Unlocking past fire components and testing if they triggered tipping points in tropical ecosystems – TipTropFire

Tropical forests and savannas are crucial for global biodiversity, carbon storage, and local livelihoods, yet their responses to climate change remain difficult to predict. A key challenge lies in resolving the mechanisms controlling their distribution and stability. While some theories suggest that hydroclimate (rainfall, relative humidity) primarily determines forest-savanna transitions, the Alternative Ecosystem States (AES) framework posits that these ecosystems can coexist under similar climatic conditions, with fire acting as a key factor in maintaining or shifting between them. However, long-term empirical evidence testing fire’s role in these transitions remains limited. Resolving this debate is crucial for anticipating how tropical landscapes will respond to future environmental changes and for guiding conservation and management strategies.
This project seeks to test whether fire alone can drive forest-savanna transitions in West and Central Africa, or if other factors—such as climate variability or human activity—play a dominant role. If fire is a primary driver, ecosystem transitions may not be directly reversible, meaning fire management could be crucial for landscape stability and for restoration. Studying these transitions requires millennia-scale paleoecological data to capture long-term ecological dynamics, yet current methods lack the ability to provide quantitative reconstructions of past fire regimes.
To address this, we will integrate paleoecological reconstructions, FTIR-based fire proxies, and dynamic modeling to quantify past fire regimes and their role in vegetation change. Specifically, we will develop high-resolution fire reconstructions using Fourier Transform Infrared (FTIR) spectroscopy of charcoal from lacustrine sediments, coupled with fire modeling, to infer fire temperature, frequency, burned area, and ignition source. We will then compare fire histories with vegetation and climate records over the past 3,000 years, establishing timing of fire, vegetation and climate changes. Finally, a dynamic model of tree cover will be developed using these reconstructions to assess whether changes in fire regimes alone can trigger forest-savanna transitions or if additional factors must be considered. These simulations will also examine whether AES exist with respect to tree cover and if observed vegetation changes align with critical transitions and tipping points.
By combining quantitative paleo-fire reconstructions with dynamic vegetation modeling, this project will provide a new methodological framework for detecting past fire-driven transitions. Our findings can ultimately be used to improve predictions of tropical ecosystem responses to global change and inform future management and conservation strategies.