Comprehensive analysis of flooding in unmanaged catchments

Abstract

Environmental models are intended to represent or simulate natural environmental processes, often in mathematical or statistical terms. Sensitivity analysis is a modelling tool that is essential in proper use of a model because it enables the model user to understand the importance of variables and the effects of errors in inputs on computed outputs. A few studies have reported sensitivity analyses of flood routing parameters. In this study, sensitivity analysis of the parameters of flood routing with the implicit numerical method is investigated through mathematical, statistical and field measurement procedures in three unmanaged catchments of the Persian Gulf region. The results indicate that the importance rankings of various parameters on output results are peak inflow, roughness coefficient, bed slope, bed width, river length, side slope, weighting factor and base flow. The results also show that the effects of input parameter errors on the output results are significant in some cases, such as the peak inflow on the peak outflow and volume of flooding, the river length on the time to peak, the roughness coefficient on the velocity, and the peak inflow, roughness coefficient and bed width on depth.l sweep and ejection events at the edge of floodplain for d/b equal to 0·1. Furthermore, the transition probability of internal, external sweep and external ejection events for d/b = 0·1 decreased at the intersection of the floodplain and main channel. Although scaling of the experimental results requires further study for application to wide rivers, these findings may be useful in practical applications for narrow rivers.