Methodology for land use change scenario assessment for runoff impacts: A case study in a north-western European Loess belt region (Pays de Caux, France): A case study in a north-western European Loess belt region (Pays de Caux, France)

International audienceChanges of agricultural land use often induce changes in hydrological behavior of watersheds. Hence, effective information regarding runoff responses to future land use scenarios provides useful support for decision-making in land use planning and management. The objective of this study is to develop a methodology to assess land use change scenario impacts on runoff at the watershed scale. This objective implies translating qualitative information from scenarios into quantitative input parameters for biophysical models. To do so, qualitative information from scenarios should be quantified and spatialized. The approach is based on the combination of 2015 local land use change scenarios (SYSPHAMM method) based on local stakeholders expertise, a model of spatio-temporal allocation of crops to fields (LandSFACTS model) and a watershed runoff model (STREAM model). The study was conducted for regions underlain by silty loamy soils scattered across Northern Europe. It was applied on the Saussay watershed in Upper Normandy (France). The approach is illustrated through runoff assessment of one of the land use change scenarios (characterized by the ending of the set-aside obligation and the disappearance of dairy farming). This scenario appeared relevant for local stakeholders. The methodology presented suggests that assessing local land use scenarios in terms of runoff requires taking into account crop allocation diversity allowed by farmers' decision rules. This requirement accounts for runoff variability at the watershed outlet since crops spatial distribution throughout the watershed, depending on farmers' specific decision rules (i.e. cropping systems), strongly condition runoff phenomenon. Besides, choices regarding scenario implementation (quantification and spatialization) should to be made according to those cropping systems. Accordingly, taking into account crop allocation diversity due to farmers' cropping systems shows that there is a variability in terms of runoff at the watershed outlet (from 19 478 m3 to 35 004 m3 for the winter period and a low-intensity rainfall event for example). This variability can then be explored with local decision makers with the aim of finding solutions reducing runoff risks. The proposed approach provides a useful source of information for assessing the responses of surface runoff of future land use changes. Such scenarios providing impact assessment on runoff should encourage both local policy makers and local actors to actively discuss the future of land use in Upper Normandy