The rainfall-induced landslides in the Nepal Himalaya extend tremendous damage to lives, property, infrastructure, and environment, particularly in the monsoon season. This paper particularly discusses the issues related to the rainfall-induced landslides in the Nepal Himalaya. All geological zones of Nepal were evaluated from the landslide occurrence perspective. To evaluate the landslide processes and associated hazards in Nepal, the rainfall-induced landslides were assessed from three perspectives: hydrological and slope stability modelling, rainfall threshold of landslides, and landslide hazard. This paper is an outcome of these evaluations and exclusively covers the major issues of rainfall-induced landslides in the Nepal Himalaya.
The role of large rivers in conveying sediments from continents to oceans has long interested geoscientists but until relatively recently the quantitative description of sediment cascades within large river catchments had been somewhat limited. Renewed interest in global biogeochemical cycles has refocused attention on questions of how large rivers respond to disturbance imposed by changing climate and tectonic conditions or from human-induced influences on sediment production and conveyance. Some progress has been made in examining the contemporary and spatial dynamics of sediment delivery in some Asian river basins, where the exposure of populations to water and sediment-related disasters is acute. This paper considers how these essentially desk-top studies of sediment cascades can be linked effectively to field-based studies in order to contribute to explaining and predicting sediment disasters. It focuses on current understanding of response to disturbance in large rivers and considers the conceptual challenge of representing connectivity and discontinuity in sediment cascades.
Deep catastrophic landslides (DCLs) sometimes lead to large-scale debris flows with serious impacts on human life and infrastructure. However, no adequate information about DCL-triggered debris flows, such as the topography of eroded and deposited areas or the grain size distribution, exist. We compiled published data and obtained additional new data for the topographic characteristics and grain size distributions of 10 recent DCL-triggered debris flows in Japan. We compared these data with previously published data of small-scale debris flows, steep-slope failures, and large-scale debris flows. We examined the effects of topography and DCL volume on erosion and deposition due to debris flow as well as on grain size distribution. The longitudinal gradient of the lower end of the deposited area decreased with increasing landslide volume, and about half of DCL-triggered debris flows deposited material where the longitudinal gradient of the lower end of the deposited area was less than 2°. However, the minimum longitudinal gradient of the eroded section due to debris flow was not affected by the landslide or the debris flow volume. We found that the travel distance of debris flow, including DCL-triggered debris flow, might also be a function of landslide and/or debris flow volume and that the grain size of debris flows triggered by DCLs spanned more than eight orders of magnitude.
In steep mountainous regions, landslides may include both soil and underlying weathered bedrock (hereafter, —deep catastrophic landslides•). The method for assessing susceptibility to deep catastrophic landslides, originally developed for landslides caused by heavy rain, was tested in this study against historical landslides caused by the Iwate and Miyagi inland earthquake of 2008. The method proved to be capable of independently identifying catchments in which deep catastrophic landslides occurred with fair accuracy.
In this study, two methods were used to examine the influence of rainstorms on landslides. The first method used mean rainfall intensity and duration (I-D), and the second used a distributed-landslide conceptual model based on the distribution of shallow-infiltration water and a safety factor estimated using infinite slope-stability analysis. These two methods were applied to the Utsunomiya University forest in Funyu, Tochigi, Japan. Two indices were used to discriminate between rainstorms with and without landslides: one indicated the effect of short-duration events on landslide occurrence, and the other reflected the effect of the duration of an entire rainfall event on landslide occurrence. Although the indices derived using the two methods produced similar estimated effects of rainstorms on landslides, confirming the validity of both approaches for the estimation of rainfall properties, the two indices were better correlated with the distributed-landslide conceptual model than with the I-D, suggesting the superiority of the former model.
This study analyzes the landslide event at Hsiaolin Village during Typhoon Morakot in 2009. This landslide event resulted in 400 deaths. This event indicates that extreme rainfall conditions can alter the disaster mechanism of a landslide. The extremely high intensity and cumulative rainfall events may cause large-scale and complex landslide disasters. To study and understand the landslide event a combination of field investigations and numerical models are used. The landslide area is determined by comparing topographic information from before and after the event. Physiographic parameters are determined from field investigations. These parameters are applied to a numerical model to simulate the landslide process. And the landslide process was considered as two scenarios, a single event or a double event. For the single event, the main erosion area was the initial condition of the landslide movement. In Scenario II, the landslide process was divided into two parts. Finally, the simulation results were compared with investigation data. It shows the time of landslide movement is about 100 seconds. The deposited volume was well simulated. The deposited sediment formed a natural dam in the main channel. The minimum height of the dam is about 30 ~ 35 meter. The length of the dam is 1,100~1,600 meter.
On July 21, 2009, Hofu City in Yamaguchi, Japan recorded a precipitation of 256 mm/day which was the all-time high in the observation record. This rain triggered a disaster of debris flow containing large amount of fine particle sediment at the Tsurugi River located in a granite zone which reached houses and Route 262 at the valley exit and killed two people. Unlike pebble-type debris, the debris flow occurred in this disaster contained fine particle sediment, and high concentration of the debris reached the valley exit. Debris flow containing large amount of fine particle sediment like this is said to result in flows with higher concentration than pebble-type debris flow. Thus, hydraulic model experiments were conducted in this study to observe equilibrium concentration of debris flow in various waterway gradients using the sediment from the collapsed areas at the Tsurugi River which contained large amount of fine particle sediment. The experiment used collapsed sediment collected in the Tsurugi River where large amount of sediment was actually discharged. Using the outcome of these experiments and Takahashi’s equilibrium debris concentration formula as a basis, the authors suggested a method for calculating the concentration of debris flow and the amount of discharged debris which takes account of the effect of fine particle sediment while incorporating the concept of boundary grain size which defines fine particle sediment. This study found that conventional calculations for the amount of discharged debris targeting pebble-type debris flow may underestimate the amount of discharged debris in basins where debris flow containing fine particle sediment might occur.
Geological setting is crucial to the occurrence of slope failure and landslides, resulting in prolonged suspended sediment runoff; in this sense, local geology can affect the magnitude of suspended sediment yield. To clarify the suspended sediment sources in a watershed with various geological units, hydrological monitoring and fingerprinting techniques using natural radionuclides were conducted during a heavy rainfall event in August 2010 in the Nukabira River watershed, a tributary of the Saru River, in northern Japan. A GIS analysis for slope failure and landslide areas was also conducted to investigate the distribution of potential suspended sediment sources. During the rainfall event, the dominant sources of suspended sediment were found to be areas consisting of metamorphic rock (31%), sedimentary rock (30%), and accretionary sedimentary rock (24%). The highest sediment yield was found in metamorphic rock, which is consistent with the dense distribution of slope failure and landslides identified by the GIS analysis. Active landslides can introduce a significant amount of sediment from the hillslope to the stream channel, providing a source of fine sediment. In the area of the accretionary basalt block, there was an inconsistency between the sediment yield and the density of slope failure and landslides, perhaps attributable to the lack of a fine particle fraction in the sediment deposited on the bare slope along the stream channel and riverbed due to the resistance of bed rock to slaking and weathering. These results indicate that natural radionuclides can be used in order to elucidate the suspended sediment sources and sediment yield of various geological units within a watershed.
In the processes of designing reservoirs, planning river channels and managing downstream river environments, it is very important to clarify the mechanism of transport for suspended sediment in basins and to quantitatively evaluate the amount of sediment discharge. In this study, year-round monitoring and calibration were conducted using high-range turbidimeters at five observation points in the Mu and Saru river basins, and a practical estimation equation for the evaluation of suspended sediment (SS) concentration (mg/L) as determined from the measured turbidity value (ppm) at each point was proposed. The estimation equation is expressed as a simple linear function of discharge Q, and is able to accurately reproduce water sampling results and the turbidimeter measurements taken at the observation points. The characteristics of sediment discharge in the Mu and Saru river basins were also examined based on SS concentration values obtained from the year-round monitoring in 2010 using the simple estimation equation proposed. The study results revealed that the sediment discharge process and amount varied among the measurement points in the two basins, and that there were clear seasonal variations among the snowmelt, summer and other seasons in both basins.
In January 2011, a magmatic eruption occurred in Shinmoedake Volcano, which is part of the Kirishima volcanic group and is located in Kyushu, southwestern Japan. We employed C-band synthetic aperture radar (SAR) data acquired by RADARSAT-2 to examine ground surface changes caused by the volcanic activities (ash fall and volcanic bombs) for the prediction of sediment-related disaster. Change detection images recorded by the before-eruption and after-eruption SAR data revealed regional and local changes, especially around the craters and distribution area of volcanic ejecta. The Pauli color-coded image was classified into bare ground, forest area, and built-up area on the basis of the differences in the reflecting features, enabling us to detect a distribution of volcanic ejecta. Although our method cannot yield quantitative features, it is suitable for mapping widely distributed volcanic ejecta. Because SAR is an all-weather sensor, we found that our method of using SAR data was more efficient for investigating the volcanic eruption than image comparison using optical sensor data.
Indonesia is extremely vulnerable to climatic hazards and the frequency of disasters is gradually increasing. In the Kelara watershed of South Sulawesi, shallow landslides have been occurring every year in the last six years due to increases in rainfall intensity. As a result, roads and the local inhabitants often suffer from incidences of shallow landslides causing loss of life and wealth. The objective of this study is to determine rainfall thresholds for shallow landslides in the Kelara watershed. Thirty-six shallow landslides occurring from 2005 to 2010 were studied to analyze rainfall thresholds. Records of rainfall data and history of shallow landslides were collected from the Ministry of Public Works of the government of Indonesia. The threshold, defined by the lower envelope of the plots representing shallow landslides-triggering rainfall events, is expressed as I = 31D-0.72. The threshold rainfall for shallow landslides is very important information for warning developed in the study area.
Many of the dams and reservoirs in Japan have been operated for more than thirty years and some of them face sedimentation problems. In addition research and investigation of climate change is ongoing and its impact has been clarified recently. Therefore the authors try to predict future reservoir sedimentation considering impact of climate change by use of a sediment runoff model developed by Egashira and Matsuki. The authors consider that their sediment runoff model is convenient because a topographic model of the dam’s basin for numerical analysis is relatively easy to provide. However, the model does not have a sediment supply process from mountain slopes to river channels. This is an issue and applicability of the model is considered to be limited. In this paper the authors carry out correlation analysis between water inflow to reservoirs and reservoir sedimentation for several dams and analyze applicability of the model from the viewpoint of relationship between correlation coefficient and characteristics of the dam’s basin. As a result, the authors reach a conclusion that large specific sedimentation rate and gentle river bed slope near the reservoir entrance are important characteristics for applicability of the model. In the end the authors carry out simulation analysis of reservoir sedimentation against one of the dams and confirm acceleration of future reservoir sedimentation due to the impacts of climate change.
A freeze-thaw experiment was conducted to investigate the destruction mode of weathered bedrock in different lithologies. The experiment was conducted on bedrock samples of weathered granite (WGr), weathered granite porphyry (WGp), weathered shale (WS), weathered sandstone (WSa) and weathered rhyolite (WR). Because of freeze and thaw action, porosity of the bedrock samples increased and weight of the samples decreased. When porosity of WGr, WGp and WS increased greater than 0.43, 0.1 and 0.27, respectively, small pieces were detached from the main sample bodies. In contrast, freezing and thawing caused only slight increases in the porosity of WSa and WR and production of fine sediment from the surface. Depth of the sediment productions of WSa and WR due to one freeze-thaw cycle was equivalent to less than 1.0 and 0.2 mm, respectively. Hence, these rocks are exposed to only slight damage, which are at the ground surface. The experimental results were also used for developing a simple destruction model.