After a volcanic eruption, urgent field surveys should be conducted because the susceptibility of debris flows increases. However, under the condition of volcanic activity, it is difficult to survey eroded areas near the craters. Thus, some indirect surveys using topographical information are necessary to sophisticate the disaster prevention system in volcanic regions. This study estimated a deep learning model to predict eroded areas by inputting the angle and Laplacian of a 100m times 100m area, discussed the quantitative accuracy of the estimated model. The analytical results revealed that the deep learning results smoothed by median filter could reproduce the training data with high accuracy and precision.

View full abstract

A questionnaire survey was conducted in the Jinnai District, Kumamoto City, where several relocations occurred due to river improvement following damage caused by the Northern Kyushu Rain Storm Disaster in July 2012. In their responses to the survey, many of the remaining households reported that they were dissatisfied with the number of relocated units, especially those who were affected by the disaster. In addition, it is thought that a large number of the remaining households continue to worry about flood damage because the river channel has approached the residences of the remaining households even after the renovation. In addition, among the relocated households, half wanted to remain when they were asked to relocate at the beginning, but following negotiations, many of them were convinced to relocate. The reason for this may be that the relocated households received sufficient compensation. Other findings suggest that a certain number of relocated households have selected inundationrisk locations as relocation destinations and are unfamiliar with the post-relocation area.

View full abstract

The use of former edition maps for disaster risk learning is effective, but requires proficiency of instructors. In this study, we developed a spatiotemporal GIS with many layers in addition to the former edition maps, and applied it to a special exhibition on the Typhoon Vera (the Isewan Typhoon) in 1959, Japan. And we evaluated the effectiveness of spaciotemporal GIS from the system logs and the questionnaire survey to the visitors. The results showed that a certain percentage of users used the spaciotemporal GIS for scenariobased learning, and this percentage increased in conjunction with the special exhibition. At the same time, the results of the questionnaire survey showed an improvement in the user’s satisfaction. From these results, the effectiveness of the method was confirmed.

View full abstract

Reclaimed valleys may become deformed during large earthquakes. Therefore, in this study, we examined the validity of the method for extracting the deformed area of the reclaimed valleys where shifts in the landforms occurred during the 2016 Kumamoto and 2018 Hokkaido Eastern Iburi earthquakes. The deformation was detected by interferometric synthetic aperture radar (InSAR), and the validity of the findings were determined by comparing the results of field surveys and distribution maps of the reclaimed valleys that were created using the Structure from Motion and Multi-view Stereo (SfM-MVS) technique. Results showed that in the case of small incident angle of the radio wave, the sites detected as localized phase change areas or non-interference areas in the SAR interferograms were consistent with the area of the actual deformation in the reclaimed valleys, indicating the effectiveness of interferometric SAR during disaster response after an earthquake at detecting reclaimed valley deformation. At one site in a reclaimed valley, the degree of change in elevation could be estimated from the difference in the digital surface models generated by SfM-MVS technique from aerial photographs taken before and after the earthquake.

View full abstract

Several stone bridges were damaged during the 2016 Kumamoto earthquake. The most severe damage was the collapse of the stone walls. It is reported that the condition of filling materials has a strong influence on the stone wall collapse of the Futamata Fukura Bridge. With this evidence, this study focuses on the size effect of filling stones on the seismic strength of stone bridges. We carried out shaking table tests using two simplified stone bridge models with different crushed stone sizes. It was observed that the model filled with larger crushed stones could resist higher acceleration than the model filled with smaller crushed stones. The numerical simulations using the refined DEM were conducted, and the same tendency that larger crushed stones improve seismic resistance was obtained.

View full abstract

When bridges in mountainous rivers are blocked with accumulated woody debris, a high flow depth leads to overflow and causes damage. Recent studies have focused on straight trunk shape. However, woody debris showing shape with curved trunk have been reported, but the trunk shape influence is not clear. In this study, we conducted field survey in Higashi- Hiroshima after 2018 disaster focusing on curved trunks woody debris, and proposed two representative curved trunks; single wavelength and half wavelength with different curving degree. We conducted laboratory experiments using woody debris model with straight trunk and two types of curved trunk. We also changed the mixing ratio of curved and straight trunk. From experiment results, curved trunk caused blocking in smaller woody debris comparing with straight trunk in both shape and in all mixing ratio. We applied the experimental results to examine the blocking possibility with logistic regression analysis. From the results, half wavelength with large curving degree showed higher possibility of blocking comparing with single wavelength.

View full abstract

This study is an analysis of the actual relationship between the awareness of flood / sediment disaster risk and the resident evacuation behavior focusing on the evacuation behavior of residents who did not evacuate while recognizing disaster risks. We performed various cross-tabulation analysis using the questionnaire survey data obtained in 3 cities and 1 town (Takayama city, Gero city, Seki city, Shirakawa town) in Gifu prefecture in the July 2020 heavy rain disaster. As a result, it was clarified that the confirmation of hazard maps and the awareness of disaster risks were rarely linked to appropriate evacuation behavior of residents.

View full abstract

In this study, physical experiments were performed to investigate the influence of two key morphological factors in paddy fields, i.e., paddy ridges and irrigation channels, on sediment deposition during fluvial inundation. The inlet boundary was given constant water and sediment discharges to simulate embankment failure conditions. Hydraulic parameters show that sediment came into the floodplain in the form of suspended load. By changing the outlet boundary and channel outlets conditions, three groups of experiments were conducted. The results show that, under the high-water-level scenario (with a wall at the outlet boundary), sediment mainly deposited along the main flow path, and the influence of paddy ridges was only constrained in its surroundings because of the three-dimensional flow characteristics. The operation of the channel outlets effectively changed the water depth and main flow direction and accordingly influenced the sedimentation distribution on the land surface. Under the shallow-water-depth scenario (open outlet boundary), the ridges on the bed retained more sediment by decreasing the bed-load transport rate. The intersection of the channel and main flow path was blocked by the bed load from the land surface. Hence, setting ridges close to the channel could remarkably relieve sedimentation. The experimental results are helpful for better understanding the mechanism of sediment deposition on paddy fields with complex topography and can provide data for numerical model verification.

View full abstract

The purpose of this study is to investigate the sustainability of the effect of disaster preparedness education and to clarify the factors that promote the diffusion on families. Disaster preparedness education classes were conducted for fifth grade students (ages 10–11) at 15 elementary schools in Miyagi and Fukushima Prefectures, and a questionnaire survey was conducted on childrenʼs awareness of disaster prevention and their behavior at home. The results were as follows. a) For most survey items, the studentsʼ scores were higher after the class than before, but their scores had dropped one month afterward. b) Regarding the “Residential Area Disaster Risk Assessment,” the scores raised by classes in coastal areas remained at the same level after one month. c) The students in coastal areas scored higher than students in inland areas in overall. d) The diffusion effects of disaster preparedness education to families were positively related to the willingness to learn about disaster prevention, the self-efficacy of disaster mitigation, and the prediction of appropriate evacuation behavior of family members during disasters, and negatively related to the fear of natural disasters. These results suggest the importance of continuing disaster preparedness education and the need for intervention for children in the inland areas. In addition, the factors for diffusing the effect of disaster preparedness education to the home were discussed.

View full abstract

The “when-to-do” conflict refers to a situation in which there is general agreement on the necessity of an activity but public backlash is expected if the timing of the activity is inappropriate. Economic recovery is undoubtedly necessary after a large-scale disaster such as Nankai megathrust earthquake, but so are search and rescue activities and humanitarian logistics. Nevertheless, economic activities in disaster-affected areas are often fraught by “when-to-do” conflicts, especially in the early stage of the recovery process, when people tend to demand focus on disaster relief. To manage these conflicts, it is important to identify the timing for the commencement of diverse social and economic activities following a disaster (we call this post-disaster timelines [PDTs]). We surveyed residents’ PDTs in Aichi prefecture, Japan, and observed significant differences among residents. The structure of the differences could be classified as “concentrated type,” “quasi-concentrated type,” and “ dispersed type.” These findings carry significant implications for managing“when-to-do” conflicts.

View full abstract

In this study, we analyzed the position data of the victims and 1km mesh rainfall observation data in the heavy rains of July 2018, Typhoon No. 19 of 2019, and the heavy rains of July 2020, and tried to estimate the number of victims using various rainfall indicators. As a result of analyzing the relationship between rainfall indicators such as 3, 6, 12, 24, 48, 72 hour rainfall and soil water index and the number of victims, it was found that the “ratio to historical record”, which is the ratio to the maximum value observed in the past, is highly related to the occurrence of victims. Since the rainfall indicator that corresponds well to the occurrence of victims differs depending on the case of heavy rainfall, we used the maximum value of past maximum ratio of 7 indicators and were able to roughly estimate the number of victims caused by heavy rain disasters.

View full abstract

This study presents ash-fall hazard maps for an eruption of Sakurajima volcano of similar scale to the Taisho eruption based on tephra transport simulations accounting for daily variations in wind fields over a 1-year period. Existing hazard maps indicate deposition of ash on the eastern side of Sakurajima by westerly winds, which reflects the mean annual wind direction. This study analyzed ash deposition by implementing daily-base computations with no prior statistical processing. The results reveal that the maximum-class deposits occur around Kagoshima Chuo Station focused in August, when the wind direction is not annually predominant one. The resultant hazard map based on the maximum thickness of the deposits differs from existing maps and provides useful information for creating disaster evacuation plans. Further hazard maps are proposed based on the frequency of the ash-fall exceeding a particular thickness and on a highest 1% deposit thickness, and we discuss suitable methods of expressing ways responding to the users’ class and object.

View full abstract

It is important for high-speed trains to early slow down and stop in the event of an earthquake. Shinkansen system is equipped with an earthquake early warning (EEW) system that transmits warnings based on seismometer observations. In recent years, an ocean bottom seismometer (OBS) observation network has been installed offshore in the Pacific Ocean. Subduction-zone earthquakes, such as the 2011 off the Pacific coast of Tohoku Earthquake, may be detected earlier if OBS can be utilized for the EEW system. Therefore, we developed a method to utilize OBS for the EEW system, in which we proposed a method of setting an OBS threshold value and a method of preventing false warnings. Furthermore, we evaluated its intended effects. The East Japan Railway Company started utilizing OBS in the Boso offshore area in November 2017, and in the offshore area between Ibaraki and Kushiro in January 2019.

View full abstract

In large-scale disasters, human resource management in disaster response was inefficient due to uncertainty in knowing the work volume. In this research, we conducted a questionnaire survey on disaster-affected local governments after Kumamoto Earthquake. A prediction equation of work volume for shelter management was developed after classification by local government size and degree of damage and elimination of outliers. The prediction equation shows a different trend between the ordinance-designated city and other cities, which is assumed to reflect local government size and characteristics between urban and rural areas. This formula will help local governments assess work volume on shelter management before a disaster, which leads to efficient disaster response and prevents unnecessary chaos.

View full abstract