicknell's Thrush in the 1930s, he called the species "one of the least understood passerine birds in North America." Confusion over the species' taxonomic status was not resolved until 1995 (ADU 1995), while questions still remain with regards to breeding ecology, population levels and trends, winter distribution, and habitat requirements in eastern Canada.
Bicknell's Thrush is a strict ecological specialist, confined to high elevation, mature spruce-fir forests. The species also occurs in similar habitat throughout eastern Canada. It is the only breeding songbird endemic to the subalpine forests of northeastern North America (Rimmer and McFarland 1996). As such, it qualifies as a potentially valuable, long-term indicator of the health of subalpine avian populations and their associated forest habitat.
A significant portion of the Canadian population of this rare species also occurs in "nontypical" habitat - young stands of regeneration resulting from logging activities. Such stands may prove to be less vulnerable to some of the pressures currently facing subalpine habitat. While the species still appears to be selecting for certain structural characteristics within regenerating stands (Holmes and Nixon 1997, Rompré et al. 1997), thereby reducing potential habitat, these stands are more accessible to improvement by direct management. Should productivity studies show that regeneration does, in fact, provide suitable and not "sink" habitat, Canadians may be in the unique situation of being able to increase amounts of breeding habitat for a rare bird through appropriate forest management.
Concern regarding the status of Bicknell's Thrush has been expressed nationally and internationally. The Canadian Wildlife Service considers the species to be one of the highest priority species for research, monitoring and conservation (Dunn 1997), and it is currently the top Neotropical avian conservation priority in the American Northeast (Rosenberg and Wells 1995). A proposed status of "vulnerable" has been put forth by BirdLife's Globally Threatened Species Programme, using IUCN "Red List" criteria, and the species has also been listed as a moderate to high priority species under the Partners in Flight WatchList program (Muehter 1998).
Bicknell's Thrush is one of the rarest songbird species in North America, with a total population estimated at 5000 - 15 000 pairs. Realistic estimates for the Canadian population vary between 2000 and 5000. Distributional data suggest that populations are stable in core areas, but several historic peripheral populations have disappeared.
Of all the provinces, Quebec appears to have the largest population, estimated at about 50% or more of the total Canadian population. Few distributional changes appear to have occurred, other than a possible decline along the north shore of the St. Lawrence. The Maritimes population is divided mainly between New Brunswick (generally estimated at between500-700 pairs) and Nova Scotia (estimated at 300 pairs). The Maritimes population has undergone some distributional changes. Populations have been extirpated from Seal Island and Mud Island, off the southwest coast of Nova Scotia, and from Cape Forchu, on the mainland. Overall, the core population in the Maritimes appears to be relatively stable, with some fluctuations and declines occurring among peripheral populations.
In all parts of Bicknell's Thrush range, typical breeding habitat is under pressures through large-scale forestry operations, acid rain, ski resort development, construction of communication and microwave towers, and heavy recreational use. And, while the effects of global warming are as yet unrealized, geographically restricted, high elevation forests are particularly vulnerable. Recent observations of the species in regeneration forests has been seen as a positive development, but current research suggest that only limited amount of such forest is actually suitable for the thrush. The species' wintering habitat is also threatened, probably more so than the breeding habitat. Current trends in land use on the wintering ground suggest that critical areas of habitat may be insufficient to support forest-dwelling songbirds such as Bicknell's Thrush within the decade.
In summary, Canadian populations of Bicknell's Thrush do not appear to be in significant decline at this point, but it might be at risk due to its low numbers, patchy distribution, low reproductive potential and rangewide habitat pressure. The species' unique occurrence in regeneration in Canada may prove beneficial; alternatively, such habitat may prove unsuitable. Additionally, habitat destruction and modification on the wintering grounds is occurring at tremendous rates.
Many people and organizations have generously contributed their research, time, and support to this report, including Ghislain Rompré, Gilles Seutin and Veronique Connolly at McGill University, Yves Aubry and Jean-Pierre Savard, Canadian Wildlife Service - Quebec Region, Dan Busby, Canadian Wildlife Service - Atlantic Region, Tony Diamond, University of New Brunswick, and Chris Rimmer and Kent McFarland, Vermont Institute of Natural Science. A special thanks is owed to Steve Holmes, Canadian Forest Service - Sault Ste. Marie, for his valuable suggestions, editorial comments, and logistical support.
I would also like to thank the following people for sharing their knowledge and observations of Bicknell's Thrush with me: Henri and Reginald Ouellet, Tony Erskine, Norman Famous, Brian Dalzell, David Christie, Jean Gauthier, Nev Garrity, Joe Marshall and Jonathen Atwood.
Much of the logistical support for this report was provided by the Environmental Research and Assessment Project, Canadian Forest Service, Sault Ste. Marie.
AOU. 1957. Checklist of North American Birds. 5th ed. American Ornithologists' Union, Washington, D.C.
AOU. 1995. Fortieth supplement to the AOU checklist of North American birds. Auk 112:819-830.
Able, K.P. and R.R. Noon. 1976. Avian community structure along elevational gradients in the northeastern United States. Oecologia 26:275-294.
Adamus, P.R. 1983. Atlas of breeding birds in Maine, 1978-1983. Maine Dept. Inland Fisheries and Wildlife.
Allen, E.C. 1916. Annotated List of Birds of Yarmouth and Vicinity, Southwestern Nova Scotia. Trans. N. S. Inst. of Sci. 14:67-95.
Allen, G.M. 1902. List of the birds of New Hampshire. Proc. Man. Inst. Sci. 4:40-183.
Anonymous. 1995. Una Estrategia para la Conservation de la Biodiversidad de la Republica Dominicana: 1994-2003.
Arbib, R. 1980. The blue list for 1980. Amer. Birds 33:830-835.
Arendt, W.J. 1992. Status of North American Migrant Landbirds in the Caribbean: a summary. In Ecology and Conservation of Neotropical Migrant Landbirds (J.M. Hagan and D.W. Johnston, Eds.). Smithsonian Press, Washington, D.C.
Atwood, J.L, C.C. Rimmer, C.H. Darmstadt, and L.R. Nagy. 1994. An expanded investigation of the population status of Bicknell's Thrush in the northeastern United States. Manomet Observatory for Conservation Sciences and Vermont Institute of Natural Science.
Bailey, I.W. 1924. Studies on the spruce budworm, part 2. Cdn. Dept. Ag. Tech. Bull. 37:58-90.
Battles, J.J., AJ. Johnson, T.G. Siccama, AJ. Friedland, and E.K Miller. 1992. Red spruce death: effects on forest composition and structure on Whiteface Mountain, New York. Bull. Torrey Bot. Club 119:418-430.
Bicknell, E.P. 1882. A sketch of the home of Hylocichla aliciae bicknelli, Ridgway, with some critical remarks on the allies of this new race. Bull. Nuttall Ornith. Club 7: 152-159.
Blais, J.R. 1983. Trends in the frequency, extent and severity of spruce budworm outbreaks in eastern Canada. Cdn. J. For Res. 13:539-547.
Bond, J. 1956. Check-list of the birds of the West Indies. Acad. Nat. Sci. Philadelphia, Pennsylvania.
Brewster, W. 1906. Birds of the Cambridge Region. Memoirs of the Nuttall Ornith. Club 4:386-387.
Christie, D.S. 1980. Avifaunal survey: Fundy National Park. New Brunswick Museum, N.R. no 1993. Survey of breeding birds in Fundy National Park, 1992. Vol. 1. Canadian Parks Service Contract No. FNP92-004.
Cory, C.R 1891. On a collection of birds made on the islands of Anguilla and Cay Sal or Salt Cay, Bahama Islands, by Mr. Cyrus S. Winch, during May, 1891. Auk 352.
Davenport , E.B. 1903. Birds observed on Mt. Mansfield and the west end of Stowe Valley at the base of the mountain, in the summer of 1902. Wilson Bull. 10:77-86.
Davis, M.B. and D.B. Botkin. 1985. Sensitivity of cool-temperate forests and their fossil pollen record to rapid temperature change. Quaternary Res. 23:327-340.
Dilger, W.C. 1956. Adaptive modifications and ecological isolating mechanisms in the thrush genera Catharus and Hylocichla. Wilson Bull. 68:171-199.
Dunn, E.H. 1997. Setting priorities for conservation, research and monitoring of Canada's landbirds. Can. Wild. Servo Tech. Rep. Ser. No. 293, Ottawa, Canada.
Erkamo, V. 1952. On plant biological phenomena accompanying the present climatic change. Fennia 75:25-37.
Erskine, A.J. 1980. A preliminary catalogue of bird census plot studies in Canada, part 4. Canadian Wildlife Service Progress Notes 112.
Erskine, A.J. 1984. A preliminary catalogue of bird census plot studies in Canada, part 5. Canadian Wildlife Service Progress Notes 144.
Erskine, A.J.1992. Atlas of breeding birds of the Maritime provinces. Nimbus Publishing Limited, Nova Scotia.
Erskine, A.J. Unpublished ms. Numerical status of breeding "fog-forest thrushes."
Erskine, J.S. 1955. Seal Island. (Nova Scotia) J. Education 5:3-13.
Faxon, W. 1889. On summer birds of Berkshire County, Mass. Auk 1:42.
Faxon, W. 1895. Turdus aliciae bicknelli and Otocoris alpestris praticola as summer residents of Berkshire County, Mass. Auk 12:392-393.
Faxon, W. and F.H. Allen. 1888. Birds of the White Mountain Region, N.H. Auk 5: 153.
Hohn, H. 1980. Possible climatic consequences of a man-made global warming. International Institutefor Applied Systems Analysis, Laxenburg, Austria.
Foster, J.R and W.A. Reiners. 1983. Vegetation patterns in a virgin subalpine forest at Crawford Notch, White Mountains, New Hampshire. Bull. Torrey Bot. Club 110:141-153.
Gillet, L.B. 1935. The thrushes of our woods. Yale Review 24: 743-759.
Godfrey, W.E. 1986. The birds of Canada, revised ed. Nat. Mus. Nat Sci., Ottawa, Canada.
Hardy, Y., M. Mainville, and D.M. Schmitt. 1986. An atlas of spruce budworm defoliation in eastern North America, 1938-80. U.S.D.A. (Forest Service) Miscellaneous Publication No. 1449.
Holmes, S.B. and E.A. Nixon. 1997. Habitat requirements of Bicknell's Thrush Catharus bicknelli in eastern Canada, progress report 1.Unpublished ms.
James, R.D. 1991. Annotated checklist of the birds of Ontario. 2nd ed. Life Sciences Miscellaneous Publications, Royal Ontario Museum, Toronto.
Job, H.K. 1905. Wild Wings. Houghton Mifflin and Co., Boston and New York.
Kettela, E.G. 1983. A cartographic history of spruce budworm defoliation from 1967 to 1981 in easternNorth America. Canadian Forestry Service Information Report DPC-X-14.
Kibbe, D.P. 1985. Gray-cheeked Thrush. In The atlas of breeding birds of Vermont (S.B. Laughlin and D.P. Kibbe, Eds.). University Press of New England, Hanover, N.H.
Lamb, H.H. 1977. Climate: present, past and future. Vol. 2. Methuen, London.
Langille, I.H. 1884. Bicknell's Thrush. Auk 1:268-270.
Macoun and Macoun. 1909. Catalogue of Canadian Birds. Dept. of Mines, Geological Survey Branch.Government Printing Bureau, Ottawa, Canada.
Marshall, J. and R. Clapp. Unpublished ms. Preliminary survey of Bicknell's Thrush, 29 June - 4 July 1984.
Martin, T.E. and I.I. Roper. 1988. Nest predation and nest site selection of a western population of the Hermit Thrush. Condor 90:51-57.
Mc Farland, K.P. and C.C. Rimmer. 1996. Investigations of Bicknell's Thrush (Catharus bicknelli) in the Northeastern United States. Progress Report 1996. Vermont Institute of Natural Science. Woodstock, VT.
Miller-Weeks, M.M. and D. Smoronk. 1993. Aerial assessment of red spruce and balsam fir condition in the Adirondack Region of New York, the Green Mountains of Vermont, the White Mountains of New Hampshire, and the mountains of western Maine, 1985-86. USDA Forest Service, Tech. Rep. NA-TP-16-93.
Morse, D.H. 1979. Habitat use by the Blackpoll Warbler. Wilson Bull. 91:234-243.
Muehter, V.R (Ed.). 1998. WatchList Website <http://www.audubon.orgibirdlwatch/>. Version 97.12. National Audubon Society, New York, NY.
Nixon, E.A. 1996. A distribution survey of Bicknell's Thrush (Catharus bicknelli) in New Brunswick. Canadian Wildlife Service/Canadian Forest Service.
Nolan, V. 1978. The ecology and behavior of the Prairie Warbler Dendroica discolor. Ornithol. Monogr. 26:1-595.
Noon, B.R 1981. The distribution of an avian guild along a temperate elevational gradient: the importance and expression of competition. Ecol. Monogr. 51:105-124.
Ottenwalder, I.A. 1989. A summary of conservation trends in the Dominican Republic. Biogeography of the West Indies.
Ouellet, H. 1991. Is Bicknell's thrush a Gray-cheeked Thrush? [Abstract] American Ornithologists' Union 1O9th Stated Meeting, Montreal, Quebec.
Ouellet, H. 1993. Bicknell's thrush: taxonomic status and distribution. Wilson Bull. 105:545-572.
Ouellet, H.1995. Bicknell's Thrush. In The breeding birds of Quebec (J. Gauthier and Y. Aubry, Eds.). Canadian Wildlife Service, Quebec Region, Environment Canada.
Palmer, R.S. 1949. Maine birds. Harvard Univ. Mus. Compo Zool. Bull. 102.
Peters, H.S. and T.D. Burleigh. 1951. The birds of Newfoundland. Houghton Mifflin Company, Boston. Petersen, W.R 1990. Life on the edge. Sanctuary 30:7-8.
Peterson, J.M.C. 1988. Gray-cheeked Thrush Catharus minimus. In The atlas of breeding birds in New York State (R.F. AndrIe and J. R Carroll, Eds.). Cornell University Press, Ithaca, New York.
Pulliam, H.R 1988. Sources, sinks, and population regulation. The American Naturalist 132:652-661.
Rappole, J. 1995. Ecology of Migratory Birds: A Neotropical Perspective. Smithsonian.
Reiners, W.A. and G.E. Lang. 1979. Vegetational patterns and processes in the balsam fir zone, White Mountains, New Hampshire. Ecology 60:403-417.
Ridgway, R 1882. Descriptions of two new thrushes from the United States. Proc., U.S. Nat. Mus. 4:374-379.
Rimmer, C.C. and K.P. McFarland. 1994. Population studies of Bicknell's Thrush (Catharus bicknelli) on Mount Mansfield, Vermont. Annual report: Vermont Monitoring Cooperative.
Rimmer, C.C. and K.P. McFarland. 1995. Investigations of the population status of Bicknell's Thrush (Catharus bicknelli) in the northeastern United States and the Dominican Republic. Vermont Institute of Natural Science. Woodstock, VT.
Rimmer, C.C. and K.P. McFarland. 1996. Investigations of High Elevation Bird Communities in the NortheasternUnited States and the Dominican Republic. Progress Report 1995. Vermont Institure of Natural Science. Woodstock, VT.
Rimmer, C.C., J.L Atwood, and L.R Nagy. 1993. Bicknell's Thrush: a northeastern songbird in trouble? Bird Observer 21:84-89.
Rimmer, C.C., J.L Atwood, K.P. McFarland, and L.R Nagy. 1996. Population density, vocal behavior, and recommended survey methods for Bicknell's Thrush. Wilson Bull. 108:639-649.
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. Progress Report 1994-1997. Vermont Institute of Natural Science, Woodstock, VT.
Rimmer, C.C., J.E. Goetz, and KP. McFarland. 1998. Bird observations in threatened forest fragments of Sierra De Neiba, Dominican Republic. EI Pitirre 11:38-39.
Rompré, G., V. Connolly, Y. Aubry, J.P. Savard, and G. Seutin. 1997. Distribution, abundance, and habitat requirements of Bicknell's Thrush (Catharus bicknelli) in Quebec. Progress report 1. McGill Univ., Montreal.
Rosenberg, K V. and J. V. Wells. 1995. Importance of geographic areas to neotropical migrant birds in the Northeast. Cornell Lab of Ornithology.
Sabo, S.R 1980. Niche and habitat relations in subalpine bird communities of the White Mountains of New Hampshire. Ecol. Monogr. 50:241-259.
Sabo, S.R. and R.T. Holmes. 1983. Foraging niches and the structure of forest bird communities in contrasting montane habitats. The Condor 85:121-138.
Seutin, G. 1991. The Gray-cheeked Thrush: One Species or two? An analysis of mt DNA. [Abstract] American Ornithologists' Union 109th Stated Meeting, Montreal, Quebec.
Seutin, G. 1998. Bicknell's Thrush: A pristine highland forest specialist or a second-growth lover? Paper presented at the Canadian Pulp and Paper Association general meeting, Montreal, 1998.
Siccama, T.G. 1974. Vegetation, soil, and climate on the Green Mountains ofVermont. Ecol. Monogr. 44:325-349.
Sicama, T.C., M. Bliss, and H.W. Vogelmann. 1982. Decline of red spruce in the Green Mountains of Vermont. Bull. Torrey Bot. Club 109: 163-168.
Sladen, P.W. 1988. Some new records and observations of birds in the Virgin Islands. Amer. Birds 42:1227-1231.
Squires, W.A. 1976. The birds of New Brunswick. 2nd ed. Monog. Ser. New Brunswick Mus. 7:1-221. Sweet, D.W. 1906. Bicknell's Thrush on Mt. Abraham. J. Maine Ornith. Soc. 8:81-84.
Terborgh, J.W. 1977. Bird species diversity on an Andean elevational gradient. Ecology 58: 1007-19.
Terborgh, J.W. 1980. The conservation status of neotropical migrants: present and future. In Migrant Birds in the Neotropics: Ecology, Behavior, Distribution, and Conservation (A. Keast and E.S. Morton, Eds.). Smithsonian Institution Press, Washington, DC.
Thayer, J.E. 1907. Nesting of Bicknell's Thrush. Oologist 24:120-122.
Todd, W.E.C. 1963. Birds of the Labrador Peninsula and adjacent areas. Univ. Toronto Press, Toronto, Ontario.
Torrey, B. 1889. Bicknell's Thrush breeding in Vermont. Auk 6:194.
Townsend, C. 1917. In Audubon's Labrador. Auk 34: 145
Tufts, H.P. 1909. Breeding note/!. Macoun's Cat. of Canadian Birds. Dept. of Mines. Geol. Sur. Tufts, RW. 1986. Birds of Nova Scotia. 3rd ed. Nova Scotia Museum, Halifax, Nova Scotia.
Veit, R.R and W.R Petersen. 1993. Birds of Massachusetts. Massachusetts Audubon Society, Lincoln, Massachusetts.
Vogelmann, H.W. 1982. Catastrophe on Camel's Hump. Nat. His. 91:8-14.
Vogelmann, H.W., T.G. Siccama, D. Leedy, and D. Ovitt. 1968. Precipitation from fog moisture in the Green Mountains of Vermont. Ecology 49: 1205-1207.
Wallace, G.J. 1939. Bicknell's Thrush, its taxonomy, distribution, and life history. Proc., Boston Soc. Nat. Rist. 41:211-402.
Wallace, G.J. 1949. Bicknell's Thrush. In Life histories of North American thrushes, kinglets, and their allies (A.C. Bent, Ed.). U.S. Nat. Mus. Bull. 196:199-217.
Warrick, R.A., H.H. Shugart, M.J.A. Antonovsky, J.R Tarrant, and C.J. Tucker. 1986. The effects of increased CO 2 and climatic change on terrestrial ecosystems. In The Greenhouse Effect, Climatic Change, and Ecosystems (B. Bolin, B.R Doos, J. Jager, and R.A. Warrick, Eds.). John Wiley & Sons, New York.
- Date Modified: