十二月三十一日
成果報告30 份 成果報告光碟10 片
表C：全程計畫預期進度甘梯圖。
表D：期末須交付之各項工作項目。 2-2 節 (p83-p102)
4. 利用所得之地表速度場及應變率，對台南及高屏
3. 完成資料庫整合，持續建立資料庫（收集並處理
摘要
變異常事件，新竹網發現13 個應變異常事件，嘉義網則有 15 個應變異常事件，其中 5 之間，機率為36.3％，46.3％，61.5％；小崗山斷層往北延伸，總長 31 公里，以 Poisson 模式評估未來30 年、50 年及 100 年發生地震之機率，規模 MW 6.7~6.8，機率為 6.0 中，2009/12/19 M6.9 的花蓮地震和 2011/3/20 M5.8 的台東外海地震顯現了地下水位、
井下應變儀、地化氣體的同步異常。同天並伴隨著長微震事件，顯示異常觀測和長微震 可能具有相同機制。這兩個地震的異常記錄在嘉義網和新竹網被觀察到，而地震皆為東 部外海、非鄰近嘉義網及新竹網，並且規模小於六的地震亦對應到異常，說明異常記錄 的空間分布(廣度)和地震規模、離地震的距離可能不相關，或者是我們沒有足夠的同步 異常統計資料來說明異常觀測和地震參數的關係。
Abstract
The main goals of this project are to: (1) explore the anomalous crustal deformation phenomena by analyzing the time series from 72 continuous GPS and 13 borehole strainmeter stations; (2) determine the accumulation of current crustal strain pattern revealed from principal strain rate, shear strain rate and rotation rate by GPS velocity filed and calculate the coupling coefficient, slip deficit and slip rate along the fault plane by inversion of geodetic measurements; (3) compare the strain rate from comtinuous GPS and strainmeter measurements and assess the parameters for the change of strain rate; (4) study the anomalous observation from geochemical monitoring stations with the comparison of continuous GPS and strainmeter measurement; (5) assess the potential of fault activity of the Chishang fault and Hsiaokangshan fault based on the earthquake probability model developed in USA and Japan ; (6) build and maintain the database for access borehole strainmeter records and set up database security. The goal of this proposal is to gather new information about fault behavior in Taiwan using the new-installed GPS and strainmeter stations. Along with multidisciplinary studies of plate-boundary tectonics, observation of earthquake activity, numerical simulations, and active fault systems, we aim at integrating these informations into a comprehensive understanding of fault activity and earthquake probability, seismogenic structures, fault mechanics and information of possible earthquake precursory.
For the 2012 final report, we focus on the topics of (1) examinations of results of GPS and precise leveling data that has been measured up-to-date; and (2) analysis of surface deformation in Southern Taiwan. We first process the measurements from compaign-mode GPS in Taiwan area and 29 precise leveling lines. The GPS velocity field in Hengchun area is consistent with the GPS velocities from 2002 to 2011.
Precisions of 90.0% leveling sections of 29 precise leveling lines are less than 2.0 mm per kilometer, and all sections are within the closure. In terms of analysis of GPS and leveling measurements and results of fault model, the fault slip rates of the Liuchia fault, Chukou fault, and Chutouchi fault are 8.6 ± 1.3 mm/yr, 6.9 ± 1.6 mm/yr, and 29.6 ± 2.7 mm/yr, respectively. The locking depths of these three faults are 1.2 ± 0.4 km, 0.4 ± 0.3 km, and 5.6 ± 0.4 km, respectively. Assuming the fault lengths of the Liuchia fault, Chukou fault, and Chutouchi fault are 23 km, 27 km, and 25 km, respectively. The accumulative moments and maximum moment magnitude of these three faults are 1.44 ± 0.46 × 1023 dyne-cm (Mw 6.66 ± 0.28), 0.41 ± 0.17 × 1023 dyne-cm (Mw 6.76 ± 0.19), and 2.10 ± 0.22 × 1024 dyne-cm (Mw 6.71 ± 0.28), respectively.
The persisted data procrssing on almost all the borehole strainmeter stations in Chiayi Hsinchu and Taipei networks are carried out to maintain the security of database for facilitating the analysis of time series of strainmeter data. No monitoring data are available in LMMT and ECT stations due to the unfunctionable status. The CH1 in RST is out of work. Since July 2012, the interrupt and abnormal spike are obsevered in CH0 of CINT station, further maintainance is expected. Consequently, all the raw data are processed to the components of strain of each channel, and
combined them into areal and shear strains for each station. We also establish the correction model for the influence of air pressure recorded in borehole-strainmeter data by using the observations during several large air-pressure variations. After removing the groundwater and air pressure effects, the residual strain data show clear tidal effect, thus the tidal effect could be furhter modeled and removed. The residual strain after removing all the above-mentioned environmental effect could be used for searching pre-seismic strain anomaly. From 2009 to November 2012, We find 3 strain anomalies in Taipei Network, 13 strain anomalies in Hinchu Network, 15 strain anomalies in Chiayi network respectively. In addition, 5 strain anomalies are found bouth in the Hsinchu and Chiayi networks.
For assessment of fault parameters and logic trees of probability analysis for active faults, three expert consult meetings were held by using the Chelungpu fault as a case example to demonstrate how to establish the logic trees for probability analysis. First of all, we collect the fault parameters of the Chihshang and Xiaogangshan faults. Based on BPT model, the probabilities for a Mw 6.7~7.4 earthquake occurred along the Chihshang in 30, 50 and 100 years are 36.3%, 46.3%
and 61.5% respectively. The Xiaogangshan fault could extend northward with a fault length of 31 Km. On the base of Poisson model, the probabilities for a Mw 6.7~6.8 earthquake occurred along the Chihshang in 30, 50 and 100 years are 6.0%, 9.8%
and 18.5% respectively.
We have searched for possible precursory anomalies for magnitude greater than 5 earthquakes (M>5), using geochemical, hydrological observation, borehole strainmeter, and seismological observations. During the study period of Jan.2009 to Oct. 2012, there are 112 M >5 events occurred in Taiwan. Two earthquakes appear to reveal the synchronized anomalies from hydrological, geochemical, and borehole strain signals, whereas ambient tremors are also observed. These two earthquakes are not local to stations nor both bigger than 6, therefore we argue that the observed anomalies are a function of earthquake magnitude and source-to-station distance.
KEYWORDS: observation of surface deformation, Time series, Borehole Strainmeter, Analysis in Potential Activity of Faults