Formulation and procedure for in situ stress back-analysis from borehole strain changes measured during nearby underground excavation

Estimation of in situ stresses based on back-analysis of measured stress changes and displacements has become an alternative to the direct stress measurement methods.In order to help users conduct own measurement and analysis,this paper presents in detail a field stress back-analysis approach directly from borehole strain changes measured during nearby underground excavation.Essential formulations in major steps and the procedure for the entire analysis process are provided to allow users to follow.The instrument for borehole strain change measurement can be the CSIR or CSIRO stress cells and other borehole strain cells that can measure strains on borehole walls.Strain changes corresponding to the stress changes at a borehole location are calculated in borehole environment.The stress changes due to nearby excavation can be calculated by an analytical model for a single circular opening and simulated by a numerical model for non-circular and multiple openings.These models are based on isotropic,homogeneous and linear elastic assumptions.The analysis of borehole strain changes is accomplished by multiple linear regression based on error minimization and an integrated process provides the best-fit solution directly to the in situ stresses.A statistical technique is adopted for screening outliers in the measurement data,checking measurement compatibility and evaluating the reliability of analysis results.An application example is included to demonstrate the practical application and the analysis procedure.

Abnormal strain changes observed by a borehole strainmeter at Guza Station before the Ms7. 0 Lushan earthquake

Several days before the MsT. 0 Lushan earthquake, the YRY-4 borehole Strainmeter at Guza Station recorded prominent abnormal changes. The strain anomalies are very striking on the smooth background of several years＇ recording after the Wenchuan earthquake. However, because construction in the town of Guza has been undergoing rapid development in recent years, many factors have interfered with observations at the station. Whether or not the observed strain changes before the Lushan earthquake were affected by any of the sources of interference becomes a question that must be answered. Among the likely sources of interference, apartment construction, sportsground reconstruction, and tunnel cutting can be excluded by analyzing the morphological characteristic of the anomalies. The two remaining most possible sources are road construction in front of the station and the water level change of the nearby Dadu River caused by water filling into and discharging from an upstream reservoir. Through field investigation, comparison of the correlation between the strain and the seismographic recordings, comparison of the correlation between the strain and the Dadu River flow recordings, and analysis of the strain anomaly characteristics, we conclude that the abnormal changes observed at Guza Station cannot be attributed to either of these two sources but should be related to the Lushan earthquake.

Developments of borehole strain observa- tion outside China

Borehole strain observation is playing an increasingly important role in the study on the crustal movements. It has been used by many countries such as China, USA, Japan, Peru, Australia, South Africa, Iceland and Italy, in re- search fields of plate tectonics, earthquake, volcanic eruption, dam safety, oil field subsidence, mining collapse and so on. Borehole strainmeter has been improved rapidly and tends to get more and more components included in one probe. Based on observations by this kind of instruments, studies on seismic strain step, slow earthquake, earthquake precursor and volcanic eruption forecasting have made remarkable achievements. In the coming years, borehole strain observation is going to become one major geodetic means, together with GPS and InSAR.

The strain seismograms of P- and S-waves of a local event recorded by four-gauge borehole strainmeter

At a sampling rate of 100 samples per second,the YRY-4 four-gauge borehole strainmeters（FGBS） are capable of recording transient strains caused by seismic waves such as P and S waves or strain seismograms. At such a high sampling rate, data from the YRY-4 strainmeters demonstrate fairly satisfactory self-consistency. The strain tensor seismograms demonstrate the senses of motion of P waves, that is, the type of seismic wave travels in the direction of the maximum normal strain change. The observed strain patterns of S waves significantly differ from those of P waves and should contain information about the source mechanism. Spectrum analysis shows that the strain seismograms are consistent with conventional broadband seismograms from the same site.

The Application of Multiquadric Function Fitting to Borehole Strain Time Series Data Processing

Based on the existing continuous borehole strain observation,the multiquadric function fitting method was used to deal with time series data. The impact of difference kernel function parameters was discussed to obtain a valuable fitting result,from which the physical connotation of the original data and its possible applications were analyzed.Meanwhile,a brief comparison was made between the results of multiquadric function fitting and polynomial fitting.

Study on relationship between features of strain anomaly obtained by borehole strain meter at Wushi station in Xinjiang and strong earthquake: Taking Jiashi MS = 6.8 earthquake as an example

The borehole strain meter at Wushi station recorded obvious anomaly before Jiashi MS= 6.8 earthquake occurred on February 24, 2003. Its features are as follows. j Anomaly types are complete. The trend anomaly, short-term anomaly, short-imminent anomaly and exponential anomaly appeared 19 months, 56 days, 4 days, and more than 1 month before the event, respectively; k Anomaly is large in magnitude. The maximal magnitude of strain anomaly is 1.7×10?5, which is rare in the past 20-year observation records at Wushi station; l Strain rate fluctuates sharply with obvious alternation of tension and compression. According to the magnitude of strain anomaly, time of expo- nential anomaly appearance and regional features of recorded anomaly, we could predict the magnitude, occur- rence time and potential region to a certain degree.

Earthquake-related anomalies recorded by two borehole bulk-strain meters

This paper gives a report on borehole bulk-strain data recorded at the seismostation Nanton$ in Jiangsu Province from January 1 , 2003 to December 31 , 2008. While the data at Nantong showed much noise possibly due to site instability and no earthquake-related changes, the data at Liuhe showed anomalous changes that began one to several weeks before three earthquakes of magnitude 4.0 to 6.0 at eplcentral distances up to 400 kin.

Strain data recorded in Xinjiang and their bearing on Tianshan’s uplift/shortening and Tarim basin’s rotation

Based on the continuous strain data recorded in Xinjiang since 1985, we discuss the mechanisms of Tianshan＇ s uplift and Tarim basin＇ s clockwise rotation. The results indicate ： 1 ） The principal - compression directions in Tianshan are nearly NS, and their intersection angles with regional structures and mountains are nearly perpendicular, which is in accordance with Tianshan＇ s uplift and crustal shortening. 2）The principal compressions around Tarim basin tend to facilitate the regional faults＇ left-lateral strike-slip movements and the basin＇ s clockwise rotation. These phenomena of uplift/shortening and rotation are fundamentally the re- suits of India plate＇ s northward push on Euro-Asia plate, and the associated Pamir arc ＇ s rapid northward movement and regional blocks＇ interaction.

Co-seismic changes of well water level and volume strain meter in capital area and its vicinity,due to the Nov.14,2001 Ms8.1 Kunlun Mountain earthquake,China

The Kunlunshan Mountain Ms8.1 earthquake, occurred in Nov.14, 2001, is the first event with magnitude more than 8 in the China earthquake monitoring history, specifically at the beginning of digital techniques in precursor monitoring networks. Any investigation of recorded data on this earthquake is very important for testing the operation of the digital monitoring networks and understanding the preparation, occurrence, and adjustment of stress/strain of strong continental earthquakes. In this paper we investigated the coseismic response changes of well water level of groundwater and volume strain meter of bore hole in digital earthquake monitoring network of Capital area and its vicinity, due to the Nov.14, 2001 Ms8.1 Kunlun Mountain earthquake. The responding time, shapes or manners, amplitudes, and lasting time of well water level and strain-meters to seismic wave are studied in comparison. Then we discussed the possibility that the response changes of groundwater to strong distant earthquakes can be understood as one kind of observing evidence of stress/strain changes induced by distant earthquake.