Using high level validation to increase lake ecosystem model reliability

Lake ecosystem models are essential for capturing ecosystem response to climate change and for studying eutrophication. Model ability to capture ecosystem dynamics is limited for various reasons. Aiming to increase confidence in simulations of future scenarios, the current study adapts a recently established four level validation framework, named CSPS (Conceptual, State, Process and System validation). Lake ecosystem model WET was calibrated for lake Kinneret and validated using the CSPS framework, inspecting 18 state variables, 11 processes and 15 emergent properties. Validation showed that most processes and emergent properties were correctly simulated with some weaknesses revealed such as light attenuation and particle dynamics. Modifying the model to overcome those weaknesses will result in increased model reliability. This conclusion would have been impossible without the process validation. We conclude that higher validation levels are necessary to receive reliable results when running process-based lake models on future scenarios.

Automated summary

Lake ecosystem models are important for understanding the effects of climate change and eutrophication. However, these models have limitations in capturing ecosystem dynamics. In this study, the lake ecosystem model WET was calibrated and validated using the CSPS framework, revealing strengths and weaknesses in the model's ability to simulate various processes and emergent properties. Modifying the model based on these findings will improve its reliability, especially when simulating future scenarios.