This paper examines the efficiency of an index derived from a distributed landslide conceptual model (DLCM) at demonstrating the influence of rainstorms on shallow landslides compared with two other indices : the soil-water index (SWI) and antecedent precipitation index (API). SWI is derived from a three-layer tank model, and its absolute value is defined as the sum of the water depths in each tank. API is used for estimating the influence of antecedent precipitation on landslides based on the half-life period. Both of these indices have been applied widely and used for an early warning system in Japan. The validity and efficiency of both have been confirmed based on many actual operations. The DLCM is based on the vertical and shallow lateral infiltration flow due to three-dimensional topography, and a safety factor is estimated using infinite slope stability analysis. The calculation of the vertical infiltration flow in this model is incorporated into the concept of the half-life period in API. The index derived from this model shows the temporal fluctuation of an unstable area. The three indices were applied to the western area (approximately 10 ㎢) of Izu-Oshima Island, Tokyo, Japan, where typhoons Wipha (2013) and Ida (1958) caused severe sediment disasters. The comparison of the index derived from the DLCM with other two indices gave the following results : Result 1 : The very strong positive correlation between the DLCM-derived index and two other indices was confirmed, supporting the efficiency of this index. Result 2 : Applying the DLCM index to the early warning system, the official announcement of a sediment disaster alarm would be set to the occurrence and disappearance time of an unstable area. This methodology is simpler than the current early warning system in Japan.

During sediment disaster, multi-hazard (e.g. landslide, debris flow, flood, inundation, and strong rainfall) can make evacuation difficult or dangerous. In order to clarify the safety of the evacuation, the estimation of the occurrence timing of each phenomenon is very important. Japan Meteorological Agency and prefectures cooperatively operate the sediment disaster alert system using the snake line and critical line on the two-dimensional plane whose horizontal and vertical axis are the soil water index and cumulative rainfall in 60 minutes. However, the alert does not contain any information on the kind of hazards or the evacuation possibility. In order to evaluate the risk of the multi-hazard, we have developed the multi-hazard simulator SiMHiS based on the landslide prediction model and water and sediment runoff model in the watershed scale. In this simulator the risk level of the 3 categories of multi-hazard related to rainfall, landslide, and flood can be obtained spatially and temporally. However, the landslide prediction model in the simulator has a problem with the decreasing process of water content. To solve the problem, before employing the simulator, we modified the numerical model using the theory of saturated seepage flow. In this study, we applied the improved simulator employing the several rainfalls whose maximum intensity and duration are different. The estimated risk level of the hazards was arranged on the same two-dimensional plane used in the sediment disaster alert system in Japan. As a result of the arrangement, the boundary diagram of the probable phenomenon is obtained on the plane at overlaying the critical lines of the flood, landslide and rainfall related hazards. Using the diagram, the kind of hazards and evacuation possibility can be estimated moment by moment. Also, an additional information related to the probable phenomena and evacuation can be supplied with the sediment disaster alert employing the diagram.

We started using cooperation methods with the intention of providing sediment disaster warning information nearly 10 years ago. However, setting a critical line using this method is problematic in the case of heavy rains. Also, it has been pointed out that the meaning of any such critical line is unclear. In this study, we examined a range of values at a radius of 4 km from the Okayama Prefecture Tamano Observatory. We analyzed rainfall data to determine when the critical line was crossed, and found that some heavy rains occurred within the critical region. Due to this fact, and noting the shape of the functions using the RBFN method, which is central to the cooperation method, we proposed basing the response surface on a circle. When such a surface was considered, the critical line was not crossed during times of heavy rain at which no disaster occurred. Additionally, there is a problem with the proposed cooperation method that makes the RBFN line unclear. To solve this problem, we calculated the RBFN value for each rainfall incident. Then, we evaluated the RBFN value in terms of probabilities corresponding to the yearly probability of crossing the critical line. For example, it is possible to set the critical line in terms of the probability of it being exceeded over 20 years. Our proposed critical line differs to the previous version with respect to short-term rainfall measures. Our formulation enables us to calculate the probability of the critical line being exceeded. This calculation takes into account the soil conditions and the effect of 60 minutes of continuous rainfall.

Based on the sediment-related disaster prevention law established in 2000, special sediment-related disaster hazard areas are being created across the country. However, analyses of methods to implement a policy from this law regarding the relocation of houses from sediment disaster special alert areas have not been completed until relatively recently. In Japan, damage from sediment disasters is a concern due to heavy rains. Therefore, it is necessary to rationally consider housing relocation as a preventative measure to protect residents from sediment disasters. This study analyzed past results, administrative issues, and the considerations of residents related to housing relocation. In addition, support programs for relocating houses in special sediment-related disaster hazard areas were examined. Finally, these analyses were used to examine the correspondence necessary for housing relocation and proposed methods of supporting the policy for existing disqualified houses.

In steep mountainous regions, not only soils but also weathered bedrocks are sometimes sliding simultaneously. These landslides often move rapidly and trigger debris flow and sometimes induce landslide dam. In this study, these landslides are referred to “deep-seated rapid (catastrophic) landslide” (hereafter DCL). In the last decade, several methods for assessing DCL susceptibility have been proposed. However, it is still difficult to assess (1) probable scale of future DCL and (2) the area prone to DCL. Here we proposed new method for assess DCL induced hazard of each village. In this method, we surrogate the potential hazard induced by DCL as the number of historical disasters due to DCL. The number of historical disaster was estimated by the number of ancient landslide scars, area of scars and distance between scars to villages.

It has been recognized that a linear-shaped mesoscale convective system (LsMCS) causes the localized, torrential, and stationary rainfall. In this paper, the characteristics of LsMCS during sediment disaster events are investigated using radar data which has 5 minutes time interval and 1 km horizontal resolution. To extract the LsMCS from radar data, a contiguous regions of convective cell group is identified with more than 20 mm/h of rainfall intensity and more than 300 k㎡ of the area. The shape of the convective cell group is approximated as an ellipse and the motion vector is calculated from the change of position within the consecutive radar data using the correlation coefficient method. The characteristics of the shape, motion and rotation were analyzed using radar data in sediment disasters caused by LsMCS. As the result, it was found that the criteria for detecting LsMCS are as follows. (1) The aspect ratio (major axis/minor axis) of ellipse is more than about 2.0. (2) The motion vector of the ellipse is rotated in the counterclockwise direction from the long axis of the ellipse and the angle of rotation is less than about 22.5 degree. (3) The weighted center of convective cell group is shifted to the windward side as compared to the center of ellipse.

Typhoon No. 10 which made landfall in the area around Ofunato in Iwate Prefecture on the 30 th August 2016 caused extensive damage due to rivers flooding and sediment disasters in many municipalities, towns and villages on the eastern side of the Kitakami Mountains including Miyako and Kuji, isolation of communities due to roads being cut off and landslides. The number of landslides that occurred in Iwate Prefecture reached 155 (as investigated by the Ministry of Land, Infrastructure, Transport and Tourism) and the Japan Society of Erosion Control Engineering organized a research team immediately after the disaster with the aim of investigating the processes of generation, flow and deposition of sediment and clarifying what actually occurred during the disaster. Our findings are published here.

Soon after three typhoons (No. 7, 11, and 9) hit Hokkaido in August 2016, Typhoon No. 10, Lionrock brought a large amount of orographic rainfall on the eastern slope of the Hidaka Range from August 29 to 31, which caused substantial discharge of sediment. Nine tributaries of the Tokachi River, from the Panke-Shintoku in the north to the Totsutabetsu in the south, discharged the huge volume of sediment by debris flows and experienced bank erosion and flooding, which caused damages in residential area, railway, expressway and national highway and so on. For example, the Pekerebetsu River discharged at least 0.63 million ㎥ of sediment, the largest amount in approximately last 50 years. Bank erosion and flooding caused aggradation of the river bed and blockage at bridges (many drift wood were included in the sediment) and washed away several houses in the center of Shimizu Town. The sites of the substantial sediment discharge (e.g., Totsutabetsu, with the total precipitation of 530 mm, and Nissho, with 367 mm) coincided not with an area of the heaviest precipitation (Notsuka, with 713 mm) but with granite geology. The sediment was mainly composed by granite boulder and fine material which came from gently sloping periglacial terrain.