在三維環境中繪製高解析度地質圖：以臺灣北部雙溪地質圖幅為例

Abstract

傳統地質製圖係以野外調查為主，但野外工作中卻存在一些限制，例如：植被覆蓋或部分區域難以到達，使得傳統製圖缺乏露頭與位態資料，僅能靠地質學家以零星的調查資料運用作圖法推測並建構出區域地質樣貌。本研究運用空載光達影像來克服傳統野外的限制，其高解析度之數值高程模型 (DEM) 被用來作為地表特徵辨識的基底，幫助我們建立一套使用立體化DEM在電腦三維環境中的地質製圖流程。研究區域範圍為經濟部中央地質調查所出版之五萬分之一雙溪地質圖幅，該區域屬臺灣北部褶皺逆衝帶，以漸新世至中新世沉積岩、輕度變質岩以及基隆火山群之火成岩組成。本研究運用1公尺與2公尺解析度數值高程模型資料在三維立體環境進行岩層線形與構造判釋，輔以野外調查與過去文獻收集與整合描繪出地表的岩層與斷層分布，利用所繪製之特徵線段構成的回歸面計算走向與傾角，並針對研究區域重新繪製地質圖。本研究整合室內與野外工作共產製出四張新一代地質圖，分別為五萬分之一雙溪圖幅，以及兩萬五千分之一和平島、雙溪及澳底圖幅，並且利用製圖結果繪製兩條新的地質剖面。高解析度的地表資料有助於提升地質圖中的資訊，在立體圖層中進行製圖時能更精準地描繪並建立地層邊界與構造之GPS位置。除了探索與建立地質製圖之方法，本研究也進一步討論此方法之限制與問題以及不同地層之地形特徵對判釋與描繪的影響，再針對各個小區域進行新、舊地質圖比較與討論。利用高解析度DEM判釋構造、岩層與地形上的關係，加上標靶野外與文獻之輔助，使我們能夠補足與克服傳統地質製圖的限制，重新檢視構造幾何與精進現有的地質圖。新的地質圖中具備更多地質資訊，有助於拓展地質圖的應用範圍與實用性，期待將來本研究發展之製圖方法與成果能夠作為地質科學與工程地質調查更好的應用。

Traditional geological mapping is based on field survey, but there are some limitations for field work, such as vegetation cover or difficult areas to reach leading to lack of outcrop observation and orientation data. Traditional mapping mainly relies on geologists interpreting sporadic data to infer and construct regional geological structures. Airborne LiDAR images are used in this study to overcome the limitations of traditional field survey. The LiDAR-derived high-resolution digital elevation model (DEM) is used as a basis for identifying surface features, helping us to establish a geological mapping process by using stereoscopic topography in computerized three-dimensional environments. The study area is located in northern Taiwan and covers the Shuanghsi Sheet, which was published by the Central Geological Survey, MOEA. This area belongs to the fold-thrust belt in northern Taiwan, containing Miocene and low-grade metamorphic Oligocene sedimentary rocks and igneous rocks of the Keelung volcano group. In this study, 1m- and 2m-resolution DEMs were used to analyze the lineaments of rock formations and structure in three-dimensional environments. Supplementing by field surveys and published literature, we interpreted and traced the distribution of the beddings and faults on the surface. We also calculated the strike and dip of the regression plane formed by the characteristic line, and finally re-mapped the geological map of the study area. This study integrated laboratory and field work to produce four new geological maps, one for Shuanghsi sheet in 1:50,000, and three for 1:25,000 scale: Hepingdao, Shuangsi and Audi. Two new geological profiles were produced from the mapping work. High-resolution surface data help to improve the information in the geological map. When mapping in a stereoscopic layer, it can accurately depict and establish the GPS positions of the boundary of structures and formations. In addition to establishing methods for geological mapping, this study also discusses the limitations and problems of the method and the influence of the topographic features of different strata on the interpretation and tracing. The comparison of new and old geological maps were discussed and demonstrated by small areas. The use of high-resolution DEM to interpret the relationship between structures, formations and topography is aided by targeted field checks and references. It enables us to overcome and make up for the limitations of traditional geological mapping, and reexamine the structural geometry and refine the existing geological map. The new geological map shows more geological information, which helps to expand the practicality and potential applications of geological maps. It is expected that the mapping results and the new methods developed in this study will be able to serve as a renewed application for geological survey and engineering purposes.