Effects of spatial resolution on a raster based model of flood flow

The scaling properties of a simple raster-based flood flow model are investigated. Models of resolution varying from 1000 to 10 m are tested and predictions compared with satellite observations of inundated area and ground measurements of floodwave travel times, with a calibration strategy being used to determine channel friction coefficients. The optimum calibration is found to be stable with respect to changes in scale when the model is calibrated against the observed inundated area, the model reaching maximum performance at a resolution of 100 m, after which no improvement is seen with increasing resolution. Projecting predicted water levels onto a high resolution DEM improves performance further, and a resolution of 500 m proves adequate for predicting water levels. Predicted floodwave travel times are, however, strongly dependent on model resolution, and water storage in low lying floodplain areas near the channel is identified as an important mechanism affecting wave propagation velocity. A near channel floodplain storage version of the model is shown to be much more stable with respect to changes in scale when the model is calibrated against floodwave travel times, and shown to represent the retardation of the floodwave caused by water storage near the channel. The model cannot be calibrated to give both acceptable travel times and inundated area, and in this respect performance is poor.