臺灣東部井下應變儀陣列觀測資料處理與初步分析

Preliminary Analysis of the Borehole Strainmeter Data in Eastern Taiwan

張淵淑(Yuan-Shu Chang)

指導教授 ： 陳于高

共同指導教授 ： 許雅儒(Ya-Ju Hsu)

國立臺灣大學/理學院/地質科學研究所/碩士(2012年)

https://doi.org/10.6342/NTU.2012.02808

摘要

中研院地球所於2002年開始陸續在臺灣東部裝設Sacks-Evertson類型井下應變儀，目前共設置11個測站。井下應變儀測量微小的地殼活動訊號，其精度可達10-12量級。但也因儀器的高敏感度，使觀測記錄會受到埋設地點周圍環境因子變化影響，干擾儀器所記錄到的地殼變形訊號。因此在使用井下應變儀的應變記錄探討區域性地殼變形及斷層活動之前，釐清環境因子對於應變記錄的影響為一項重要的前提工作。本研究嘗試分析裝設在海岸山脈中部Sacks-Evertson類型井下應變儀陣列所觀測到的應變資料，探討此區域的應變變化與環境因子之間的關係。之後，在釐清環境因子對於應變影響程度後，本研究刻意避開環境因子的影響時間，再比較數月時間尺度下GPS與井下應變儀記錄，以探討井下應變儀記錄的穩定度。
　　井下應變儀的觀測資料須經過一系列資料處理，才能變成可應用的數據。首先，不連續之應變觀測資料要被串接成連續的應變時間序列，並利用兩個指數函數移除應變資料中井孔回復及水泥乾涸效應造成的變形量。另外，又由於大氣變化對應變記錄有所影響，使用線性函數計算井下應變儀記錄對大氣壓力變化的響應並修正之。本研究利用SPOTL計算理論潮汐模型，比較應變觀測記錄中潮汐振幅與理論潮汐模型之振幅，將應變資料單位由電壓讀數（count）轉換為真實的應變（strain），最後使用交互耦合矩陣完成三分量井下應變儀記錄的校正。
　　初步結果顯示，體應變與面應變常記錄到伴隨著颱風侵臺氣壓劇降的地殼永久壓縮應變，於2006年後加入的剪切應變記錄變化一同比較，發現剪切應變在颱風期間的變化並不顯著。若是颱風期間造成的氣壓或降雨量之變化進而造成斷層活動，則剪切應變也會同時記錄到顯著的永久變形訊號，且體應變與面應變會有壓縮和伸張訊號出現。根據上述概念，可以推論在2006年之後颱風侵臺期間，井下應變儀所記錄之壓縮應變和斷層活動較無關聯性。在強降雨時亦出現體應變與面應變產生永久壓縮應變但剪應變並未出現顯著變化。假設井下應變儀埋設點附近的岩石大致為均質分布，因降雨產生的水文荷重會直接反應在岩石的體積變化，產生壓縮應變，但此一荷重對剪切應變的影響較小，因此在剪切應變訊號上對於雨量的荷重反應較不顯著。本研究亦針對臺灣東部乾季時的體應變記錄進行分析，並與GPS的膨縮率比對，發現兩者的變化一致。此結果表示若能避開水文效應，井下應變儀在數個月尺度下的應變記錄應該是可靠的。因此未來若能有效移除水文的影響，便可使用井下應變儀的應變記錄，探討地體構造運動的變化。

關鍵字

井下應變儀 ； 應變變化 ； 臺東縱谷 ； 降雨 ； 氣壓

並列摘要

The institute of Earth Sciences, Academia Sinica (IESAS) has deployed eleven Sacks-Evertson-type borehole strainmeters in eastern Taiwan since 2002. The high sensitivity borehole strainmeter enable us to detect crustal strain in the scale of 10-12. Due to such high precision, the recorded strain is affected by environmental disturbances as well. In order to extract tectonic signals from borehole strainmeter data, it is of importance to clarify the influences of environmental factors on borehole strainmeter records. This study analyzed the datasets recorded by Sacks-Evertson borehole strainmeters installed in the central Coastal Range of eastern Taiwan. We first remove the effects of hole relaxation and grout-curing as well as strain steps due to valve resets and data gaps in the raw data using two exponential functions and a linear rate. Since it has been well known that atmospheric pressure would provide a significant influence on the crustal strain, we accordingly compute the barometric pressure admittance using a linear function. The SPOTL program is then applied in calculating the theoretical tide to obtain tidal admittance. The data processing method for single-component and three-component strainmeters are similar, except that the three-component instrument requires a coupling matrix to convert strain records at three gauges into the engineering shear strain and area strain.
Our preliminary results show the contractional strain may be affected by strong precipitation. Although the contractional signals associated with heavy rainfall are often recorded in dilatometer and areal strain, the recorded shear strain in 3-component strainmeter does not show significant changes at the same time. This finding indicates the hydrological loading due to rainfall only affects the volumetric strain and presumably has little influence on shear strain given assumption of isotropic materials. If the contractional signals are generated by fault activities, both shear strain and volumetric strain should record significant signal changes in the same time; nevertheless, shear strain doesn’t show significant signal changes while volumetric strain and areal strain have permanent strain changes. Thus, we eliminate the possibility of tectonic-origin motions.
We further compare month-scale volumetric strain from borehole strainmeters with surface dilatation rate derived from GPS velocity, in particular in the dry season. It shows correlated signal changes, suggesting that the borehole strainmeter record in month-scale may be a good index to recognize the crustal strain when the impact from the hydrologic influence is insignificant.

並列關鍵字

borehole strainmeter ； strain changes ； Longitudinal Valley ； precipitation ； barometric pressure

參考文獻

Agnew, D. C. (1997), NLOADF: A program for computing ocean-tide loading, J. Geophys. Res., 102, 5109-5110, doi:10.1029/96jb03458.

Agustsson, K., A. T. Linde, R. Stefansson, and S. Sacks (1999), Strain changes for the 1987 Vatnafjoll earthquake in south Iceland and possible magmatic triggering, J. Geophys. Res., 104, 1151-1161, doi:10.1029/1998JB900027.

Chang, C. P., J. Angelier, and C. Y. Huang (2000), Origin and evolution of a melange: the active plate boundary and suture zone of the Longitudinal Valley, Taiwan, Tectonophysics, 325, 43-62, doi:10.1016/S0040-1951(00)00130-X.

Chang, C. P., T. Y. Chang, J. Angelier, H. Kao, J. C. Lee, and S. B. Yu (2003), Strain and stress field in Taiwan oblique convergent system: constraints from GPS observation and tectonic data, Earth Planet. Sci. Lett., 214, 115-127, doi: 10.1016/S0012-821X(03)00360-1.

Chung, L. H., Y. G. Chen, Y. M. Wu, J. B. H. Shyu, Y. T. Kuo, and N. Y. N. Lin (2008), Seismogenic faults along the major suture of the plate boundary deduced by dislocation modeling of coseismic displacements of the 1951 M7.3 Hualien–Taitung earthquake sequence in eastern Taiwan, Earth Planet. Sci. Lett., 269, 416-426, doi: 10.1016/j.epsl.2008.02.035.