Adetailed field observation was conducted to elucidate mechanisms of freeze and thaw process and following sediment yield on a bare slope in Tanakami Mountains, southern part of Shiga Prefecture. During the observation period (December 2004-April 2005), air, surface and subsurface(10, 25, 50 cm depth)temperatures were continuously measured with 10 min interval. The sediment which was yielded on the bedrock was collected and weighted once in a week from a plot area(plot 1). For another plot area(plot 2), the sediment was left on the bedrock until the end of observation period. From the observation results, it was indicated that the freeze and thaw cycle occurred repeatedly, and the frozen zone(temperature<0°C)reached to the depth of 10 cm. The sediment yield occurred due to the active freeze and thaw process, and stopped at the end of the observation period when the freeze and thaw cycle also stopped (April, 2005). Total amounts of the sediment yield on plot l and 2 were 108 and 44 kg/m2/year respectively. This difference indicates that the sediment cover on the bedrock surface mitigated the effect of freeze and thaw process. These observed results were simulated by simplified thermal conductivity analysis method. Comparison between observed and simulated results suggested that multiple experiences of the freeze and thaw cycles are necessary for the bedrock to become sediment. It was confirmed that the estimation of the sediment yield due to freeze and thaw process is possible by this numerical simulation method.
The relationship between the ratio of forested area (F) and suspended sediment (SS) discharge was examined after eliminating the influence of geography as represented by the average slope gradient in a basin (G). The research area comprised 75 basins nested in the upper part of the Agatsuma River basin (707.9 km2). Using manual sampling, 617 samples were collected between 1994 and 1998, and SS concentrations were determined using the gravimetric method. Water discharge was measured at the time of sampling to estimate SS discharge. After classifying all the data into 3 and 4 groups according to F and G, respectively, sediment rating curves for each group were obtained using the least squares method. By comparing the sediment rating curves, the SS discharge in the basins was found to decrease with higher F, even with scattered data, while no clear relationship was found between SS discharge and G, which means that the forested area had a stronger influence on SS discharge than the geographic characteristics of the research area. Furthermore, we ascertained that the SS concentration converged on a lower value where F was near 1, resulting in especially low SS discharge in wholly forested basins. From these results, we conclude that it is difficult to evaluate SS discharge emphasizing only F, although F is one of the important factors that strongly affects SS discharge. The difficulty, we think, is caused by the existence of fields, which have a strong negative correlation with F in these basins, and which generate SS according to their local geographic characteristics. It is also possible that the relationship between sediment yields and geographic characteristics, which has often been highlighted in previous reports, can be applied only in basins where F is sufficiently high, because most of the research areas in such studies have been limited to upper mountain areas.
The change in the landslides largely affects the sediment yield of a mountainous catchment, however quantitative studies analyzing these changes at the scale of forested mountainous catchments still remain scarce. This paper aims to study the long-term change in the landslide area in Ikawa Lake catchment located at the upper Ohi river basin in Shizuoka Prefecture, central Japan from the year 1964 to 1997. Landslide data at the interval of about five years from 1964 to 1997 were derived from the visual interpretation of aerial photographs. The extracted data were mapped using the GIS software. The analysis showed that the year 1964 had the largest area of landslides whereas the year 1975 had the highest number of landslides. The landslides covered about 2 percent of the area of the Ikawa Lake catchment on average. The landslide area in general is in decreasing trend, the average annual rate of decrease being 195 m2 /km2. The area of a landslide becomes approximately half in about 10 years after its occurrence. The amount of continuous rainfall was found mostly related to landslide occurrence, which caused an episodic increase in landslide area. From the analysis of the aerial photograph, it was found that deforestation and road construction in the study area from 1964 to 1997 caused many new landslides, which is now in decreasing trend.
It is necessary to estimate immediately a flood hydrograph, peak discharge in particular for appropriate preparedness downstream when landslide dams are formed. Some studies have been carried out, however they were seldom applied to real landslide dam outbreaks because there aren't field data to be compared. The flood mark data of Takaiso-yama Landslide dam outbreak happened in 1892 was found and analyzed recently. A flood hydrograph estimation method based on erosion and sediment transport equations developed for debris flow was applied to the flood. Flood depth was well explained, estimated flow velocity was larger than reported values though. The winding channel may be a reason of the disagreement.
Asimplified dynamic cone penetrometer is a useful apparatus for the measurement of the hardness of slope materials at many points and depths of the soil layer on steep mountains. Therefore, some physical and mechanical properties of soils, such as the unit weight and the shear strength parameter whose measurement generally needs much time and effort, were estimated by the simplified dynamic cone penetrometer hardness(Nc)in many previous works. However, the estimated values often change depending on grains size reflecting lithology. In the present study, based on the results of field measurement of physical and mechanical properties of hillslope materials in six granitic mountains, we deduced the dry unit weight of gnus(γd, gf/cm3)and the angle of internal friction(φsat, degree), the cohesion(Csat, gf/cm2)and the hydraulic conductivity(K, cm/sec)of grus under saturated water content from two parameters of Nc-value and grain size of soil as following regression equations: γd=1.228Nc0.135CC2.00.102 φsat=1.377NcCC0.5+26.61 Csat=4.049Nc-46.70CC0.0625+71.13 log10K=-0.616log10NcCC2.0-0.115-1.872CC2.0-0.115 where CCD is the content for coarser grains than D mm. The values of CC0.0625, CC0.5 and CC2.0 can be easily measured by sieving method, so that these parameters can be easily estimated using the above equations.
The large earthquake on October 23, 2004 named Chu-etsu earthquake caused a number of landslides in the Imokawa basin, Niigata Prefecture. The landslides resulted in a large amount of unstable debris in the Imokawa River. Because the Imokawa basin is well-known as a heavy snowfall area, we should study the method of prediction of snowmelt to prevent the secondary damage by the snowmelt flood. In this study, the simplified and practical heat balance method was applied to predict the amount of snowmelt in the Imokawa basin. This method requires only the simple meteorological forecasting data of weather, temperature and precipitation and the topographic data of the basin. To test its applicability for the short-term prediction, daily snowmelt was predicted and compared with the actual data obtained by the on-site measurement in the Imokawa basin during the winter in 2004105. The results show that the proposed method can provide the short-term forecasting of snowmelt which is precise enough to prevent the disasters induced by snowmelt.
This paper was written with the objective of introducing bank side erosion and sediment transport equations into a simulation model of the outburst of a landslide dam to improve the projection accuracy when the mass removal of the soil occurs. The targeted model was applied to a drainage basin containing a landslide dam. The following elements were calculated : the peak discharge, the change of the river width, the rate of erosion, and the fluctuation of the riverbed at the site of the landslide dam. It was concluded that the peak discharge downstream of the landslide dam would be 10 times the volume of the inflow. Furthermore, the model was able to better express observed phenomena such as how sites where the river width is narrow or flow volume large are eroded more heavily. In the future, the authors intend to perform a sensitivity analysis of this model to clarify its characteristics based on the supporting data, and create a better sediment discharge forecasting system for the outburst of landslide dams.
To develope a method for assessing the potential of slope failures during earthquakes, we proposed a simple method to map the potential of slope failures during earthquakes using Digital Elevation Model and potential earthquake wave data. In the developed method, slope gradient, mean curvature and peak ground acceleration were used to evaluate the relative potential of slope failures. Here we examined the applicability of the method to Niigataken Chuetsu Earthquake, 2004. The analyses showed that the proposed method is an useful tool for assessing the relative potential of shallow slope failures triggered by Niigataken Chuetsu Earthquake, although the proposed method is not effective to predict the location of deep seated landslide during earthquake.