Advancing Coastal Hydrodynamic Modeling: Integrating D-Flow FM into the NextGen Framework for Lake Champlain's Water Level Predictions

The Office of Water Prediction (OWP) is advancing the Next Generation Water Resources Modeling Framework (NextGen) to enhance Total Water Level (TWL) forecasting for US coastal regions. The NextGen framework enables the integration of various coastal models across different domains, all connected to a unified channel network. Among these, the Semi-implicit Cross-scale Hydroscience Integrated System Model (SCHISM) and D-Flow Flexible Mesh (D-Flow FM) have been deployed to simulate TWL for ten coastal regions across the US. OWP has recently developed Basic Model Interface (BMI) applications for these model formulations to run in a coupled fashion under the NextGen framework for the Lake Champlain domain.

In this study, we focus on optimizing the D-Flow FM model for the Lake Champlain domain, which is driven by inflows from 68 river sources. We conducted a five-and-a-half-month long calibration period, comparing model outputs to observations from four USGS and Environment Canada stations within the lake. Different inflow strategies, including open discharge boundaries and polygonal lateral forcing techniques, were evaluated. The polygonal approach, in particular, effectively manages TWL by utilizing natural bed slopes to direct inflows, reducing the need for extensive bathymetric adjustments. This method improves the model’s ability to handle broader wetting and drying thresholds, crucial for robust operational forecasts in NextGen's coastal modules.

These advancements are expected to lead to more efficient and stable production runs, while improving the accuracy of water-level predictions. Future applications of these findings are aimed at enhancing TWL forecasting accuracy across other coastal model implementations within the NextGen framework.