Installation of first GPS receivers
on Papandayan, July 1999

The University of New South Wales (UNSW) in collaboration with the Volcanological Survey of Indonesia (VSI) and the Institute of Technology Bandung (ITB) successfully installation the first low-cost GPS-based volcano deformation monitoring system in Indonesia.

Craig Roberts returned to Bandung, West Java, in July 1999. "The surface of a volcano deforms as magma moves below the surface. Our goal was to install three GPS receivers on Gunung Papandayan and another away from the zone of deformation, at its base. By constantly measuring the baseline distances between the antennas, we hope to determine any deformation of the volcano. This is not a new idea but the challenge was to still maintain centimetre level accuracy with a system costing perhaps a tenth of the price of existing GPS designs."

The equipment consisted of four GPS receivers & antennas, radios & antennas, solar panels and all manner of cables and gadgets weighing a total of 180kgs. A few days of preparation was spent purchasing ancillary equipment in the dusty hardware stores of Bandung before travelling onsite to Gunung Papandayan.

Ony Suganda, a vulcanologist from the VSI, sacrificed two and a half weeks of his time to help install the GPS system. Hasan Abidin, senior lecturer at the ITB and co-investigator for this project, lent his technician and three of his best final year Geodesy students to assist.

All of the monuments installed in August 1998 were found to be in good condition. Seven local farmers were employed to help carry all the equipment to the sites and to install the slave station systems.

Setup of volcano monitor station with telemetry mast, solar panels and container.

GPS package and batteries in container.

At each site a concrete pad was built, large enough to house the GPS/PC module, radio and a battery inside a cut-off 44-gallon drum. The solar panels were mounted on four PVC pipe struts. The GPS antenna was elevated 1.5m above the ground surface on a fibreglass pole and 5m high antennas were erected for radio communications. At Kawah, the most trafficked station, a fence was erected around the installation to guard against vandalism. The remaining two slave stations were considered sufficiently remote to not require this added protection. (It remains to be seen how wise this decision is.)

Building the volcano monitor station at Kawah (Indonesian for "crater").

Setup of the volcano monitor station at Kawah.

"It took a full four days to install the three slave stations on the volcano and the base station at the VSI observatory, which is approximately 8 kms away. I started the survey at the slave stations, and I returned like an expectant father to the base station to see if we had successful radio communications.

The year or so of planning and testing paid off as all stations began sending data as intended. The bubble quickly burst, however, as various technical and engineering problems stopped the survey a day later. Some of these problems have since been addressed, and some remain unsolved. Such is the nature of field-based research."

The team returned to Gunung Papandayan after six days of analysis and repairs in Bandung to restart the survey and to collect more data. In this short time, the effects of corrosion from sulphurous gases could already be seen on the pitted shank of a previously shiny padlock. Experts at the VSI advise that the 44-gallon drums covering the GPS/PC modules will only last about three months in such conditions!

The power system at the base station is also a challenge. Although the system is running on AC power, frequent blackouts of up to six hours duration, and spikes and surges in the power mean that extra measures need to be taken to ensure that continuous DC power is supplied to all components so as to avoid stoppages in the survey.

The collaboration between the VSI, ITB and UNSW has been refreshing. All groups have different skills, which augers well for cooperation. Members from the VSI and ITB have been trained to use the system so that any in-country changes can be easily carried out. Several presentations describing the current system were made at the VSI in Bandung, which generated a lot of interest amongst their volcano experts.

The focus now is on demonstrating the longevity of the system in a harsh volcanic environment, and acquiring consistent and reliable data. This is the first step toward providing a feasible continuous deformation monitoring system for use in forecasting volcanic behaviour from ground deformation techniques. Future hardware and software improvements will be reported here.