Sensibilidad del coeficiente de Manning en la estimación de los niveles de crecida para el mapeo de inundaciones en un río de la región interandina de Ecuador

Abstract

Input data, parameters, and the structure of mathematical models introduce uncertainty in the results; thus, sensitivity analysis is required when implementing a model or as a preliminary step. This study evaluated the sensitivity of Manning’s coefficient based on landcover types in the floodplains and the main channel, for calculating water levels and defining flood extent. The mike 11 one-dimensional [1-d] model was used in this study on a five-kilometer stretch of the Santa Barbara River in southern Ecuador. Six hundred Monte Carlo simulations were run for each flood event, and the results were used to assess the robustness of the 1-d model. This analysis revealed that Manning’s coefficient was sensitive to changes: there was a difference of up to 1.02 m between maximum and minimum water levels. The resulting flood maps, however, revealed almost imperceptible differences in flood extent. These findings demonstrate that the recommended Manning’s mean value is reliable for flood hazard mapping of Mountain Rivers in the inter-Andean region