Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/86751
標題: 都會區住宅邊坡崩塌潛感評估模式初探-以臺北市為例
Establishing a New Susceptibility Model for Landslides in Urban Areas: A Case Study of Taipei City
作者: Po-Chuan Chi
Che-Wei Shen
Shu-Yeong Chi
Li-Yuan Huang
Shih-Yuan Lin
Sin-Yuan Pan
紀柏全
沈哲緯
冀樹勇
黃立遠
林士淵
潘信元
關鍵字: Buffer strips;SWMM;Vollenweider model;Nonpoint Source Pollution;都會區;罕崩塌區;人工邊坡效應;極端降雨事件;山坡地住宅社區
Project: 中華水土保持學報, Volume 45, Issue 4, Page(s) 217-224.
摘要: 
目前針對廣域崩塌評估多採多變量統計為之,假設坡面為自然邊坡型態,考量地形、地質及水文因子建立崩塌潛感模式,然若研究區域為高度開發的都會區 (urban area) 或罕崩塌區 (rare event),常面臨崩塌樣本過少,以及無法量化人工邊坡效益之限制,以致無法獲致具代表性且正確率良好的崩塌預測模式。本研究以臺北市山坡地為研究範圍,採用極端降雨事件 (2001 年納莉颱風) 並納入人工邊坡不利條件機率之考量,建立「都會區住宅邊坡崩塌潛感評估模式」。分析結果顯示,考量人為因子時 (包含建築物密度、道路密度及人工邊坡因子) 潛感模式可得到最佳之正確率 76.9%,其中又以人工邊坡不利條件機率對於模式之影響最為顯著。此外,本研究提出以崩塌斜坡單元涵蓋率之概念研提兼具合理性與學理依據之潛感分級門檻,透過其他五起颱風事件之驗證,本研究所提出之分級門檻與原則可適用於 2012 年前重大颱風事件下之潛感分級。

Recent studies of regional landslide susceptibilities have been approached through multivariate statistics, taking only natural causes into account (viz. terrain, geology or hydrology variants). They offer, however,limited predictive accuracy for predicting rare events in urban areas and the effects of man-made construction. Thus,this study establishes a new susceptibility mode, taking into consideration the effects of construction (including density of structures, density of roads, and disadvantageous index of man-made structures). The rainfall of Typhoon Nari,a well known extreme rainfall event in Taipei city, was used as the triggering factor in the susceptibility model. The results show that the accuracy reached 76.9% for both scenarios with natural causes and the effects of construction.Among these factors, the most crucial role was played by the disadvantageous index of man-made structures in the model. This study also suggests a set of thresholds for susceptibility classification and verifies its suitability when appliedto other typhoon events after 2000.
URI: http://hdl.handle.net/11455/86751
Appears in Collections:第45卷 第04期

Files in This Item:
File SizeFormat
86750-1.pdf1.89 MBAdobe PDFView/Open
Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.