Identification of significant shorebird areas: thresholds and criteria

Abstract

Aim Conservation managers designate significant areas for shorebirds based on imperfect data. Significant wetlands for migratory shorebirds have usually been identified on the basis of whether they exceed certain thresholds, defined either by total abundance (usually 20,000 waterbirds) or percentage of a population (usually 1.0%). We evaluate the performance of existing criteria and determine if lowering thresholds would improve shorebird conservation without adding unreasonable numbers of significant sites. Location Australia. Methods We evaluated the best available data, which is used by managers to designate significant areas, to describe the effect of lowering thresholds on the number of significant sites identified and the number of shorebirds these sites support using a range of thresholds in existing criteria. We also investigated factors which may explain interspecific differences evident when lowering thresholds. Results When the threshold for total abundance was lowered from 20,000 to 2000 shorebirds, an additional 45 shorebird areas, holding 65% more shorebirds, were identified. When thresholds for the percentage of a population criterion were lowered from 1.0 to 0.1%, an additional 86 shorebird areas were identified which held 29% more shorebirds. The proportion of a species population counted within wetlands identified as significant by the application of criteria varied widely between species. The percentage of population criterion always identified a network of areas that included more individuals of each species than the total abundance criterion at all threshold levels tested. The percentage of species populations found in networks of significant areas showed greater increase as thresholds were lowered for species that were abundant, widespread and well represented at existing thresholds. Main conclusions Our results suggest lowering thresholds will substantially increase the number of shorebirds in identified significant areas. However, some species will remain under‐represented, partly because of interspecific differences in distribution and inadequate sampling of some shorebird habitats.

Ecology and conservation of Australia’s shorebirds

“…This suggested that the interaction of coarse environmental variables can reflect population trends measured in other ways, and that one can begin to make inferences about population trends across remote interior Australia using modelling approaches such as those developed here. For shorebirds, these models helped improve our estimates of total population size, which has been the foundation on which abundance thresholds to identify important habitats have been based (DEH, 2006;Clemens et al, 2010). Model predictions also indicated that over the decades some species experience periods of far greater contractions in suitable habitat than others, consistent with other recent work (Runge et al, 2015a).…”

“…Existing metrics on the importance of a wetland for shorebirds rely on counts of the number of birds that use the habitat (DEWHA, 2009;Clemens et al, 2010), but high spatial and temporal variation in the availability of ephemeral habitats has left no clear understanding of how, where or when to take conservation action related to ephemeral wetlands that would be sufficient to protect these species. As Piersma (2007) points out, people throughout the globe are conducting unintended massive scale experiments on the impacts of habitat loss and degradation on shorebird populations.…”

A pragmatic approach for comparing species distribution models to increasing confidence in managing piping plover habitat

“…Models generally predicted habitat in areas characterized by wide sandy beaches or overwash fans, with sparse vegetation and access to low-energy shorelines (Maslo, Leu, et al, 2016;. The similarity in model predictions when viewed as a binary value and not a raw probability value underscores the importance of the selection and application of suitability threshold values in species distribution and other habitat models (Clemens, Weston, Haslem, Silcocks, & Ferris, 2010). When viewed as a binary measure of suitability, we observed core areas of habitat predicted by both models (e.g., Figure 3-5; and 196.3 ha of habitat in the New Jersey and New York study areas, respectively.…”

An assessment of the efficiency of protection status through determinations of biodiversity hotspots based on endemic bird species, Taiwan

“…Studies have found that in protected habitats, plants and animals are more than twice as likely to move toward recovery as species in unprotected habitats (Myers, 1988;Myers et al, 2000). However, identifying critical habitats and hotspots remains highly challenging because the comprehensive data on species distribution and abundance that are required for such identification are often lacking (Clemens, Weston, Haslem, Silcocks, & Ferris, 2010;Clemens, Herrod, & Weston, 2014). To overcome this limitation, species distributions are predicted by either (1) associating individual species occurrences with known habitat preferences, constructing relationships between environments and species distribution at the individual species level, and then combining distribution information from different species (Austin, 2002;Elith et al, 2006;Araújo & New, 2007;Elith & Leathwick, 2009) or (2) using a cross-covariance matrix to find linear combinations of environments and a group of species' distributions that have a maximum correlation with each other at the community level Johnson & Wichern, 2007).…”