Prothonotary warbler (Protonotaria citrea) COSEWIC assessment and status report: chapter 7

• Previous page
• Table of contents
• Next page

Population Sizes and Trends

Search effort

Because of the relative inaccessibility of its swamp forest habitat, the Prothonotary Warbler is not particularly well surveyed by the roadside-based Breeding Bird Survey (Robbins et al. 1986; Flaspohler 1996). Hence, specialized search effort is required to accurately reveal its presence and numbers. Breeding bird atlas surveys and associated site and regional inventories offer the best information on its distribution and numbers in Canada. A second Ontario atlas, which has just completed five years of effort, provides the best information on distribution and trends in Canada. The two atlas periods (1981-85 and 2001-05) involved comparable search effort: about 124,000 person hours were logged in the first atlas versus over 148,000 hours in the second. In addition, intensive surveys for the species have been conducted annually since 1997 by the national recovery team, so current information on the species in Canada is undoubtedly more comprehensive than historically.

Abundance

Based on extrapolations made from Breeding Bird Survey data, the continental population of Prothonotary Warbler is estimated at about 900,000 pairs (Rich et al. 2004), over 99% of which reside in the U.S., mostly in the southeastern states. Intensive surveys in Canada from 1997 to 2006 have seldom located more than 20 mated pairs (McCracken et al. 2006). In addition to paired birds, from 5-14 unmated territorial males have been located annually since 1997. The most recent population survey conducted in 2005, suggests a current population size of between 28 and 34 adults, including unmated males.

Fluctuations and trends

Although the Breeding Bird Survey is not particularly suited for monitoring Prothonotary Warbler population trends (i.e., flagged as data with a deficiency; Sauer et al. 2005) it is the only such source of information at the continental scale. Breeding Bird Survey results suggest that the continental population has experienced a significant long-term decline of –1.3%/year between 1966 and 2005 (p = 0.02; N = 474 routes; Sauer et al. 2005), or a population decline of about 40% overall. Results from the last 10 years (1996-2005) indicate a non-significant increase of 0.13%/year (p = 0.89; N = 283 routes). The long-term declines appear to have been quite widespread, including the southeastern core of the Prothonotary Warbler’s breeding range.

In Canada, trend information cannot be determined from Breeding Bird Surveys because sample sizes are too small. Trend information, therefore, comes from surveys that target Prothonotary Warblers and from the Ontario Breeding Bird Atlas.

Intensive surveys conducted annually since 1997 have found between 8 and 20 pairs (McCracken et al. 2006), down from an estimated maximum of 40-80 pairs during the period 1981-1986 (McCracken 1984, 1987). Based on information from these surveys, the population of Prothonotary Warblers in Canada has decreased from an estimated 40 pairs in 1995 (Page 1996) to 8 pairs in 2005, which amounts to an 80% decrease in population size in the last decade.

In Canada, the species is best known from Rondeau Provincial Park. The Rondeau population was estimated at about 100 pairs during the early 1930s. Following the removal of many dead trees killed by Dutch elm disease, the population was drastically reduced (see McCracken 1984). In all likelihood, the 100-pair estimate was a broad extrapolation, and likely overly optimistic (A. Woodliffe, pers. comm. 2006). Nevertheless, as recently as the early 1950s, Nickell (1969) reported that the Prothonotary Warbler was one of the most numerous species in forested sections of the park. In 1981, the population was estimated at 20-25 pairs (McCracken 1984). Intensive annual surveys of the park were conducted from 1997-2005, during which time the number of known pairs has ranged from a high of 13 pairs in 2000 to a low of 3 pairs in 2005 and 2006 – the lowest on record (McCracken et al. 2006; J.D. McCracken, unpubl. data).  

Despite the population decline, the Prothonotary Warbler’s distribution in Canada has not changed significantly over the past two decades. During the first Ontario Breeding Bird Atlas in 1981-85, the species was reported from a total of 15 10x10 km squares versus 16 squares during the present atlas (2001-05).

At many locations in Canada (e.g. Point Pelee, Holiday Beach, Long Point region, Hamilton), Prothonotary Warbler populations exhibit a pulsating pattern, with breeding activity periodically blinking on for a number of years and then off. Recolonization of formerly occupied sites, rather than colonization of new areas, appears to attest to the species’ rather narrow habitat tolerances.  

Rescue effect

Although there is no direct evidence of immigration of Prothonotary Warblers from the U.S., some immigration must take place, particularly from nearby breeding stations in northern Ohio, and potentially from more scattered populations residing in Michigan, Pennsylvania, and New York. A population viability analysis for the Canadian population was conducted by Tischendorf (2003), who used metapopulation and individual-based, spatially explicit models that were based on parameters for fecundity, juvenile and adult survival, density dependence, initial population size (24 pairs), 1000 simulation runs, dispersal, demographic stochasticity, and environmental stochasticity. In the absence of immigration, all models showed that the Canadian population had a >90% probability of extirpation within 100 years. However, results predicted that immigration of at least one female per year from potential source populations in the U.S. was sufficient to eliminate the extirpation risk.

Rescue effect depends, however, on the probability that immigrants successfully locate mates and breed, which is entirely density-dependant. Small amounts of immigration are unlikely to sustain small, fragmented, isolated populations of migrant landbirds (McCracken 1998b; Ward 2005). Tischendorf’s (2003) modelling procedures did not consider the fragmented and isolated nature of the Prothonotary Warbler population in Canada, and so may be overly optimistic.

The probability of rescue effect also largely depends upon the density and distribution of a species in adjacent jurisdictions and how well it is faring there. Breeding bird atlas maps from the Great Lakes states (e.g. Eaton 1988; Peterjohn and Rice 1991; Walkinshaw 1991) reveal that occurrences of Prothonotary Warblers within 100 km of Ontario are extremely scattered and localized, suggesting that there are few reliable source populations for Canada. This suggests that if Prothonotary Warblers were to disappear from Canada they would not likely be replaced by birds from the U.S.