Spatiotemporal drivers of crop pests and pathogens abundance at the landscape scale
The agricultural landscape has been recognized as a relevant scale for the control of bioagressors (pest and pathogen) epidemics. In this study, we assess the consistency of the impacts of semi-natural areas and host crop fields in the landscape on bioagressor abundance in cultivated fields. Our approach relied on the analysis of systematic French epidemiosurveillance data spanning 9 years of observations over two third of the French territory. 30 majors bioagressors of arable crops (wheat, barley, maize, potato, oilseed rape and sugar beet) were jointly studied with landscape composition as described by official CAP data and woody vegetation maps. Beyond the simple correlation between landscape composition and bioagressors abundance, we controlled for the presence of the host crop the former years in the plot and for bioagressors former prevalence. Our findings highlighted that landscape simplification around a same crop does not in general favor epidemic outbreaks during the current growing season, but rather that its harmful effect lie in the season ahead. The effects of semi-natural areas were, however, very contrasted between organisms. This study used available tools and data to quantify automatically in an interpretable way the general impact of landscape components. Such findings have relevant implications for the design of bioagressor management strategies at the landscape scale. ResumeLe paysage agricole est reconnu comme une echelle pertinente pour la gestion des bioagresseurs, insectes arthropodes et maladies cryptogamiques. Dans cette etude, nous avons evalue la coherence et la regularite de limpact des espaces semi-naturels et cultures hotes dans le paysage sur labondance des bioagresseurs dans les parcelles cultivees. Cette approche a repose sur lanalyse de donnees nationales depidemiosurveillance regroupant 9 ans dobservations. 30 bioagresseurs du ble, du colza, de lorge, de la pomme de terre, du mais et de la betterave ont ete parallelement etudies a laide de cartes dassolement des cultures (RPG) et de la vegetation arboree (BD Topo (R)). Au-dela des simples correlations entre la composition du paysage et labondance de bioagresseurs, nous avons pris en compte leffet de la surface en culture hote et labondance des bioagresseurs de lannee precedant les observations. Les resultats ont mis en evidence quune importante surface dune meme culture hote dans le paysage ne favorisait pas systematiquement les epidemies de bioagresseurs durant la meme annee mais plutot durant la suivante. Les effets du paysage etaient quant a eux tres inconsistants. Cette etude controle des bioagresseurs a lechelle du paysage.