Publications des agents du Cirad


Rainfall-human-spatial interactions in a salinity-prone agricultural region of the Western Australian wheat-belt

Asseng S., Dray A., Perez P., Su X.. 2010. Ecological Modelling, 221 (5) : p. 812-824.

To explore the complexity of temporal and spatial dynamics of an agricultural landscape under various external and internal driven scenarios, SimKat, an agent-based model, has been developed with the simulation computer program CORMAS. This model combines simplified bio-physical processes of land cover, regional dry-land salinity changes, rainfall impact on productivity, farm profitability and the complexity of land-use decisions of individual farmers in a dry-land agricultural catchment (no irrigation). In this model, simulated farmers formulate individual decisions dealing with land-use changes based on the combined performance of their past land cover productivity and market returns. The willingness to adapt to market drivers and the ability to maximize returns vary between farmers. In addition, farmers in the model can demonstrate various attitudes towards dry-land salinity mitigation as a consequence of experiencing and perceiving salinity on their farm, in the neighborhood or across the entire region. Consequently, farmers can adopt land cover strategies aimed at reducing dry-land salinity. Aggregating the simulated individual behavior of farmers combined with historical rainfall and market price records, reproduced similar aggregated trends of land-cover changes, regional salinity change and farm number decline as observed in the last 20 years in the subject region (Katanning, Western Australia). Using the model in an initial exploratory study on the impact of rainfall variability and change highlighted the importance of average rainfall decline and the widespread willingness of farmers to adapt perennial vegetation in their farming systems to combat regional dry-land salinity. Rainfall decline, as one of the observed consequences of climate change in this region, can also lead to prolonged sequences of dry seasons in the future. Adaptation by farmers to sequences of dry and wet seasons, rather than an average trend in rainfall, seems to be critical for farm survival in this region. Intensifying cropping during wet seasons to maximize farm returns can increase vulnerability in subsequent periods of dry seasons, in particular where alternative income from pasture and sheep production has been lost in the adaptation process. (Résumé d'auteur)

Mots-clés : changement climatique; agriculture; modélisation des cultures; australie

Thématique : Méthodes mathématiques et statistiques; Economie et politique foncières; Météorologie et climatologie

Documents associés

Article de revue