Introduction

Lake Mendota, the largest lake in Madison, Wisconsin, exhibits complex thermal dynamics with pronounced summer stratification and reverse winter stratification under ice cover. The lake monitoring benefits from extensive, high-resolution data collected by the North Temperate Lakes Long-Term Ecological Research (NTL-LTER) project.

In this study, DSI developed a hydrodynamic model of Lake Mendota’s thermal dynamics using the EFDC+ Explorer Modeling System (EEMS). Given that the choice of turbulence models, such as the Mellor-Yamada models, the k-ε model, or k-ω model, plays a critical role in simulating the vertical lake temperature dynamics, the primary objective of this study is to assess the model’s sensitivity to different turbulence closures, thus aiming to provide a valuable reference for future thermal calibration efforts.

Model Configurations

The EFDC+ model setup for Lake Mendota utilizes a high-resolution 3D grid comprising 1,405 horizontal cells (49 rows by 58 columns), as shown in Figure 1. The model uses a Sigma-Zed coordinate system with 30 layers (Figure 2), which allows for precise stratification modeling across varying depths. This structuring approach provides robust coverage from the lake’s surface to its bottom topography, ensuring accurate simulation of hydrodynamic and thermal processes.

The model runs for one year, from January 1st, 2020, through January 1st, 2021, to capture the annual cycle of thermal stratification. The EFDC+ Full Heat Balance model was selected to simulate the water surface heat exchange. Regarding the vertical mixing, four different models were tested, including Mellor-Yamada (EFDC Original Approach), Mellor-Yamada (GOTM Approach), k-ε (GOTM Approach), and k-ω (GOTM Approach).