Spatial autocorrelation of species diversity and distributions in time and across spatial scales
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Aim
Spatial autocorrelation (SAC), also known as aggregation, is a notable property of species distributions and diversity; it reflects species niche and dispersal, has conservation significance, and affects ecological models. Yet, we know little about spatial and temporal patterns of SAC in empirical data. Here, we assess SAC in both observed species distributions and species richness, quantifying its magnitude and prevalence over large extents and across spatial resolutions. We also assess its dynamics over the past 50 years.
Location
Czechia, Europe, New York State, Japan
Time period
1972 - 2017
Major taxa studied
Birds
Methods
We analyzed four temporally replicated gridded bird atlases, each aggregated to multiple grain sizes. To measure SAC in species distributions, we used the Join count statistic (JC) and its deviation from the expectation under a random distribution. We assessed temporal changes in JC and their relationship with changes in occupancy, given their close association. We used Moran’s I to measure SAC in species richness.
Results
Both species distributions and diversity were positively autocorrelated across all regions, periods, and grains, and the magnitude of autocorrelation mostly decreased with increasing grain. We found that the temporal change of JC varied across species and regions, with zero average trends in Moran’s I, JC, and occupancy. However, when JC and occupancy were considered jointly, we found systematic temporal shifts: contracting species became more aggregated (compact) while expanding species became more fragmented (disjoint).
Main conclusions
Stronger SAC at finer grains suggests greater predictability of diversity and distributions at these scales. Despite zero average change in occupancy or SAC, their coupled shifts highlight the importance of considering both jointly. We found long-distance dispersal (rather than advancing edge) and vulnerability of isolated populations to extinction as the major drivers of range dynamics in temperate birds.
