Thesis Type: Postgraduate
Institution Of The Thesis: Orta Doğu Teknik Üniversitesi, Faculty of Engineering, Department of Civil Engineering, Turkey
Approval Date: 2015
Student: ARTUR NİMAEV
Supervisor: SEVDA ZUHAL AKYÜREKAbstract:
Hydrodynamic computational modeling plays a vital role in assessment and management of flood risks. In particular, flood modeling in urban environments is especially important due to high damage to infrastructure and property as well as losses of human lives. Many numerical two-dimensional schemes, as a consequence, have been developed to perform simulations of urban flood inundation benefiting from recent technological advancements in topographic surveying techniques. To understand sensitivity of model outputs to different hydraulic modeling approaches; namely, Flow-limited, Adaptive, Acceleration and Roe numerical solvers of varying complexity in terms of representing shallow water equations of Lisflood-FP and 2D full-dynamic Mike 21 models were evaluated in this study. Furthermore, 25 cm, 50 cm and 1 m resolution rasters based on terrestrial LIDAR data of town center of Terme located in Samsun, as well as four different roughness parameters representing urban surface conditions were compared to see how resolution and surface friction may impact simulation results using simplified inertial solver of Lisflood-FP model. The results indicate that among 4 solvers of Lisflood-FP model, only Acceleration numerical scheme provided consistent results to be used in practical applications. Also, compared to the Mike 21 hydraulic model, Acceleration solver generally predicted similar test results, except for the areas of ponding. Increasing DEM resolution resulted in more rapid flow propagation due to more detailed representation of the topography such as sloping alignment of a road. However, it was shown that the use of LIDAR data in flood modeling studies obtained from high frequency terrestrial laser scanners is limited due to sophisticated processing techniques. Moreover, the Acceleration solver correctly predicted sensitivity to friction as water was conveyed faster in models with lower Manning’s coefficients.