Comparison of Hydraulic Simulations Using Various Topobathymetry Digital Elevation Models

The accuracy of hydraulic simulations and flood predictions for a lake–river system highly depends on the accuracy of input data of topobathymetry. Although digital topographic elevation and bathymetry models are available in most cases for Canadian water bodies, there are two major concerns: (i) a lack of ready-to-use topobathymetry digital elevation models (TB-DEMs) for hydraulic modeling; (ii) the inaccuracy of available data, which inevitably causes uncertain predictions. This study used two TB-DEMs of different resolutions for hydraulic simulations of the Lake of Two Mountains in Montreal, Quebec. The hydraulic model used is TELEMAC-2D, a free source-code module that provides the opportunity to develop new models in future studies of floods. One of the two TB-DEMs has a 1-m resolution. The other TB-DEM is developed by integrating bathymetry data from a NONNA-100 database with topographic data from a DEM of a 30-m resolution from the U.S. Geological Survey (USGS); this TB-DEM has a 100-m resolution. The simulation results are compared between the TB-DEMs. The 1-m TB-DEM slightly improves the prediction accuracy, compared to the 100-m TB-DEM. Nevertheless, a finer resolution TB-DEM should be used to capture details of flood inundation such as roads with high elevations (relative to the adjacent waterbody) and the barriers between wet (lakes or rivers) and dry (land) areas. The use of a finer resolution TB-DEM mesh incurs higher computing costs, which hinders real-time simulations and flood forecasting. This study proposes a reliable strategy for hydraulic simulations of river–lake–shore floods. Future efforts may lead to the development of a real-time simulation and flood warning system.