Amazonian Landscapes in Transition – ALT

Amazonian forests are of major significance to climate change, being a large stock of carbon and essential to other major ecosystem services, including the provisioning of forest resources. However, nearly a quarter of Amazonia’s surface has been converted into human-modified landscapes, and a major drought-related threat is suggested by models. An ecosystem shift would have serious and adverse impacts on ecosystem services of both natural and production forests, and more generally on Amazonian socio-ecosystems. Forest multifunctionality and the vulnerability to climate change are yet to be fully accounted for in tropical forest management plans. A fundamental question is whether Amazonian forests will be able to withstand the simultaneous impacts of climate and local anthropogenic disturbances. This question is of global relevance, as France is directly involved in negotiations on the potential of its forests to both offset carbon emissions and mitigate the biodiversity crisis.
Amazonian Landscapes in Transitions (ALT) seeks to address this key challenge in French Guiana. The approach is to calibrate detailed forest dynamic models to generate maps of ecological indicators that will account for forest vulnerability, in addition to current forest indicators. Such territorial model-based estimates of ecological indicators for French Guiana will be associated with uncertainty maps. This strategy depends on the consolidation of a solid knowledge based of accurate forest inventories, including trees in the forest understory, which condition the future of forest regeneration, on information on plant functional diversity, and on patterns and processes of tree mortality.
ALT proposes to perform virtual experiments aimed at exploring the processes underpinning the regeneration potential of natural and production forests, using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) an innovative vegetation demographic model, and the individual-based spatially explicit forest dynamic model TROLL. The two models are complementary, and their calibration requires similar input data (plant physiological traits, climatic and edaphic conditions, and ecosystem parameters). The models will be upscaled to provide maps of ecological indicators across 80,000 km2 of forest.
ALT will seek to fill key knowledge gaps for the modeling of tropical forests. First, understanding understory tree dynamics is essential to predicting the sustainability of production forests. Building on the large experience of the consortium, ALT will set up detailed forest understory inventories and will decipher the spatial distribution of saplings with respect to soil conditions and microclimate. Second, there is a need to improve the representation of tropical tree functional diversity in models. ALT will expand trait databases already collected by the scientific community to include understory species. Third, landscape patterns of canopy tree mortality remain poorly documented. Repeated-pass airborne lidar scanning offers a detailed description of canopy gap formation, and it will be complemented with lower resolution radar remote sensing, which provides near-real time canopy changes across the entire territory. ALT will thus harness the potential of remote sensing to unravel the patterns of tree mortality.
Finally, ALT will explore the impact of management strategies on ecological indicators, based on prescribed forestry operations and socio-economic scenarios. Our results will be directly relevant to the nationally determined contributions of the UNFCCC, and as contributions by France to upcoming reports to the IPBES.