Use of localized descriptive statistics for exploring the spatial pattern changes of bird species richness at multiple scales

We used localized descriptive statistics (e.g., localized mean, standard deviation, and correlation coefficient), local Moran’s I and a nonparametric test to explore the spatial patterns, variations, and autocorrelations of bird species richness changes, and the spatial association between bird species richness changes and three climatic variables in New York State at multiple spatial scales. We found that the spatial distributions of species richness between two time periods (1980–85 and 2000–05) were significantly different. The local spatial autocorrelation index revealed that the spatial patterns of bird species richness changes were spatially non-random and scale-dependent. Using localized correlation coefficients we found that the associations between bird species richness changes and three climatic variables were spatially heterogeneous and also scale-dependent. Our results illustrate that localized descriptive statistics are useful tools for characterizing bird species richness changes across time and space, visually describing geographic patterns of species diversity changes, and identifying scale-dependent hot spots of those changes.

► We explore the spatial heterogeneity of bird species richness in New York State.
► We used localized descriptive statistics and local spatial autocorrelation.
► Species richness between two time periods is spatially significantly different.
► Species richness changes are spatially non-random and scale-dependent.