Spatially-explicit biodiversity models help unpack the varying effectiveness of Agri-Environment Measures

Agri-Environment Measures (AEM) are the primary policy tools under the European Union’s Common Agricultural Policy to combat farmland biodiversity loss, yet their effectiveness is highly variable. This synthesis paper summarizes findings of a doctoral thesis investigating drivers of variation in modelled biodiversity responses to AEM: (1) the use of different land-use intensity (LUI) metrics, (2) the types of AEM and their species-specific scale of effect, (3) landscape structural complexity across different regions. First, virtual species (i.e. species with known species-environment relationships) were used to explore how using alternative LUI metrics in biodiversity models influences the estimated species-AEM relationships. Second, bird observations for the Mulde River Basin in Germany were used to model farmland bird responses to different AEM across scales. These relationships varied across species and AEM type, but were generally strongest at the landscape level as compared to locally. Lastly, landscape-moderated effects of AEM on red-backed shrike (Lanius collurio) occurrences were analyzed across three study regions in Germany, Spain and Czechia. Positive shrike-AEM associations were stronger in structurally simple compared to complex landscapes, but this effect was inconsistent across regions. These findings exemplify species-, scale- and landscape-dependent AEM effects, and support AEM’s spatial targeting and regional tailoring.