Résumé
Implementing operational tools to assess and prioritize the impact of land use on the ecological quality of rivers means relying on the availability of water quality data from monitoring networks. In addition, these tools must account for the nested spatial scales and dynamics, in particular to ensure a good estimate of upstream-downstream relations. The analysis of nested spatial scales and temporal dynamics is necessary to ensure a good overview of these relationships. We have developed a three-stage methodology: data collection and pretreatment, detection of spatial dependence (or not) of water quality scores measured for two successive stations, and modeling the relationship between water status and pressure. Our results, based on two sub watersheds of the Saone (France), identify the human pressures influencing the most the ecological quality of rivers and show that the spatial dependence is not systematic. We subsequently consider extending our approach to other water quality parameters and the Saône basin as a whole.