2-D Fine Grid Hydrodynamic Modeling For More Accurate Floodplain Mapping In Southern California

摘要

In general, there are three major factors that determine the accuracy of 2-D hydrodynamic modeling results as far as their use for floodplain delineation is concerned. Those are: (a) hydrology-related factors that are the drivers of the hydrodynamic models (empirical models such as Rational Method, and process-based or physical models that account for both temporal and spatial variability such as HSPF, HEC-HMS); (b) accurate representation of site conditions (surface topography, channel morphometry (geometry), control structures and infrastructures); and (c) boundary conditions (fluxes in and out of the model domain). All of these factors affect accuracy of modeling results and consequently the extent of floodplain maps, especially in areas with significant urban topographic features and variation in land use, and areas that lack adequate storm water conveyance. The traditional approach of 1-D Steady State modeling is by nature more conservative, simpler to conduct, requires lower level hydrology data (generally Q_(100) and Q_(500) peak flows) and it is readily accepted by the regulatory bodies (e.g. FEMA and the States), although it is often less accurate. On the other hand, the Unsteady State 1-D and 2-D modeling approach requires a higher level of hydrologic data (time series of discharges or hydrographs at many locations), higher levels of output analysis and more rigorous model calibration, and often more effort in convincing the regulatory bodies that the results are more realistic. Calibration can be very challenging when 2-D models are applied; depending on the accuracy of topographic information, complexity of the physical domain, model grid size, as well as number/location of inflows in the model set up. Significantly, the small simulation time step and interpolation of time series boundary inputs, the execution time can be very long even with highspeed computers. In this paper, we will describe the results of 2-D fine grid hydrodynamic modeling efforts using the FEMA-approved FLO-2D software to more accurately delineate floodplain extent in comparison with conventional 1-D HEC-RAS models in Ventura County, California. Although the level of data requirements for the 2-D model is more extensive and the computational time is substantially longer than that of the 1-D model, the results are much more accurate and corroborate with the weight of evidence in recent flooding events. This approach is also gaining more acceptability by the local entities submitting LOMC/LOMR/CLOMR and the regulatory body requirements because of animation capability, graphical user interface, and use of GIS layers to depict flooding extent to which they can easily relate.