It is important to take appropriate non-structural measures based on the Sediment-Related Disaster Prevention Act in order to reduce the number of victims from large-scale debris flow disasters―i.e., debris flow disasters induced by the failure of natural-dams or the deposits of volcanic ash-fall（hereinafter called “volcanic debris flow”）.Accordingtothe Act, the national government is responsible for providing “Emergency Information on Sediment-related Disasters （EISD）” to the municipal and prefectural governments, which is supposed to be issued on the basis of “Emergency Investigation”. Basically, EISD provides information about the areas that are hazardous and the periods for people in those areas to stay away. Technologies need to be developed to improve the precision of EISD, in particular information about the periods, in order to enhance the non-structural measures. The volcanic debris flows of active volcanoes, such as, Sakurajima and Mt. Shimmoe in the Kirishima mountain range, are fine examples that cause great concern to this day. The periods in which higher probability of volcanic debris flow occurrence are determined not only by rainfall intensity but also by multiple factors, including the physical and chemical characteristics of ash-fall, quantity of ash-fall, topography, surface geology, etc. This study, which aims to enhance the precision of EISD on volcanic debris flows, proposes a statistical analysis method （SAVER） to be applied to the analysis of the relationship between the occurrence of volcanic debris flow and the index taking into account both of the rainfall and the ash-fall intensities. From the results of this study, it became possible to assess the risk of volcanic debris flow occurrence considering the effects of the ash-fall deposits. The method also enables the establishment of non-linear critical lines for volcanic debris flow risk that can be modified according to the intensity of ash-fall. Thus, the critical lines are highly compatible with the criteria for issuing the “Warning Information on Sediment-related Disaster” and, hopefully, the lines can also be used for determining the time of termination for “Emergency Investigation”.
Shinmoe-dake in Kirishima volcano erupted violently on January 26, 2011 and spewed a large amount of volcanic ash and pumice over surrounding area, especially eastward and southeastward, in the following several days. We observed the states of the tephra fall deposits and traces of surface runoff at a lot of hillslopes around Mount Takachiho in May and June, 2011. The observation aimed at making clear some features of surface runoff along hillslopes covered with tephra fall by the eruptions. The result shows that traces of surface runoff are likely to occur on hillslopes where the surface is covered with volcanic ash densely. On the other hand, there are many forest slopes where no trace is found in spite of deposition of volcanic ash. According to visual observation, we observe that fallen leaves had formed gap structures into original surface layer and keep the surface permeable. The structure seemed to make it difficult to cover the surface of the ground with thin volcanic ash fall deposit. However, the permeability-holding effect of the gap structures seemed to be dismissed by the thick deposit of pyroclastic materials. It might be thought there are some necessary conditions to generate surface runoff. We estimate that one of them is existence of volcanic ash in the surface layer and another is thick deposit of pyroclastic materials. We also compared the distribution of traces of surface runoff with distribution of torrents with significant sediment discharge during rainfall in June and September, 2011. They are coincident to each other. In order to evaluate the post-eruption sediment discharge from the volcanically disturbed watersheds, however, more quantitative studies on hillslope hydrology and sediment dynamics are necessary.
It is considered that the extent of damage due to the sediment-related disaster induced landslides were influenced the scale of landslide. Meanwhile, the Scale of landslide was controlled by scale of slope. So, we investigated relationship between slope relief and scale of landslide using 3871 landslides data due to the Iwate and Miyagi inland earthquake in 2008. And we were identified 3871 landslides by two types of sediment movement. At result, landslide susceptibility was confirmed by using slope relief. Also, in the landslide area ranges from 100-1,000 m2 to 5,000-50,000 m2, the window size indicating the highest landslide mesh ratio is 3×3 in most cases. In the range of 10,000-100,000 m2 on both types of sediment movement, however, the window size of 5×5 when the cover ratio is about 0.5 to 0.7 indicates the highest landslide mesh ratio. The assessing of landslide susceptibility to earthquake using slope relief is not influenced by types of sediment movement.
It is well known that volcanic ash often generates surface flow on hillslopes as it decreases the infiltration rate of surface soil. Volcanic ash itself has a grain size of sand and has non-cohesive properties so it commonly has a high rate of water infiltration in laboratory tests. In spite of this, hillslopes covered by volcanic ash often generate surface flow, which results from the low infiltration rate of surface soil. Ordinarily, one of the reasons given for this low infiltration rate is the soil crust which develops on the surface of soil as a result of compression by drops of rainfall. This paper describes the vegetation and soil properties of Sakurajima, a volcanic island in southern Japan, where surface water and debris flow frequently occur. Laboratory tests on samples of the volcanic ash soil and in situ testing of surface flow were carried out. The test results demonstrated that slope surfaces covered by leaf litter or moss had very low infiltration rates, generating surface flow. To compound this, surface water was found to transport the ash and its deposition may have led to the formation of soil crust layers through a grading effect in the deposition process, giving the surface soil a low infiltration rate. However, if the slope has a large amount of living vegetation, such as Japanese pampas grass, surface flow rarely occurs. The presence of vegetation greatly improves the infiltration rate, limiting surface flow.
We examined effects of juvenile pyroclastic materials produced by the 2011 eruption of the Shinmoedake Volcano in the Kirishima Mountains on runoff and sediment discharge. The four characteristics of juvenile pyroclastic materials are thickness, grain size, porosity, and solidification. Pyroclastic materials consist of fine ash and coarse pumice, distributed mainly around the east and southeast areas of Shinmoedake, respectively. The thickness of the fine ash and the pumice deposits is more than 1 cm (maximum 3 cm) and more than 10 cm, respectively. The pumice particles are vesicular. Solidification materials were not reportedly found in the ash fall, so it is unclear whether the ash fall contained any solidification materials. Hyetographs and hydrographs of the eight rainfall events corresponded well. Although direct runoff related to total rainfall of less than about 100 mm was low in all the watersheds, direct runoff related to total rainfall of more than about 200 mm increased in the two watersheds whose headwater is located in the Shinmoedake and whose runoff coefficient was higher than that of the other watersheds. The two watersheds with a high runoff coefficient had a relatively thick ash fall deposit because of their closer proximity to the Shinmoedake Volcano compared to the other watersheds. In addition, these two watersheds were almost completely covered by fine ash. These facts suggest that the infiltration capacity in the watersheds temporarily decreased, and hence, there was a high level of discharge during rainfall events. The sediment movement after the 2011 eruption of Shinmoedake was only due to the high rainfall intensity (>30 mm/h). Because the pyroclastic materials were mainly porous pumice and the deposit of fine ash was thin, decrease in infiltration capacity was small. Therefore, sediment discharge was not triggered by low rainfall intensity events. To evaluate the potential for sediment discharge after an eruption, it is essential to investigate the infiltration capacity and the thickness of the ash fall, as well as the characteristics of juvenile pyroclastic materials considered in this study.
This is a prompt report of the observation results obtained by new equipments installed into the Arimura river in Sakurajima, Japan for field measurements of normal stress and cross section of moving debris flow. The normal stress was continuously obtained by the force plate designed by McArdell et al. (2007). The cross sections of debris flow were continuously obtained by a laser scanner. The data obtained in a debris flow event occurring on June 21, 2012 show that volumetric sediment concentration of the debris flow was around 40% at the time of the peak flow and decreased to around 30% with decreasing flow rate. The data obtained by the equipments show that normal stress and cross sectional area of debris flow observed were considered to be precise enough because values and temporal change of the data can provide reasonable explanations to the phenomena observed through video images. The data obtained by the field observations in Sakurajima are expected to be utilized not only to improve the method to design sabo dams but also to enable us to set more appropriate parameters of numerical simulations for obtaining debris flow inundation area.
The author briefly reviews the articles, reports, notes and presentation papers on debris movements or related issues around active volcanoes that have been published in the Journals and the annual workshops proceedings of JSECE. Research themes on sediment movements of two third are on debris flows and lahars that seems to appear the most frequent events in volcanic area are focused. The trend of the presentation numbers in the annual workshops seems to be followed the occurrences of volcanic eruption and related sediment hazards in these forty years. Particularly the Unzen Volcano eruption, 1990-1995, gave a lot of opportunities on researches in active volcanoes. The author classifies the agents on sediment movement initiation that are characterized in active volcanoes. Volcanic events such as pyroclastic flows and so on can be the trigger of sediment movements directly and juvenile pyroclastic deposit can be one of the mechanical factors of remobilizing sediment in the vicinity of active volcanoes. To enhance studies in volcanoes, the author makes a point the necessity of forecasting technique that will support the disaster prevention programs in active volcanoes in tomorrow. Field surveys can detect historical or geological eruptive events and sediment movements in detail. Some information about materials, processes of sediment movements and estimated volume are the sources of establishment for the sediment scenarios. To complete these matters, structural representation is helpful to understand events in active volcanoes in space and time series.