Many researchers have pointed out the existence of rapid water flow in the forest soil layer, which can not be explained by Darcy's law. The time of the slope failure estimated from the unsaturated-saturated percolation analysis and stability analysis is likely later than the time of actual failure. One reason for this discrepancy can be due to the assumption that the permeability of the soil layer is uniform. To accurately forecast the time of slope failure, we need to examine the 3-dimensional position of the water paths in the upper soil layers. In this paper, a procedure is proposed to make clear this 3-dimensional position of water paths through the analysis of temperature distribution in the upper soil layer. The outline of the procedure is as follows : Vertical earth temperature distribution in the holes bored by the simple penetration test is detected with an instrument equipped with 14 thermocouples. Data gained in the aligned holes yield isothermal chart. The disturbance appeared in this isothermal chart should indicate suggestions the existence of the water paths. The procedure was applied on a slope where the actual water paths were confirmed by excavation. By using this procedure, the 3-dimensional position of water path was grasped with the maximum error of 40 cm.
The volume of fine particle sediment, measured over several days using simple sampling tanks at the mouth of small mountainous catchments, was analyzed in relation to rainfall and landform attributes. The relationships between the effective cumulative rainfall calculated from daily and antecedent rainfall and the sediment volume sampled for several days were shown by power functions with high correlation coefficients. However the power functions are determined primarily by increases in sediment at 1400mm in effective cumulative rainfall and 600mm and/or 900mm in effective event rainfall. Because catchment areas addressed here were remarkably small and slopes were steep compared with other catchments, exponents of the power function were smaller than the results from measurement data from other catchments. These results indicate that overland flow might influence the sediment yield from the surface of small, steep catchments.
The purpose of this study is to clarify changes in surface runoff characteristics and sediment discharge on a granule ash-covered slope. Three years after the latest eruption of Unzen volcano, we made an observation of surface runoff and sediment discharge on a slope of the volcano. We scattered fine-grained volcanic ash in one-third area of the slope in June, 1999. As a result of the ash-covered treatment the following points were found. (1) It was considered that the volcanic ash covered on the study slope was washed away from the slope by surface runoff as suspended sediment. (2) The peak discharge and total amount of surface runoff after the treatment increased under the approximately similar rain conditions. (3) The decrease in the surface runoff and sediment discharge resulted from disappearance of the volcanic ash on the surface of the slope. It is concluded that the runoff characteristics of the study slope had changed with the covering and the movement of fine-grained volcanic ash.
Debris flows have occurred at Nishinogaito-stream and Kotaki-stream, in Fujiwara town, Mie prefecture on July 17th, 2002. Heavy rainfall of 39mm/h (50mm succession) was the most effective trigger for the occurrence of the debris flows which reached to the lowest check dam of each stream, and some of them were flowed over the lowest dam as mudflow. There was no human damage by the debris flows according to the existence of extended series of check dams after the previous debris flows occurred on Aug. 19, 1999.
After the eruption of Miyakejima Volcano in July 2000, debris flows occurred in torrents whose source areas had been mantled by the volcanic ash. The debris flows in the torrents were generally composed of not only newly deposited fine-ash but also preexisting bold rocks and scoria. In the upper slope of the torrents, numerous gullies and rills had been dissected. A large portion of sediment discharge from the torrents seems to be attributed to the gully erosion. Therefore, it is important to know gully development characteristics to predict debris flow occurrence and its volume at volcanically-disturbed torrents. In order to investigate gully development characteristics, the authors gathered three sets of aerial photographs taken in Aug. 2000, Nov. 2000, and June 2001, and investigated gully development processes in these periods, which were from the onset of the eruptions to Aug. 2000, from Aug. 2000 to Nov. 2000, and from Nov. 2000 to June 2001, respectively. As a result of the surveys, the following were clarified. 1) The development rate of gullies was largest in the first period. The rate rapidly decreased in the second and third period. 2) The gully network in one sectioned slope had been fully developed in the first period. On the other hand, the width of the gullies kept increasing even in the second period. It shows that the gullies get longer at first and subsequently widened in the gullies. 3) Most of gullies had been formed along the topographic depressions recognized in the topographic map made before the eruption. It seems that the gully development is subject to the original topography in the case that volcanic ash deposit is not thick enough to change the original topography. 4) Weak positive correlation exists between width of gully and drainage area. These findings infer the capabilities to predict volume of sediment discharge prior to volcanic eruption.