{"title":"“莫拉克”台风后台湾柳尾试验林滑坡与地形因子的空间关系","authors":"Shiang-Yue Lu, Chao-Yuan Lin, Liang-Shin Hwang","doi":"10.7075/TJFS.201112.0088","DOIUrl":null,"url":null,"abstract":"Typhoon Morakot produced copious amounts of rainfall and triggered enormous landslides in southern Taiwan when it hit Taiwan on 8~9 August 2009. The Liukuei Experimental Forest (LEF) managed by the Taiwan Forestry Research Institute was located in the peak rainfall region and inescapably incurred severe landslides and road destruction. This paper presents a statistical approach to study spatial relationships between landslides and their geographic factors in the LEF after typhoon Morakot. The normalized difference vegetation index (NDVI) of satellite images SPOT2 and SPOT4 photographed before and after typhoon Morakot were compared to extract landslide patches. Those extracted patches, consisting of 10 x 10-m digital elevation model (DEM) data and the related coverage of vegetation of LEF, were examined using the WinGrid computer software for the topographic analysis. Results indicated that the extremely heavy rainfall was the main factor that triggered landslides during the typhoon. There were 204 locations which experienced identifiable landslides, and their total area was about 804.49 ha. The elevation and types of vegetation cover were not factors related to the failure of the slopes. Steepness, aspect, and distances to roads and streams were factors that contributed to instability of the slopes. About 70.2% of the total area of landslides occurred on slopes with steepness in the range of 60~80%, and this showed that steep slopes are prone to collapse. Landslides with distance of the lower edge to a stream of < 10 m occurred in 49 locations, and their areas accounted for 33.6% of the total landslide areas. In addition, 34 landslides (24.7% of all landslides areas) occurred within 50 m of a forest road, and the average area of those landslides was 5.79 ha. The deeply concentrated surface flows resulting from the extreme rainfall in streams and on road surfaces striking against the foot of hillslopes that contained saturated soil water were the main factors causing slides of large area. This also indicated that landslides in the LEF were closely related to stream systems, and forest road construction. The aspect of slope was not the main factor related to landslides in most cases. However, slopes with southerly and southeasterly aspects directly faced the strong winds and heavy rainfall brought by typhoon Morakot, and more landslides occurred on those slopes than on slopes with other aspects.","PeriodicalId":22180,"journal":{"name":"Taiwan Journal of Forest Science","volume":"114 1","pages":"399-408"},"PeriodicalIF":0.0000,"publicationDate":"2011-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Spatial Relationships between Landslides and Topographical Factors at the Liukuei Experimental Forest, Southwestern Taiwan after Typhoon Morakot\",\"authors\":\"Shiang-Yue Lu, Chao-Yuan Lin, Liang-Shin Hwang\",\"doi\":\"10.7075/TJFS.201112.0088\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Typhoon Morakot produced copious amounts of rainfall and triggered enormous landslides in southern Taiwan when it hit Taiwan on 8~9 August 2009. The Liukuei Experimental Forest (LEF) managed by the Taiwan Forestry Research Institute was located in the peak rainfall region and inescapably incurred severe landslides and road destruction. This paper presents a statistical approach to study spatial relationships between landslides and their geographic factors in the LEF after typhoon Morakot. The normalized difference vegetation index (NDVI) of satellite images SPOT2 and SPOT4 photographed before and after typhoon Morakot were compared to extract landslide patches. Those extracted patches, consisting of 10 x 10-m digital elevation model (DEM) data and the related coverage of vegetation of LEF, were examined using the WinGrid computer software for the topographic analysis. Results indicated that the extremely heavy rainfall was the main factor that triggered landslides during the typhoon. There were 204 locations which experienced identifiable landslides, and their total area was about 804.49 ha. The elevation and types of vegetation cover were not factors related to the failure of the slopes. Steepness, aspect, and distances to roads and streams were factors that contributed to instability of the slopes. About 70.2% of the total area of landslides occurred on slopes with steepness in the range of 60~80%, and this showed that steep slopes are prone to collapse. Landslides with distance of the lower edge to a stream of < 10 m occurred in 49 locations, and their areas accounted for 33.6% of the total landslide areas. In addition, 34 landslides (24.7% of all landslides areas) occurred within 50 m of a forest road, and the average area of those landslides was 5.79 ha. The deeply concentrated surface flows resulting from the extreme rainfall in streams and on road surfaces striking against the foot of hillslopes that contained saturated soil water were the main factors causing slides of large area. This also indicated that landslides in the LEF were closely related to stream systems, and forest road construction. The aspect of slope was not the main factor related to landslides in most cases. However, slopes with southerly and southeasterly aspects directly faced the strong winds and heavy rainfall brought by typhoon Morakot, and more landslides occurred on those slopes than on slopes with other aspects.\",\"PeriodicalId\":22180,\"journal\":{\"name\":\"Taiwan Journal of Forest Science\",\"volume\":\"114 1\",\"pages\":\"399-408\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2011-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Taiwan Journal of Forest Science\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.7075/TJFS.201112.0088\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"Agricultural and Biological Sciences\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Taiwan Journal of Forest Science","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.7075/TJFS.201112.0088","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Agricultural and Biological Sciences","Score":null,"Total":0}
引用次数: 2
摘要
台风莫拉克于2009年8月8日至9日登陆台湾,带来了大量降雨,并在台湾南部引发了巨大的山体滑坡。台湾林业研究所管理的柳尾实验林位于降雨高峰区,不可避免地发生了严重的山体滑坡和道路破坏。本文采用统计方法研究了莫拉克台风后左风区滑坡及其地理因子的空间关系。利用莫拉克台风前后卫星影像SPOT2和SPOT4的归一化植被指数(NDVI)进行对比,提取滑坡斑块。利用WinGrid计算机软件对提取的10 × 10 m DEM数据及其植被覆盖度进行地形分析。结果表明,台风期间的特大暴雨是引发山体滑坡的主要因素。有204个地点经历了可识别的滑坡,总面积约804.49公顷。高程和植被覆盖类型与坡面破坏无关。坡度、坡向以及与道路和溪流的距离是导致斜坡不稳定的因素。约70.2%的滑坡发生在陡度在60~80%之间的斜坡上,这表明陡坡容易发生崩塌。滑坡下缘距离小于10 m的滑坡有49处,滑坡面积占滑坡总面积的33.6%。此外,在森林道路50 m范围内发生滑坡34次(占滑坡总面积的24.7%),平均滑坡面积为5.79 ha。极端降雨在河流和路面上撞击含有饱和土壤水的山坡脚下而形成的深度集中的地表流动是造成大面积滑坡的主要因素。这也表明,东部地区的滑坡与水系和森林道路建设密切相关。在大多数情况下,坡向并不是影响滑坡的主要因素。而南向和东南向坡面直接面对台风莫拉克带来的强风和强降雨,其滑坡发生率高于其他坡面。
Spatial Relationships between Landslides and Topographical Factors at the Liukuei Experimental Forest, Southwestern Taiwan after Typhoon Morakot
Typhoon Morakot produced copious amounts of rainfall and triggered enormous landslides in southern Taiwan when it hit Taiwan on 8~9 August 2009. The Liukuei Experimental Forest (LEF) managed by the Taiwan Forestry Research Institute was located in the peak rainfall region and inescapably incurred severe landslides and road destruction. This paper presents a statistical approach to study spatial relationships between landslides and their geographic factors in the LEF after typhoon Morakot. The normalized difference vegetation index (NDVI) of satellite images SPOT2 and SPOT4 photographed before and after typhoon Morakot were compared to extract landslide patches. Those extracted patches, consisting of 10 x 10-m digital elevation model (DEM) data and the related coverage of vegetation of LEF, were examined using the WinGrid computer software for the topographic analysis. Results indicated that the extremely heavy rainfall was the main factor that triggered landslides during the typhoon. There were 204 locations which experienced identifiable landslides, and their total area was about 804.49 ha. The elevation and types of vegetation cover were not factors related to the failure of the slopes. Steepness, aspect, and distances to roads and streams were factors that contributed to instability of the slopes. About 70.2% of the total area of landslides occurred on slopes with steepness in the range of 60~80%, and this showed that steep slopes are prone to collapse. Landslides with distance of the lower edge to a stream of < 10 m occurred in 49 locations, and their areas accounted for 33.6% of the total landslide areas. In addition, 34 landslides (24.7% of all landslides areas) occurred within 50 m of a forest road, and the average area of those landslides was 5.79 ha. The deeply concentrated surface flows resulting from the extreme rainfall in streams and on road surfaces striking against the foot of hillslopes that contained saturated soil water were the main factors causing slides of large area. This also indicated that landslides in the LEF were closely related to stream systems, and forest road construction. The aspect of slope was not the main factor related to landslides in most cases. However, slopes with southerly and southeasterly aspects directly faced the strong winds and heavy rainfall brought by typhoon Morakot, and more landslides occurred on those slopes than on slopes with other aspects.
期刊介绍:
The Taiwan Journal of Forest Science is an academic publication that welcomes contributions from around the world. The journal covers all aspects of forest research, both basic and applied, including Forest Biology and Ecology (tree breeding, silviculture, soils, etc.), Forest Management (watershed management, forest pests and diseases, forest fire, wildlife, recreation, etc.), Biotechnology, and Wood Science. Manuscripts acceptable to the journal include (1) research papers, (2) research notes, (3) review articles, and (4) monographs. A research note differs from a research paper in its scope which is less-comprehensive, yet it contains important information. In other words, a research note offers an innovative perspective or new discovery which is worthy of early disclosure.