How is broad-scale distribution pattern in different animal groups explained by climate, natural habitats and land-cover while accounting for spatial autocorrelation?
|Year of publication||2013|
|Type||Appeared in Conference without Proceedings|
|MU Faculty or unit|
|Description||Evaluation of the relationships between the structure of animal assemblages and habitat distribution crucially depends on accurate habitat description as well as on considering issues connected with spatial autocorrelation. In our study, we focus on evaluation of relative effects of field-mapped natural habitats and climatic variables on broad-scale distribution patterns of mammals, birds, reptiles, amphibians and butterflies native to the Czech Republic. We also compare performance of field-mapped natural habitats with habitat description based on land-cover data derived from remote sensing. To quantify and test the effect of each environmental variable, we used redundancy analysis, where spatial autocorrelation was accounted for by spatial filters created by Moran’s eigenvector maps. Significant environmental variables were finally used in variation partitioning to quantify the amount of variation explained by sets of climatic, habitat and land-cover variables. Our results showed that individual climatic variables explained generally more variation than natural habitat types, but if we accounted for spatial autocorrelation the relative importance of natural habitats increased whereas importance of climatic variables generally decreased. Variation partitioning revealed that shared effects of climate, natural habitats and land-cover explain most variation in distribution patterns of all groups. However, natural habitats always performed better than land-cover and climate.|