Inconsistent effects of landscape heterogeneity and land-use on animal diversity in an agricultural mosaic: a multi-scale and multi-taxon investigation

Context

The landscape heterogeneity hypothesis states that increased heterogeneity in agricultural landscapes will promote biodiversity. However, this hypothesis does not detail which components of landscape heterogeneity (compositional or configurational) most affect biodiversity and how these compare to the effects of surrounding agricultural land-use.

Objectives

Our objectives were to: (1) assess the influence of the components of structural landscape heterogeneity on taxonomic diversity; and (2) compare the effects of landscape heterogeneity to those of different types of agricultural land-use in the same landscape across different taxonomic groups.

Methods

We identified a priori independent gradients of compositional and configurational landscape heterogeneity within an agricultural mosaic of north-eastern Swaziland. We tested how bird, dung beetle, ant and meso-carnivore richness and diversity responded to compositional and configurational heterogeneity and agricultural land-use across five different spatial scales.

Results

Compositional heterogeneity best explained species richness in each taxonomic group. Bird and ant richness were both positively correlated with compositional heterogeneity, whilst dung beetle richness was negatively correlated. Commercial agriculture positively influenced bird species richness and ant diversity, but had a negative influence on dung beetle richness. There was no effect of either component of heterogeneity on the combined taxonomic diversity or richness at any spatial scale.

Conclusions

Our results suggest that increasing landscape compositional heterogeneity and limiting the negative effects of intensive commercial agriculture will foster diversity across a greater number of taxonomic groups in agricultural mosaics. This will require the implementation of different strategies across landscapes to balance the contrasting influences of compositional heterogeneity and land-use. Strategies that couple large patches of core habitat across broader scales with landscape structural heterogeneity at finer scales could best benefit biodiversity.