地震與降雨引致邊坡大規模順向滑動的震動訊號特性分析

Seismic Signals Analysis of Earthquake and Rainfall Induced Large-Scale Dip Slope Failure

吳昕孺(Hsin-Ju Wu)

指導教授 ： 林美聆

國立臺灣大學/工學院/土木工程學研究所/碩士(2021年)

https://doi.org/10.6342/NTU202102892

摘要

台灣位於板塊交界，造山運動盛行，地質情況複雜，又因颱風與地震發生頻繁，經常引發邊坡災害，進而造成生命及財產之損失，為了減少遺憾發生，監測與維護成為了一項重要的課題，然而無論是何種邊坡災害，皆會使地表產生震動，因此若能從地震測站中對震動訊號進行分析，以最快速的方式辨別不同因素引發之崩塌訊號及崩塌行為，即可有助於監測與應變規劃。

本研究以1999年集集大地震引發之草嶺與九份二山崩塌和2009年莫拉克颱風所引發的小林村獻肚山崩塌為研究案例。三個案例均為大規模順向滑動，在地震案例中，以強震測站資料進行分析，首先以二維Newmark滑動塊體法分析初始破壞、達臨界位移與撞擊之時間點，再將其時間點與STFT頻譜比對觀察，水平方向訊號之頻率範圍於初始破壞約為0 ~ 60 Hz、達臨界位移約0 ~ 80 Hz、持續滑動約0 ~ 20Hz、撞擊約0 ~ 30 Hz，而垂直向頻率範圍較水平向寬約5 ~ 10 Hz，由初始破壞與達臨界位移時段發現其訊號有呈現平台狀的特徵，且初始破壞至持續滑動的過程主要以S波為主，撞擊訊號則以P波為主，各滑動歷程之訊號可傳遞距離不一，以持續滑動訊號傳遞最遠，然而其距離仍會受地形與地質構造有所限制。本研究亦進行愛氏強度分析，結果發現其主要能量累積之方向有指向震央或平行崩塌滑動方向之趨勢。

在降雨案例中，以寬頻地震測站進行分析，分析結果以甲仙測站為主，其測站之愛氏強度能量主方向與崩塌方向平行，且在0.5 ~ 6 Hz之帶通濾波下可明顯看見主要崩塌滑動之波型，時間約在06:16:15 ~ 06:17:45，並判定崩塌滑動之主要訊號是以S波之方式傳遞，與地震案例一致；此外，亦發現在0.02~0.05 Hz的帶通濾波下，可以看到有較原始波型早出現的明顯振幅訊號，然而尚無法確定此頻帶之訊號為何，後續之撞擊訊號亦尚無法準確判定。

關鍵字

邊坡順向滑動 ； 滑動行為特性 ； 震動訊號 ； Newmark 滑動塊體分析 ； STFT時序頻譜分析 ； 愛氏強度

並列摘要

Taiwan is located at the boundaries of tectonic plates, which leads to orogenic movement. Thus, the geologic condition is fragile and complicated. Moreover, typhoon and earthquake occur frequently, which often cause serious landslide events. To avoid more casualties and economy loss, monitoring and mitigation measures have become an important issue. However, no matter what kind of slope movements occurred, ground vibration would be generated. Therefore, analyzing seismic signals from ground motion stations; identifying signals corresponding to landslide and behaviors will be helpful for mitigation strategy.

This research focuses on three field cases of large scale dip slope failure in Taiwan, i.e. Tsaoling and Chiufenershan landslide induced by 1999 Chichi earthquake; and Shiaoling landslide induced by 2009 typhoon Marakot. In the earthquake induced cases, records of strong motion station were used. First, 2D Newmark sliding block method was applied for estimating the timing of initiation, critical displacement, and landing impact. And compared to those of STFT spectrum. The horizontal signal shows the frequency range of 0~60 Hz at initiation, 0~80 Hz at critical displacement, 0~20 at sliding, and 0~30 Hz at impacting. The frequency range of vertical signals are 5~10 Hz higher than horizontal signals. Moreover, it was found that a platform-shaped signal occurred when reaching initiation and critical displacement. These processes and sliding produced more S wave while impact produces P wave signal. In addition, the possible propagation distance of signals in all process are different. Sliding signal can transfer the farthest, but the distance will be affected by topography and geologic structures. Arias intensity analysis were also conducted. The result shows that the main energy cumulating direction is possible to point to the epicenter or parallel to the sliding direction.

In the rainfall case, broadband seismic station was used and the results mainly focused on SGSB station. The main energy cumulating direction of arias intensity is parallel to the sliding direction. In addition, the main sliding signal can be seen significantly under 0.5 ~ 6 Hz bandpass filter and the main sliding time is around 06:16:15 ~ 06:17:45. The sliding signal is also produced by S wave. Moreover, it was discovered that under 0.02~0.05Hz bandpass filter, a signal with significant amplitude can be found before the main landslide wave occurred. However, the factor contributing to this signal is not yet confirmed. The impact signal in this case cannot be identified accurately yet.

並列關鍵字

Dip slope failure ； Sliding motion characteristics ； Seismic signal ； Newmark sliding block analysis ； STFT time-frequency spectrum analysis ； Arias intensity

參考文獻

[1] Allen, J. (1982). Applications of the short time Fourier transform to speech processing and spectral analysis. In ICASSP'82. IEEE International Conference on Acoustics, Speech, and Signal Processing (Vol. 7, pp. 1012-1015). IEEE.

[2] Chang, K. J., Taboada, A., Chan, Y. C. (2005). Geological and morphological study of the Jiufengershan landslide triggered by the Chi-Chi Taiwan earthquake. Geomorphology, 71(3-4), 293-309.

[3] Chen, W. F., Liu, X. L. (1990). Limit analysis in soil mechanics. Elsevier.

[4] Chen, T. C., Lin, M. L., Wang, K. L. (2014). Landslide seismic signal recognition and mobility for an earthquake-induced rockslide in Tsaoling, Taiwan. Engineering geology, 171, 31-44.

[5] Crespellani, T., Madiai, C., Vannucchi, G. (1998). Earthquake destructiveness potential factor and slope stability. Geotechnique, 48(3), 411-419.