Multi-scale evaluation of a 3D lake model forced by an atmospheric model against standard monitoring data

Evaluating a three-dimensional lake model requires large datasets of many variables, including velocity fields, that are seldom available. Here we discuss how to assess the performance of a model at multiple scales (in time and space) with data from standard monitoring systems, i.e., mostly limited to water temperature. The modeling chain consists of a lake hydrodynamic model (Delft3D-Flow) forced by an atmospheric model (WRF, Weather Research and Forecasting). The two models are tested on the case study of Lake Garda (Italy), where a comprehensive dataset of atmospheric and water temperature observations is available. Results show that a consistent picture of the inherent dynamics can be reproduced from a heterogeneous set of water temperature data, by distilling information across diverse spatial and temporal scales. The choice of the performance metrics and their limitations are discussed, with a focus on the procedures adopted to manage and homogenize data, visualize results and identify sources of error.

Automated summary

The paper discusses how to evaluate the performance of a three-dimensional lake model using limited water temperature data at multiple scales, and demonstrates through a case study that a consistent picture of the inherent dynamics can be extracted from diverse spatial and temporal scales of water temperature data.