作者: Katja M. Raatikainen , Risto K. Heikkinen , Miska Luoto
DOI: 10.1007/S10531-008-9462-9
关键词: Generalized additive model 、 Grassland 、 Range (biology) 、 Biodiversity 、 Habitat 、 Microclimate 、 Boreal 、 Species richness 、 Ecology 、 Biology 、 Ecology (disciplines) 、 Ecology, Evolution, Behavior and Systematics 、 Nature and Landscape Conservation
摘要: Using species and environmental data from an extensive grassland area in south-western Finland, we investigated the effect of patch connectivity, management local habitat variables on occurrence spring-flowering vascular plants their richness boreal agricultural landscapes. Generalized linear models (GLM) variation partitioning were used to study explanatory power three groups combined contributions six plant species. additive (GAMs) associated cross-validation tests evaluate predictability patterns. Present-day connectivity important predictors for studied In addition, factors, especially radiation, accounted major fractions patterns Hybrid including all variable had higher predictive capability than partial models. However, performance separate single-species varied considerably between Exclusion radiation or hybrid decreased performance, suggesting that these factors are particular importance at northern range margins. When developing conservation planning Northern Europe, attention should be paid well-connected networks patches large, steeply-sloped with a favorable microclimate.