Modelling forest biodiversity using remote sensing

Forests are among the most diverse ecosystems on our planet. Their various layers, from the soil to the canopy, contain thousands of organisms, including plants, birds, mammals, microbes, and protists. In particular, microbial and infaunal communities show remarkable diversity and play a critical role in key ecosystem services such as nutrient cycling and carbon sequestration. However, assessing the diversity of these communities, especially across large spatial scales, remains challenging due to the intensive fieldwork and laboratory analyses required. Recent advances in remote sensing and machine learning now enable the upscaling of point-sample biodiversity data to landscape scales. Remote sensing products are especially valuable for such assessments as they capture broad spatial patterns and provide consistent monitoring of environmental changes over extensive areas with high efficiency. These products offer detailed insights into forest canopy characteristics, including biochemical properties (e.g., chlorophyll and water content) and biophysical attributes such as leaf area index (LAI), thereby shedding light on habitat conditions. This thesis topic aims to integrate microbial and/or infaunal occurrence data with remote sensing products to explore the diversity of mixed forest ecosystems at landscape scales.