| 摘要: | 自然災害的強大破壞力與特性一直都是阻礙發展中國家經濟與社會發展的主要威脅。對於海地這個世界上極度貧窮的國家而言，未受控管的都市化現象、缺乏政策的都市規劃及空間管理都會加劇自然災害對人類的傷害。近來遙測技術的發展與其應用，可使人們能利用遙測影像來評估城市地區可能受到自然災害侵襲的地區，並且取得自然災害的分佈和判釋災害等級與其他變因間之關係。在2010年1月12日海地地震的事件中，國際組織便利用遙測的資料作為評估災害程度與災後重建工作之用，如聯合國(UN)。本論文利用變遷偵測法與目視檢測法，分析由1979至2009年間之海地衛星影像，取得災害程度與建築物土地年齡間之資料。主要分析方法為下列三個步驟：1）變遷偵測分析和變遷類型分類，2）檢測建築物土地年齡，3）檢測建築物土地年齡和受地震破壞之建築物受損程度。本研究經由分析後顯示以以下結果： 遙感資料顯示：在某些特定時期，尤其是1984至1989年，資料顯示當時為維繫農地與木材能源之生產進而破壞森林地； 在首都太子港與其附近區域有顯著的都市化現象。事實上在2009年，太子港約有20.19％的面積為建築用地，Cité Soleil和Delmas各有11.02%及6.75%的面積為建築用地。 首都地區的建築物用地是最古老的，平均年齡約為28.83年、Delmas的為25.67年、Carrefour的為21.5年、Pétion – Ville的為20.25年、Cité Soleil的為19.65年。 來自於太子港地區建築物用地年齡與建築物損害程度的數據分析，顯示兩者有顯著性的關係。未來可進一步分析建築物年齡與災害程度的關係。 基於以上這些結果我們建議： 主管單位可有效利用地理資訊系統和遙測資料，評估和減輕自然災害的影響，作為未來對災害風險評估有更好的準備; 為了有更好的土地管理與宜居空間，政府當局應採用綜合管理計劃，作為減輕災害對人類生命和財產的影響，須包含： 預測、預報和預防 適當的土地使用規劃 保護措施的應用和建設的規則 發展緊急救援計劃和危機管理 Natural disasters have always been and still are a handicap to economic and social development and are characterized by their great destructive power on lives and properties worldwide especially in developing countries. Haiti, one of the poorest countries in the world, is characterized by an uncontrolled urbanization and a lack of policies for urban planning and space management that have sharpened the population vulnerability to natural disasters. Recent advancements in remote sensing and its application technologies made it possible to use remotely sensed imagery data for assessing vulnerability of urban areas, for capturing damage distribution due to natural disasters and for determining some relationships between damage levels and certain other factors. In the case of January 12, 2010 earthquake in Haiti, remotely sensed information have been used by international organizations like the United Nations for determining different damage levels and the preparation of the Post-Disaster Needs Assessment which are being used currently for reconstruction efforts. This thesis allowed giving information about the trends of damage grades and build-up land ages obtained by change detection methods and visual interpretation from 1979 to 2009. With an adequate scientific literature, the methodology followed 3 main steps: 1) change detection analysis and change types classification, 2) determination of build-up land ages and 3) determination of relationships between build-up land ages and building damages levels caused by the earthquake. Actually, this study has revealed the following facts: remotely sensed information shows, in some specific periods, notably from 1984 to 1989, there is evidence of destruction of forest lands for subsistence farming and wood energy production paving the way for unplanned construction activities;it is demonstrated as well a significant urbanization in the capital city, Port-au-Prince, and in its metropolitan region. In fact, in 2009, 20.19% of the area of Port-au-Prince is detected as build-up land, Cité Soleil and Delmas have respectively 11.02% and 6.75% of their areas assigned to build-up land. In the other municipalities the percentage is lower;build-up areas in the capital city are the oldest, their average age is 28.83 years old followed by Delmas whose buildup lands average age is 25.67 years old. Moreover, in Carrefour, build-up areas are 21.5 years old; in Pétion-Ville they are 20.25 years old and in Cité Soleil they are 19.65 years old; damage grades are scattered oddly inside each municipality, information about damage localization is not available.further studies will show that most of the heaviest damages are undergone by the oldest buildings and most of the youngest buildings undergo the lightest damages. Based on these results, it is recommended that: an efficient use of GIS and remote sensing information by authorities to monitor, assess and mitigate natural disasters in the country, and to study risk and vulnerability of the population for a better preparation; for a better management of the livable space, governmental authorities should apply an integrated plan of management aiming mitigation of disasters impacts on human lives and properties which consists in: Forecasting, prediction and prevention information The proper planning of land use The adoption of protective measures and rules of construction Development of contingency plans and crisis management. |
