Disturbance increases negative spatial autocorrelation in species diversity

• Published in: Landscape ecology Vol. 32; no. 4; pp. 823 - 834
• Main Authors: Biswas, Shekhar R., MacDonald, Rebecca L., Chen, Han Y. H.
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01-04-2017; Springer Nature B.V
• Subjects: Biodiversity; Biomedical and Life Sciences; Boreal forests; Clearcutting; Deforestation; Disturbance; Ecology; Environmental Management; Landscape Ecology; Landscape/Regional and Urban Planning; Life Sciences; Nature Conservation; Plant communities; Research Article; Riparian ecology; Species diversity; Species richness; Sustainable Development; Timber industry; Canada, Alberta; Moran’s coefficient; Riparian community; Ecological processes; Spatial pattern; Clearcutting
• ISSN: 0921-2973; 1572-9761
• DOI: 10.1007/s10980-017-0488-9

Context
Ecological processes that shape diversity and spatial pattern of ecological communities are often altered by disturbance. Spatial patterns (spatial autocorrelation) in species diversity are thus expected to change with disturbance.
Objective
When examining spatial patterns, ecologists traditionally lump positive and negative spatial autocorrelation into the overall spatial autocorrelation. By contrast, here we aim to understand disturbance effects on both positive and negative spatial autocorrelation of species richness and evenness, which may be related to environmental filtering and restricted dispersal, and to competition, respectively.
Methods
For 8 years, we monitored the spatial autocorrelation in species richness and evenness of riparian plant communities in both uncut control and experimentally clearcut sites in the boreal forest of Alberta, Canada. The overall spatial autocorrelation for each of these two indices of diversity was separately decomposed into the components of positive and negative spatial autocorrelations through eigendecomposition of the spatial weighting matrix.
Results
Negative spatial autocorrelation in richness and evenness were more pronounced in the clearcut than uncut sites, although positive spatial autocorrelations in all indices of diversity remained unchanged. Effect of disturbance was not detected on the overall spatial autocorrelation.
Conclusions
Disturbance increases negative spatial autocorrelation in species richness and evenness, with a stronger increase in evenness than richness, which underscores the importance of competition in structuring post-disturbance riparian communities. Our results also highlight the need for assessing positive and negative spatial autocorrelation and different aspects of diversity separately in understanding disturbance effects on the spatial pattern, or identifying processes from patterns.