Carabids are important biological control agents of weeds and other pests in agricultural fields. The carabid community is built upon direct and indirect ecological effects of landscape complexity, field management intensity and biotic components that in interaction make any prediction of community size and composition challenging.

We analyse a large-scale sample of 60 European cereal fields using Structural Equation Modelling to quantify the direct effects of field management intensity and the surrounding landscape, and their indirect effect via biotic components, on carabid diversity.

Our results highlight that direct and indirect effects of increasing landscape complexity, mediated by trophic resources, mainly affect carabids positively. Field management intensity only ever affects carabids through indirect effects that are generally negative, by suppressing standing weeds and weed seeds.

Indirect effects on granivore carabid species depended on weed seed availability whereas omnivores depended on the availability of both weed seeds and animal prey.

Synthesis and applications: A consideration of both the direct and indirect effects of landscape and field management is necessary for predicting carabid communities and with interactions. These effects, mediated via trophic resources, supports the diversity and abundance of carabid communities and their provision of ecosystem services. Our results show that promoting crop diversity and connectivity to semi-natural habitats will directly enhance carabid communities in farmland by manipulating their migration from source habitats. A reduction in field management intensity will preserve local standing weeds and weed seeds, and indirectly support carabid communities. These local and landscape modifications could contribute to improve the natural regulation of pests and weeds by carabids.

Read the material and method of the article entitled "Direct and indirect effects of landscape and field management intensity on carabids through trophic resources and weeds".

This dataset consists of 4 files:
-a weather data file (Carbonne_JAE_Weather_Data)
-a file containing biological, landscape and management data (Carbonne_JAE_Data)
-a file describing each variable (Carbonne_JAE_Data_Labels)
-a file with the R code (Carbonne_JAE_R_code)