Site selection and preparation

Installing Sensors

• Bury seismometers for noise reduction and stable ground coupling, typically 0.5-0.8 metre depth but deeper the better.
• Ensure the sensor is leveled correctly, typically the sensor can be placed on a well leveled paver to make this easier.
• Orient the sensor correctly using a compass, paying special attention to the north direction and accounting for declination.
  • When using a compass to orient the sensors, ensure it is kept away from metal objects or structures that could interfere with its magnetic field.
  • It is recommended to take a picture of the sensor's orientation next to the compass in case there are questions or issues later.
• Hold the sensor or sensor covering securely while infilling and compacting the hole to ensure the setup is kept in the correct position (level and oriented).
• Burial styles can vary depending on sensor type, soil, wetness/humidity and the duration of the experiment. See the Sensor Protection section below for more detail.

Importance of locking sensor feet

The Trillium Compact 120s and 20s sensors have three adjustable feet for leveling. It is critical to "lock" these feet in place by spinning the locking disk upwards towards the sensor, as tight as possible. This reduces "wobble" which shows up in both low and high frequency signal. It is also a good idea to keep the length of the three feet as small as possible to maximize stability.

Setting up Data Logger

• Install data loggers or recorders compatible with the sensors.
• Establish a reliable power source, such as solar panels, batteries, or local grid connection. Use a compass to align solar panels facing north in the southern hemisphere for optimal sunlight exposure.
• Set up a GPS antenna to provide accurate time synchronization for the seismic data.
• Ensure the GPS antenna has a clear view of the sky for optimal signal reception.
• Calibrate sensors and data acquisition systems for accuracy.
• Test for sensitivity, noise levels, and overall performance.
• More information for logger setup can be found on the ANU Seismic Data Loggers page.

Setting up Fencing

• You'll have to know a priori where north is as that is where you want to point the solar panel towards (in the Southern Hemisphere, anyway)
• Pound in the star picket fence post well away from the sensor hole, and slightly north of it. You will want to then put the fence through the star picket so that it is on the NORTH side. This lets techs "flip up" the fence from behind for easy access.
• Use wire to secure the fence to the post and also the solar panel to the fence.
• Place the solar panel as high on the fence as possible to reduce any interference from grass and weeds. When securing the solar panel to the fence with wire, make the wire as tight as possible to reduce "rattle" in the wind. Test yourself. It's usually a great idea to use a pair of pliers to make the final twist in the wire so that it is really tight.

Fire Protection & Security

• In fire-prone areas, clear a perimeter around the installation to reduce fuel sources. Assume the area will catch on fire-- will your site survive?
• Consider using fireproof covers for the logger. This will have the added benefit of reducing soil contact, especially dirt getting into SD card slots. A permeable cover will avoid trapping rainwater -- a problem associated with plastic tarps, which damages the logger and attracts insects/animals (ants, centipedes, snakes, etc.). We are still testing materials for suitability in the field (e.g., safe and tolerable degradation in the environment).
• We recommend adding a soil layer for insulation. This seems to keep loggers cooler, reduce fire damage, and discourage interference by people passing by.

Metadata & Site Logs

• Documenting site installs and service information is a mandatory requirement of your ANSIR agreement. You are expected to keep proper site logs... trust us, it's for your own good. Failure to do so may result in ban for future loans.
• Document the installation process, including sensor types, serial numbers, orientations, high quality latitude/longitude coordinates, elevation, and system configurations, along with fire safety measures implemented.
• Draw a map, or at least take a google/open maps screenshot with some drawn annotations so others can find the site.
• Record essential metadata for seismic data interpretation.

Sensor protection

Sensors can be covered with a PVC pipe to help prevent degradation of the components (i.e. sensor casing and cable connections), however, some sensors are made for direct burial (no protection). When installing a sensor, only use these two options. Do not try to protect the sensor via any other means such as a plastic bag or moisture absorbers & desiccants.

Step-by-step field installation guide (with images)

1. Dig a hole roughly 80cm deep and wide enough to place a paver in.	2. Level the bottom of the hole. A paver or brick can be used to create a flat base for the device. Draw a line on the paver, orient it towards North and level it.	3. Place the seismometer on the paver and ensure it is levelled and oriented (remember to account for the declination in your area).
4. Depending on the model of the seismometer, place a protective PVC tub over it.	5. Fill in the hole whilst stabilising the sensor and packing the dirt firmly.	6. Once the hole is filled and levelled, drive a stake (star pickets work well) into the ground next to the seismometer, and fit the wire panels over it, driving pegs into the north-facing side.
7. Attach the solar panel to this side, and the GPS to the stake (wrap the wire around the stake once or twice so it stays in position if it is detached).	8. Connect the seismometer, solar panel, and GPS to the logger and turn the logger on. See "setting up data logger" and "ANU seismic data loggers" page for more details. Set the logger to record and ensure it starts recording.	9. Cover the logger with a fire blanket, starting by placing it on the top, so the opening is underneath the logger.
10. Once all the sides of the blanket are tucked under the logger, cover the logger with rocks or dirt.	11. Fold the other metal grating panel over and peg it into the ground. If the ground is too hard, a wire can be used to fix the metal grates to each other. This will also help reduce noise from the rattle of the cage prevent animals from getting to the logger.	12. A completed site. Rocks and sticks can also be piled on the ends of the cage if there is a risk of animals tampering in the area.

Experimental Sand Burial:

Direct burials are used for Trillium Compact Postholes as they are specifically designed to be corrosion resistant. Other sensors such as the Trillium Compact 120, Trillium Compact 20, and 3D Lites are not built with this degree of corrosion resistance. Currently, the best prevention for moisture trapping and corrosion is burying them with a PVC covering. However, in particularly wet environments, some moisture can still collect within these coverings. One experimental method that is being trialled in hopes of providing better drainage is the sand burial. Steps 1, 2, and 3 remain the same, though once the sensor is set with the correct leveling and orientation, the subsequent infill steps differ. Once the open PVC tube is placed around the sensor, infill the hole around the tubing, holding it in place (image A). Once the hole is back-filled level with the tubing, securing it in place, begin filling the interior of the PVC tube with sand. Hold the sensor in place and ensure the sand is packed tightly around it (image B). Once the sand fills the PVC tube, gently remove the PVC tube while holding the sensor in place (pliers may be required to grip the tube). The last deviation for a sand burial from the standard burial is continuing to place some sand above the sensor, and dirt (or other local substrate around that sand pocket) until level with the surface (image C).

A)

B)

C)

Comparison of installation methods on Black Mountain (Canberra): 

The seismic equipment was left in the field for approximately 6 months (April – October) through winter. There was much rain throughout the experiment period, with the last rain occurring one day prior to excavation, and heavy rain occurring the week before. Note that more temperature fluctuations may occur throughout summer, and thus more chances for condensation and corrosion to occur, therefore this experiment should be conducted again over a summer period. 

Upon excavating all setups, there appeared to be little difference in the moisture levels. 

The first method (PVC) had collected a small amount of moisture. The seismometer was a little bit wet, while the interior of the PVC housing had condensation lining the walls. The paver beneath was also quite damp, however, no water had pooled at the bottom of the seismometer. 

The other two setups (both using the sand method) were also quite moist. Upon excavation, the coarse river sand was damp (though not soaked). There had been no infiltration of mud from the surrounding area into the coarse river sand, suggesting that all the moisture had come straight down the column of sand, and not flowed in from the sides. Underneath the feet of the seismometer (SN3660) from the full setup, much condensation had been collected. There were no feet on the seismometer of the second sand setup and thus no space for condensation underneath. The tops of both seismometers were covered in damp sand, though there appeared slightly less moisture than the top of the sensor in the PVC housing. An accurate assessment of moisture content was difficult as one seismometer was covered in sand, and the other showed a clear view of how much surface condensation there was.  

Additionally, when unwrapping the cloth adhesive from the PVC burial, there appeared to be a glob of hydrated glue from the adhesive. This was not present on the connections of the two sand burial seismometers, despite using the same cloth adhesive. This may suggest that moisture lingers for a longer period in the PVC housing (as suspected), having time to be absorbed into the adhesive more.  

Upon inspection of the connectors of each seismometer, there was little condensation within each, and no sand had appeared to infiltrate the connections in the sand buried equipment. 

Key findings: 

The PVC burial method resulted in a small amount of moisture collecting on the top of the seismometer and within the PVC casing, and the development of a hydrated glob around the connector.  

The sand burial method resulted in slightly less moisture collecting around the top of the seismometer (subject to error of perception), though no glob of hydrated adhesive, suggesting quicker drainage did occur. The sand burial did however allow for more moisture to collect on the underside of the seismometer compared to the PVC burial, though as the connector is atop the seismometer, this result may be irrelevant. 

It should be noted that there was no mud present in the column of sand, suggesting little to no lateral seepage of moisture. 

Lastly, all connectors contained a small amount of condensation, were free of sand or dirt (thanks to the cloth tape), and no corrosion was noticeable. 

Figure 1: Typical station setup (LPR-200 & 10w solar panel)

Figure 2: Flipping up the back of the fence onto the support picket to work comfortably