Archived 2011-2012 topics: Bahama Warbler (Dendroica flavescens): newly split and Near Threatened?

Initial deadline for comments: 31 January 2012.

Bahama Warbler Dendroica flavescens has been split from Yellow-throated Warbler D. dominica following an assessment of its taxonomic status, including analyses of the morphology, vocalisations and genetics of D. (dominica) flavescens and continental D. dominica (McKay et al. 2010) and a subsequent decision by the American Ornithologists’ Union (AOU 2011). D. flavescens is endemic to Grand Bahama and Abaco (Great Abaco and Little Abaco) in the Bahamas, where it appears to be restricted to mature Caribbean Pine Pinus caribaea forest (McKay et al. 2010).

There are estimated to be 900-1,000 km2 of pine forest remaining on Grand Bahama and Abaco combined, although the remaining area of mature pine forest is apparently not known. D. flavescens has been estimated to occur at densities of 3.5-17.1 individuals/km2, suggesting a total population of 3,150-17,100 individuals, although more recent density estimates suggest a population of only 3,150-3,500 individuals (McKay et al. 2010 and references therein, W. K. Hayes in litt. 2011). Surveys in the pine forests of Grand Bahama found the species to be patchily distributed and existing at low densities (Lloyd and Slater unpubl., Lloyd et al. 2008). Its population is thought to have declined since the onset of major logging of the pine forests in the mid-20th century (W. K. Hayes in litt. 2011). Some areas of pine forest, such as those located west of Freeport, have been fragmented by roads and residential development, while their structure and composition have been altered by changing fire regimes, human activity and hurricanes (Lloyd and Slater unpubl.). Caribbean Pine forests are at high risk of further development and renewed logging (McKay et al. 2010, White 2011). This situation is probably exacerbated by the presence of freshwater recharge and supply areas in the pine forests of Grand Bahama coupled with the conversion of the Bahamas to a reverse osmosis water supply, suggesting that these areas may be prone to further development (W. K. Hayes in litt. 2011).

The species’s Extent of Occurrence (EOO) is estimated at c.2,300 km2 (see range map). It is important to note that the EOO is a measure of the spread of extinction risk and not the extent of suitable habitat (ESH), and is thus far less precise compared to the ESH, although areas of sea are excluded because they are completely unsuitable. This estimate for the EOO meets the range size threshold for Endangered under criterion B1 of the IUCN Red List, and there may be continuing declines in the area, extent and/or quality of habitat, thus potentially fulfilling one of the subcriteria (B1b(iii)). However, there is no indication that the species’s population is severely fragmented or restricted to fewer than 11 locations (fewer than 6 to qualify as Endangered) (subcriterion B1a).

Draft BirdLife range map for Bahama Warbler (click on map to see larger version)

In terms of the IUCN criteria, a ‘location’ is defined according to the most important threat, and is regarded as a geographically or ecologically distinct area in which a threatening event can rapidly affect all individuals of the species in question. The major threat facing the species is the logging of pine forest for timber or development, a threat that would normally be expected to affect discrete portions of habitat during each event, thus the species should probably be regarded as occurring at more than 10 locations. For the purpose of Red List assessments, a species’s population is regarded as ‘severely fragmented’ when over 50% of suitable habitat exists in patches that are too small to support viable populations and separated by distances several times greater than the average long-term dispersal distance of the species. There is currently insufficient information on the level of fragmentation in the species’s habitat and population, thus further information is requested.

With a population that probably includes fewer than 10,000 mature individuals, which may be inferred to be undergoing a continuing, but unquantified decline, the species might qualify as Vulnerable under criterion C2; however, information is required on the species’s sub-population structure. For the purposes of Red List assessments, sub-populations are defined as geographically or otherwise distinct groups between which there is little or no demographic or genetic exchange, i.e. typically one successful migrant per year or less. Information is required on the level of movement between the populations on Grand Bahama and Abaco. If they are considered sub-populations according to the IUCN definition, then the proportion of the total population present in each sub-population should be estimated.

There are an estimated 600 km2 of pine forest on Grand Bahama, although the effective area of suitable habitat may be 10-15% lower owing to the effects of recent storms (W. K. Hayes in litt. 2011). With the total area of pine forest being put at 900-1,000 km2, the implication is that Grand Bahama holds 51-67% of the total population. Following surveys in April 2007, Lloyd et al. (2008) estimated a population of 2,116 individuals (95% CI: 1,239-3,614) on Grand Bahama, based on a density of 3.6 birds/km2. Even at this density, it seems unlikely that either of the two island populations number fewer than 1,000 individuals, although clarification is needed on this, and it should be noted that all population extrapolations so far do not take into account the remaining area of mature pine forest, which has apparently not been estimated. If the level of movement means that the species has one sub-population, it would probably qualify as Vulnerable under criterion C2a(ii).

Based on current information the species would probably qualify as Near Threatened on the basis that it appears to nearly meet the thresholds for Vulnerable under criteria B1a+b(iii), and probably also criterion C2a(i). However, further information is requested, particularly on the level of habitat fragmentation and the likely sub-population structure. Comments would be welcomed on the estimation of the total population size and the likely number of mature individuals. Information is also requested on the severity of threats faced by the species, with emphasis on whether the extent and/or quality of habitat is in decline and whether a continuing population decline can be inferred in the present or projected to occur in the future.


AOU (2011) Fifty-second supplement to the American Ornithologists’ Union Check-list of North American Birds. Auk 128: 600-613.

Lloyd, J. D., Slater, G. L. and Metcalfe, A. E. (2008) Taxonomy and population size of the Bahama Nuthatch. Report for the National Geographic Society Committee for Research and Exploration, Grant No. 8146-06. Available at

McKay, B. D., Reynolds, M. B. J., Hayes, W. K. and Lee, D. S. (2010) Evidence for the species status of the Bahama Yellow-throated Warbler (Dendroica “dominica” flavescens). Auk 127: 932-939.

White, T. (2011) The False Kirtland’s: A cautionary tale. Birding, November 2011: 34-39.

This entry was posted in Americas, Archive, Caribbean and tagged . Bookmark the permalink.

3 Responses to Archived 2011-2012 topics: Bahama Warbler (Dendroica flavescens): newly split and Near Threatened?

  1. I suspect there are no further data available to resolve the issues raised. I’m not aware of anyone currently studying the bird.

  2. John Lloyd says:

    The reference to Lloyd et al. (2008) can be updated to:

    Lloyd, J. D., and G. L. Slater. 2011. Abundance and distribution of breeding birds in the pine forests of Grand Bahama, Bahamas. Journal of Caribbean Ornithology 24:1-9.

    The information is the same, but the peer-reviewed article is probably a better source than our unpublished report.

    Concerning the estimation of total population size, I think that the density estimate presented in Lloyd and Slater (2011) is reasonably robust. One potential concern from the analysis presented in that paper is that we grouped multiple species together to estimate the detection function, which assumes that all of the species in the group share the same detection function. I’ve done the same analysis using only data from singing, male Bahama Warblers, and the results are nearly identical: probability of detection was 0.52, with a 95% CI ranging from 0.41-0.67. In the paper, we used 0.54 as the detection probability for Bahama Warbler. So, I’m reasonably confident that the estimate of density in Lloyd and Slater (2011) is reliable, with the caveat that distance sampling, like any approach to correcting for detectability, carries with it a number of other assumptions. I’m also reasonably confident that the population estimate for Grand Bahama is reliable given that we sampled randomly from all of the existing pine forest. What I am unsure of, however, is whether our estimate is also reasonable for Abaco. Perhaps Bill Hayes or some of the other folks that have worked on both islands could comment on the extent to which the pine forests on the two islands are similar.

    A more problematic issue regarding our population estimate is the assumption that one female exists for every singing male in the population. This is almost certainly not the case, but whether the actual ratio is different enough from 1:1 to matter is not clear. Related to this, it is also important to note that our estimate, even after correcting for detectability, doesn’t account for non-territorial males, which don’t sing and which therefore weren’t available for detection. In some bird populations, these floaters can account for a significant proportion of total population size. I think the bottom line is that we may have underestimated population size on Grand Bahama but that we are probably correct to the order of magnitude. I would also agree that it is unlikely that fewer than 1,000 birds exist on Grand Bahama, although again I don’t know whether this is also true for Abaco.

    Regarding fragmentation of habitat on Grand Bahama, I think that the forest is relatively unfragmented for Bahama Warbler. The forests are covered by a network of roads (Google Earth images show this well), and a few large clearings, but mostly the pines are in large, well-connected patches. However, were any of the existing subdivisions ever built out, the situation would change dramatically. For example, if development proceeded at Lucaya Estates, much of the pine forest in the central part of the island could become uninhabitable by Bahama Warbler. From my opinion, the severity of the threat faced by Bahama Warbler depends largely on whether the existing subdivisions are built out; if they are, I would predict a substantial loss of habitat and a concomitant decline in numbers of Bahama Warbler. Until there is some certainty about the fate of the pine forest in the existing subdivisions, I would suggest using a conservative approach and assume that about 50% of the existing habitat is at risk of loss. If timber harvest occurs, then the amount of habitat at risk would go up dramatically. One way to quantify the risk is to look at the 2010 Forestry Bill. None of the forest in the central portion of the island is included in the forest reserve system, and so is presumably at risk of further clearing. The pine forests on the eastern side of the island are identified as Forest Reserves, which are designated both for resource extraction and conservation of endemic flora and fauna. So, essentially all of the pine forest is vulnerable to some form of extractive (e.g., timber harvest) or destructive (e.g., housing developments) use in the future. I believe that the same is true for Abaco; looking at the maps included in the Forestry Bill, I don’t see any areas where pine forest is side aside for wildlife conservation. Bottom line is that I don’t see any strong administrative protection for habitat of the Bahama Warbler.

    Regarding population declines, there is some evidence of large declines between 1969 and 2007. Emlen (1977. Land bird communities of Grand Bahama Island: the structure and dynamics of an avifauna. Ornithological Monographs 24:1-129) estimated densities of Bahama Warbler as 14 individuals per square km, assuming that he counted only 20% of the birds present. Even with the difference in estimated detectability, the density estimates in Lloyd and Slater (2011) are about 30% lower, although Lloyd and Slater (2011) sampled over a larger area than did Emlen. Although certainly not definitive, there is certainly some evidence that population size has decreased substantially over the past several decades.

    I think that additional information from Abaco would be helpful, as would data that would let us evaluate the accuracy of our population estimates. We’d also benefit from having a better idea of what forest conditions this species requires, as well as its tolerance for fragmentation, which would in turn help us determine whether our population-size extrapolations are reasonable. Related to this, it would be helpful also to determine how Bahama Warbler might respond to timber harvest, perhaps by reviewing existing literature on Yellow-throated Warblers in the southeastern United States.

  3. John’s information is helpful. I can’t speak for his surveys, but in my own on Grand Bahama, I found the highest densities of birds, including Bahama Warblers, in the Lucayan Estates area on central portions of Grand Bahama. The forest to the east was less productive, and the more fragmented forest in the vicintiy of Freeport and Lucaya was probably more depauperate. To my knowledge, there are no surveys from the pine forests west of Freeport, but much of that forest now appears to be lost due to hurricane damage (primarily from seawater intrusion, I suspect). Thus, island-wide extrapolations have limits, but I’d agree with John there are likely to be >1,000 individuals on Grand Bahama.

    I don’t know of anyone having population data from Abaco. I visited the island once but did not do surveys. David Lee found a locally high concentration of birds during a winter survey at one location, but we can’t conclude much from that.

Comments are closed.