Follow-up and modeling of the land use in an intensive agricultural watershed in France

Abstract

In intensive agricultural regions, monitoring land use and cover change represents an important stake. Some land cover changes in agro-systems cause modifications in the management of land use that contribute to increase environmental problems, including an important degradation of water quality. In this context, the identification of land-cover dynamics at high spatial scales constitutes a prior approach for the restoration of water resources. The modeling approach used to study land use and cover changes at a field-scale is adapted from a vector change analysis method generally applied to assess land cover changes from regional to global scales. The main objective of this study is to identify vegetation changes at the field scale during winter, in relation with crop successions. Magnitude and direction of the vector of changes with remote sensing data and GIS, calculated on a small watershed located in Western France for a six-year period (1996-2001) indicate both intensity and nature of observed changes in this area. The results allow to qualify accurately (i.e. at the scale of the field) the type of changes, to quantify them and weigh up their intensity. Then, all the results are integrated in a probabilistic model to build-up a short time land use prediction.