General Lake Model (GLM) Multi-Lake Comparison Project – Phase 1 Physics

In this study a network of 25 “Lake Modellers” from around the world have come together to “road-test” an open source, collaboratively coded 1-D lake model, GLM (General Lake Model) over 20 different lakes spread over 10 different countries. We aim to over-come the difficulty in comparing model applications from different sites where different model approaches are adopted, and support a detailed comparative analysis. In particular we were motivated to understand how our model assumptions stand up across a diverse range of lakes. GLM adopts a flexible Lagrangian layer structure which allows for layers to change thickness by contracting and expanding in response to inflows, outflows, mixing and surface mass fluxes. The model accounts for the surface fluxes of momentum, sensible heat and latent heat, energy fluxes to account for shortwave and longwave radiation and algorithms to account for ice and snow The surface mixed layer deepening rate is determined by a balance between available kinetic energy (from wind stirring, shear production etc.) and potential energy required to undergo mixing. Any number of lake inflows and outflows can be specified and are applied daily. We chose a range of metrics to assess model performance at each site, including the use of Lake Analyser (Read et al. 2011) to compare metrics of stratification. Once calibrated, a comparative analysis was then conducted between lake model application, in relation to sensitivity analysis, calibration parameters, validation model skill assessment and stratification metrics. A set of statistical evaluations relates patterns of model fit and stratification metrics against lake characteristics. The outcome of the study is an improved knowledge of the strength and limitations of the GLM, a standardized calibration and globally relevant parameter set, and improved understanding of characteristic drivers and forcing patterns of stratification in lakes. The analysis highlights how the ability of the model to capture stratification varies based on the lake morphometry and the prevailing climate. Having a globally validated model will allow us in the future to better undertake global change assessments of lake and reservoir systems. Many more people and lakes than listed here are involved.