"Comparison of 2D Hydrodynamic Models in River Reaches of Ecological Importance: Hydro_AS-2D and SRH-W
"The use of two dimensional hydrodynamic models has become ubiquitous and indispensible tool for the study of natural rivers. Such models are especially useful when modeling results are required on scales relevant for ecological processes, where local details of velocity and depth distributions are significant especially in river reaches of ecological importance.
The main objective of this Master’s Thesis was to compare two hydrodynamic models with respect to modeled parameter outputs (water surface elevation, flow depth, velocity, etc.), accuracy, computational time, and their relevance of application for scales of ecological importance. The two models used were Hydro As-2D and SRH-W. The spatial interpolation techniques Krigging and triangulation with linear interpolation were also compared for their accuracy of interpolation. Additionally, a sensitivity analysis was performed for the two models to investigate the effects of mesh resolution and the exclusion of boulders from the bathymetry data on model outputs. Three representative case study reaches with different bed morphologies and flow characterstics were selected for the study. Two of the reaches were from the Black Forest in Germany while the third is a reach from the Austrian Alps.
Model calibration was performed by changing the bed roughness values such that observed and predicted water surface elevations had close agreement. The same number of roughness zones and bed roughness values were specified for both models to compare their outputs.
The results of the case studies indicated that both Hydro As-2D and SRH-W predicted the observed water surface elevations quite well. The velocity outputs of the two models were also comparable. There was not any significant difference found between the accuracy of the two spatial interpolation methods. The sensitivity analysis showed that SRH-W is more sensitive to mesh resolution than Hydro As-2D and the inclusion of the boulders affected velocity outputs to a greater extent than the water surface elevation or water depth. It was also found that in order to capture the complex velocity patterns around the boulders, high resolution bathymetry data and the use of a finer mesh was absolutely necessary.