General Design

Gladwin Tensor Strainmeters (GTSMs), the instruments that will be used by PBO, have a precision of 1 part per billion over short periods. The GTSM uses differential capacitive plate transducers to measure change in the borehole diameter. As the borehole deforms the plates move relative to each other causing a change in capacitance proportional to the change in distance.

Figure 2. Drawing of a PBO borehole strainmeter installation.

The GTSM has three strain gauges oriented 120 degrees apart. The independent measurements of change in length along each axis can be combined to obtain three other strain components that describe the horizontal strain tensor: the areal strain, and gamma 1 and gamma 2 shear strains. Each gauge is contained in a separate module about 10 cm in diameter and the three modules are stacked above each other within the strainmeter. The entire instrument is about 254 cm long and weighs 50 kg.

Installation

The PBO Strainmeter Design Specifications Document v.1.4 (July 2004) contains the full technical details regarding PBO borehole strainmeter installation; here, we give a brief summary.

PBO strainmeters will be installed in regions of scientific interest, e.g., near active faults. The actual location will be determined by various factors such as local geology, access, security and landowner permission. For the instrument to measure tectonic strain it should ideally be placed in unfractured rock. A drill rig is required to drill the borehole necessitating vehicular access. PBO will enter into an agreement with the landowner, either public or private, to ensure security and access to the site.

The borehole is drilled to depths of about 150 m and cased using a steel pipe. The final 50 m are then drilled and core taken to examine the rock type. When suitable rock is found the final section is reamed and the bottom 4 m filled with grout. The strainmeter is then lowered in to the grout and the grout left to harden. Expansive grout is used to couple the strainmeter to the borehole walls so that it does not shrink as it hardens.

1. The borehole is initially drilled using a tri-cone drill bit. For the first 20 feet a 10-inch bit is used and then an 8-inch drill bit from the rest of the drilling.

2. Hollow steel casing is inserted as the hole is drilled. The steel casing prevents rock or debris from collapsing into the hole. Drilling can progress at a rate of 25 feet per hour.

3. Samples of the cuttings are taken every 20 feet to get an idea of the rock type the borehole is being drilled in. The samples are caught in a small sieve from the stream of cuttings pumped out of the hole.

4. The cuttings are washed and saved for later analysis.

5. The borehole must be vertical to within 5 degrees. Several times during drilling the drill bit and pipe are removed from the borehole and inclinometer is lowered in to measure any deviation from the vertical.

6. When rock is encountered the tri-cone drill bit is removed and a hammer drill-bit is used to pound through the rock.

7. After the target depth has been reached ~500 feet the drill bit and drill pipe are removed from the hole and logging begins.

8. Logging yields information about the geophysical properties and rock type the hole is drilled in. First a three-legged caliper is lowered into the hole to measure the diameter of the borehole from top to bottom. The caliper is lowered down at a rate of 6 feet per minute.

9. An acoustic televiewer is lowered in to make a 360 degree scan of the borehole walls. The televiewer moves through the hole at a rate of 3 feet per minute The strength and density of the rock are measured using a full wave sonic tool.

10. A natural gamma tool is used to detect naturally formed boundaries in the rock. Self-potential and resistivity tests are also performed to identify rock boundaries

11. After the borehole has been logged the next step is to case it. The casing is made up of 20-foot sections of 6-inch diameter steel pipe. Each piece is lowered into the borehole leaving about 18 inches above the floor of the drill rig.

12. The next piece of casing is welded to the pipe in the hole. The process is repeated until the entire hole is cased. Care is taken to ensurethe casings are welded together vertically.

13. Cement is then delivered to the bottom of the hole through the drill rig pipe, which acts as a hose. The cement is forced upwards to fill the gap between the walls of the 8-inch borehole and the 6-inch diameter casing.

Once the borehole has been cased the bottom of the borehole is cored another 30 to 100 feet. Coring involves extracting cylinders of rock from the bottom of the borehole. The core will be examined to identify the rock type at the bottom of the borehole and to find the best depth at which to install the strainmeter. When suitable rock is found the final section is reamed and the bottom 4 m filled with grout. The strainmeter is then lowered in to the grout and the grout left to harden. Expansive grout is used to couple the strainmeter to the borehole walls so that it does not shrink as it hardens.

Other scientific instruments installed in PBO strainmeter boreholes will include a 3-component seismometer, a pore pressure monitor and, in some cases, a two-component tiltmeter. The seismometer will be installed 6 m above the strainmeter and cemented in place. The position of the pore pressure monitor depth will vary from site to site and will be sand packed. The borehole will be filled with cement to within 50 m of the surface and the tiltmeter installed. The upper 50 m of the borehole will be left open for installation of future instruments. The electronics that control the strainmeter, environmental sensors, and power supply and telemetry system are housed in an enclosure on the surface.

Mini-clusters of strainmeters will be installed at some sites. These involve drilling two or more strainmeter boreholes within 100 m of each other and installing one strainmeter in each hole. This system allows scientists to study how closely located strainmeters respond to a geophysical signal and will allow discrimination of locally-generated signals from tectonic signals.

Data and Data Products

The PBO Data Management Plan describes in detail the management and generation of PBO strainmeter data products. Here, we give a summary.

Metadata

Metadata are ancillary data about each strainmeter station, instrument operation, and data analysis, which will be collected and stored in the PBO Operational Database (POD). The POD will contain such metadata as station locations, site descriptions and photographs, equipment types, network state of health, and other similar information. All visits made to a given station and any change in station configuration will be documented in the POD. Derived data products,such as the tidal admittance and the scale factors required to convert single gauge measurements to areal and shear strain, are also considered metadata and will be retrievable from the POD.

Data Acquisition and Quality-checking

Borehole strainmeter (BSM) data will be buffered on-site and downloaded in near realtime to PBO HQ via direct Internet connections. BSM data will be transferred using a secure data transport system, likely based on scp or the Antelope or Earthworm software, to a central quality-checking system at PBO HQ. BSM data will be quality-checked at PBO HQ before being passed on both to the Strainmeter Archives and Strainmeter Data Analyst, using such tests as metadata conflict resolution, data completeness, and the like.

Data Analysis and Products

Once these data have been vetted, they will be analyzed by the PBO Strainmeter Data Analyst. A future document on borehole strainmeter data analysis will fully describe PBO methods of analyzing strainmeter data, but for now, we give the following summary.

PBO will provide raw data, metadata, and two levels of processed data from borehole strainmeters. All these data products will be available from the Strainmeter Archives and the EarthScope Data Access System in either SEED or XML format.

Level 0 will be raw strain gauge and environmental data, updated at least hourly. Level 1 data will be automatically cleaned strain gauge and environmental data, updated at 24-hour intervals and scaled to strain units using manufacturer's calibrations. Level 2 data will consist of verified strain gauge and environmental data, areal and shear strain, the amplitudes and phases of the main tidal constituents observed in the data, an atmospheric response coefficient, and corrections for borehole-relaxation and grout-curing, with three sub-levels: 2a, 2b, and 2c. Level 2a and 2b data represent rapid solution data products and are updated at daily and 2-weekly intervals. Level 2c data will be updated at 3-month intervals and represents the final processed strain data.

Data processing includes:

• Identification of strain induced by the natural environment.
  Changes in the atmospheric pressure or the movement of water around the borehole can induce strain signals. A coefficient that linearly relates strain change to atmospheric pressure change will be estimated when the data are processed and pore pressure measurements will be made in all PBO boreholes to monitor and define the monitor fluid flow around the borehole.
  
  
• Calculation of the grout-curing and borehole-relaxation trend.
  The dominant signal in the first few years after the strainmeter is installed is that of grout-curing and borehole-relaxation. As the grout hardens the strain gauges detect a change in borehole diameter and, because the hole was drilled in the Earth a pre-stressed medium, the borehole will deform with time as the surrounding rock exerts a compressive force. The combined effect will be estimated and provided for the user to remove.
  
  
• Calculation of a correction for the Earth tides.
  As the Earth rotates in the gravitational field of the Sun and Moon it deforms in a periodic way. The deformation is called the solid Earth tides. Water movement on the surface of the Earth causes an additional periodic loading; this is called the ocean loading effect. The total deformation is referred to as the theoretical tides. The Earth tides are easily measurable by the strainmeter and are a useful indicator that the strainmeter is working as it should be. To model tectonic strain variations though, it is necessary to remove the tidal signal. In processing, the amplitudes and phases of the main tidal constituents are estimated from the data and a tidal correction is produced.
  
  
• Detection of spurious spikes and offsets in the data.
  These are often caused by power glitches, up-hole electronic problems, or during data transmission.
  
  

Level 0 and Level 1 strain data will be available in Standard for the Exchange of Earthquake Data (SEED) format. SEED format has been developed by the Earth science community to facilitate the exchange of information between research groups. The SEED files will contain the strain measurements and instrument response information. Documentation and software to manipulate SEED files can be found at http://www.iris.washington.edu/manuals/manuals.htm/.

Level 1 and Level 2 strain data will be available in XML format. The XML files will be self-contained consisting of header and data sections. The header section will contain general instrument response information such as scale factors used to convert the raw strain measurements into geophysical units and SEED codes to identify the corresponding SEED data channels. The data section will contain the date and time of each data point, the gauge, areal and shear strain measurements, and corrections for atmospheric pressure and the solid Earth tides. Any outliers or offsets found in processing will be recorded in the XML file. No data are deleted, but instead the potential outlier will be flagged as such, giving the user the option to discard the data value or not. The magnitude and time of any offsets will be recorded in the file so the user can choose to leave the step in the data or remove it.

Data Analysis Software

We will add further information to this section in the future.

Data Access

We will add further information to this section in the future.

Data Download

PBO has been developing and testing processing codes and procedures for PBO borehole strainmeters by cleaning mini-PBO data; the cleaned data are also useful for mini-PBO operations.

» Download the cleaned data

For further details, please contact Kathleen Hodgkinson.

Related PBO Publications

• Plate Boundary Observatory Data Management Plan (June 9, 2004)
• PBO Strainmeter Update (May 26, 2004)
• PBO Strainmeter Design Specifications v.1.4 (July 2004)
• PBO Strainmeter Report (May 25, 2004), Figures Section 2 and 3, and Figures Section 4-8