Premium
Species characteristics associated with extinction vulnerability and nestedness rankings of birds in tropical forest fragments
Author(s) -
Feeley K. J.,
Gillespie T. W.,
Lebbin D. J.,
Walter H. S.
Publication year - 2007
Publication title -
animal conservation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.111
H-Index - 85
eISSN - 1469-1795
pISSN - 1367-9430
DOI - 10.1111/j.1469-1795.2007.00140.x
Subject(s) - nestedness , guild , ecology , extinction (optical mineralogy) , habitat , range (aeronautics) , extinction debt , local extinction , habitat fragmentation , geography , dominance (genetics) , population , trophic level , habitat destruction , abundance (ecology) , biology , demography , biological dispersal , gene , composite material , paleontology , biochemistry , materials science , sociology
Following habitat fragmentation, species are predicted to go locally extinct from remnant patches in a predictable order due to differential extinction vulnerabilities. This selective species loss will result in nested distributions of species such that species found in depauperate patches will also tend to be found in larger, more speciose patches. Therefore, it should be possible to determine the relationship between species‐specific characteristics and extinction vulnerability by comparing the order in which species are nested [i.e. nestedness ranking (NR)] with various natural history characteristics available from the literature and/or collected in the field. In this study, we investigate the relationship between the NRs of 41 resident forest‐interior bird species inhabiting recently isolated landbridge islands in Lago Guri, Venezuela, with a large number of natural history characteristics collected from the literature (regional abundance, body length, habitat specificity, trophic guild, sensitivity to disturbance, range size) and from the field (local population density). In a comparison of the best regression models generated using just variables available through the literature (i.e. no local population density) with the best model generated using all possible variables, we found that the inclusion of field‐based data significantly improved the amount of variation explained. The best overall model ( r 2 =0.40, P <0.001) included body size, habitat specificity, zoogeographic distribution (a measure of range size) and local population density as predictors of NR. Understanding the factors that influence extinction vulnerability has important implications for conservation and could be used to help direct management efforts.