|DC.description||National Central University||en_US|
|dc.description.abstract||由於台灣地狹人稠，山坡地的開發勢必無法避免，而西部走廊山區位於褶皺地區，所以順、逆向坡之分佈極廣，潛在地質災害之發生機率亦很高。因此不論是一般相鄰山坡地建築開發或是公共建設的拓展，面臨選址或是選線的初步設計規劃階段，對於順、逆向坡確切位置的掌握實為一重要的課題。本論文主旨在於利用 GIS 建立一個廣域地區順向坡至逆向坡之評估架構。
本論文實際研究場址為台灣北部橫貫公路復興至巴稜路段。經由蒐集已發表的地質圖、航空照片、數值地形、物性影像等。進行室內資料處理，主要工作包括航空照片立體對人工判釋順向坡、地層位態多項式趨勢面分析、接著由程式根據差異角度自動萃取順向坡。接著進行現地查核，與程式自動萃取所得之順向坡範圍相互比較，進行野外查證以確定順、逆向坡位置及其範圍。最後依照現地查核地層位態角度比對結果，提出本研究區域單組弱面控制邊坡破壞指標分佈圖，進行成果輸出。成果比較顯示，在復興路段對於大規模順、逆向坡判斷良好，但在榮華大壩過後地層變化極大的區域，其準確度不佳，此為多項式迴歸產生極大殘差值所致。故在最後使用另一種迴歸方法『Kriging』與原方法作一比較。另外在研究範圍區域限制方面，亦縮小了研究區域的複雜性另行比較。相較於傳統人工野外踏勘的曠日廢時、以及航照判釋因人而異的不確定性。本論文建立了一套完整自動化順、逆向坡測繪模式，整合 GIS / GPS / RS 等地理資訊系統進行初步自動順、逆向坡的測繪工作，結合單組弱面控制的順、逆向坡的潛感指標給予其不同權重，以提供未來廣域山坡地開發選址、交通工程線型先期規劃的重要輔助工具。||zh_TW|
|dc.description.abstract||Because of huge population over a relatively-small plain area in Taiwan, the continuous reclamation of slope-lands is a general trend. The west-side corridor of Taiwan is located in a fairly-folded terrain of numerous bedding planes, in which cataclinal and anaclinal slopes are populated, both leading to geological hazards. Locating the probable cataclinal and anaclinal slopes is greatly valuable for siting such reclamation in the preliminary stage. However, a detailed site investigation and precise mechanical analysis is not affordable for performing this task over a large area. The objective of this thesis is to establish a preliminary framework of quantitatively rating cataclinal and anaclinal slopes over a wide region using the geographic information system (GIS) technique.
The preliminary rating system of cataclinal and anaclinal slopes in this thesis is based on the National water-soil preservation code and the kinematics analysis by comparing the attitude data of a surface slope and those of the closest outcropping bedding plane (as a only discontinuity), neglecting the contribution of others. And the adequacy of this system will be justified through the accumulative field observations and model test results. The attitude data of a surface slope were determined by GIS with input of DTM, while those of the bedding plane at any point were estimated by a polynomial interpolation method (up to an order of n) with input of real attitude measurements. A hypothesized case of one symmetrical continuous slope surface and smoothly-varied bedding planes was used to testify the above proposed framework, and it is found that the rating maps produced are very reasonable.
The target field site under investigation is the area between Fu-hsin and Ba-lin along the Northern Cross-Island Road. The grid size of DTM map used is 40m?40m. The attitude data of weak planes were collected through various sources, whose correctness were unknown; the total number of these data was much less than that of slope surfaces computed by the DTM map, and its spacial distribution was not uniform over the target area. By comparing the rating map produced with the limited number of field observations, it is found that good match appears in most sites of smoothly-changing bedding attitude, but large deviation occurs in the Zong-hwa damsite with considerable variation in bedding attitude. In the latter case, an additional effort was made in which the Kriging method was used to replace the polynomial interpolation method for estimating the bedding attitude. Subsequently, it is concluded that the proposed scheme is very useful for rating that the upper-left quarter of the study area, in which bedding attitude varies quite smoothly. In summary, the rating framework established in this thesis is a automatic, quick and objective tool for quantitatively locating cataclinal and anaclinal slopes with a sufficient number of reliable input data, and it will provide a sound reference of preliminary siting and planning in the slope-lands of Taiwan in the near future.||en_US|
|DC.subject||Northern Cross-Island Road||en_US|
|DC.title||Identification and Potential Evaluation of Cataclinal and Anaclinal Slopes over a Large Area Using GIS – Northern Cross-island Road for Example
|DC.publisher||National Central University||en_US|