| 摘要: | 近年來極端氣候對全球各地城市所造成的影響甚鉅，不僅各類危害及災害發生頻率、強度提高，都市集居及各種開發行為更使災害型態趨於複雜，各項防災規劃成為都市規劃重要考量因子之一。西元2005年世界銀行災害高風險評估報告（Natural Disaster Hotspots- A Global Risk Analysis），臺灣同時暴露於三項以上天然災害的土地面積與面臨災害威脅的人口均數為73%，而同時暴露於二項天然災害的土地面積與面臨災害威脅的人口均數更高達99%，以上兩數據臺灣均占世界各國之鰲頭，顯示臺灣居民承受極大的天然災害風險。
本研究目的以極端氣候下對都市災害類型探討，對顯著影響的氣候變遷災害因子篩選，並考量都市規劃具體策略，以設施興建與強化、土地使用之規劃管理及策略、民眾宣導及訓練、財稅制度及政府部門管理與合作等面向，作為具體的都市規劃策略指標。
為評估複雜的氣候變遷因子之影響，以及探討政策的有效性，因此透過文獻回顧與資料收集，歸納整理出適切的評價準則，並藉由專家問卷、模糊德爾菲法（Fuzzy Delphi Method，FDM），評估出氣候變遷下都市防減災指標，並且以分析網路程序法（Analytic Network Process，ANP）進行氣候變遷下都市規劃策略方案之評選，建立氣候變遷下都市規劃方法之評估模式。
研究發現「洪水與暴風」、「山崩」、「洪氾」、「海岸地區土地使用」、「洪水風險」、「都市人口密度」、「土石流保全人數」、「建築面積」、「淹水潛勢」等9項指標適宜作為都市防減災之評估標的。且在「土地使用管理規劃、「民眾宣導及訓練」、「設施興建/強化」、「政府部門管理與合作」、「財稅制度」適宜作為重要的管制策略，尤其以「土地使用管理規劃」策略最為突出。
在實證驗證上，以臺中市高鐵門戶區計畫擬定過程為探討，應用氣候變遷下都市規劃策略指標評估模式，提供都市規劃決策者客觀且具可信度之參考資訊及快速評估方法，符合現今迫切需要減緩劇烈氣候變遷下所帶來都市問題的時空環境，並可作為後續都市規劃之政策推動與相關研究之參考依據。
研究結論為非結構性減災以空間規劃管理為最重要的影響，都市規劃策略指標有： 「土地使用規劃與管理」、「基礎設施」、「資本門投資改善計畫（CIP）」、「防救災設施」、、「風險控管與災害保險」、「財稅制度」、「特別評估」、「相關資料庫建置」、「公眾教育與社區防災」「合作機制」等具體策略。整合結構與非結構性防災的規劃方法，可廣泛使用在調整新訂都市計畫方案，透過模擬與土地使用檢討，擬定實證區的都市規劃策略。透過本整合性架構，可以初步提供判斷、調整規劃方案、擬定規劃管理方針、節省規劃時間，顯示本研究極具實務應用價值。
The extreme change of climate has recently influenced numerous cities around the world. This change has resulted in a higher frequency and increased intensity of natural disasters. Habitat and development types even cause complex disaster types. In 2005, the report of Natural Disaster Hotspots-A Global Risk Analysis declared that 73% of world population and land area are exposed to more than three types of natural disasters. Ninety-nine percent of the land area and population are exposed to natural disasters worldwide. These above records are much higher than those of other countries worldwide. Taiwanese people face high risks of natural disasters.
The present study attempts to look into sustainable development policies that seek to prevent disasters through an expert questionnaire. The results of the questionnaire are further analysed with methodologies of the fuzzy Delphi method (FDM) and analytic network process (ANP). The analyses suggest that disaster prevention strategies should be prioritised in urban planning, accounting for the effects of climate change.
The study discovers that extreme floods and storms, landslides, flood risk, flood risk, urban population density, population survived by landslide, building area, and flood potential are suitable indexes for the evaluation of urban disaster prevention and mitigation. In particular, the planning of land use management, public advocacy and training, facility building and reinforcement, management and collaboration with governmental sectors, and financial and taxation system are suitable for significant regulation strategies with the planning of land use management as the most effective measure.
Using the policy-making process of the Taichung High Speed Rail Gateway Plan as a positivist examination, the index evaluation model of urban planning strategy under the climate change will provide validated and objective information and immediate evaluations that fulfill the current urban problems. Furthermore it will serves as the reference for future urban planning policy accordingly.
Most importantly, the most effective non-structural disaster prevention measures for the spatial planning and management identified in this study are land use planning and management, infrastructure, capital investment improvement plan, disaster prevention and rescue facility, risk control and disaster insurance, financial system, special evaluation, establishment of relevant database, public education and community prevention, and collaboration mechanism. The integration of planning methods in structural and non-structural disaster preventions can be applied in the revision and making of new urban planning projects through simulation and land use examinations in the research area. The time saved in the quick judgement and adjustment of planning project and drafting of management guidelines using the integrative framework provided by this research exhibits its pragmatic application value. |
