Abstract
This paper presents the impact of the choice of building representation techniques and hydrodynamic models on urban flood simulations using HEC-RAS 2-D for the Toce River physical model. To this end, eight numerical models based on previous laboratory experiments were prepared to simulate unsteady urban flooding on each side of building units. Two simplified building layouts (aligned and staggered) were examined, where models were prepared for two different building representation techniques: Building Block (BB) and Building Resistance (BR). Water depth variation computations using the BR and BB techniques were compared to the laboratory measurements and previous studies in the literature. A statistical analysis was performed using both the Root Mean Square Error (RMSE) and the Pearson Product-Moment Correlation Coefficient (PPMCC) in order to evaluate the performance of the models. A sensitivity analysis showed that the proper mesh resolution and model parameter values were obtained. As far as the BR technique is concerned, it is well-suited for representing building units in numerical simulations using high Manning coefficients. Furthermore, this study confirms the importance of the BR technique, which should help researchers in using low-resolution Digital Elevation Models (DEMs) along with open-source programs. Moreover, the study aims to produce a deeper comprehension of numerical modeling and urban flooding.