Stochastic overland flows

Abstract

The theory developed in Part I of this paper is now applied to study the probabilistic behavior of the depth at the outlet of an impermeable overland flow section under the diffusion and kinematic wave approximations. This process is excited by stochastic rainfields which are conceptualized from radar observations. The depth at the outflow section is of prime importance and the solution methodology concentrates on obtaining the evolutionary probability distribution function for this physical quantity. This theoretical distribution is then compared with the empirical distribution function obtained from a thousand Monte-Carlo simulations. The simplified theory leading to the Fokker-Planck equation is also investigated. It is observed that the ‘time window’ used for simulation purposes can affect the results. The theoretical methodology performs satisfactorily when compared to simulation results. Some of the notable features of the proposed methodology are presented and further suggestions for improvement and extension of this work are discussed.