Predictive uncertainty of watershed rainfall-runoff models using weather radar

Two hydrologic models were used in association with radar-rainfall fields – AMBER (Areal Mean Basin Effective Rainfall) and F2D (Flood 2-Dimensional). AMBER is a software application to evaluate flash flood threats computes the Average Basin Rainfall (ABR) for watersheds every 5-6 minutes in it's area of responsibility. The likelihood of flooding is established by comparing the ABR with the Flash Flood Guidance (FFG) (i.e., the ABR needed to bring a stream to bankfull stage). F2D is an event-based, kinematic, infiltration-excess, distributed rainfall-runoff model was developed to acknowledge and account for the spatial variability and uncertainty of several parameters relevant to storm surface runoff production and surface flow. Both models are compatible with raster Geographic Information Systems (GIS) and spatially and temporally varied rainfall data defined by NEXRAD radar. Principal raster data inputs include hydrography data derived from a digital elevation model (DEM), soil texture, forest type, forest density, lakes, reservoirs, land use and land cover (LULC), and impervious regions within a basin. Monte Carlo simulation and a likelihood measure are utilized to calibrate the model; allowing for a range of possible system responses from the calibrated model. Using rain gauge adjusted radar-rainfall estimates, the models were applied and evaluated to a limited number of historical events for watersheds within the Denver, Colorado metropolitan region that contain mixed land use classifications.