本研究探討 1960 年智利地震海嘯及 1867 年基隆海嘯事件，並使用 COMCOT 數值模式進行模擬與分析。分析方法採影響強度分析法(Impact Intensity Analysis, IIA)，並搭配本研究新發展之海嘯到時分析法(Tsunami Arrival-Time Analysis, TATA)，以分析可能之海嘯源及評估潛在海嘯強度。 IIA 法植基於互逆格林函數，將研究區域布設單元海嘯點源，並求解各點源 之海嘯衝擊係數，其結果可顯示各點源位置對於研究場址之海嘯影響強度。 而 TATA 法則是利用海嘯之傳播方向，分析海嘯源走向與海嘯到時之關係， 以掌握大規模之帶狀海嘯之方向性。IIA 法可以精準掌握各地點之海嘯衝擊 敏感性，然而 IIA 法並無法掌握海嘯源之方向性，因此本研究提出 TATA 法， 來加強對海嘯源方向性之掌握，並透過亞普海溝及馬里亞納海溝南端之情 境，證明 IIA 法結合 TATA 法之準確性。透過 IIA 與 TATA 法之分析，發現 1960 年之地震震央所在位置，對於台灣之影響並非最為顯著，經過交叉比 對與分析，發現 1960 智利地震位置往北偏約 500 公里，海嘯波高於基隆港 會增加 70%。類似的情況亦發生於夏威夷希洛市，其波高會增加 50%。本 研究亦使用 IIA 法分析 1867 年基隆海嘯之海嘯源。透過文獻對於基隆港與 金山之海嘯記載與描述進行交叉比對，結果顯示基隆海谷及棉花峽谷之海 底山崩與核二廠東北方之海底火山處，有滿足文獻記載之海嘯源。此外，發 現介於宜蘭縣及花蓮縣之和平海盆對於台灣北部(基隆及金山)存在著潛在 海嘯威脅，且由台灣北部各研究點之 IIA 發現八重山所發生之海嘯對台灣 影響輕微。本研究亦透過比較線性及非線性之 IIA，獲得非線性效應之強弱 分布圖，有助於了解台灣北部海域複雜海底地形所伴隨之非線性效應分布。;The purpose of this research contains two parts: first, assessing the potential threats of the tsunamis of 1960 Chilean Tsunami and 1867 Keelung Tsunami, second, locating the latter event. Both parts involve COMCOT (Cornell Multi- grid Coupled Tsunami Model) integrated with IIA (Impact Intensity Analysis), and TATA (Tsunami Arrival-Time Analysis), which accuracy has been proved by applying to the scenarios for Yap Trench and Mariana Trench. Through applying the methods of IIA and TATA, it has been found that the impact in Taiwan is not significant due to the source of 1960 Chilean Tsunami. The cross- validation shows that the height of tsunami at Keelung would have increased 70%, while the one at Hilo City would increased 50%, if the souse of 1960 Chilean Tsunami had been shifted 500 kilometers toward north referred to the source. Results of IIA at Jinshan and Keelung in northern Taiwan were used to design scenarios. It was found that the wave heights at Chilung Valley, Mien-Hua Canyon and the unnamed submarine volcano is located in the northeast of the second nuclear plant, concur with the literature. In addition, it reveals also potential tsunami hazard to the north of Taiwan caused by the movements in Hoping Basin. The map obtained by the comparison of linear and nonlinear IIA results is used to understand the distribution of nonlinear effects of the complex bathymetry in northern Taiwan.
