Remote sensing-based forest modeling reveals positive effects of functional diversity on productivity at local spatial scale

Forest biodiversity is critical for many ecosystem functions and services. Yet, it remainsuncertain how plant functional diversity influences ecosystem functioning across environmentalgradients and contiguous larger areas. We integrated remote sensing and terrestrialbiosphere modeling to explore functional diversity–productivity relationships atmultiple spatial scales for a heterogeneous forest ecosystem in Switzerland. We initializedforest structure and composition in the ecosystem demography model (ED2) througha combination of ground-based surveys, airborne laser scanning and imaging spectroscopyfor forest patches at 10×10-m spatial grain. We derived morphological and physiologicalforest traits and productivity from model simulations at patch-level to relate morphologicaland physiological aspects of functional diversity to the average productivityfrom 2006–2015 at 20×20-m to 100×100-m spatial extent. We did this for model simulationsunder observed and experimental conditions (mono-soils, mono-cultures and monostructures).Functional diversity increased productivity significantly (p < 0.001) acrossall simulations at 20×20-m to 30×30-m scale, but at 100×100-m scale positive relationshipsdisappeared under homogeneous soil conditions potentially due to the low beta diversityof this forest and the saturation of functional richness represented in the model.Although local functional diversity was an important driver of productivity, environmentalcontext underpinned the variation of productivity (and functional diversity) at largerspatial scales. In this study, we could show that the integration of remotely-sensed informationon forest composition and structure into terrestrial biosphere models is importantto fill knowledge gaps about how plant biodiversity affects carbon cycling andbiosphere feedbacks onto climate over large contiguous areas.