Diversity of methodologies to experiment Integrated PestManagement in arable cropping systems: Analysis and reflections based on a European network

tIntegrated Pest Management (IPM) aims to promote physical and biological regulation strategies thathelp farmers contain populations of pests (pathogens, animal pests and weeds) and to finally reduce thereliance on pesticides. It is based on the holistic combination of multiple management measures ratherthan on the sum of single methods, each of them having only small effects on pests reduction. Thus, toanalyse the interactions between IPM measures and to evaluate the sustainability of their implementa-tion, we require an approach considering the whole cropping system (CS), i.e. a functional entity whosecomplexity is more than the sum of its parts. A network of European experiments at the CS level wasset up recently, and aimed at sharing data and expertise to enhance knowledge of IPM. Comparison ofexisting methodologies highlighted a diversity of CS designs and experimental layouts. We deduced thatthe concept of CS itself was viewed differently among scientists, and this affected experimental proto-cols. Other differences were related to the research context and objectives. Some experiments aimed toexplore very innovative strategies and generated knowledge on both their effects on the agroecosys-tem and their ability to satisfy a set of performance targets, while others aimed to provide quick adoptable solutions for local farmers in line with the current socio-economic constraints. In some researchprogrammes, the experiment was part of the CS design process -- and tested CS were regularly revisedbased on an continuous improvement loop -- while in other cases CS were kept stable across years so as toenable the evaluation of their long-term cumulative effects. A critical aspect contributing to the diversityamong CS experiments was the distinction between a factorial design of experimental CS and systemicapproaches: factorial experiments allowed quantification of the effects of each IPM component regardlessof the consistency between components defining the CS. In contrast, systemic approaches focused on theoverall evaluation of CS designed with consideration of their consistency, hence maximising their abilityto meet the objectives. Because CS experiments represent a huge investment in terms of economics andtime, preliminary reflections of the relevance of the experimental strategy is of critical importance