台灣弧-陸碰撞造山帶的形成是極具挑戰性的科學問題，「薄皮理論」及「岩石圈碰撞」都曾經用來解釋台灣造山帶所觀測到的地球科學資料。Suppe (1987) 首先提出台灣的薄皮構造模型，認為板塊隱沒發生在厚15 公里增積岩體內。不過，隨著台灣地震網在1991 年完成，豐富的地震數據提供了更詳細的地震層析圖，有證據顯示在中央山脈地區下的地殼有50 公里厚，並發現一般速度分佈不符合增積岩體模型。儘管有先進的地震網，但對台灣向東傾的班氏帶仍然沒有確切的證據。Wu et al. (1997)考慮所有這些數據，提議一個連續的岩石圈變形模型，可能適用於台灣地體構造，這個假說認為變形不僅限於下部地殼，甚至延伸到上部地函，這中間沒有活躍的隱沒帶。目前，沒有足夠的數據來解決不同模型間的差別，也沒有足夠的資料來探討造山過程的細節。本計畫在三年時間內，進行下列工作： 1）50 部陸上及50 部海底寬頻地震儀地震觀測，2）五條陸上剖面爆炸震源廣角反射，3）七條海陸剖面之海陸聯炸， 4）六條海上剖面多波道反射震測及廣角反射，5）構造地質、定年及岩石分析，6）地形研究，7）大電電磁研究，8）大地動力模型。本計畫結合現代地質和地球物理學的技術，以台灣島為對象，仔細檢查現有的造山帶模型，實驗的設計可以達到以前沒辦法獲得的地殼與地函構造的測繪精度。透過問題的陳述、假說的測試、資料的整合及解釋，期望為台灣活躍的碰撞造山帶，建造一個完整的大地構造模型。 ; The Taiwan arc-continent collision is one the most challenged tectonic problems in the world. Both ‘thin-skinned model’ and ‘lithospheric collision model’ have been invoked to explain the first order geologic and geophysical attributes. The thin-skinned tectonics model was proposed by Suppe (1987), which has been elaborated in some detail (Suppe, 1981, 1987; Davis et al., 1984) and updated recently (Suppe et al., 2001). The theory calls for the buildup of a 15-km thick wedge during subduction of the Asian lithosphere. With the improved Taiwan seismic network in 1991, the local earthquake data became adequate for more detailed seismic tomography. The imaged crust under the Central Range can be as deep as 50 km and the general velocity structures under the Central Range are found inconsistent with those of an accretionary wedge. With the improved network there is still no evidence for an east-dipping Benioff zone in central Taiwan. Wu et al. (1997) considered all these data and proposed that a continuous lithospheric deformation model (Molnar et al., 1999) may be applicable to Taiwan. This hypothesis predicts that deformation extends not only into the lower crust but also into the upper mantle. It incorporates shortening of the whole lithosphere due to the collision of the Philippine Sea and the Eurasian plates, without active subduction of either plate. At present, the data is not sufficient yet to clearly resolve the differences between these models, nor do we have enough data to understand the details of the mountain building processes. The integrated research plan outlined in this proposal, which integrates modern geologic and geophysical techniques, is designed to make use of this “natural laboratory” to test the predictions of existing models for mountain building. Our proposed study is specifically crafted to characterize these first order aspects of mountain belt structure and structural history. The proposed research will address these questions, and through hypothesis-testing, we expect to build a comprehensive geodynamic model for the evolution of collisional mountain belts of Taiwan. The present proposal concentrates on the explosion experiment. Fourteen high-power dynamite sources and 1000 Texan records are arranged in three separated crossing-island lines, which may provide the most direct and unambiguous images of the Moho and major structures in the crust as well as the upper mantle. Details of this explosion experiment will be described. ; 研究期間 9708 ~ 9807
