TWO-DIMENSIONAL FLOOD MODELLING IN URBAN AREAS

• Main Author: Farid (NIM : 250 05 035), Mohammad
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/6484
• Institution: Institut Teknologi Bandung

Abstract: <br /> <br /> <br /> <br /> <br /> Urban flood has different characteristic with flood in nature. The basic process in nature involved rain, where the rainfall run off will flow gravitationally or infiltrated. Urban area land cover is mainly buildings, concretes or roads which can not be penetrated by water. In result, the infiltration value in urban area will be very small. Deforestation also caused the transpiration value small. Evaporation alone does not have a significant effect. In the case of urban area, it can be concluded that the infiltration, evaporation and transpiration would be insignificant. Basically, water from precipitation will flow in drainage to the main river. The topography condition of the urban area is also engineered so that the overland flow will gravitationally flow to channel. However, there are times when the channel capacity is less than the discharge load from the precipitation. This condition leads to flood. In several cities, which are crossed by river, the water from the river flooded the city occasionally due to the large discharge from the upper catchment. Flooding cases in urban area have several problems which require further research. The flow direction is not entirely depending on the topography due to buildings which blocks the flow. Buildings model a challenge in modeling urban flood. The model is developed using the full dynamic 2D St.Venant equation which is solved using Mac Cormack scheme. The flow is affected by the land cover and topography. The flood in the model can be caused by both, precipitation or upper cathcment discharge. The precipitation is added in the continuity equation as lateral discharge. Upper cathment discharge is added as boundary condition in the river nodes. Building is modeled as topography (building elevation) and givinglarge manning coefficient (0.1). The method is chosen to accommodate the precipitation on the roof of the building. The wet dry condition in the model is done by determining certain minimum flow depth. Model simulation for flood due to dam break with a building located at the front of the dam gives good comparison with experimental data. The 2 dimensional depth averaged model limitation caused the initial condition of water elevation upstream and downstream of the dam to be simplified. The building lay out is not perpendicular to the dam. This caused the model tends to be instable compare to other numerical model with finite volume method. The model is tested with cases for flood due to upper catchment discharge. A very wide channel with certain slope and given dry initial condition is modeled. The up stream is given inflow hydrograph. The model is compared with commonly known kinematic wave routing. The result shows good comparison. However, the outflow volume is somewhat smaller then the inflow volume. In the case of flood due to precipitation, the model is used to simulate rainfall on a small rectangular cathcment with channel in the middle of it. The result is compared with well known synthetically hydrographs, Rational and Nakayasu. The model result shows good comparison with the synthetically hydrographs. Urban flood modeling due to precipitation, upper cathcment discharge and both, shown commonly known pattern. The model is implemented for Batang Kuranji River Basin, Padang. This basin is chosen due to the data availability for verification purpose. The overland flow is generated by precipitation only. The precipitation is chosen from measurement data. The model result is compared to the measurement discharge at the basin outlet for the corresponding precipitation occurrence. The discharge measurement and model result shows good comparison.