Spring water that appears on the slope during heavy rain is widely known as one of the signs of slope failure. The appearance of spring water means that the groundwater level has risen to near the ground surface. In addition, soil runoff and slope erosion associated with spring water may lead to a decrease in slope strength. Therefore, spring water is one of the important factors in considering the seepage flow in the slope during heavy rain. However, in the conventional seepage flow analysis assuming uniform field, the situation of the locally appeared spring water cannot be expressed due to the spatial heterogeneity of the hydraulic conductivity. In this study, the relationship between rainfalls, permeability of slope, thickness of surface layer, and the amount of spring water was examined quantitatively by using the stochastic fractal model as a spatial distribution model of hydraulic conductivity and three-dimensional numerical seepage flow simulation. As a result, it was shown that the amount of spring water was clearly underestimated by assuming the slope to be uniform. In addition, it was shown that when the permeability of the slope is relatively low, the groundwater level tends to rise for the same amount of rainfall, so springing is likely to occur. On the other hand, when the thickness of the slope is thin and the amount of rainfall is high, the amount of spring water was larger on the higher permeability slope.