Abstract
Some field works and model experiments were carried out in order to study the geometrical morphology of gully development. (1) Statistical laws of gully morphology and (2) hydrological conditions of initial gullying by surface runoff' were discussed here. Some quantitative relations are derived as follows.
As for the morphological aspect, gullies originate totally at random at the initial stage, but in the course of their development they become to show their distinctive characteristics. The results of statistical treatment manifest that both the width and the interval of mature gullies have very limited distributions with narrow peaks (Fig. 5, 8, 12, 21, 22, 23), and that the number of gullies across the unit length of a contour line becomes constant in a specified slope (Fig. 6, 9, 13). These statistical character-istics cannot be explained at the present stage of' study by physical relations which may control the developing processes. Assuming that a criterion for the gully initiation is given by the hydrological condition that the shearing stress by surface runoff exceeds the critical scouring stress of slope, the following equation (6) was examined.
_??_ (6)
where (C/A) is a physical and hydrological constant, (f 1/3/K) is a factor which shows soil conditions and is named a soil factor, B is the angle of slope, l is the slope length, r is the intensity of rainfall, and d is the grain size. The soil factor (f 1/3/K) related to slope ero-sion was introduced by the authors as an independent factor, and the numerical values of the factor were calculated from observation data about real slopes. The calculation results show that the value of the soil factor at the initial stage of gully development seems to be larger than that at the succeeding stages, and that a gully can originate with smaller rainfall than the succeeding gully development requires (Table 2).