Disentangling Sources of Uncertainty in CLM5 Model Predictions: Water, Energy, and Carbon Fluxes at European Observation Sites

This study evaluates the performance of the Community Land Model version 5 (CLM5) in simulating key ecosystem variables, including soil moisture (SM), evapotranspiration (ET), sensible heat flux (H), net ecosystem exchange (NEE), and gross primary production (GPP), using ensemble simulations across 14 European measurement sites. Key Highlights: Comprehensive Evaluation: The study dissects the influence of perturbed soil, vegetation, and atmospheric parameters on model outputs. Performance Metrics: Results show that CLM5 captured soil moisture variability well in most sites but struggled with carbon and energy fluxes, particularly in cold and boreal ecosystems. Uncertainty Quantification: Ensemble spread effectiveness was evaluated using metrics like Coverage Percentage (CP) and Spread-Skill Ratio (SS). Insights for Model Development: Key areas for improvement were identified as structural errors and parameterization gaps in CLM5, such as energy partitioning biases and underrepresentation of snow dynamics.