Fluid transport properties and estimation of overpressure at the Lusi mud volcano, East Java Basin (Tanikawa et al., 2010)

Abstract

1. IntroductionThe Lusi mud volcano in Sidoarjo, East Java, started to erupt onMay 29th 2006 and has displaced 13,000 families. Controversysurrounds whether the mud volcano was caused by drilling of theBanjar Panji 1 gas exploration well (Davies et al., 2007; Manga, 2007;Davies et al., 2008; Tingay et al., 2008) or due to the Yogyakartaearthquake that occurredat 05:54 amon the27th May2006(Mazziniet al., 2007; Sawolo et al., 2009). Constraining (a) pore pressure insedimentary strata prior to drilling; (b) changes in pore pressure thatoccurred due to drilling the Banjar Panji 1 well and the earthquake,and; (c) potential routes for fluid to the surface are critical forresolving this debate. These areas are tackled in the recent paper byTanikawaetal.(2010),whomodelporepressuredevelopmentduringburial of sedimentary strata at the site of the eruption and measurethe permeability and porosity of outcrop samples of formations ofequivalent age and lithology as those penetrated by the well. Fromthis analysis, they conclude that overpressure developed withinspecific successions, in particular the Upper Kalibeng clays and a deepcarbonate formation. They go on to argue that the overpressure madethe clay-rich unit susceptible to liquefaction as a result of cyclicdeformation duringtheYogyakartaearthquake,and thatthisinitiatedthe mud volcano. In this discussion we begin by summarising themain conclusions made by Tanikawa et al. (2010) and then considerthe validity of their most important conclusion that Lusi is a naturaldisaster. Lastly, we reiterate the compelling evidence (Davies et al.,2008;Tingayetal.,2008)thatLusiisman-madeandwascausedbyanunderground blowout at the Banjar Panji 1 well.2. Main conclusions of Tanikawa et al. (2010)The main conclusions of the Tanikawa et al. (2010) paper werethat overpressure developed within the bluish grey clay, termed theUpper Kalibeng Formation Unit 2, and therefore, these strata wereundercompacted and susceptible to remobilisation. They consideredthe deeper carbonates (termed the Upper Kujung Formation) as themost likely source of fluids. (Note: earlier papers on Lusi (e.g. Davieset al., 2007; Mazzini et al., 2007) referred to the carbonates as theKujung Formation however, strontium isotope ratios from themindicate they are 16–18 Ma (Kusumastuti et al., 2002), and thus arenottheOligioceneKujungFormation, butrathershouldbetermedthePrupuh or Tuban Formations (Tingay, 2010) and are thus termed thePrupuh Formation for the remainder of this discussion.) The authorshypothesized that the elevated pressure within the Upper KalibengFormation Unit 2, made the sediment more susceptible to the smallamplitude static or dynamic stresses caused by the Yogyakartaearthquake, so that small changes in pressure were sufficient toinduce liquefaction. The rapid influx of gas and liquid from the deepcarbonates flowed through pre-existing pathways formed by naturalhydraulic fractures, and this explains the continuous mud eruption atLusi. In summary, Tanikawa et al. (2010) hypothesize that pore fluidpressure changes caused by the Yogyakarta earthquake causedliquefaction of undercompacted shales and fluid flow occurredthrough natural fractures.3. DiscussionThe paper provides key new information on the hydrodynamics ofLusi, and we indeed agree with several important points made by theauthors. For instance data from Banjar Panji 1 corroborates themodelling by Tanikawa et al. (2010) that the Upper Kalibeng Unit 2 isa fine grained lithology that is overpressured and undercompacted.We also agree that the most likely source of fluid is the deepcarbonates. Furthermore, we agree with the model results suggestingthat significant overpressures exist in and below the PrupahFormation carbonates. Very high magnitude overpressures wereobserved in the same section in the Porong 1 well, 6 km away fromLusi.Although we are in agreement in these areas, there are severalconclusions on how the mud volcano was triggered which are at oddswith our published research (Davieset al., 2007; Manga, 2007; Davieset al., 2008; Tingay et al., 2008) and given the scale of thishumanitarian disaster it is important that these are challenged. Themost important of these, is their reasoning that the Yogyakartaearthquake triggered the eruption rather than drilling of the BanjarPanji 1 well. Tanikawa et al. (2010) propose that stress fluctuationsinduced by the M6.3 Yogyakarta earthquake (2 days prior to theeruption) caused the mud in the Upper Kalibeng Unit 2 to losestrength and liquefy (their abstract and section 7.1). Others have also