Debris flows cause flooding and sediment deposition when they reach alluvial fan areas. In Japan, urban development has led to housing commonly being built in such areas. Therefore, in the event of a debris flow, the location of the housing seems to affect flooding and deposition. Due to the nature of the housing and the often steep topography of the areas in between, roads are generally located at lower elevations ; during previous debris flow disasters, many cases have been reported where the flows moved down along these roads. In this study, we conducted disaster verification on the debris flow events that occurred in Hiroshima, Japan, in August 2014. Then we performed debris flow numerical simulations using the Hyper KANAKO system to determine the influence of housing and roads on the Hiroshima debris flow events. We performed the simulations with and without housing. The results show that the location of housing can explain the causes of the disaster. Considering the hazard area and evacuation procedure, it seems to be advisable to take the influence of housing into consideration. Furthermore, we conducted simulations by varying the width of the main road, because in a disaster situation, the main debris flow seems to move down along this road. In the simulation, we observed that the main debris flow moved along this road when the road was widened, while part of the flow spread laterally towards other roads when the main-road width was reduced.

Previously, we proposed a model called idH-SLIDER for assessing the time and location of landslides. The method was shown to predict landslides triggered by heavy rainfall and to demonstrate the influence of each modeled parameter. As the validity of this approach has not been confirmed, this study applied the method to areas in northern Yamaguchi prefecture, and attempted to reproduce 38 years of landslide occurrences using long-term successive hourly rainfall data, including multiple rainfall events. Many shallow landslides occurred in the study region in July 2013 and July 1983. The distribution of soil thickness was estimated using simple penetration tests and LiDAR data obtained after the 2013 rainfall event. Parameters were determined from soil tests, except for the saturated hydraulic conductivity, which was scaled in increasing order of magnitude from the default value of 1 to 10,000 times. It is well known that soil conductivity varies greatly. Calculations show that short-term rainfall data from 2013 were not distinguishable between saturated hydraulic conductivities of 1, 10, and 100 times. At 1,000 and 10,000 times, a few elements were unstable, preventing the modeling of landslides. We demonstrated the prediction of landslide occurrences using data recorded over 37 years, and identified saturated hydraulic conductivity of 100 to 10,000 times as an appropriate default. This method was more accurate than previous models designed for long-term datasets of longer than 10 years. The model is applicable to various rainfall patterns and geological environments.

For integrated catchment-scale sediment managements, it can be thought that the sediment yield induced by the subsequent expanding landslides due to rainfall is one of key processes. This study presents a quantitative analysis of expanding landslides by surveying the Shukushubetsu River basin, at the foot of the Hidaka mountain range in central Hokkaido, Japan. This area recorded heavy rainfall in 2003, reaching a maximum daily precipitation of 388 mm. We extracted the original and expanding landslides from 1963 to 2008 using aerial photographs. As a result, we found that in the period 2003-2008 (after the heaviest rainfall in 2003), the expanding landslide area was 2.7 times larger than the original landslide area. Also, the expanding landslide rate from 2003 to 2008 is 4 times greater than that from 1963 to 2003, while the original landslide rate after the heaviest rainfall in 2003 was similar to that before the heaviest rainfall. Maximum daily precipitation from 2003 to 2008 was not much higher than that from 1963 to 2002. We also found that around 70% of expanding landslide extended to upward or side. Thus, we concluded that due to the heaviest rainfall in 2003 gave a impact on the stability around the landslide scars triggered in 2003, then these affected area was landslide during the relatively small rainfall.

To understand the states of the groundwater level in soil layers and the trough discharge when sediment runoffs occur on a slope, numerous real-time landslide detection sensors have been deployed in a small basin in the Rokko mountain system and the groundwater level in soil layers and the trough discharge have been observed at multiple locations for about six years. During Typhoon No.11 in July 2015, real-time landslide detection sensors were activated and the data for the groundwater level in soil layers and the trough discharge at that time was obtained. Consequently, the following was revealed : 1) The sediment runoff during Typhoon No. 11 in July 2015 was not a surface failure but erosion and outflow of the surface soil owing to the traction force of stream. 2) The groundwater level in soil layers in the sediment runoff area was the highest groundwater level, causing sediment runoffs with a difference of a few centimeters in the groundwater level. 3) The rainfall that led to the highest groundwater level in soil layers varied depending on the location of the observation hole.

Sediment transportation in mountain streams has been observed by direct and indirect methods. Direct sampling methods are thought to be desirable. However, the instability and difficulties have been a major obstacle. A more radical technological breakthrough remains a holy grail. In the meantime it is imperative to make full use of indirect methods. For quality calibration, an effective application of hydrophone systems is sought out, after 15-year field experience of semi-direct sampling. The systems capture bed load transportation, which can be turned out as sediment-related quantities. There are three chief roadblocks : (1) sampling method itself, (2) estimation of unit section sediment, and (3) cross-sectional and lateral development to obtain total sediment discharges. Comparison among estimation methods, based on full year data retrieval and in site monitoring is made. The result tentatively indicates that the application of hydrophone systems produces more robust estimation, with minor but non-negligible discrepancy with prior estimation.

Three typhoons (T 7, 11, 9) struck Hokkaido within one week in August 2016. Heavy rain brought by the typhoons induced debris flows and sediment transport along the Kurodakesawa river and Shogakkonosawa in Sounkyo, Kamikawa Town, Hokkaido. Monitoring cameras for debris flows recorded the events, which provided an opportunity to describe some features of the debris flows and sediment transport. We also conducted ground and air (UAV) survey right after the series of events to supplement the monitored data. This article documents the rainfall record, geologic and geomorphic settings of the area and the monitored and surveyed data of the debris flows and sediment transport. We also report a rapid analysis of the monitored and surveyed data on sediment sources and sinks, sediment materials (type of rocks, sizes and distribution), depositional features, flow velocity and discharge. It was clear that check dams and training channel protected residential area and hotels located in Sounkyo hot spring area.

Since 2013, National Institute of Forest Science started the sediment transport monitoring in mountain rivers to clarify effects of forest damages due to forest fire and landslide on the sediment discharge from mountainous catchment. For this monitoring project, National Institute of Forest Science collaborated with National Institute for Land and Infrastructure Management. We used hydrophone (Japanese pipe hydrophone) for bedload monitoring and turbidity meter for suspended load monitoring. We set three observation stations in 2013, and other three observation stations in 2015. We also focus effects of bedrock geology on sediment discharge, thus, we conducted the sediment discharge monitoring at catchments underlain by three different bedrock geology, such as igneous rock, sedimentary rock and metamorphic rock. We successfully observed the sediment discharge during several heavy rainstorms to date.