Forest cover and proximity to forest affect predation by natural enemies in pasture and coffee plantations differently

Biological pest control is one of the key services from which agricultural production benefits. Despite being a well-studied ecosystem service, the potential of different matrices in shaping natural enemy contributions to pest predation is not yet clear. We used an experimental approach with predation experiments to investigate whether matrices with different degrees of structural similarity to the native forest (coffee and pasture) modulated the effects of forest cover and forest proximity on predation rate from different natural enemies. The effects of landscape structure on predation rates varied with natural enemy response to matrix type. Predation rates by arthropods (the main natural enemy acting in both matrices) were higher in coffee plantations and were positively and negatively affected by landscape and local forest cover, respectively. Predation rates by birds were higher near the forest in both matrices and two times higher in coffee plantations than in pastures. Regardless of natural enemy identity, predation rates inside agricultural matrices were higher in low-contrasting matrices, and increase with forest cover and proximity to forest. Given the clear importance of matrix type in modulating the effects of forest cover and forest proximity on predation rates, we recommend that agricultural management consider reducing habitat-matrix contrasts while increasing contact and habitat cover in the landscape to improve biological pest control. Among matrix management strategies, we recommend within-farm actions that would lead to greater matrix heterogeneity and permeability, along with an increase in landscape habitat cover, such as more sustainable and wildlife-friendly agricultural systems and the restoration of protected areas within farms.

Automated summary

This study investigates the role of different matrices in shaping natural enemy contributions to pest predation. The results show that landscape structure has varying effects on predation rates depending on the natural enemy and matrix type. Predation rates were higher in coffee plantations and were influenced by landscape and local forest cover. Birds had higher predation rates near the forest and in coffee plantations. Regardless of natural enemy identity, predation rates were higher in low-contrasting matrices and increased with forest cover and proximity. These findings highlight the importance of considering matrix type in agricultural management to enhance biological pest control.