Despite meta-analysis collating local studies from around the globe, the bridge from local data to large-scale modeling is still wide open and prevents the upscaling of AFS knowledge crucial for AFS to play its part on the climate change agenda. To this regard, global land surface models (LSM) have a lot of potential but are limited by the lack of adequacy between the AFS’ complexity and their own homogeneity hypotheses. The goal of this project is to build upon plot-level AFS data to include key features of agroforestry systems in a global LSM. While model developments will be focused on implementing and testing key missing processes for our proof-of-concept Faidherbia parklands ecosystems, AFS data will be gathered both at local and global scale to prepare for future extension to other AFS. This project will thus pave the road for a process-based assessment of the climate impact of agroforestry practices.
WP1 : Data-mining for AFS modeling
WP2 : Model developments for proof-of-concept AFT