Oxford University Press, Ornithological Applications, 1(124), 2022
DOI: 10.1093/ornithapp/duab063
Full text: Unavailable
AbstractThe lack of high-quality information on data-poor species can hinder efforts to inform conservation actions via spatial distribution modeling. This is particularly true for tropical birds of conservation concern, for which ecological studies and assessments of their conservation status have received limited funding. Here we use a cost- and time-efficient protocol for assessing the distribution of range-restricted taxa and to identify priority areas for their conservation based on a sequential application of environmental niche models (ENMs) and occupancy-detection models. This approach first uses available geographical information and niche-theory to prioritize potential study sites, which can later be surveyed to obtain high-quality presence–absence data to accurately model distributional ranges with limited resources. We apply this protocol to identify priority areas for two Neotropical birds of conservation concern endemic to the Colombian Andes: Yellow-headed Brush-finch (Atlapetes flaviceps) and Tolima Dove (Leptotila conoveri). We first fitted ENMs using spatially filtered datasets containing all available records up to 2018. We then conducted field surveys across climatically suitable areas identified for both species, carrying out a total of 1,750 counts to generate input data for the occupancy models. Overall, our results suggested more extended and more continuous distribution ranges for both species than previously reported, but also identified population strongholds that are not currently represented within the national protected areas system. Both species occupied a narrow elevational belt (~1,300–2,600 m above sea level) of the Central Andes of Colombia primarily on the slopes of the Magdalena River valley, with isolated populations in the Western and Eastern Andes; these areas have undergone some of the most marked landscape transformations in Colombia. This straightforward protocol maximizes available information and minimizes costs, while allowing for estimation of occurrence probabilities for range-restricted, data-poor taxa.