Synergies and tradeoffs in natural regulation of crop pests and diseases under plant species diversification

It has been suggested that increasing plant species diversity (PSD) in agroecosystems at different spatiotemporal scales reduces the impacts of crop pests and diseases as well as the dependence on synthetic plant protection products. This principle was applied to a range of tropical case studies. These studies involved various pests and pathogens with contrasting life history traits, different cropping systems (a cereal crop in conservation agriculture, vegetable crops in rotational and trap cropping systems, perennial crops in agroforestry) and various spatial scales of PSD deployment (field and farmscape). Here we review the outcomes of these studies, and discuss the lessons learned regarding synergies and tradeoffs associated with regulation effects provided by PSD. The major points are: 1) results contributed to solve local crop pest and disease problems such as bacterial wilt on tomato in Martinique, scarab beetles and witchweed on upland rice in Madagascar, fruitworms on tomato in Martinique and okra in Niger, fruit flies on cucurbit vegetables in Reunion, mirid bugs and black pod rot on cocoa in Cameroon, berry borer and leaf rust on coffee in Costa Rica; 2) the importance of cross-cutting issues regarding green manure, cover crops or companion plants across case studies at the field scale involving below-ground and aerial processes, were highlighted, particularly that of the within-species genetic variation of these plants; 3) based on the fruitworm/tomato case study, a dynamic and spatially-explicit individual-based model was developed as a generic tool to improve understanding of system functioning by assessing infestation patterns in response to main crop/trap crop relative attractiveness, spatiotemporal deployment of the main crop/trap crop and insect behavioral traits; 4) tradeoffs were highlighted regarding pest and disease complex management, single-option pest and disease control via several pathways based on a single PSD-deployment measure and other