A thesis submitted in partial fulfillment of the requirement for the degree of Master of Engineering of the Asian Institute of Technology, Bangkok, Thailand.

Thesis (M.Eng.) - Asian Institute of Technology, 1974

This study deals with the methodology of laying down a more comprehensive way of evaluating flood damage. Since damage occurs mainly due to height of inundation. and its duration, the river stage-probability curves of floods of different durations are important tool s in understanding the character of flooding of the area considered. Out of these curves , the areas inundated by each flood were identified in the map, and the different economic activities in these areas were investigated to estimate the damage. Flood damage was divided into three categories : agricultural losses, urban losses, and social sufferings. Agricultural losses were determined by mainly considering the type of vegetation, the phase of growth, the height of submergence and the duration off loading. Urban losses were mainly due to height of inundation. Residential losses , commercial losses, industrial losses, road; losses, emergency losses, relief costs, are all types of urban losses. Sampling and statistics considerations and also field investigations are significantly needed in the proper estimate of flood losses especially for t he case of urban losses . Social sufferings indicate the people affected by floods, these form of losses are intangibles. The construction of damage probability curve had given us an area bounded by the curve itself which in reality was the mathematical expectation of annual flood damage. To achieve non-dimensionality, every element of damage axis in the damage probability curve was divided by the damage occurring at the expected annual average river stage without flood mitigation structures. Thus, modular coefficients of damages were developed. The area under the flood modular coefficient-probability curve is the dimensionless flood damage index. Considering a flood mitigation measures which aimed at lowering the river stage, the degree of protection is defined for this purpose as the percent of reduction of the untouched expected annual average river stage. Therefore for specified flood mitigation structures, the residual damage can be obtained by multiplying the damage at expected annual average river stage with control structures to the dimensionless flood damage index. In the hypothetical river basin considered, graphical flood damage models were developed considering different degrees of flood protection. The models show the response of the flood plain to different degrees of protection. With a given degree of protection, the residual damage, t he avoided damage, the relieved people and the affected people are al l revealed in the models.