Spatial non-stationarity effect of determinants regulates variation in amphibian species richness

•The factors with a higher spatial heterogeneity explain richness more effectively.•Human impacts increase the sensitivity of amphibians to environmental changes.•Climate factors act at a smaller scale than topography and anthropogenic impacts.•Neuralnetworks deepenthe understanding of ecologically...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Ecological indicators 2023-06, Vol.150, p.110268, Article 110268
Hauptverfasser: Wu, Zhaoning, Borzée, Amaël, Qian, Tianlu, Dai, Wenyu, Li, Siqing, Wang, Jiechen
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•The factors with a higher spatial heterogeneity explain richness more effectively.•Human impacts increase the sensitivity of amphibians to environmental changes.•Climate factors act at a smaller scale than topography and anthropogenic impacts.•Neuralnetworks deepenthe understanding of ecologically consistent regions. Despite the variation in amphibian biodiversity being often explained through environmental variables, the effects of spatial non-stationarity factors are too-often ignored as significant geographic characteristics, especially at large scales. Here, using a spatial regression approach through a multiscale analysis, we explored the spatial heterogeneity and scale effects in the impacts of environments on amphibian species richness across China. We showed that the impacts of variables varied with regions, and the individual scales specific to each variable were negatively correlated with effect sizes. We then demonstrated that climate variables, followed by topography, showed a high explanatory power for species richness in most areas, while the effects of precipitation and temperature were characterized by high geographical heterogeneity. In addition, elevation was positively related to local species richness in most plains, while being a main negative variable to species richness in the western highlands. The analysis of geographic heterogeneity showed that the explanatory power of most variables declined with increasing elevation. Although anthropogenic impacts contributed less than climatic variables, they significantly increased the sensitivity of amphibian species to environmental variations. Finally, to measure the aggregation pattern of heterogeneous effects of variables on species richness, we used a neural network to identify ecoregions regulated by similar variables and determined the presence of four ecologically consistent regions. Our findings provide further evidence supporting spatially variables regulators of amphibian diversity.
ISSN:1470-160X
1872-7034
DOI:10.1016/j.ecolind.2023.110268