Conclusions
5. CONCLUSIONS
· The LUSI mud volcano that erupted near Surabaya (Java) at the end of May 2006 is the dominant surface manifestation of a geo-pressured, low temperature geothermal system. It has discharged hot liquid mud close to boiling point temperature already for more than 3 years and has flooded in March 2009 an encased area of more than 10 km2. Other manifestations include wide-spread gas discharges (mainly CO2 and CH4). The hot mud rises from >1.7 km depth.
· In comparison with other mud volcanoes described in the literature (Kopf, 2002), the LUSI manifestations are anomalous because of the rather low viscosity of the discharged hot liquid mud with its high gas component that has produced an extensive ‘mud flood-plain’ (‘mud-lake’) which strictly can not be classified as a ‘mud volcano’. However, the term ‘mud volcano’ has been used for LUSI in earlier publications and is used here as well.
· The active mud discharge centre exhibits a sub-surface crater structure and occurs close to the 2.8 km deep uncompleted oil and gas exploration well (BJP-01). Two different hypotheses have been put forward to explain the triggering of the eruptive event. One hypothesis suggests that a large earthquake (M 6.2) in Central Java, which occurred two days prior to the first mud eruptions on 29 May 2006, caused fracturing within an already existing fracture zone thus triggering the event. The other hypothesis implies that a blowout, as a result of drilling problems in the unprotected c. 1.7 km long bottom section of the BJP-01 well, was the triggering event. Each hypothesis on its own is not fully supported by monitoring results.
· Several detailed monitoring studies were made during the first ten months after the first mud eruptions. They included assessments of volume and mass flow-rates, subsidence and surface movement, geochemistry, micro-gravity, and microearthquake (MEQ) surveys. Many results were affected and disturbed by soil compaction at sites located on dams and in protected enclaves. Only a few long-term GPS surveys and long-term level changes of a few engineering sites provide some representative long-term subsidence rates which are of the same order of magnitude as rates obtained from an INSAR analysis of satellite-borne radar data observed in 2006/7. Modelling of subsidence patterns points to subsidence caused by compaction of de-pressured sediments, initially at c. 0.6 to 0.7 km depth. An analysis using a half-wavelength approach produces a similar result.
· The time-lapse anomalies of micro-gravity surveys conducted in 2006 can not be interpreted because changes in station height were not recorded. The same applies to repeat gravity surveys undertaken in 2008. MEQ studies in 2006 and 2008 showed that shallow seismic events mainly occur beneath the southern sector and the southern margin of the subsidence area pointing to a triggering surface loading effect. The data do not allow a more detailed interpretation because of the rather inhomogeneous near-surface seismic velocity structure. Two shallow ground temperature surveys were carried out in 2008 and showed that anomalously low temperatures at 1 m depth (a few deg C below mean annual minimum air temperature) occur near gas discharge centres pointing to a Joule-Thompson effect by rising gases (dominantly CO2). Another short-term (20 days) monitoring survey of mud discharge temperatures of the LUSI ‘crater’ showed that the temperature of the liquid mud varied between 88 and 110 deg C with the highest temperatures occurring after a large, distant quake.
· Satellite (IKONOS) photos published at roughly monthly intervals constitute important long-term monitoring data that allow an assessment of heat and mass discharged at the crater by the always present steam plume and the upwelling mud. The published photos were all taken under almost cloudless conditions, c. 1 ½ hr before local noon. Meteorological conditions are rather constant at that time throughout the year. Heat transfer by direct steam discharge can be assessed from the steam cloud volume reduced for small variations in relative humidity. The steam losses were found to fluctuate between c. 3 MW and 150 MW during the last 3 years. Heat losses by hot mud discharge could be assessed for the first 10 months for which a few volume discharge rates are known. Two months after the first eruptions, the heat transfer by hot mud increased from c. 100 MW to c. 300 MW. It was found that the diameter of the steam cloud at mid-height and that of the gas-charged outer mud ring around the crater show some correlation that can be used for trend analysis.
· The data indicate short and long period variations which are most pronounced in steam cloud changes. Short-term variations occurring during periods of less than a week and often during one day point to some ‘slug flow’ of the liquid component. Voluminous mud discharges which can cause dam breaks and excessive mud flooding seem to follow some periodicity. Such large discharges occurred at the end of 2006, and again during the first months of 2008 and 2009. There are no clear trends which point to an overall long-term decline of steam and mud discharge activity.
· In view of the present uncertainty in predicting future activity trends of the LUSI mud volcano, installation of a well co-ordinated and regular monitoring programme is required. This should include quantitative assessment of the green-house gas discharges about which nothing is known yet. In addition, analysis of more recent radar data should be continued as well as monitoring of subsidence in non-flooded areas outside the confining dams.
ACKNOWLEDGEMENT
The input arising from several discussions with Mr. B. Istadi (PT Lapindo), Dr.Prihadi (ITB), and Mr. Soffian (BPLS) was a great help in compiling this paper. Mr. K. Vincent from CRISP (National University of Singapore) gave permission to present the IKONOS satellite photo of the LUSI mud volcano taken on 05.06.2007.
Pasted from <file:///F:\1A_LULIB_NEW\SUDARMAN\Sudarman2010a.doc>