Distributed hydrological model transferability across basins with different hydro-climatic characteristics

    loading  Checking for direct PDF access through Ovid

Abstract

In this paper, we investigated the transferability of a distributed hydrological model to better realize flood predictions at ungauged basins with different hydro-climatic characteristics. First, the transferability of a distributed hydrological model's parameters calibrated at the basin outlet in order to make consistent predictions at ungauged internal sub-basins for a range of floods with different initial conditions is investigated. It is found that the transferred model parameters can be used to make reasonable predictions at ungauged internal sub-basins for flood events with wet initial conditions. In contrast, the prediction accuracy at internal sub-basins is found to be low for floods that have initially dry conditions. By incorporating the spatial heterogeneity of the parameters, depending on the topography, we are better able to predict floods with initially dry conditions.

Abstract:

Secondly, to investigate the transferability of a distributed hydrological model for basins with different characteristics, the model is applied to three basins: the Illinois basin in the US, the Mae Chaem basin in Thailand and the Upper Kotmale basin in Sri Lanka. After analyzing the accuracy of event-based flood predictions, we found that the present model gives comparatively accurate predictions at the wet and moderate slope Upper Kotmale basin. However, the prediction accuracy at the dry and moderate slope Mae Chaem basin and at the dry and mild slope Illinois basin is low. It was concluded that the present model structure, which gives reasonable predictions for wet and steep slope basins, can be transferred to wet and moderate slope basins. Adopting a different model structure is necessary to obtain the transferability for dry and mild slope or dry and moderate slope basins. Copyright © 2011 John Wiley & Sons, Ltd.

Related Topics

    loading  Loading Related Articles