Forest to Savanna? Assessing Potential Trajectories of Restored Tropical Forests Under Climate Change

• Grant number: 25/26020-5
• Support Opportunities: Scholarships abroad - Research Internship - Post-doctor
• Start date: February 23, 2026
• End date: February 22, 2027
• Field of knowledge: Agronomical Sciences - Forestry Resources and Forestry Engineering - Nature Conservation
• Principal Investigator: Pedro Henrique Santin Brancalion
• Grantee: Fabio Antonio Ribeiro Matos
• Supervisor: David Edwards
• Host Institution: Escola Superior de Agricultura Luiz de Queiroz (ESALQ). Universidade de São Paulo (USP). Piracicaba , SP, Brazil
• Institution abroad: University of Cambridge, England
• Associated to the scholarship: 24/07707-7 - Recovery of biomass and tree diversity along a forest restoration continuum, BP.PD

Abstract

The Atlantic Forest in the State of São Paulo is a biodiversity hotspot that has undergone extensive deforestation, leaving only a fraction of its original cover, particularly in the inland seasonal forests. Tropical forests in this region are regenerating through both active and passive restoration. A central question for restoration is to understand how climatic variables and land-use legacies influence the composition, diversity and biomass of tree communities, which is crucial for predicting forest resilience under climate change. This is particularly important because, in some areas, restored forests may undergo savannisation - a transition towards more open, savanna-like conditions - driven by dry climates, nutrient-poor soils, recurrent fires, competition with C4 grasses and herbivory associated with cattle grazing. We hypothesise that the interaction among climatic, edaphic and biotic factors may drive some forests towards alternative states resembling savannas, resulting in predictable changes in their communities. To address this question, we will assess the effects of climate and land-use legacies on tree diversity and carbon-stock recovery along climatic gradients, considering restoration history and previous vegetation cover (i.e., Atlantic Forest and Cerrado). Data from permanent 900 m² plots, integrated with soil properties, landscape metrics, fire occurrence and grass cover, will be used to generate models and maps of biomass, diversity and patterns of community convergence or divergence. (AU)