Black-throated Blue Warbler

Setophaga caerulescens


Demography and Populations

Welcome to the Birds of North America Online!

You are currently viewing one of the free species accounts available in our complimentary tour of Birds of North America. In this courtesy review, you can access all the life history articles and the multimedia galleries associated with this species.

For complete access to all species accounts, a subscription is required.

Subscribe Now

Already a subscriber? Sign In
Explore Black-throated Blue Warbler distribution in BirdVis

Use our distribution analysis tool to view predicted relative abundance in specific areas of the Black-throated Blue Warbler range.

Figure 5. Relative abundance of Black-throated Blue Warbler during the breeding season.

Based on data from the North American Breeding Bird Survey, 2011–2015. See Sauer et al. (2017) for details.

Figure 6. Regional trends in Black-throated Blue Warbler breeding populations.

Based on data from the North American Breeding Bird Survey, 1966–2015 (Sauer et al. 2017). Data show estimates of annual population change over the range of the survey; areas of increase are shown in blue and declines are shown in red. See Sauer et al. (2017) for details.

Measures of Breeding Activity

Age at First Breeding

Both males and females breed in their first year but some (< 20%), usually males, remain unmated in their first breeding season (Holmes et al. 1992); unmated males are more frequent in marginal habitats (Holmes et al. 1995). No unmated females, first-year or older, have been detected in breeding areas (RTH).


Clutches typically contain 4 eggs, but range from 2 to 5 (Bent 1953b, Griscom and Sprunt 1957, RTH, NLR). In a sample of 768 nests at Hubbard Brook in New Hampshire, clutch size averaged 3.6 eggs ± 0.9 SD, with a range of 2 to 5 and both a median and mode of 4 (TSS, RTH). Clutch size in the Nantahala National Forest, North Carolina, averaged 3.7 (n = 562 nests; Stodola et al. 2013). At Hubbard Brook, females laid up to 5 clutches per season, with an average of 1.7 clutches, and a median and mode of 2; however, no more than 2 clutches were successful, the others being lost to predation or other factors (TSS, RTH). Such multiple clutches in one season represent (1) replacements for those lost to predation and other factors, and (2) second or, rarely third clutches, following successful earlier ones.

Annual and Lifetime Reproductive Success

Based on 25 years of an intensive study at Hubbard Brook in New Hampshire (Holmes et al. 1992, Sillett and Holmes 2005, Townsend et al. 2013, TSS, RTH), females laid an average of 6.0 eggs ± 2.5 SD (range 5.6–6.4, n = 457) and fledged an average of 3.1 young ± 2.1 SD (range 2.8–3.6, n = 457) per season. Lifetime reproductive success unknown, although currently under study (M. T. Hallworth, MSW, TSS, NLR, RTH, unpublished data).

Number of Broods Reared per Season

Of 411 individually-marked females monitored for their complete breeding seasons at Hubbard Brook between 1986–2010, 63% successfully fledged 1 brood per season, 20% fledged 2 broods, and one female fledged 3 broods (Holmes et al. 1992, Sillett and Holmes 2005, Townsend et al. 2013, TSS, RTH). The remaining 17% were unsuccessful in fledging even 1 brood, even though they made several nesting attempts. Thus, the majority of females in this population fledge 1 brood per season, with 0–62% fledging 2 or more, depending on the year. Double-brooded females bred on territories with greater food availability, delivered more food to their young, and made fewer feeding trips per hour (i.e., were higher quality females or bred on higher quality territories), and fledged fewer young in their first nests than did single-brooded females (Nagy and Holmes 2005b, Kaiser et al. 2015). Females provided with extra food produced more second broods, confirming the importance of food as a limiting factor affecting annual reproductive output in this population (Nagy and Holmes 2005a, Kaiser et al. 2015). In northern Michigan, the frequency of double-brooding ranged from 26–45% over 6 yr (27–44 territories monitored annually; K. R. Hall, unpublished data). Birds in the southern part of the range tend to fledge fewer young per year, largely due to reduced likelihood of double-brooding (R. Cooper, R. Chandler, K. Stodola, and M. Cline, personal communications). Studies with individually-marked females in other temperate breeding passerine species (e.g., Nolan 1978, Eliason 1986a, Petit 1989, Keast 1990, Podolsky et al. 2007, Ligi and Omland 2007) indicated that multiple brooding in Nearctic–Neotropical migrant songbirds may be more common than previously thought.

Life Span and Survivorship

Oldest recorded bird at Hubbard Brook Experimental Forest in New Hampshire, a male banded as an older (after-second-year) individual in 1984; returned each summer to breed through 1991, for a minimum age of 9 yr (RTH). The oldest record in U.S. Fish and Wildlife Service band recovery files as of 2016 is a 10-year-old female (USFWS Banding Office, unpublished data).

Few data on survival during the first year, owing to very high natal dispersal (see below). No data on survival during first weeks or months after fledging. For hatch-year males caught and color-banded in October on Jamaican overwintering grounds between 1986 and 1989, 40% (n = 26) returned to the same places the following October, while 62% (n = 13) of the males initially marked as older (after-second year) adults returned to their overwintering sites, suggesting either higher survival of older individuals or their stronger site tenacity (Holmes et al. 1992). Females (age unknown in this study) returned from one October to the next at a rate of 42% (n = 24). Survival over the major part of the overwintering in Jamaica, as estimated by persistence of color-marked birds from October to March, was 77% (n = 31) for males and 66% (n = 53) for females (Holmes et al. 1992). Similar overwinter persistence rates have been found for this species in lowland forested habitats in Puerto Rico, although at high altitudes rates averaged lower, between 28–47% (Wunderle 1995). Also, overwinter persistence in coffee plantations in the Dominican Republic were within the range of those in natural forests and did not vary significantly with plantation size (Wunderle and Latta 2000). The estimates of survival given above do not separate out the possibility of dispersal. Data are lacking on the degree to which individuals disperse when they return to either summer or overwintering habitats.

Sillett and Holmes (Sillett and Holmes 2002) analyzed data from 1986–2000 (n = 336 individuals with capture histories from New Hampshire, and 151 with capture histories from Jamaica) using Cormack-Jolly-Seber (CJS) models that account for recapture probability. Results indicate that annual survival measured on returns to the breeding grounds each spring (May to May) was 51% (± 0.03 SE) for males and 40% (± 0.04 SE) for females (recapture probabilities = 93 % (± 0.03 SE) and 87% (± 0.06 SE) for males and females, respectively). When estimated from color-marked individuals returning to the overwintering quarters in Jamaica (October to October), annual survival (43% ± 0.03 SE) and recapture (95% ± 0.04 SE) probabilities did not differ between the sexes. Similarly, annual survival and recapture probabilities of young birds (i.e., yearlings in New Hampshire and hatch-year birds in Jamaica) did not differ from those of adults, suggesting that from the time hatch-year individuals acquire territories on overwintering quarters in mid-October, they survive as well as adults within the same habitat (Sillett and Holmes 2002).

These CJS analyses of data from New Hampshire and Jamaica further showed that monthly survival probabilities over the summer (May–August) and overwintering (October–March) periods were high: 1.0 for males in New Hampshire, and 0.99 + 0.01 for males in Jamaica and for females in both locations (Sillett and Holmes 2002). Using these annual and seasonal survival estimates, survival was calculated for the intervening migration periods at 77–81% (± 0.02 SE). Thus, mortality rates were at least 15 times greater during migration compared to that in the stationary (breeding/overwintering) periods, and more than 85% of apparent mortality of this species occurred during migration.

Disease and Body Parasites

Body Parasites

Hippoboscid flies (Diptera: Hippoboscidae) are found rarely on adults (NLR). Parasitic (hematophagous) fly larvae (Diptera: Calliphoridae: Protocalliphora) are recorded for nestlings of this species in New Hampshire (P. P. Marra personal communication). They caused death of one brood and wounded nestlings in another (Rodenhouse 1986). An examination of 128 nests sampled during 2005–2007 at Hubbard Brook Experimental Forest, New Hampshire, showed that the prevalence of protocalliphorid parasitism ranged among years from 38% to 52% of nests, with lower intensity of parasitism at higher elevation and parasitism more likely earlier than later in the breeding season (K. A. Ciurej and NLR, unpublished data).


Few data. Reported to be a carrier of West Nile Virus (U.S. Geological Survey, National Wildlife Health Center, Malarial infections noted in 236 of 1,083 (21.8%) breeding adults sampled across their North American breeding range (Fallon et al. 2006). This study identified 10 genetically distinct Plasmodium lineages and 2 Haemoproteus lineages, but their geographical distributions were uninformative in terms of determining their geographic origins.

Causes of Mortality


At Hubbard Brook in New Hampshire, nestlings dying from exposure during cold, rainy periods vary from year to year, and range from 2% (Holmes et al. 1992) to about 15% (Rodenhouse 1986). No information from other localities.


Little information available to evaluate predation as a mortality source on adults. In the breeding season, < 1% of nesting females disappear from their nests, probably due mostly to predation (RTH, Sillett and Holmes 2002). Information on predation rates on eggs and nestlings, available from only a few sites. At the Hubbard Brook Experimental Forest and nearby parts of the White Mountains in New Hampshire, nest mortality, which is due mostly to predation, varies among years, sites, and habitats (Holmes et al. 1992, Holmes et al. 1995). At Hubbard Brook between 1986 and 2010, the mean proportion of nests/territory lost to depredation ranged annually from 0.04–0.76, averaging 0.27 ± 0.14 SD (n = ~30 nests per year, Sillett and Holmes 2005, TSS, RTH). Nest predation did not vary with conspecific (male) density (Sillett et al. 2004), but was higher (41%) in areas of low shrub density compared to high shrub density (30%, Holmes et al. 1995). Nest mortality data from other sites, based on Mayfield estimates, include 11–30% in New Brunswick (n = 4–10 per site, Bourque and Villard 2001), and 37–57% in North Carolina (n = 105 nests, Guzy 1995). Depredation rates were not density-dependent at Hubbard Brook (Sillett and Holmes 2005).

Human/Research Impacts

Frequently found dead at television towers during migration (Bent 1953b, Taylor 1973c). See Conservation: Effects of Human Activity.


Natal Philopatry

Return of first-year birds to their natal site appears to be very low. Of the over 5,000 nestlings banded at Hubbard Brook in New Hampshire between 1986 and 2016, only 22 have been resighted, the closest to its natal site being about 300 m, but most were resighted > 500 m, and several were 2 km or more away (RTH, TSS). This low rate of return suggests wide natal dispersal in this species, but also in part to the fact that this research at Hubbard Brook has been conducted on relatively small plots (10–150 ha) within large tracts of continuous forest. Thus, even if young birds return to within a few hundred meters of their hatch location, they may go undetected. Stable isotopes in feathers of first-time breeders along an elevational gradient in North Carolina also indicate wide natal dispersal, although natal origins of these individuals were unknown (Graves et al. 2002). These findings thus far indicate that determinations of natal dispersal will require intensive coverage over a large area, and/or the development and refinement of new techniques for obtaining such information.

Breeding Dispersal and Fidelity to Breeding Site

Adults often return to the same general area from one year to the next (Holmes et al. 1992, Graves et al. 2002, Cline et al. 2013; see also return rates above). Based on individually marked birds at Hubbard Brook in New Hampshire from 1998–2008, the mean distance between centers of breeding territories for returning males was 163 m ± 11 SE and 245 m ± 20 SE for females, and overall, 66% of males and 46% of females returned to within 150 m of the center of their previous year’s territory (Cline et al. 2013).

Overwinter Dispersal and Fidelity to Overwintering Site

Individuals of this species also return to the same overwintering locality in successive years (Holmes et al. 1992, Wunderle 1995, Wunderle and Latta 2000). Based on individually marked birds, median distances between territory centers of individuals returning to Jamaican overwintering sites from one October to the next were 28 m (range 9–52, n = 14) for males and 41 m (range 8–262, n = 10) for females (Holmes et al. 1992). These dispersal distances were significantly shorter than those reported for this species in New Hampshire breeding areas (Holmes et al. 1992, see above).

Connectivity Between Breeding and Overwintering Sites

Despite strong fidelity to both breeding and overwintering sites for many migrant species (see above), little is known about the relationship between where a given individual breeds and where it overwinters or whether individuals from different breeding areas segregate or mix during the overwintering period (Webster et al. 2002). For small Nearctic-breeding songbirds, such as the Black-throated Blue Warbler, band recoveries are too few to provide useful information on connectivity between breeding and overwintering areas. Recent studies employing analyses of stable-isotope ratios in Black-throated Blue Warbler feathers and other tissues provide some insight into this question (Chamberlain et al. 1997). Such analyses show that those individuals breeding in the northern part of the summer range overwinter mostly in the western part of the Greater Antilles (e.g., Cuba, Jamaica), while those breeding in the southern Appalachians overwinter mostly in Puerto Rico and Hispaniola (Rubenstein et al. 2002). Furthermore, the stable isotope data indicate that Black-throated Blue Warblers within local overwintering areas in Jamaica derive from a wide range of longitudes across the breeding range, which indicates considerable mixing of individuals in the overwintering grounds from geographically separated breeding localities (Rubenstein et al. 2002). Preliminary data from archival light-level geolocators recovered from males breeding in New Hampshire support the isotope findings: Of 4 light-level geolocators recovered, 2 individuals had overwintered on the island of Jamaica while one overwintered in Cuba and the other in Hispaniola (M. T. Hallworth, unpublished data).

Population Status


Using data from the North American Breeding Bird Survey (BBS), the Black-throated Blue Warbler population was estimated at 2,400,000 individuals for the United States and Canada from 2005–2014 (Rosenberg et al. 2016). Point-count abundance data collected by breeding bird atlas projects estimated populations at 1,500,000 individuals in Ontario between 2001–2005 (Cadman et al. 2007a) and 150,000 males (95% CI: 136,000–171,000) in Pennsylvania from 2004–2009 (Wilson et al. 2012).

Breeding density differs markedly with habitat (see Figure 5), varying directly with the thickness of the shrub layer (Steele 1992). In a northern hardwoods forest in New Hampshire, densities range locally from highs of 8–9 pairs/10 ha in sites with thick shrub layers to low of 2–3 pairs/10 ha where shrubs are sparse (Sherry and Holmes 1985, Holmes et al. 1986, Steele 1992, Holmes et al. 1995, RTH). In other habitats, pairs or unmated males may occur at densities of 1–5/10 ha, and may exist in isolation without neighboring conspecifics (RTH). Based on the number of individuals encountered per investigator effort, relative abundance across the breeding range was negatively correlated with latitude, being highest in the Appalachian Mountains, especially south of 40ºN. (Graves 1997b). In Quebec (in Darveau 1996a), lowest breeding densities found in coniferous stands: 1.7 pairs/10 ha in hemlock, 1.6/10 ha in red spruce, and 2.2/10 ha pairs in balsam fir; by contrast, 6.4 pairs/10 ha in a white birch stand and 10.8 pairs/10 ha in sugar maple–yellow birch; greatest variations in density were seen in heterogeneous deciduous stands (Table 2). Quantitative estimates of density are lacking from most other breeding localities.

Overwintering densities also vary with habitat, although data are limited. The only estimates available are from 2 sites in Jamaica (Holmes et al. 1989): In wet limestone forest at Copse Mountain in Jamaica, the density averaged 33 individuals (males & females) per 10 ha (range 29–37 over 3 yr), while in dry limestone habitat near Rocklands Jamaica, the average density was 12 individuals per 10 ha (range 10–13 over 3 yr).

Age ratios vary among habitats within a region (Holmes et al. 1995) and over geographical range (Graves 1997b), perhaps reflecting differences in habitat quality and/or dispersal patterns of young individuals.


From 1969 to 2016, numbers of Black-throated Blue Warblers were relatively stable in northern hardwoods forest at Hubbard Brook in north-central New Hampshire (Holmes et al. 1986, Holmes and Sherry 1988, Holmes and Sherry 2001, Rodenhouse et al. 2003, Sillett and Holmes 2005, Holmes 2007, Holmes 2011, RTH). Trends in abundance at this local site synchronous over regional spatial scales in accordance with annual changes in climate and Lepidoptera populations (Jones et al. 2003b). Based on 32 yr of mistnet captures in southeastern Massachusetts, numbers stable in spring and increasing in autumn (Lloyd-Evans and Atwood 2004). In northeastern Minnesota, 1991–2003, significant increases noted (10–11% annually; Lind et al. 2004), although BBS data from 1966–2015 show only a slight increasing trend for Minnesota as a whole (Sauer et al. 2017).

Analyses of 50 years of BBS data (1966–2015; Sauer et al. 2017) indicate a survey-wide annual increase of 1.95%. Populations in the northern parts of the range, mostly in eastern Canada (New Brunswick, Quebec, and the “Boreal Softwood Shield,” in general) seem to be increasing the most, while those in the northeastern part of the U.S. (Maine, New Hampshire, Vermont, Massachusetts, Connecticut, and the “Atlantic Northern Forest,” in general) are relatively stable or slightly increasing. The species appears to be declining in New York state and, more notably, at southern edge of range, especially North Carolina and Virginia (Sauer et al. 2017). These patterns are consistent with predictions from ongoing climate change (Rodenhouse et al. 2008a, Matthews et al. 2011). Similarly, this species showed an upslope elevational shift in distribution in the White Mountains of New Hampshire between 1993 and 2009 (DeLuca and King 2017).

Population Regulation

Factors limiting numbers of this species have been studied intensively at Hubbard Brook in New Hampshire (Rodenhouse and Holmes 1992, Rodenhouse et al. 1997, Rodenhouse et al. 2003, Rodenhouse et al. 2006, Sillett et al. 2000, Sillett et al. 2004, Sillett and Holmes 2005, Sofaer et al. 2011). The major factors affecting breeding success are food availability (Rodenhouse 1986, Holmes and Schultz 1988, Holmes et al. 1991, Rodenhouse and Holmes 1992, Nagy and Holmes 2005a, Nagy and Holmes 2005b, Nagy et al. 2007, Kaiser et al. 2015), predation on eggs and nestlings (Rodenhouse and Holmes 1992, Holmes et al. 1992, Holmes et al. 1995), and local population density (Rodenhouse et al. 2003, Sillett et al. 2004, Sillett and Holmes 2005). Indeed, these three factors, food, nest predation and adult density, explain 91% of the variation in the average number of young fledged per year within the population studied (Sillett and Holmes 2005). Thus, two biotic factors—density-independent nest predation and density-dependent fecundity—the latter in combination with climatically affected food abundance, appear to be the most critical variables influencing the annual fecundity of this species. The two most important mechanisms underlying the density dependence in breeding populations involve crowding (as revealed by experimental decreases in neighbor density, Sillett et al. 2004) and site-dependence (an inverse relationship between population size and mean territory quality; Rodenhouse et al. 1997, Rodenhouse et al. 2003, Rodenhouse et al. 2006, McPeek et al. 2001). Crowding and site-dependence mechanisms appear to act simultaneously, but at different spatial scales to regulate local abundances of this species (Rodenhouse et al. 2003).

The long-term demographic data from Hubbard Brook indicate in several ways that the population is regulated (Rodenhouse et al. 2003). First, time series analysis of year-to-year changes in abundance indicates strong density-dependence during this period (Rodenhouse et al. 2003). Second, annual fecundity is significantly negatively related with warbler density (see above), indicating density-dependent reproductive performance. Third, a stochastic projection matrix model, parameterized with field data for Black-throated Blue Warblers at Hubbard Brook, demonstrates that the negative feedback on annual fecundity is sufficient to regulate this population (Sillett and Holmes 2005). Last, breeding success, i.e., warbler fecundity, is positively correlated with the number of yearlings recruited into the breeding population in subsequent years (Holmes et al. 1992, Sillett et al. 2000, Townsend et al. 2016). This relationship between nesting success one year and recruitment (and thus abundance) the next holds despite mortality on migration and potentially during the overwintering period (Sillett and Holmes 2002). If changes in breeding habitat, e.g., through fragmentation or other forms of degradation, lead to changes in breeding productivity, as has been shown for nest predators and parasites (e.g., see Wilcove and Robinson 1990), this could eventually lead to lower population abundances. Thus, events during the breeding period appear to be of particular importance in maintaining population sizes of this species (Holmes et al. 1992, Sherry and Holmes 1992b, Sherry and Holmes 1995, Rodenhouse et al. 2003, Sillett and Holmes 2005).

Black-throated Blue Warbler populations may also be influenced by events during the overwintering period season (Holmes et al. 1989, Sillett et al. 2000, Rubenstein et al. 2002). During the overwintering period, high site tenacity and low dispersal distances indicate this species' strong attachment to particular places and habitats (Holmes et al. 1992). In addition, territorial individuals that disappear during the overwintering period are replaced by others (Holmes et al. 1989), suggesting that habitats there might be saturated. The latter has been confirmed for American Redstart in Jamaica through experimental removals (Marra et al. 1993). Thus, if much overwintering habitat is destroyed or degraded, and if suitable areas are not available nearby, then overwintering mortality rates may increase and overwintering habitat could at that point become limiting. No data exist at the present time to indicate whether or to what degree this scenario is happening for this species, or for most other species of Nearctic–Neotropical migrant (Sherry and Holmes 1995, Newton 2004). Nevertheless, declines in abundance of breeding populations in the Southern Appalachians may be linked to severe habitat degradation in parts of the overwintering range, particularly Haiti and perhaps other parts of the eastern Greater Antilles (Rubenstein et al. 2002).

Recent studies on American Redstart have found that overwintering habitat may in fact influence bird physiological condition and subsequently their reproductive success, thus showing a cross-seasonal effect (Norris et al. 2004a, Reudink et al. 2009a). Similarly, stable-isotope ratios in claws of migrating Black-throated Blue Warblers were used to identify habitats occupied during the preceding overwintering period and their relationship to the warblers' physiological condition during migration (Bearhop et al. 2004). More intensive studies on the habitat relations and survivorship of Black-throated Blue Warblers and other migrant passerines in overwintering habitats and during migration are needed to determine the importance of the non-breeding period to year-round population dynamics (see Faaborg et al. 2010a, Faaborg et al. 2010b).

Recommended Citation

Holmes, R. T., S. A. Kaiser, N. L. Rodenhouse, T. S. Sillett, M. S. Webster, P. Pyle, and M. A. Patten (2017). Black-throated Blue Warbler (Setophaga caerulescens), version 3.0. In The Birds of North America (P. G. Rodewald, Editor). Cornell Lab of Ornithology, Ithaca, NY, USA.