Ecological filtering shapes the impacts of agricultural deforestation on biodiversity.

The biodiversity impacts of agricultural deforestation vary widely across regions. Previous efforts to explain this variation have focused exclusively on the landscape features and management regimes of agricultural systems, neglecting the potentially critical role of ecological filtering in shaping deforestation tolerance of extant species assemblages at large geographical scales via selection for functional traits. Here we provide a large-scale test of this role using a global database of species abundance ratios between matched agricultural and native forest sites that comprises 71 avian assemblages reported in 44 primary studies, and a companion database of 10 functional traits for all 2,647 species involved. Using meta-analytic, phylogenetic and multivariate methods, we show that beyond agricultural features, filtering by the extent of natural environmental variability and the severity of historical anthropogenic deforestation shapes the varying deforestation impacts across species assemblages. For assemblages under greater environmental variability-proxied by drier and more seasonal climates under a greater disturbance regime-and longer deforestation histories, filtering has attenuated the negative impacts of current deforestation by selecting for functional traits linked to stronger deforestation tolerance. Our study provides a previously largely missing piece of knowledge in understanding and managing the biodiversity consequences of deforestation by agricultural deforestation.

Predation Pressure in Tea (Camellia sinensis) Plantations in Southeastern China Measured by the Sentinel Prey Method.

Tea (Camellia sinensis) is an important food product with thousands of years of human use. Being a non-washable food, no pesticide residues are allowed, which increases the importance of natural means of plant protection. Predation, a component of natural pest control, is an important contributor to this, but its level and sustainability are not known in most of the areas of tea production. We quantified predation intensity using the artificial sentinel prey method in a tea-growing landscape containing remnants of the original forest vegetation in Fujian Province, China. The most common predators were chewing arthropods (49.8% of predation events) and birds (48.1%). Overall, predation rates at the edges of forest fragments (18.9% d-1) were lower than either in fragment interiors (25.4%d-1) or in the surrounding tea plantations (19.2-24.1%d-1). Arthropod predation was higher inside, and at the edge of, forest fragments than within plantations, and generally decreased with increasing distance from a fragment edge, indicating limited spillover of arthropod predators from the native habitat remnants to the cultivated matrix at the local scale. Bird predation, though, showed a different trend: it was lower on the inside of forest fragments than in the tea planation, and bird attack rates increased at increasing distances (up to 40 m) from the forest fragment edge. We also found a reciprocal relationship between attack rates by birds and arthropods, suggesting intra-guild predation. Measures protecting arthropod natural enemies could increase the combined pest suppression effect, contributing to pesticide-free tea production in China.