Cost of best management practices to combat agricultural runoff and comparison with the local populations’ willingness to pay : Case of the Austreberthe watershed (Normandy, France)

Land Use and Cover Changes (LUCCs) significantly increase the frequency of mudflows in the silty areas of north-western Europe and particularly in the NUTS 3 Seine-Maritime region (France). Predicting the effects of a range of possible LUCCs helps local authorities choose policies that can help to mitigate the risks to which local populations are exposed. In this article we build scenarios for changes in farming systems, with a 2015 horizon and with 2007 as the initial situation. These scenarios are assessed through combined biophysical and economic approaches. Two scenarios for the disappearance of dairy farming are chosen. One scenario has no public-action program (StopMilk), while the other one has a program based on the funding of best management practices (StopMilk-E). These scenarios are assessed at the small watershed scale (7 km2) in terms of both changes in farming systems and effects on runoff (use of the STREAM model). Finally, the economic evaluation of additional costs of StopMilk-E is extrapolated at the level of the Austreberthe watershed syndicate (214 km2), using the French Land Parcel Identification System (LPIS) with a spatially referenced database of cropping plans. StopMilk leads to a significant increase in runoff whereas the local public-action program proposed under StopMilk-E reduces runoff to below the 2007 level. The Austreberthe watershed residents’ willingness to pay for a 15-year program is around €395,000/year, which is not sufficient to balance the cost of the modified farming practices (€640,000/year over the first 3 years). Funding of the practices would require either subsidies from a higher level (Europe), borrowing by the community, and/or a more selective approach in the type and the location of farming practices to be funded. The method used for calculating costs at the watershed syndicate level shows the huge potential of new databases like LPIS (available in France since 2006) for the assessment of environmental issues.