Fire resilience of Amazonian forests: past, present and future

Objective

The challenge. Despite a global decline in burned area, total CO2 emissions from fires have increased due to more fires in previously largely fire-free and therefore fire-sensitive forests. Existing knowledge of fire impacts comes from fire-adapted ecosystems and cannot be easily translated to fire-sensitive forests. Particularly the Amazon has seen recent increase in fire frequency and intensity, with devastating local effects on the forest, regional-scale air pollution, and global CO2 rise. Aim. I aim to understand fire resilience – i.e. resistance to fire and post-fire recovery – of fire-sensitive ecosystems by studying the Amazon. I will assess fire resilience at multiple temporal scales: 1) the past to understand long-term recovery and full disturbance-recovery cycles, 2) the present to experimentally understand mechanisms underlying fire damage and mortality, and 3) the future to Amazon resilience to future fire regimes. Approach. I will integrate across temporal scales and their different disciplines. For past fire resilience (last 10,000 y), I will use paleo pollen and charcoal records. For present fire resilience, I will build a monitoring network in burned forests and perform a unique fire experiment. For future resilience, I will use the previous steps to adapt a dynamic forest model and simulate future fire resilience. I will quantify fire regime as fire frequency and intensity, which are fundamentally different aspects of fire regime but rarely tested simultaneously. To assess how climate drives fire resilience, I will study seasonally dry and wet evergreen tropical forests. To understand mechanisms underlying tree mortality and recovery, I will use physiological measurements and plant functional traits to understand species and community resilience in the past, present and future. Impact. This project will advance our understanding of resilience of fire-sensitive ecosystems, and provide the knowledge needed for future safeguarding of these ecosystems