Premium
Disentangling the effects of geographic distance and environmental dissimilarity on global patterns of species turnover
Author(s) -
Qian Hong,
Ricklefs Robert E.
Publication year - 2012
Publication title -
global ecology and biogeography
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.164
H-Index - 152
eISSN - 1466-8238
pISSN - 1466-822X
DOI - 10.1111/j.1466-8238.2011.00672.x
Subject(s) - jaccard index , ecoregion , geographical distance , ecology , latitude , multidimensional scaling , distance matrices in phylogeny , principal component analysis , beta diversity , geography , macroecology , distance matrix , biology , statistics , biodiversity , mathematics , population , demography , bioinformatics , geodesy , cluster analysis , sociology , algorithm
Aim To distinguish the effects of geographic distance and environmental dissimilarity on global patterns of species turnover in four classes of terrestrial vertebrates (mammals, birds, reptiles and amphibians). Location Six hundred and sixty terrestrial ecoregions across the globe. Methods We calculated species turnover between each pair of ecoregions, using the Jaccard index ( J ). We selected seven variables to quantify environment in each ecoregion, and subjected the environmental values to a principal components analysis. For each realm, we applied multiple regression analysis relating the natural logarithm of the Jaccard index (ln J ) to geographic distance alone and in combination with differences in the environment variables measured as principal components (PC). We used partial correlations to partition variance in ln J between unique contributions of distance and environmental PC scores, the covariation between distance and environment, and unexplained variance. To examine the latitude and species turnover relationship, we regressed ln J on latitude with distance between ecoregions being included as a covariate. Results The natural logarithm of the Jaccard index (ln J ) decreased significantly with increasing geographic distance for all vertebrate classes in each zoogeographic realm, and the slopes of the relationships per 1000 km ranged from −0.251 to −1.043. With environmental differences included in the analysis, both geographic distance and environmental differences were substantial predictors of ln J for every combination of taxon and realm. On average, the unique contribution of geographic distance to variation in species turnover between ecoregions was about 1.4 times that of the environmental differences between ecoregions. Species turnover generally decreased with increasing latitude when controlling for geographic distance. The value of ln J for each vertebrate class was highly and positively correlated with those of the other vertebrate classes. Main conclusions Our analyses suggest that both dispersal‐based and niche‐based processes have played important roles in determining faunal similarities among vertebrate assemblages at the spatial scale examined. Furthermore, reptiles and amphibians exhibited greater distance‐independent faunal heterogeneity among ecoregions and greater turnover among ecoregions with respect to geographic and environmental distance than birds and mammals.