COMPARISON OF GERMAN AND SWISS RAINFALL SIMULATORS - INFLUENCE OF PLOT DIMENSIONS

The construction and operation of a rainfall simulator becomes easier with decreasing plot size. We studied the effect of plot size on the erosion dynamics and whether it is preferable to decrease plot width or plot length. 22 simulated rainfalls were carried out with Veejet 80 100 simulators and on plots of considerably differing dimensions. The plot width has to cover a representative width of the field. With the common 3 m wide farm machinery, a plot width of at least 1.5 m containing one wheel track is necessary. Narrower plots are not suited for erosion experiments on agricultural land. The influence of plot length can be described with the LS factor of the USLE down to a plot length of 4.5 m and with the RUSLE for interrill plots (0.75 m). With decreasing plot size, the runoff started later (up to 20 min) and the afterflow ended earlier, This is not only a result of the plot length (r = 0.78), but mainly of the plot size (r = 0.92). The large time lag on small plots complicates the interpretation of the results. We recommend to disregard the rain erosivity during the time lag for the determination of USLE parameters. This is in accordance with the procedure of Wischmeier & Mannering (1969). Plot sizes have to be selected particularly carefully if the fate of soluble substances is to be studied.