Canada warbler COSEWIC assessment and status report: chapter 7

Population Sizes and Trends

Search effort and monitoring programs

The Breeding Bird Survey (BBS)

The Breeding Bird Survey is a large-scale program that monitors breeding bird populations across North America (Sauer et al. 2005). Breeding bird abundance data are collected by volunteers at 50, 400-m radius stops spaced 0.8 km apart along permanent 39.2-km routes (Downes and Collins 2007). In Canada, the surveys usually take place in June during the breeding season of most forest birds and are conducted from 0.5 hour before to approximately 4.5 hours after sunrise.

The BBS provides relatively good coverage of those areas of the breeding range where Canada Warbler abundance is high (Figure 2).

The Partners in Flight population estimate database provides an estimate of the proportion of the global breeding range (80% of which is in Canada) of the Canada Warbler, in lat/long degree blocks, that is sampled by the BBS and it also rates BBS coverage in relation to species abundance. These data suggest that 54% of the global breeding range of the Canada Warbler is sampled by the BBS and that 36.4% of the population (by abundance) has good coverage (67-100% of range covered), 18.1% has fair coverage (33-67% of range covered) and 39.2% has poor coverage (10-33% of range covered). Finally, relative abundance estimates from the BBS and the Ontario Breeding Bird Atlas (see below) point counts indicate that the highest relative abundances for the Canada Warbler are found in areas sampled by the BBS (P. Blancher pers. comm. 2008). BBS analyses also weight trends by abundance, so in the case of the Canada Warbler, the trend is most strongly influenced by BBS routes in the east where densities are highest.

Figure 2. Relative abundance of Canada Warblers, based on Breeding Bird Survey abundance data calculated for each lat/long degree block from 1987-2007, in relation to the portion of the breeding range sampled by the Breeding Bird Survey. Grey areas = not sampled by BBS, white areas = sampled, but no Canada Warblers (P. Blancher pers. comm. 2008).

Figure 2. Relative abundance of Canada Warblers, based on Breeding Bird Survey abundance data calculated for each lat/long degree block from 1987-2007, in relation to the portion of the breeding range sampled by the Breeding Bird Survey.

Étude des populations des oiseaux du Québec (ÉPOQ)

In Quebec, the ÉPOQ database, which manages the bird checklists that have been regularly produced by thousands of volunteers since 1955, is used for determining Canada Warbler population trends in Quebec (Cyr and Larivée 1995). The ÉPOQ database covers all regions located south of the 52nd parallel during all seasons (Cyr and Larivée 1995).

The main disadvantage of this method for surveying the Canada Warbler is that it covers mostly inhabited regions in the south (i.e., St. Lawrence Lowlands), with less coverage of more northerly parts of the range. Moreover, the method does not control for the number of observers per checklist, weather conditions, spatial variation in search effort over time, but only for the total number of observation hours (Cyr and Larivée, 1995).

Ontario Breeding Bird Atlas (OBBA)

The OBBA compares the distribution and abundance of breeding birds between 1981-1985 and 2001-2005, and is an important source of information on the status of the Canada Warbler in Ontario (Bird Studies Canada 2006a). The data are gathered by volunteers who visit 10 x 10-km plots at least three times during the breeding period (Bird Studies Canada 2006a). The percent change in the distribution of the Canada Warbler in Ontario over a period of 20 years is then calculated by comparing the percentage of the 10 x 10-km squares/blocks with breeding evidence in the first atlas period to the percentage of squares/blocks with breeding evidence in the second atlas period, adjusting for observation effort (Bird Studies Canada 2006a).

This method is considered adequate for estimating Canada Warbler abundance in Ontario, because of the large number of samples gathered during the two periods and the standardized methodology used (Bird Studies Canada 2006a). In addition, this program generally covers the entire breeding range of the species in Ontario (Bird Studies Canada 2006a).

Ontario Forest Bird Monitoring Program (FBMP)

The Ontario Forest Bird Monitoring Program began in 1987 and is coordinated by the Canadian Wildlife Service, Ontario Region. Its objective is to document forest bird population trends and the relationships between the birds and their habitat during the breeding period in relatively unfragmented forest landscapes (Canadian Wildlife Service 2006). Volunteers use point counts to survey birds in both large forest areas and forest fragments.

One of the limitations of this program for monitoring Canada Warblers is that the species occurs at relatively few sites, making the interpretation of long-term trends difficult. However, this method is more likely to detect Canada Warblers than other monitoring programs, because sampling is done primarily in unfragmented forest areas (Canada Wildlife Service 2006).

Canadian Migration Monitoring Network

This program is designed to monitor populations of migratory passerines at a series of 23 monitoring stations across Canada (Bird Studies Canada 2006b). Fall migration, in particular, can provide information on population trends for birds breeding in the boreal forest and farther north, as they move from their northern breeding grounds to their southern wintering grounds. The primary activities carried out at these stations are bird banding and daily visual counts of birds during spring and fall migration periods.

The main source of bias with this monitoring program is its assumption that the number of birds observed at these stations is proportional to the actual number of birds migrating on the days monitored. A second limitation is that most stations have been operating for less than 10 years, so cannot provide long-term trend information.

Canadian Breeding Bird (Mapping) Census Database

This database provides information on the density of breeding birds in plots located in every Canadian province and territory, except Prince Edward Island (Kennedy et al. 1999). Experienced observers use “spot-or territory-mapping” methods to assess the density of breeding birds in each plot (Kennedy et al. 1999). This sampling method is one of the most accurate for estimating forest bird density (Bibby et al.2000). One of the main limitations of this program is most censuses are several decades old, and therefore data do not reflect current densities.

Project for the Prediction of Bird Presence and Density for Quebec (PPPDAQ)

Recently launched by the Canadian Wildlife Service, Quebec Region (J.-L. Desgranges and P. Agin unpubl. data), the objective of this project is to model bird presence and density at the ecoregion and ecozone scales, by forest habitat type and by tree family using learning algorithms. The database contains 3,306 point counts obtained using the IPA technique primarily in Quebec and the adjacent provinces and states taken from published and unpublished studies by various sectors (academia, government, private and other). With respect to the Canada Warbler, the density of breeding pairs was estimated for four contexts: 1) Boreal Shield – disturbed mixed forest – birches; 2) Boreal Shield – well-drained mixed forest – maples; 3) Mixedwood Plains – poorly drained mixedwood forest; and 4) Atlantic Maritime – poorly drained mixedwood forest – ash).

The main limitation of this program is that Canada Warbler density is obtained from several sampling stations and then extrapolated to large areas (i.e., the ecoregions scale), which can result in overestimating the species’ density in little used or unused forest stands. The proposed models also assume that Canada Warbler densities are homogeneous across a given spatial scale, yet it is known that the species has a highly grouped distribution and generally responds to finer-scale environmental variables than those used in the models. The major advantage of this program is that it covers several spatial scales.

National Boreal Bird-Habitat Modelling (NBBHM) Project

The National Boreal Bird-Habitat Modelling Project is a joint initiative of the Western Boreal Conservation Initiative (WBCI) and the University of Alberta’s Boreal Ecosystems Assessment for Conservation Networks(BEACONs). Its objective is to evaluate habitat relationships and distribution patterns of birds in Canada’s boreal. In collaboration with researchers from universities, government agencies and industry, the NBBHM Project is developing spatially explicit models that will be used to predict the distribution and abundance of boreal birds with the objective of incorporating this information into decision-making processes. This research program is particularly important in estimating the relative abundance of the Canada Warbler in several ecodistricts of the boreal forest.

Abundance

According to abundance estimates obtained from the BBS, Canada Warbler populations throughout North America currently total 1.4 million individuals (Rich et al. 2004). In Canada, the population is estimated at approximately 1.2 million breeding adults (85% of North American population in Canada, PIF Landbird Population Estimates Database 2007), or approximately 600,000 breeding pairs. The accuracy of these estimates is considered adequate (accuracy of 3 on a scale of 1 to 6, see Appendix B in Rich et al. 2004).

Recent Canada Warbler abundance estimates for Ontario based on the Ontario Breeding Bird Atlas between 2001 and 2005, suggest a population of 900,000 individuals (i.e., 450,000 breeding pairs) in that province (Blancher and Couturier 2007). This estimate is considered to be more accurate than the BBS estimates because the atlas has numerous samples gathered in large interior tracts of forest (P. Blancher pers. comm. 2008).

Another estimate of population size can be derived from the Ontario Breeding Bird Atlas. If Ontario, with its 450,000 pairs, accounts for roughly 33% of the Canadian breeding range, Quebec and the Maritimes another 33% and the remaining provinces the final 33% (see Figure 1), then a Canadian population would be approximately 1,350,000 breeding pairs or 2.7 million adults based on extrapolations from the Ontario numbers.

The most comprehensive surveys of Canada Warbler breeding densities based on the Canadian Breeding Bird Census (BBC) suggest that the highest densities are in the eastern parts of the range, including Ontario, Quebec, and New Brunswick (Table 1). High densities have also been reported from Manitoba (Table 1) and southern Northwest Territories, with densities up to 0.65 pairs/ha (Machtans 2006).

Breeding densities in Quebec, based on the PPPDAQ, range from 0.18 to 0.4 pairs/ha and are generally lower than those estimated by the BBC. Densities are higher in the Boreal Shield (disturbed mixed forest–birch), Canadian Shield (well drained mixed forest–maple), and Mixedwood Plains (poorly drained mixed forest) ecozones, each with 0.04 pairs/ha (J.-L. Desgranges and P. Agin unpubl. data). The density of the species is believed to be lower in the Atlantic Maritime ecozone (poorly drained mixed forest–ash), at only 0.02 pairs/ha.

The NBBHM suggests that the relative abundance of the Canada Warbler is higher in east-central Alberta, in several ecodistricts in southwestern Northwest Territories and in the area of the Ontario-Manitoba border east of Lake Winnipeg (D. Mazerolle unpubl. data). The abundance of the Canada Warbler is lower throughout northeastern British Columbia (D. Mazerolle unpubl. data).

Table 1. Summary of breeding densities per ha in Canada according to the Canadian Breeding Bird (Mapping) Census Database (Kennedy et al. 1999)
Province Ecoregion Density/ha SD Number of plots
British Columbia
Hay River Lowland 0.08  
1
Alberta
Slave River Lowland 0.11 0.05
2
Saskatchewan
Mid-Boreal Uplands 0.03  
1
Manitoba
Interlake Plain 0.23 0.12
3
Manitoba
Lac Seul Upland 0.07  
1
Manitoba
Mid-Boreal Uplands 0.38  
1
Ontario
Abitibi Plains 0.35 0.36
10
Ontario
Algonquin-Lake Nipissing 0.26 0.15
15
Ontario
Frontenac Axis 0.001  
1
Ontario
Lake Erie Lowland 0.28 0.25
2
Ontario
Lake Nipigon 0.18 0.14
4
Ontario
Lake Timiskaming Lowlands 0.25  
1
Ontario
Manitoulin-Lake Simcoe 0.16 0.07
9
Ontario
Rainy River 0.10  
1
Quebec
Appalachians 0.52 0.30
5
Quebec
Southern Laurentians 0.14 0.15
2
Quebec
St. Lawrence Lowlands 0.33 0.30
9
New Brunswick
Appalachians 0.23 0.16
4
New Brunswick
Maritime Lowlands 1.24  
1
New Brunswick
Northern New Brunswick Uplands 0.16 0.05
2
New Brunswick
Saint John River Valley 0.01  
1
New Brunswick
Southern New Brunswick Uplands 0.09 0.06
2
Nova Scotia
Annapolis-Minas Lowlands 0.25  
1
Nova Scotia
Fundy Coast 0.04 0.05
2
Nova Scotia
Southern Nova Scotia Uplands 0.001 0.00
3

Fluctuations and trends

Breeding Bird Survey

The best available estimate of Canada Warbler trends comes from the BBS, which covers a significant portion of the species range and has reasonable precision for detecting significant change. Long-term BBS data show a significant decline of 4.5%/year between 1968 and 2007 (Figure 3, Table 2). This corresponds to an approximately 85% decline in the population over the last 38 years. In the most recent 10-year period (approximately three generations), 1997-2007, BBS data show a significant decline of 5.4%/year (Table 2), which amounts to a loss of 43% of the population in the last 10 years. Trends for Ontario, Quebec and the Maritimes, which include over 60% of the breeding range and most of the breeding population, all show declines on the long (1968-2007) and short (1997-2007) term (Table 2).

Figure 3. Annual trend indices (log scale) for the Canada Warbler in Canada between 1968 and 2007 according to BBS data (from Downes and Collins in preparation).

Figure 3.Annual trend indices (log scale) for the Canada Warbler in Canadabetween 1968 and 2007 according to BBS data (from Downes and Collins in prepartion) (See long description below)
Long description of Figure 3

The best available estimate of Canada Warbler trends comes from the BBS, which covers a significant portion of the species range and has reasonable precision for detecting significant change. Long-term BBS data show a significant decline of 4.5%/year between 1968 and 2007

Trends can also be examined in relation to Bird Conservation Regions, with analyses based on routes through ecological regions, rather than provinces or territories. With the exception of the Boreal Taiga Plains, which includes less than 20% of the population (P. Blancher pers. comm. 2008), the trends for the Boreal Softwood Shield, Boreal Hardwood Transition and Atlantic Northern Forest regions, all show declines on the long and short term (Table 2).

Table 2. Annual indices of population change for Canada Warbler populations based on Breeding Bird Surveys (from Downes and Collins in preparation)
Region Period Index P Number of BBS routes
Canada
1968-2007 -4.5 P<0.05
293
Canada
1997-2007 -5.4 P<0.05
186
Ontario
1968-2007 -2.4 not significant
90
Ontario
1997-2007 -2.4 not significant
54
Quebec
1968-2007 -4.3 P<0.05
89
Quebec
1997-2007 -7.8 not significant
55
New Brunswick
1966-2007 -5.4 P<0.05
36
New Brunswick
1997-2007 -1.4 not significant
23
Nova Scotia
1966-2007 -3.1 P<0.05
30
Nova Scotia
1997-2007 -20.0 P<0.05
19
Boreal Taiga Plains
1968-2007 3.3 not significant
36
Boreal Taiga Plains
1997-2007 1.2 not significant
26
Boreal Softwood Shield
1968-2007 -5.3 P<0.05
28
Boreal Softwood Shield
1997-2007 no trend information available no trend information available no trend information available
Boreal Hardwood Transition
1968-2007 -3.0 not significant
101
Boreal Hardwood Transition
1997-2007 -4.1 not significant
67
Atlantic Northern Forest
1966-2007 -4.0 P<0.05
97
Atlantic Northern Forest
1997-2007 -4.8 not significant
62

Ontario Breeding Bird Atlas

A comparison of the species distribution in Ontario from the first (1981-1985) to the second (2001-2005) atlas period shows an overall non-significant decline of 15% in the number of occupied squares, with declines in all five regions of Ontario between the two atlas periods. Declines in occupied squares ranged from significant declines in the Southern Shield and Carolinian Life Zone of 10% and 36%, respectively, to a non-significant decline of 17% in the Northern Shield, where most of the population occurs (Bird Studies Canada 2006a).

Étude des populations des oiseaux du Québec (ÉPOQ)

The ÉPOQ database shows a significant long-term decline in the abundance of Canada Warblers of 5.0%/year (P £0.001; Figure 4) in Quebec between 1980 and 2005.

Figure 4. Annual indices of population change for the Canada Warbler in Quebec between 1980 and 2005 according to ÉPOQ data (from Larivée 2006).

Figure 4. Annual indices of population change for the Canada Warbler in Quebec between 1980 and 2005 according to ÉPOQ data(from Larivée 2006)
Long description of Figure 4

A comparison of the species distribution in Ontario from the first (1981-1985) to the second (2001-2005) atlas period shows an overall non-significant decline of 15% in the number of occupied squares, with declines in all five regions of Ontario between the two atlas periods. Declines in occupied squares ranged from significant declines in the Southern Shield and Carolinian Life Zone of 10% and 36%, respectively, to a non-significant decline of 17% in the Northern Shield, where most of the population occurs (Bird Studies Canada 2006a). The ÉPOQ database shows a significant long-term decline in the abundance of Canada Warblers of 5.0%/year (P £0.001; Figure 4) in Quebec between 1980 and 2005.

Ontario Forest Bird Monitoring Program

The trends calculated from this program show a non-significant decline of 4.6%/year (n = 40 sites) between 1987 and 2005 for all of Ontario.

Canadian Migration Monitoring Network

One station, Long Point Bird Observatory, has long-term trend information on Canada Warblers during the fall migration. Information from this station shows a non-significant decline of 0.2%/year between 1967 and 2005. Four additional stations have trend information for fall migration during the last 10 years (Table 3). All but one show declines in Canada Warbler populations (Table 3).

Table 3. Summary of Canada Warbler population trends from the Canadian Migration Monitoring Network during fall migration (from Bird Studies Canada 2006b)
Region Period Annual index P Monitoring station
Lake Erie
1967-2005 -0.2 not significant Long Point Bird Observatory
Southern Manitoba
1993-2005 -6.08 P<0.05 Delta Marsh Bird Observatory
Southern Saskatchewan
1993-2005 2.85 not significant Last Mountain Bird Observatory
Lake Superior
1995-2005 -8.1 P<0.05 Thunder Cape Bird Observatory
Central Alberta
1994-2005 -3.53 not significant Lesser Slave Lake Bird Observatory

In summary, data from the Breeding Bird Survey, which provide trend information on Canada Warbler populations across Canada, show significant long- and short-term declines. These declines are particularly obvious in the core of the breeding range in the eastern part of the country. Declines are also evident in a variety of regional surveys such as the Ontario Breeding Bird Atlas and the Étude des populations des oiseaux du Québec, although themagnitude of the decline varies with the survey.

Rescue effect (immigration from an outside source)

In the event of the extirpation of the Canadian population, immigration of individuals from a number of U.S. states bordering Canada is possible. Immigration is likely to be limited, however, because the species is also showing significant declines across its U.S. range (BBS: 1966 – 2006:-1.8%/year, P = 0.00, n = 306 routes; Sauer et al. 2005).

Page details

Date modified: