La spatialisation de Services Ecosystémiques à l'échelle locale dans un contexte de déforestation : quelles approches statistiques et quelles données ?

Abstract

The development of approaches to quantify and spatialize ecosystem services (ES) are greatly encouraged by the Millenium Ecosystem Assessment (MEA) and the REDD (Reducing Emissions from Deforestation and forest Degradation). Such approaches could allow the establishment of more appropriate and effective practices and policies for environmental management. Mapping of ES can be seen as a major management tool for places with high environmental issues. A study is conducted at the scale of three areas in the Brazilian Amazon affected by the deforestation front. Two biophysical processes delivering ES, water infiltration in soil and carbon storage in vegetation, were mapped with a spatial resolution of 30 x 30 m. Statistical models were developed in aim to link field measurements of biomass and infiltration rate in 135 sampling points and remote sensing data. These models provide an estimation of the processes for the sites and able us to identify the factors that influence them. The resulting maps facilitate the understanding of the spatial distribution of these processes and can be used as a basis to map and/or study trade-offs and synergies between the ES.

Keywords

• deforestation
• ecosystem service
• physical process
• mapping
• statistical model
• Brazilian Amazonia