6533b86ffe1ef96bd12cd8f7

RESEARCH PRODUCT

Multiscale analysis of the effects of the landscape context and of the agricultural practices on the pest pressure and on the use of insecticides in annual crops and polyculture farming

subject

description

In the context of the need to reduce the dependence of agricultural systems on the use of plant protection products, the identification of factors affecting the capacity of farmers to reduce the use of pesticides is a central issue. It has now been shown that, in certain production contexts, this reduction is possible without economic loss for the farms, by modifying the cropping systems. It can be hypothesised that this capacity to reduce pesticides will depend on local pest pressure, which is itself potentially affected by the regional and local landscape context of cropping systems. In this thesis, we quantified the effects of landscape context on pest pressure and pesticide use levels in field crops, at different scales and spatial resolutions, by mobilising two national databases, AgroSyst, which describes the cropping systems implemented on the 2000 farms of the national DEPHY-Ferme network, and Epiphyt, which gathers data from the national pest monitoring network and cartographic databases to characterise the agricultural and landscape context of the pest observation points and cropping systems of the DEPHY-Ferme network. . In the first chapter, we quantified at the national level the inter-regional variability of biotic pressures observed in 2018 in 181 small agricultural regions. This variability is explained by meteorological differences between regions but also by the major agricultural and landscape characteristics of each small agricultural region, with pressures inversely related to the proportion of grassland in the region. In the second chapter, we analysed the effects of regional pressure and landscape characteristics of the commune on the level of insecticides in different types of cropping systems. We show that insecticide use increases with regional pest pressure but decreases with increasing heterogeneity of the surrounding landscape. This landscape effect fades when pest pressure is high and differs according to the type of cropping system studied. The third chapter asks the same question in the specific case of insecticide use in oilseed rape, by integrating the effect of agricultural practices into the analysis. The results confirm the fact that insecticide use decreases when landscape heterogeneity increases but the effect of regional pest pressure is no longer detected. In addition, a higher insecticide use in oilseed rape was observed in cropping systems with tillage. The results obtained in this thesis support the hypothesis that the regional and/or local landscape context affects pest pressure and insecticide use levels. It therefore seems important to integrate these territorial levers in the strategies implemented to reduce pesticide dependency.