Environmental thresholds for drought- and heat-related tree mortality

Projektbeschreibung

Bessere Einblicke in die Baumsterblichkeit bei Trockenheit und extremer Hitze

Eine wichtige ökologische Frage lautet, wie Bäume häufigere Dürreperioden und extreme Hitze überleben können. Die weltweite Baumsterblichkeit ist angestiegen, insbesondere während der schlimmsten Dürreperiode in Europa seit 500 Jahren, die im Jahr 2022 stattfand. Obwohl unser Verständnis der Ursachen begrenzt ist, sind wir heute in der Lage, die Muster und Arten des Baumsterbens in großen Gebieten zu analysieren. Das ERC-finanzierte Projekt DRYTREE zielt darauf ab, die Umweltfaktoren zu erforschen, die das dürre- und hitzebedingte Baumsterben auf Artniveau in verschiedenen Waldbiomen beeinflussen. Es wird Fernerkundung und tiefes Lernen vorantreiben und Daten zum Baumsterben analysieren, um wichtige artspezifische Schwellenwerte und Schwachstellen zu ermitteln. Diese Forschung zielt darauf ab, das Verständnis für die Ursachen der Baumsterblichkeit zu verbessern und Forstverwalter und politische Entscheidungsträger beim Schutz gefährdeter Wälder zu unterstützen.

Ziel

One of the most pressing ecological questions is how trees and forests can survive in increasing intensity and frequency of droughts and extreme heat. Wide-spread drought-related tree mortality has been witnessed globally over the last decade and in 2022, Europe suffered from the worst drought in 500 years. Understanding and predicting the impact of drought and heat on tree mortality is limited due to lack of knowledge on the environmental conditions that lead to tree mortality.

For the first time, it is feasible to quantify spatial and temporal tree mortality patterns and capture tree structure and species over large geographic regions at the individual tree-level. This enables me to aim to uncover environmental thresholds and key environmental drivers of drought- and heat-related tree mortality at the species-level for various forest biomes. Use of state-of-the-art remote sensing and deep learning methods allows me to capture where, when and what kind of trees (species, structure) have died for tens of millions of trees to increase our understanding of spatial and temporal tree mortality patterns. My approach uses laser scanning data to provide detailed tree 3D characterization and calculation of tree position within tree community (competition) and the landscape (water availability, microclimate). Then, combining these variables with information on tree xylem vulnerability, pest insects, soil temperature and climate, we can ultimately reveal species-specific environmental thresholds and key drivers of drought- and heat-related tree mortality. This research will open new horizons, bringing ecophysiology, remote sensing, forest ecology and entomology together, and developing methods to quantify drivers of tree mortality in much greater depth than has been possible to date. From these findings, we will be able to inform forest managers and policy makers which forests are at risk for increasing the resilience of future forests.