Using a planted tree biodiversity experiment to assess remotely sensed biodiversity-ecosystem functioning relationships

Forest biodiversity plays a critical role in ecosystem functioning, yet it is undergoing rapid and often severe declines worldwide. Understanding how biodiversity loss alters ecosystem processes requires spatially and temporally explicit data on species composition, productivity, and physiological functioning. Because repeated field sampling is costly and logistically challenging, remotely sensed data have become an increasingly valuable tool for monitoring biodiversity and forest functioning at large scales. However, uncertainties remain about how well remotely sensed spectral and thermal traits capture ecological processes at the scale of individual forest plots.

In this study, we used the BIOTREE-Kaltenborn and Bechstedt tree diversity experiments in Germany—where species richness and species composition are known and controlled—to evaluate how biodiversity and species identity shape remotely sensed measures of canopy functioning. We collected airborne imaging spectroscopy, thermal data and sun-induced fluorescence (SIF) at 1-m resolution, and assessed the relationships of these remotely sensed canopy traits with ecosystem functioning, particularly tree productivity measured as cumulative basal area.

Canopy temperature tended to increase with higher tree species richness, and some species consistently exhibited warmer canopies than others. Species identity also played an important role in shaping the relationships between productivity and spectral traits. Overall, SIF was often positively associated with productivity, although this pattern varied among species. In contrast, common greenness indices (NDVI, EVI, CIre) did not always show the expected positive relationships with productivity, suggesting that additional structural or physiological factors may influence these signals. Other indices, such as the indicator for canopy water content, NDWI, or the Chlorophyll Carotenoid Index (CCI), showed mixed or weak associations, depending on the species considered.

Across species, SIF generally decreased with increasing canopy temperatures, which suggests reduced photosynthetic efficiency under warmer conditions. Greenness-related indices were generally positively related to SIF, with some indices showing stronger associations than others.

Our results demonstrate that species identity and functional traits mediate remotely sensed signals of forest functioning. These findings highlight the importance of considering species-specific spectral responses when using remote sensing to monitor biodiversity-driven variation in forest ecosystem functioning.