Search

This study added the dams and retain basin according to their dimensions measured with UAV onto the original 5m-resolition DEM to compare the effect of mitigation structures to debris flow hazard. The original and the modified DEMs were both applied to simulate the consequences by using RAMMS::Debris Flow (RApid Mass Movement Simulation) model.Hazard map is the best tool to provide the information of debris flow hazard in Taiwan. It has an important role to play in evacuating the residents within the affected zone during typhoon season. For the reason, debris flow hazard maps become a useful tool for local government to execute the evacuation. As the mitigation structure is constructed, the intensity of debris flow hazard reduces.The Nantou DF190 debris flow potential torrent is located in central Taiwan. In 1996 when Typhoon Herb stroke, 470,000 cubic-meter of debris were washed out and deposited in 91,200 square-meter area (Yu et al., 2006), and the event caused the destruction of 10 residential houses with 2 fatalities. After the event the Soil and Water Conservation Bureau constructed a 100-meter long sabo dam and sediment retain basin with capacity of 60,000 cubic-meters. In order to compare the difference of affected zone before and after the construction of mitigation structures, the study applies RAMMS to simulate the above-mentioned event.The result shows when large-scale debris flow occurs, most of the sediments still overflow and deposit on the fan with shape similar to the 1996 Typhoon Herb event. However, the intensity has reduced significantly with 50% less in area, several meters less in inundation depth and 50% less in flow velocity approximately. The comparison shows the effect of mitigation structures and could provide valuable information for debris flow hazard mapping.Key Words: Debris flow, RAMMS, Hazard map, Mitigation, Taiwan

In this study, several debris flow physical vulnerability curves and the even-based inundation depth were applied to a mountainous community hit by debris flow in 2015 to estimate the various possible loss ratio of each building. Then the comparison between estimated possible loss ratio and loss ratio determined by expert in the field is made to map out the distribution and deviation.Vulnerability is commonly related to the consequences of natural hazard. For debris flow hazard these consequences are generally measured in terms of losses (Fuchs et al., 2007). In risk management vulnerability is an essential component for analyzing natural hazard risks (Lo et al., 2012). It is expressed on a scale from 0 (no damage) to 1 (total loss) and increasing with the intensity of hazard.The Taoyuan DF034 debris flow potential torrent is located in northern Taiwan. In 2015, during Typhoon Soudelor the rainfall caused a shallow landslide which was transformed instantly into a debris flow. 13,000 cubic-meter of debris were washed out and deposited in 5,200 sq-meter area. Because of the evacuation before debris flow event, only 15 residential houses were inundated and no one was injured fortunately. In order to understand the inundation depth, the field investigation was executed shortly after the event. The building dimension, floor, structure type, location, and inundation depth were well documented and the loss ratio of each building was determined by expert as well.The comparison of loss ratio based on inundation depth and impact pressure between Kang and Kim (2016), Papathoma-Köhle et al. (2015) and Lo et al. (2012) is made. The result shows building characters and debris flow velocity affect the loss ratio significantly.Key Words: Debris flow, Vulnerability, Loss ratio, Taiwan

In quantitative risk analyses for natural hazards, vulnerability can be expressed as the ratio of reconstruction, replacement or reproduction expenses due to a damage caused by a certain process intensity and the original value of the element at risk exposed. To discuss the building vulnerability under debris flow events, the ratio is mostly related to debris flow inundation height, building materials and building values. Different types of buildings would resist to the impact of debris flows differently, resulting in different damage levels even under the same inundation height. After debris flow events, the damages to a building include the content loss and the structure loss, which is also variable due to the individual building conditions. This study proposes a flowchart to establish building vulnerability curves through estimating the damages to buildings after debris flow hazards. The losses of content and structure are firstly calculated separately to obtain the loss ratios with respect to original buildings. Secondly, by combining the content and structure loss ratio, the building vulnerability function is derived. In this paper, the original building content value was obtained from governmental statistic records and was based on the market price, and the structure value was received from a regional architecture office. The losses resulting from debris flow impacts were synthetically derived following field surveys. To combine the content and structure losses, a unit building with a floor area of 60 m2 was assumed. The result shows that due to a higher percentage of content value compared with the total building value, the loss ratio resulting from debris flows in Taiwan is higher compared with European studies, in particular with respect to high-frequency but low-magnitude events. The concept of obtaining building vulnerability is particularly suitable for regions where well-documented building loss records are unavailable.

Typhoon Megi (2010) had brought intense rainfall and hit the northeastern part of Taiwan. It caused a widespread rock fall, debris slide, rock slide, and debris flows along highway Route 9. The multi-hazard susceptibility map based on logistic regression approach and rock mass rating system (RMR) was adopted to evaluate the landslide susceptibility in this study. Firstly, the multi-hazard susceptibility model categorized the hazard potential for different geological hazards into low, moderate, and high level. Secondly, the study performed a geomechanical classification system for rocks according to field investigation along the vulnerable road section. The result reveals that a semi-quantitative hazard matrix combining with the both evaluations can assess the geological hazard appropriately. In addition, the high-resolution DEM extracted from airborne LiDAR and multi-temporal remote sensing images had been analyzed to examine the morphological change and geological structure for the landslide location with a high susceptibility. Finally, the detailed landslide susceptibility with three levels on highway is proposed and verified.

In order to understand the possible loss of building structure to debris flow, the establishment of vulnerability is necessary. With vulnerability, the loss can be calculated. This study followed HAZUS damage state probabilities concept to establish the fragility curves of building structure under debris flow impact. It indicates the relationship of damage state probability and spectral displacement related to velocity and inundation height. In this study, the assuming inundation height and velocity were taken to establish the fragility curves, and fragility curves were integrated into a unitary curve as a proxy of vulnerability of building structure with inundation height as X-axis and loss ratio as Y-axis. In this study, the brick structure type building in Taiwan was chosen to demonstrate the concept. The vulnerabilities and possible (predicted) loss ratio of brick structure could be established under various combinations of inundation height and velocity (ranging form 2 to 20 m/s).

In natural hazard research, risk is defined as a function of (1) the probability of occurrence of a hazardous process, and (2) the assessment of the related extent of damage, defined by the value of elements at risk exposed and their physical vulnerability. Until now, various works have been undertaken to determine vulnerability values for objects exposed to geomorphic hazards such as mountain torrents. Yet, many studies only provide rough estimates for vulnerability values based on proxies for process intensities. However, the deduced vulnerability functions proposed in the literature show a wide range, in particular with respect to medium and high process magnitudes. In our study, we compare vulnerability functions for torrent processes derived from studies in test sites located in the Austrian Alps and in Taiwan. Based on this comparison we expose needs for future research in order to enhance mountain hazard risk management with a particular focus on the question of vulnerability on a catchment scale.

In this study, we demonstrate the capability of combining remote sensing and GIS technique to identify and map out the possible landslide and debris flow hazards along the South Link Railway in Taiwan. Using high resolution aerial photography, terrain models, landslide inventories, the disaster prone areas of 3 types of mass movements were identified. In total, 48 slopes hold high landslide susceptibility; 35 torrents have high debris flow susceptibility; the rockfall from 11 sites would reach the railway line. The result shows consistencies with hazard records and could provide vital information for railway line improvement, hazard prevention and mitigation purposes.