Sensitivity Analysis of a Hydrodynamic Model for Jamuna River using HEC-RAS

Abstract

Hydrodynamic model setup for a specific area is a difficult task due to various influencing factors such as geographical, climate etc. Sometimes, it is difficult to determine actual field values of numerous parameters of hydrodynamic models for dynamic rivers like Jamuna. Though hydrodynamic modelling is a complex task, but if calibrated and validated properly then the model can be used for the application of flood forecasting and other types of water resource studies. Assessment of water level and discharge on different stations at vulnerable areas in a river requires upstream discharge and downstream water level, and that can be achieved by hydrodynamic modelling. In this study, a hydrodynamic model is developed for downstream high flood risk areas along the Jamuna River (Bahadurabad Transit to Aricha) of 147 km length using HEC-RAS software. The most important criterion for any modeling task is the identification of the most influential parameters on model outputs. Sensitivity analysis is an appropriate and efficient platform for identifying key parameters that have a major impact on multiple modeling stages such as calibration and validation. Sensitivity analysis was carried out on six calibration parameters in HEC-RAS software, including manning's n, storage area elevation tolerance, water surface calculation tolerance, theta for warm up, theta, maximum number of iterations of built hydrodynamic model. The analysis results revealed that among all the parameters, manning's n was found to be the most sensitive. The findings of this paper will aid in the selection of significant input parameters for hydrodynamic models are chosen. This model can also be used as a basis for future water resources planning and management.