1.Introduction The extensive use of pesticides currently constitutes a major environmentaland public health issue (Inserm, 2013; Berardi Tadié and Bonvarlet, 2019).several indicators have been used to evaluate pesticide pressure: TFI,NODU and QSA. Other generic, simple and flexible indicators are used inthis context: IRSA and IRTE (Mghirbi et al., 2015) to assess the potential riskof phytosanitary products and their impacts on human health and non-target organisms. The objective of this study is to analyse plant protectionpractices according to cropping and production systems(conventional/integrated and organic) to assess and manage the riskassociated to diffuse phytosanitary pollution at field level located in thesouthwest of France. 2.Materials and Methods This study aims to define a methodological framework for using andenhancing a database of agricultural phytosanitary practices collectedbetween 2009 and 2019 in four French departments (Gironde, Tarn-et-Garonne, Gers and Hérault) to assess the impact of these practices in termsof phytosanitary pressure (TFI), health risk (IRSA) and environmental risk(IRTE) calculated by the EToPhy software (Le Grusse et al., 2014). The impact of crop treatment practices was analysed in two stages: (1) aglobal correlation analysis between the different indicators and studyingthe variability of the indicators and sub-indicators for each crop, across alldepartments, (2) analysing the phytosanitary practices of crops betweendepartments, followed by an interregional comparison of the south andsouthwest of France. Based on this analysis, a repository of phytosanitarypractices was developed to compare the use of pesticides between cropsand production systems (conventional/integrated and organic) accordingto a typology of phytosanitary practices which makes it possible to define 3levels of crop treatment practices (low input, medium input, high input). 3.Results The results of the global correlation analysis between the differentindicators according to the crops show a medium or even strongcorrelation between phytosanitary pressure (TFI) and risk (IRSA and IRTE).Overall, the more the TFI increases, the greater the risk to human healthand the environment, but at equivalent TFIs, risk levels may vary greatlydepending on the products used. These results lead us to the analysis of thevariability of crop protection practices and their relationship with climaticfactors to justify the choice of the crop treatment practices applied at farmlevel in the different departments. The crop repository shows the variability of phytosanitary practicesaccording to the cropping and production system (conventional/integrated,organic) for each crop. The results show that arboriculture consumes themost pesticides, especially apple trees. In addition, organic farming poses ahigher risk to the environment than conventional farming due to theexcessive use of copper and sulphur. The analysis of crop treatmentpractice types makes it possible to deal with the most toxic products tohuman health and the environment by determining the contribution of eachproduct to risk and pressure, and the target that it corresponds to. 4.Discussion Agri-environmental indicators are used to build tools for the analysis andmanagement of phytosanitary practices. These tools make it possible tostudy the relationship and the variability between the pressure and thepotential risk of pesticides according to crops. They also help de finepriorities for the implementation of other levers to reduce the use ofpesticides and improve the health and environmental performance at farmlevel. This work also shows the importance of developing a repository thattakes stock of the difference within phytosanitary practices between cropsand production systems at a regional and departmental level. 5.References -Berardi Tadié B., Bonvarlet C. (2019). Évaluation des pesticides et risquespour les pollinisateurs : procédures obsolètes et conflits d’intérêts. Paris(France) : Pollinis. 78p -Inserm. (2013). Pesticides. Effets sur la santé. Paris (France) : Expertisecollective, Inserm. 1014p. -Le Grusse P., Mandart E., Bouaziz A. et al. (2014). Gestion de la toxicité enzone Ramsar (TRam) : rapport final. 68 p. Rapport scientifique duProgramme Pesticides : Programme Évaluation et réduction des risques liésà l’utilisation des Pesticides. APR Pesticide 2009. -Mghirbi O., Ellefi K., Le Grusse P. et al. (2015). Assessing plant protectionpractices using pressure indicator and toxicity risk indicators: analysis oftherelationship between these indicators for improved risk management,application in viticulture. ESPR, vol. 22, n. 11, p. 8058-8074.