The greater storage capacity of a sabo dam, the more incoming sediment the dam traps. A permeable type dam that is made of steel pipes is a good countermeasure to capture incoming debris flow because it is easy to prevent this type of the dam from loosing its capacity. In order to trap incoming debris flow, it is necessary that big boulders transported by the debris flow block open spaces of a permeable type dam up. Blockage occurs as a result of boulders interlocking each other between the steel pipes. In order to estimate the volume of the sediment that will be captured by the dam, it is necessary to develop a simulation model that takes into consideration the irregular movements of boulders. In this study, we first conducted experiments using flume and dam models. Then, we made a model that traces the movements of each boulder by using the Distinct Element Method. The experimental results were reproduced by using the model well.
Some aspects of present channel works are evaluated from the viewpoint of conservation of ecology. The fundamental theory needed to be considered for river management and conservation of ecology, during the planning of channel works have been described. The following investigations were carried out. 1) The present methods of planning and construction of channel works are reviewed. 2) Based on data analysis of presently existing channel works and hydraulic experiments, the influence of cross dykes on the formation of sand bar was evaluated. 3) The effect of size of wing of the cross dyke on the formation of inner bed topography was examined. The results of the hydraulic experiments proved the formation of inner bed topography and sand bar when cross dykes spacing and cross dyke wing are properly considered. Based on these results, a river management planning was proposed.
Because sediment input volume from branch to main stream is the key knowledge for estimating input-output balance of sediment budget along river channel, this study was focused on the temporal changes in sediment volume supplied from small catchments. Sediment volume was monitored in the concrete sand-pocket(4m×2m×1m)set at the mouth of four small catchments. Each catchment is the area between 2.48 and 10.14 ha and channel slopes of them between 0.093 and 0.249. Sediment volume measurement was conducted at 17 times during three years. The influence of catchment area, such as channel slope, landslide area and its occurrence years, on sediment volume supplied from each catchment was examined. The proportion of the duration rainfall against the duration sediment volume at each measuring period, sediment supply rate, yields regression line as a exponential curve except for the periods of less rainfall in winter season. This indicates that the sediment supply rate defined as the duration sediment volume per the duration rainfall is applied to any periods, for example, several days or a year. Since the annual sediment supply rate, Vy, decreases according to years after landslide occurrence year, T, the relationship between both is represented as Vy=α⋅exp(-γ⋅T). Here landslide area was likely to response to landslide occurrence as the initial condition, α. Catchment area, relief and averaged channel slope were not likely to influences on decreasing of annual sediment supply rate, γin such small catchments.
The hydraulic model experiment is used often to make the design of Sabo works. But reports to confirm the reliability of the hydraulic model experiment are not so much. Because the plan flood occurrence is rare in the field. In this report, the hyetograph and the maximum rainfall in September 15-16, 1998 at Uono river were compared with the plan. And the peak discharge during this flood was guessed by the maximum rainfall and the catchment area around Uono river. Those were shown to almost equal the plan from this study. The hyerograph was guessed from data of water level on Oogawara gaging station at the confluence of Daigenta river and Uono river during this flood. And the disaster occurrence time and location during this flood were compared with the experiment results under the same arrangement condition of Sabo facilities. The trend of those coincided with the experiment results. We compared the flood scale and flood situation of field with the plan flood and the experiment results. Those in the field almost equaled the plan and the experiment. And it was shown that the hydraulic model experiment was useful to make the design of Sabo works from this report.
Many sabo works have been provided in the Hirakawa River, Nagano prefecture in Central Japan. These sabo works have led to prevent occurring debris disasters leaving human lives and properties, as the socioeconomic effect thereof. This study dealt with the Hirakawa River, as a case study area to evaluate the socioeconomic effect. The study conducted in 1985, estimated the total cost (C) of these sabo works in 1947 to 1985 at 7, 083 million yen. The direct benefit of the prevention to debris disasters (B1) is estimated to be 15, 867 million yen, the indirect benefit of that (B2), 952 million yen, the enhanced land use benefit (B3), 16, 258 million yen, and the amenity benefit (B4), 360 million yen. The ratio B 1/C of 2.24 and B (B1+B2+B3+B4)/C of 4.72 were derived from the study.
It has been reported that impermeable type Sabo dams, filled with sediment, trapped a considerable volume of woody debris, despite the fact that little woody debris could be trapped under the present design criteria. Authors presumed the dominant factor to determine the woody debris trapping rate was the ratio between the wood length in the woody debris and the spillway width. Therefore, “wood mixture rate, the wood longer than the spillway width which occupied in the total volume of the woody debris” (we simplify this term and write as “wood mixture rate”) strongly related the woody debris trapping rate. Following this idea, we conducted some flume experiments using various lengths of wood models. Consequently, woody debris movement in the sediment trapping zone were classified into 4 patterns (A to D), then it was confirmed that 2 patterns produced high woody debris trapping rates. One was trapped on the sediment before reached the spillway (pattern A) and this pattern could be seen when the discharge was small. The other one was trapped by clogging the spillway (pattern C) and this pattern could be seen when the discharge and the wood mixture rate were high, simultaneously. When the woody debris was trapped by pattern A or C, sediment was also trapped behind the woody debris. However, the trapping rate of pattern A was controlled by its discharge and flow time, while pattern C produced stable sediment trapping rate. Furthermore, it was confirmed that the “wood mixture rate” was an important factor to determine the woody debris moving patterns and the trapping rates.