Do habitat association models have any generality? Predicting skylark Alauda arvensis abundance in different regions of southern England

摘要: Much of conservation science is based upon determining the use by organisms different resources. However, field data used to construct habitat association models generally come from a small number sites covering fraction area interest. It important therefore assess generality those for species occurring over large geographical areas. In this paper we test describing skylark Alauda arvensis abundance across farmland in southern England relation crop type, structure and (i.e. height surrounding boundaries). Skylarks responded most predictors measured similar ways three regions differing farming practices (arable-dominated, pasture-dominated mixture two). Most regional differences associations could be related speed development. For example, sowing cereals spring, much lauded strategy increase populations, likely have less an effect where cereal development slow than it fast. studies that explicitly performance model developed one place elsewhere presence/absence models. We adopt more sensitive novel approach using counts. found regression equations region performed poorly when tested as direct predictor 1:1 relationship) on other regions. territories observed any were positively correlated with territory numbers predicted built regions, so good relative abundance. The results suggest that, at least, conservationists should confidence advocating management strategies habitat-association extrapolating their our warn against make absolute quantitative predictions elsewhere. Decision-makers beware way.