Numerous communities and countries around the globe face significant flood risk to life and property on a regular and frequent basis. Flood insurance is one of the ways to recover from devastating flood events. Catastrophe risk modeling firms develop large-scale country-wide probabilistic flood models to help the insurance companies assess and price flood risk. Verisk Analytics has developed such physically-based flood models for about 16 countries around the world. Here we discuss the hydraulic aspect of one such model for Canada including fluvial and pluvial flood mapping. Verisk Analytics’ flood model for Canada covers a vast area of about 2.5 million sq. km and more than 98% of its population residing in southern lands bordering the United States. Hydraulic modeling for such a large area is by any measure a monumental task to be accomplished in a limited time. This article outlines the framework for this Canada Flood Model and provides insights into the development of the flood mapping hydraulics component. As with any flood model, the accuracy of the elevation dataset has a substantial and direct impact on the usefulness of model results. Details about our in-house 1-dimensional modeling systems, including processes to address DTM artifacts, automation in river cross-section generation process, and ways in which software, hardware, as well as remote sensing is discussed. The 1-D model parameters for roughness of the terrain as well as the amount of flow between the true bathymetry and the DTM were calibrated to river gages rating curves. The 1D hydraulic model was run at 20 return period flows for every river segment within the model domain. Flood extents were created using the 1D results as internal boundary conditions to the 2D shallow water equations to produce a quasi-steady state solution. Pluvial flood mapping corresponding to various return period excess runoff quantiles was also done using 2-dimensional hydraulics modeling factoring in surface roughness and drainage capacity of the urban drainage systems.
