Skip to Main content Skip to Navigation
Journal articles

Démarche de validation régionale par avis d'experts du modèle MESALES d'estimation de l'aléa érosif

Abstract : Erosion risk has been estimated at the European level through the PESERA program (Kirkby et al., 2003), as well as at the national level with the help of a cartographic model referred here as the MESALES model (Le Bissonnais et al., 1998 ; Le Bissonnais et al., 2002). The latter model integrates the erosion parameters (land use, crusting, slope, erodibility and climate) into a decision tree which forms a hierarchy of the input factors and weights their classes. The objective of the present work was to call upon expert knowledge to validate, at the regional scale of Brittany (NW France), the erosion risk evaluation map derived from the MESALES model at the national scale (Colmar, 2006). Interviews with soil experts and farming consultants allowed to delineate areas over- or underestimated by the national scale model and to suggest improvements of the input data. As to their analysis the erosion risk predicted by the model is underestimated in the eastern part of the region and does not match with their expertise in some other areas. In a second step we increased the spatial and/or temporal resolution of the topographic and the land use data. Both methods improved the results and we were able to produce a new regional map of erosion risk for Brittany.
Document type :
Journal articles
Complete list of metadata

https://hal.inrae.fr/hal-02669230
Contributor : Migration Prodinra <>
Submitted on : Sunday, May 31, 2020 - 2:36:35 PM
Last modification on : Friday, June 4, 2021 - 2:37:52 PM

Identifiers

  • HAL Id : hal-02669230, version 1
  • PRODINRA : 48757

Citation

A. Colmar, Christian Walter, Yves Le Bissonnais, Joël Daroussin. Démarche de validation régionale par avis d'experts du modèle MESALES d'estimation de l'aléa érosif. Etude et Gestion des Sols, Association Française pour l'Etude des Sols, 2010, 17 (1), pp.19-32. ⟨hal-02669230⟩

Share

Metrics

Record views

30