Calculations have been made in order to evaluate the effect of rainfall movement on peak discharge. The 25 extreme events have been selected from a long historical rain series from a single rain gauge. The rains have been simulated to move with the recorded speed and direction of the rain across an artificial, elongated catchment positioned in 8 different directions. A real catchment layout was also investigated. The paper investigates the effect of longitudinal movement. Laterally, the rain is assumed to be uniformly distributed.
Calculations were made with the MOUSE-program, modelling runoff using the dynamic wave approach. An additional program was developed to simulate the moving rain input. Peak discharge in the sewer system with and without rainfall movement was treated statistically in terms of return period and compared.
The movement of the rain is not important from a statistical point of view (up to 5 %), but for individual events the effect may be significant (at least up to 19 %). The result is restricted to the effect of longitudinal variation in time and space. The effect of variation lateral to the movement may be much bigger, increasing with size of catchment.
The artificial catchment with flow direction towards NE gives the greatest increase in the Qmax and vice versa with the catchment orientated towards SW. This is due to wind direction from the SW being the most frequent in Denmark. The effect of rainfall movement is smaller for the real catchment with sewers in many different directions.