The largest 1% of tropical trees store half the forest’s carbon, but no one knows exactly what kills them when they die. This matters because the land carbon sink—the ability of forests to absorb CO₂ from the atmosphere—depends on these giant trees staying alive. When they die, their stored carbon returns to the air. Current monitoring rarely captures the actual cause of death for these trees, leaving a critical gap in climate models. The Gigante project will fill that gap by combining frequent satellite and drone imagery with on-the-ground field surveys across 7,500 hectares of tropical forest. Researchers will identify recently dead giants, determine what killed them (drought, wind, lightning, disease), and link those deaths to climate and forest structure data. If successful, the project will produce validated models that predict where and why giant trees die, and how those deaths affect carbon storage across tropical forests worldwide. This is fundamental science with a direct application: improving the accuracy of global carbon cycle models that underpin climate policy and emissions accounting. Better predictions of forest carbon dynamics help governments and conservation organisations make informed decisions about land use and climate mitigation.
View original technical description
The land carbon sink depends on the persistence of giant tropical trees. The largest 1% of trees store half the carbon in tropical forests and their deaths release this carbon back to the atmosphere, but we do not know what kills these trees because their deaths are rarely described. A novel sampling strategy is needed to effectively monitor the life and death of giant tropical trees. Gigante will integrate remote sensing and frequent field surveys to answer: (1) What kills giant trees and how do their mortality rates vary over space and time? (2) What are the risk factors underlying variation in giant tree mortality rates? and (3) How does giant tree mortality risk influence pantropical carbon stocks? We will locate giant tree mortality events using multi-platform, high-frequency remote sensing of 7,500 ha across five tropical forest super sites. These data will facilitate targeted field surveys using detailed state-of-the art protocols to assign proximate agents of mortality to recently dead trees in an unprecedently large field study. We will integrate these data with information about climate, topography, canopy structure, and tree traits to validate mechanistic models of tree mortality risk. Finally, combining these risk models with forest plots and satellite LiDAR, we will evaluate how drivers of giant tree death predict spatial variation in forest dynamics, structure, and carbon storage. The validation of geospatial relationships will allow us estimate the contributions of giant tree mortality to pantropical forest carbon stocks.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know