Proceedings World Geothermal Congress 2010 Bali, Indonesia, 25-29 April 2010 1 Monitoring of LUSI Mud-Volcano - a Geo-Pressured System, Java, Indonesia Manfred P Hochstein, Sayogi Sudarman SGGES and IESE, University of Auckland, Priv, Bag 92019, Auckland, New Zealand Trisakti University, Jakarta, Indonesia Keywords: Surface discharge, boiling fluids, geo-pressured system, low-T system, heat and mass discharge rates. ABSTRACT The LUSI mud-volcano is the dominant surface manifestation of a geo-pressured, low temperature geothermal system in Java. It started with an eruption of hot fluids and mud at the end of May 2006, close to an uncompleted, 2.8 km deep oil exploration well, two days after a major earthquake. The mud discharges from a concealed crater have been irregular with inferred long-term volume rates between 1 and 1.5 m 3 /s and have continued ever since. The flooded area increased from c. 1 km 2 in June 2006 to c. 6 km 2 in May 2007; it displaced c. 40,000 people. The area has been enclosed and divided by a network of huge dams. The discharged hot liquid mud came initially from depths around 1.7 km where formation pressures of c. 30 MPa and temperatures of 100 deg C (gradient of c. 40 deg C/km) prevail. Pore pressure decrease in the upper sediments caused the initial subsidence over an area of the order of c. 7 km 2 after one year. The resulting subsidence bowl has accommodated only part of the extruded mud whose spreading is constrained by the huge dams. Monitoring of subsidence and mud discharge began in June 2006 but ceased in April 2007 when sites were flooded. Satellite records (INSAR analysis of radar data) have been used to monitor subsidence of the not flooded region during that period. The rather short (< 2 km) wave length of subsidence during the first year points to shallow (< 0.5 km) source depths of de-compressed sediments. Interpretation of IKONOS satellite photos allows an assessment of heat discharged by the central steam plume and the upwelling mud column. The heat loss by steam discharge over the crater has fluctuated between c. 3 and 150 MW, losses due to hot mud discharges were always > 200 MW. Discharge trends during 2008 pointed to an overall irregular decrease of steam and mud losses which, however, was followed by a renewal of discharge activity in February 2009. The rate of gas discharges (mainly CO 2 and CH 4 ) from the central crater and gas discharges over the flooded area have not been measured yet. Micro-earthquake and micro gravity surveys were started in 2006 and repeated in May 2008 together with a ground temperature survey. 1. INTRODUCTION The presently active LUSI mud-volcano is the dominant, central, surface discharge feature of a geo-pressured, low temperature geothermal system. It started with a major eruption of hot fluids and mud on 29 May 2006, close to the 2.8 km deep Banjar Panji-1 (BJP-1) exploration well, drilled by PT Lapindo near Sidoarjo in E Java (for location see Figure 1). Minor eruptions occurred a few days later along a c. 1 km long segment within a probably c. 250 m wide, NE trending fracture zone. Only the first eruption centre, c. 200 m SW of the BJP-1 well, has remained active. At the start of eruptions, the 2.8 km deep well had encountered a sequence of over-pressured and under-compacted Pleistocene sediments below 1.3 km depth. The well was not cased below 1 km depth which facilitated rupturing within the fracture zone; an M 6.2 earthquake had occurred on 27 May 2006, about 200 km away in Central Java. Figure 1: Location map of LUSI mud volcano near Sidoarjo (Java) Gas (CH 4 , CO 2 , minor H 2 S), hot water, and hot mud at boiling point ascended to the surface, initially with a flow- rate of c. 0.5 m 3 /s. The discharge fluctuated and increased on average to > 1 m 3 /s after 2 months with fluctuations reaching an inferred peak discharge rate of c. 1.5 m 3 /s in December 2006. The terrain around the discharge centre is rather flat (3 to 10 m a.s.l. within a radius of a few km). The size of the flooded areas increased rapidly from c. 1 km 2 in June to 6.5 km 2 in May 2007. The flooded area has been encased and contained by an extensive dam building program. Monitoring and assessment of discharge rates, subsidence (using GPS), and fluid characteristics began in June 2006. Surveillance included micro gravity-, resistivity- and micro seismic surveys at irregular periods up to April 2007. Some results of this first monitoring period have been published. The aim of the first publications (Mazzini et al., 2007, Davies et al., 2008.) was to prove or disprove the importance of a possible triggering effect of the drilling operation. Attempts to predict future subsidence and mass accumulation were based mainly on earlier monitoring data collected until May/June 2007 (Istadi, pers.com.). Monitoring efforts declined after May 2007 but were taken up again in April/May 2008 with new micro-gravity- and micro-earthquake surveys, augmented by a few shallow ground temperature studies. The mud discharge from the central vent has continued unabated since June 2006. Since most monitoring studies were made at irregular intervals, often by different groups from different organisations, it has been proposed to continue with a more coherent and regular monitoring programme of important parameters and features of the mud-volcano with the aim of detecting trends which might allow a prediction of short-