A volcano in Java has been expelling more than 100,000m³ of toxic material a day for three years. Remote monitoring is now being used to investigate the rate of flow and the impact it will have on the surrounding community. Report by Chris Meikle and David Shilston.
The Lumpur-Sidoarjo (Lusi) mud volcano first erupted on 29 May 2006, during the drilling of an exploratory gas well in the Porong district of Java, Indonesia. Over the past three years, a continuous flow of hot toxic mud has reached rates of between 100,000m³ and 160,000m³ a day, and has covered an area of about 640 hectares, despite attempts by local authorities to stem the flow with large earth dams.
So far, 13 people have lost their lives and more than 30,000 people have been displaced from their homes by Lusi’s eruption. The mud flow has penetrated houses, factories, farmland and the Surabaya-Gempol toll road. It has endangered a railway line, the surface water drainage and irrigation network, eco systems, water and gas pipelines,and fibre-optic cables connecting Surabaya to Eastern Indonesia.
The origin of the mud volcano is the subject of considerable controversy and there are two hypotheses. The first is that the eruption was the result of an underground blowout caused by the drilling of the Banjarpanji-1 gas exploration well into very deep over-pressured strata. The second theory is that it was triggered by an earthquake two days earlier, 300km east of Porong, which re-activated the local Watukosek fault system, which acted asconduit from the over-pressurised strata. Within the scientific and academic communities, the well blowout hypothesis is widely supported as being the most likely.
There has been much political sensitivity regarding its cause, and considerable concern within the local community. People wanted to know what impact the erupting mud was going to have, how long it would last and what mitigating measures would be implemented.
Atkins’ UK-based geohazard team was asked by the owner of a site near the mud volcano to advise on the risk that Lusi posed to its assets. An important prerequisite of this request was confidentiality, which meant that, following an initial site visit by Atkins, no site-based monitoring would be sanctioned.
Atkins therefore offered a geohazard monitoring and assessment programme using sophisticated remote sensing techniques, which eliminated the need to undertake on-theground topographic surveys. In May 2007 Atkins made an initial site visit with sub-consultant GeoPressure Technology to explore the mud volcano and its surroundingarea, to assemble information about the owners’ assets, and to make an initial assessment of the risks and hazards.
Lusi had been erupting for about a year at the time of the visit, and best estimates were that it would continue to do so for 10 to 40 years. Judgements about Lusi’s behaviour and the risk posed to the owners’ site in the months and years to come were necessarily tentative. The consultant therefore suggested to the owners that a limited p rogramme of hazard monitoring using remote sensing would best meet their requirements for regularly updating the initial risk assessment.
Atkins, with subconsultant satellite mapping specialist Fugro NPA, designed and implemented a hazard monitoring programme that made use of two types of satellite remote sensing imagery: very high resolution (VHR) optical imagery and differential interferometric synthetic aperture radar (DifSAR).
Remote sensing is a reliable and independent source of information for monitoring the surface development and impact of Lusi’s eruption. The physical consequences of Lusi have been observed and recorded using regularly acquired VHR optical images from the IKONOS and Quickbird satellites.
Comparisons made by Atkins between images taken every one to three months have enabled assessments to be made.
During analysis of the VHR optical imagery, the consultant has focused mainly on the distribution of mudflows from the vent, the construction and performance of the earth dams, changes in behaviour of the nearby rivers, changes in land use, distribution of surface water flooding, and lateral migration of the mud volcano vent.
All observed data was recorded within a geographical information system (GIS) geodatabase, which enabled data to be synthesised and thematic maps to be extracted.
Changes in ground surface profile were observed and recorded using DifSAR imagery. DifSAR is a technique for measuring ground surface movements by comparing two or more radar image (InSAR) observations, separated in time.
The resultant “differential interferogram” contains sequences of ground surface movement contours (interferomatic fringes) that are directly proportional to half the satellite’s radar wavelength: approximately 118mm for the Alos Palsar satellite SAR Data (L Band) used in the Lusi geohazard monitoring.
The DifSAR results are actually movements of the ground surface in the direction of the radar pulse from the satellite and, due to the angle of incidence of the radar pulses, they are approximations to the vertical movement of the ground surface.
Within the scientific and academic communities, the well blowout theory is widely supported as being the most likely
DifSAR pairs have been used at approximately six-month intervals, starting from 19 May 2006 (pre-eruption), to construct a three-dimensional model of ground surface deformation associated with Lusi.
The principal findings of Atkins’ monitoring are that the sustained eruption of Lusi has formed a primary area of subsidence surrounding the mudflow extent.
The subsidence is most severe near the vent, where it is difficult to monitor as it is within the constantly changing mudflow.
To the west of the mudflow there is a zone of secondary subsidence, and there is a zone of lesser magnitude uplift to the north east. So far, the immediate affects of Lusi (specifically the mud flow extent and ground surface deformation) have not had an impact on the clients’ site, and Atkins’ initial site-based risk assessment remains valid.
The large volume of erupted mud is being partly managed by pumping into the Porong River to the south of Lusi. On occasions, the river’s capacity to transport mud has almost been exceeded, and the mud is having damaging
effects on the river and local ecosystems.
It is also possible that continued pumping of mud into the Porong might result in the flooding of adjacent ground by river water due to overtopping and/or failure of its flood defence embankments.
The principal method for managing the continuing mud flow of approximately 100,000m³ a day is by the construction of large earth ams. The earth dams have been subjected to repeated stability failure, subsidence and overtopping, but they have managed to retain the mud inundation at its current extent over the past year.
On the northern side of Lusi, the earth dams have prevented the mud from reaching the minor Ketapang River, which is an important part of the drainage and irrigation system for the local community. Overflows of muddy water appear to have intermittently reached the river, but have not caused any blockage that would be detrimental for surface water drainage.
More recently, significant earth dam failures and subsidence have occurred close to the mud volcano vent, and the containment and direction of the mud into the Porong River has been partially compromised. Increased subsidence at the vent might indicate significant changes in the physical processes and stresses in the ground beneath Lusi.
Since its initiation on 29 May 2006, Lusi’s eruption has had, and continues to have, a devastating effect on the Porong district of Eastern Java, and there are no signs that the mud flow is decreasing. The flow of mud and ground surface deformation has caused significant physical, environmental, economical and social damage to thelocal and regional communities. The remaining lifespan of Lusi’s eruption is still not known, but best estimates are that it will continue to erupt for decades.
For now, a continued programme of regular remote monitoring of Lusi’s behaviour - specifically the mud flow extent and ground surface deformation - such as the one designed and implemented by the team led by Atkins, will provide a valuable resource for updating geohazard and risk assessments, and allow site owners within the vicinity of Lusi to better manage their assets.
Chris Meikle is an engineering geologist and David Shilston is technical director for engineering geologyat Atkins’ geotechnics team.