A regional analysis of fault reactivation and seal integrity using the ...

60 Reynolds et al.FIGURE 2. Bight Basin correlation chart showing the relationships between sequence stratigraphy and basin phases,modified after Totterdell et al. (2003). Distribution of source, reservoir, and seals in the petroleum systems are modifiedafter Totterdell et al. (2000), and modeled timing of onset of expulsion (arrows) are modified after Struckmeyer et al.(2002).

Analysis of Fault Reactivation and Seal Integrity Based on Geomechanical Modeling 61(Hammerhead system). Oil and gas expulsion from theAlbian to Campanian source rocks is modeled to havebegun during the Cenomanian to Maastrichtian, althoughexpulsion continued, in phases, into the Tertiaryand up to the Holocene in certain parts of thebasin (Struckmeyer et al., 2001).The exploration wells drilled to date encounteredonly minor oil and gas shows, and the majority of intervalsintersected are immature for oil generation.However, the source rock quality in the deeper partsof the basin is inferred to range from good to excellent,and modeling shows that adequate maturitieswere reached to generate and expel liquid and gaseoushydrocarbons (Struckmeyer et al., 2001).EMPIRICAL EVIDENCE FORFAULT SEAL FAILURE IN THEBIGHT BASINIdentifying the principal exploration uncertaintiesin the Bight Basin region is difficult. Several observationsindicate that poor fault seal integrity may representa key exploration risk in the Bight Basin. Based onpetrophysical analyses of fluid-inclusion samples, Rubleet al. (2001) demonstrated the presence of a 15-m (49-ft)-thick paleo-oil column in Callovian to Kimmeridgiansands in the Jerboa-1 well in the Eyre subbasin (Figure 3).The hypothesis is that Jerboa-1 was charged, probablyfrom a Late Jurassic–Early Cretaceous petroleum systembut was later breached during a period of faultreactivation during the Late Cretaceous. Vertical migrationof hydrocarbons into the Tertiary sequencesmay also have occurred subsequent to this Late Cretaceousbreaching (Ruble et al., 2001). The fault reactivationand subsequentvertical migration may havebeen the result of the far-fieldeffects of the collision of theAustralian and Asian plates inthelateTertiary.Other empirical indicatorsof seal failure are presentthroughout the Bight Basin.These include the presenceof numerous gas chimneysin the Duntroon and Cedunasubbasins, some of which correlate spatially with watercolumn geochemical sniffer anomalies and observationsof oil slicks across the Bight Basin using synthetic apertureradar data (Struckmeyer et al., 2002). Other indicatorsinclude the well-known presence of asphaltitestrandings in the region (Sprigg and Wooley, 1963;Edwards et al., 1998) and colloquial evidence for a relationshipbetween the timing of earthquakes and theoccurrence of major strandings in the area. Overall,earthquake activity has been typically focused in theDuntroon and eastern Ceduna subbasins. Most of theearthquakes in the Bight Basin are shallow, with epicentersin the upper 10 km (6 mi) of the sedimentarysection, and many are within the top 5 km (3 mi), wellwithin the syn- or postrift sections section. Magnitudesare typically in the range 2–3.5, with some events ashigh as 4.6 have been recorded on the Couedic Shelf.In addition, a series of earthquakes with magnitudesranging between 4.2 and 5.2 were recorded along thefar southern edge of the Recherche subbasin. The empiricalevidence for fault seal failure prompted us toanalyze the stress field in the Bight Basin and investigatethe likelihood that mapped fault arrays in thesubbasins in the region will undergo sufficient structuralreactivation to induce fault seal failure.METHODOLOGY FOREVALUATING FAULTREACTIVATION IN THEBIGHT BASINWe have used the results of the in-situ stress fieldanalysis in the Bight Basin (Reynolds et al., 2003) and appliedthe fault analysis seal technology (FAST) techniqueFIGURE 3. Seismic line throughthebreachedtraptestedbytheJerboa-1 well in the Eyre subbasin,western Bight Basin. Reactivationfaults stop at the baseof the Tertiary (modified afterTotterdell et al., 2000).

62 Reynolds et al.to determine the risk of fault reactivation of the LateJurassic Sea Lion faults and the Turonian–Santonian Tigerfaults. In the FAST technique, the risk of fault reactivationis determined using the stress tensor (Mohr circle)and fault rock strength (failure envelope). Brittlefailure is predicted along optimally oriented faults ifthe ratio of shear to effective normal stress exceedsthe coefficient of frictional sliding, which can be illustratedwhen the Mohr circle intersects the failureenvelope. All fault orientations plot within the Mohrcircle, and those closest to the failure envelope are atgreatest risk of reactivation. The horizontal distancebetween each fault plane and the failure envelopeindicates the increase in pore pressure (P) required tocause reactivation and is used as the measure of thelikelihood of fault reactivation in the FAST technique. Asmall P infers a high likelihood of reactivation, anda large P infers a low likelihood of reactivation. TheP value for each plane can be plotted on a stereonetas poles to planes. The risk of reactivation of any preexistingfault orientation is then read from the stereonet.A composite Griffith–Coulomb failure envelopehas been assumed in this study. No fault rockfailure envelopes are available for the area, and thus,a cohesive strength of 5 MPa and friction angle of 0.6have been assumed. For a more detailed discussion onthe FAST methodology, see Mildren et al. (2002) andMildren et al. (2005).The assessment of fault reactivation risk is commonlyappliedatprospectscale,wherethein-situstressfield is better constrained than at basin scale, and depthconvertedfault geometries are available from seismicinterpretation. However, assessing fault reactivation ata regional, basin scale can follow similar principles, butmore care should be exercised in evaluating the reactivationrisk, because more uncertainty is placed on theresults when basing the interpretation on just one set ofparameters averaged and extrapolated over large areas.Fault reactivation predictions presented herein maynot be accurate in areas where the regional stress fieldhas been perturbed by local structures, such as faultsand contrasting material properties. The inclusion oflocal stress perturbations is beyond the scope of thischapter and is not actually possible at the current levelof knowledge of the area.IN-SITU STRESS IN THEBIGHT BASINThe most important step in assessing the risk ofstress-induced fault reactivation that could lead to faultseal failure is to construct a well-constrained geomechanicalmodel. The geomechanical model consists ofthe in-situ stress field, fault rock properties, and porepressure. The in-situ stress field is made up of threeprincipal stresses, which are assumed to be the verticalstress, the maximum horizontal stress, and the minimumhorizontal stress. This assumption is reasonablegiven the generally flat-lying seabed surfaces of sedimentarybasins. To define the in-situ stress field, theorientation and magnitude of the three stress componentsmust be determined.The in-situ stress field was determined from assessingthe drilling and logging data acquired from thenine open-file exploration wells drilled in the BightBasin (Reynolds et al., 2003). Of these nine wells, six areclustered in a tight group in the Duntroon and easternCeduna subbasins. To gain a better understanding ofthe regional stress field in the region, additional wellsfrom the adjacent Polda Basin were included in thisstudy (Figure 1). Stress information from the Australianstress map database was also included in thestudy. The reader is referred to Reynolds et al. (2003)for a more detailed description of the stress field determinationin the Bight Basin.The water depth across the Bight Basin ranges fromless than 100 m (3300 ft) on the shelf to more than5000 m (16,400 ft) in the deepest parts of the Recherchesubbasin. This variation poses a significant problemwhen attempting to analyze the in-situ stress field forthe entire basin. At any location and depth in the basin,the stress caused by the weight of the water columncontributes to the magnitude of the in-situ stressfield. To overcome the problem associated with waterdepth, effective stress (total stress minus pore pressure)has been used in this study, instead of the total stress,which is more typically used.Stress OrientationsThe maximum horizontal stress (S Hmax ) orientationin the Bight Basin was determined by interpreting boreholebreakout directions in logs from a four-arm dipmeter(high-resolution dipmeter tool, HDT) from fourwells and image log data (Formation Microscanner, FMS)from two wells. The average S Hmax orientation calculatedfrom the six wells is 1308N. The reader is referredto Reynolds et al. (2003) for a more detailed descriptionof the stress field determination in the Bight Basin.The S Hmax orientations for the Bight Basin and surroundingregions are plotted in Figure 4. Stress orientationscould only be determined for the wells that arelocated on the eastern side of the Bight Basin. The averageS Hmax orientation of 1308N for the available wellsin the Bight Basin is consistent with S Hmax orientationsin the Otway Basin farther to the east (Figure 4).Stress trajectories, which are essentially regionally averagedstress orientations, for the Australian stress fieldhave been calculated by Hillis and Reynolds (2000) andplotted in Figure 4 to obtain a better understanding of

Analysis of Fault Reactivation and Seal Integrity Based on Geomechanical Modeling 63FIGURE 4. Stress map of Australia (A–D quality) including the new Bight Basin stress data. Stress trajectory map fromHillis and Reynolds (2000) has been plotted to highlight the regional trends across the Australian continent. The S Hmaxorientation for the Bight Basin is reasonably consistent with the stress trajectories in the region. Enlargement: Stressmap of the Bight Basin showing A–D quality stress indicators. Orientation of vector represents the S Hmax orientation,and length of vector represents the data quality. Wells with no data or E quality data are represented by a dot. NF =normal faulting stress regime; SS = strike-slip faulting stress regime; TF = reverse (thrust) faulting stress regime; U = unknownstress regime.the regional stress field over the entire Bight Basin. Thepreviously calculated stress trajectories on the easternside of the Bight Basin are consistent with the new S Hmaxorientation determined from the wells in the region. Onthe western side of the Bight Basin, the stress trajectoriesindicate a more east–west orientation, which is causedby the influence of the east–west S Hmax orientation inthe Perth region to the west (Hillis and Reynolds, 2000;Reynolds and Hillis, 2000). Because of the lack of availabledata in the western Bight Basin, we are unable toverify if the S Hmax orientation rotates to an east–westorientation in the western part of the Bight Basin. Assuch, we have used the average S Hmax orientation of1308N for the entire Bight Basin. Additional drilling inthe western Bight Basin would allow the true stressorientations in that area to be determined.

64 Reynolds et al.Vertical Stress MagnitudeThe vertical or overburden stress (S v ) at a specifieddepth can be equated with the pressure exerted by theweight of the overlying rocks and expressed asS v ¼Z zrðzÞgdzð1Þ0where r(z) is the density of the overlying rock columnat depth z, and g is the acceleration caused by gravity.Vertical stress magnitudes were determined using densitylog data for a total of 10 wells in the Bight andPolda basins. The density logs were initially filteredfor bad hole conditions using the bulk density correctiondensity (DRHO) and the caliper data. Verticalstress calculations require that the density log be integratedfrom the surface (here sea level, assuming thewater column has a density of 1.03 g/cm 3 ). However,the density logs are not commonly run from the surface.The average density from the surface to the topof the density log run can be estimated by convertingcheck-shot velocity data to density using the Nafe–Drake velocity-density transform (Ludwig et al., 1970).To account for the variation in water depth, the verticalstress profiles have been calculated as effectivevertical stress (S v 0 ), assuming normally pressured sediments(Figure 5). The effective vertical stress in theBight Basin is closely approximated by the power lawfunctionS v 0 ¼ 10:46z 1:179ð2Þwhere effective vertical stress is in megapascals and z isthe depth in kilometers below seabed.Minimum Horizontal Stress MagnitudeA total of seven leak-off tests and eight formationintegrity tests were performed in four wells over theBight region. Formation integrity tests are not reliableindicators of the minimum horizontal stress,becausenofractureiscreatedatthewellborewall,and hence, they were not used to constrain the minimumhorizontal stress magnitude. The reliable leakofftest pressures were plotted along with the formationintegrity tests as effective stress magnitudes tocompare wells in varying water depths. The lowerbound to the effective pressures from the leak-off testssuggests that the effective minimum horizontal stress0(S hmin ) gradient is approximately 6 MPa/km. Becauseof the lack of leak-off data, especially below 2000 m0(6600 ft), the S hmin gradient for the Bight Basin cannotbe well constrained. Nevertheless, it is clear fromFIGURE 5. Effective vertical stress magnitudes from seabedfor the Bight Basin. A power law function has been used toapproximate the effective vertical stress.0the results obtained that the magnitude of S hmin is less0than that of S v (Figure 6). Hence, the Bight Basin is0 0 0either in a strike-slip faulting (S hmin < S v < S Hmax )or0 0normal faulting (S hmin < S Hmax < S 0 v ) stress regime.Maximum Horizontal Stress MagnitudeThe magnitude of S Hmax is the most difficult componentof the stress tensor to quantify. Many of themethods commonly applied for constraining S Hmaxcould not be applied in the Bight Basin because of a lackof relevant data. The occurrence of borehole breakoutsand drilling-induced tensile fractures could not be usedto constrain S Hmax because drilling-induced tensile fracturewas not present in the image logs, and rock strengthdata were not available. Hydraulic fracture test-basedtechniques could not be applied because no extendedleak-off tests or minifracture tests have been undertaken.Nonetheless, broad limits can be placed on S Hmaxbased on the frictional limits to stress beyond which

Analysis of Fault Reactivation and Seal Integrity Based on Geomechanical Modeling 65optimally oriented, preexisting fault (Sibson, 1974;Jaeger and Cook, 1979; Zoback and Healy, 1984).The frictional limit to stress is given byS 10S 30 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2ðm 2 þ 1Þþmð3Þwhere m is the coefficient of friction that the crust cansupport until an optimally oriented preexisting faultslips to regulate the stress magnitudes; S 1 0 is the effectivemaximum principal stress; and S 3 0 is the effectiveminimum principal stress.For a typical value of m = 0.6,S 10S 30 3:12ð4ÞFIGURE 6. Effective stress-depth plots for the Bight Basin.0S hmin is represented by the lower bound to effective0pressures from leak-off tests, and S Hmax has been determinedfrom frictional limits. S v has been calculated using0the power function described by equation 2. LOT = leakofftest; FIT = formation integrity test.faulting occurs. The magnitude of the effective maximumhorizontal stress (S Hmax ) was calculated to re-0move the effect of the water depth. The magnitude of0S Hmax can be constrained by assuming that theratio of the maximum to minimum effective stresscannot exceed that required to cause faulting on anThis relationship can be used to estimate the magnitudeof S Hmax in seismically active regions (Zoback0and Healy, 1984) and could provide an upper bound0to S Hmax in the relatively seismically passive regionssuch as the Bight Basin.0In the Bight Basin, S hmin is most likely less than0the effective vertical stress, implying that S hmin = S 0 3 .0The frictional limits to S Hmax have been determinedfollowing equation 4 and are shown in Figure 6, assumingnormally pressured sediments. The maximum0S Hmax gradient, based on frictional limits, is 18.7 MPa/0 0 0km, implying a strike-slip faulting (S hmin < S v < S Hmax )0 0stress regime. A normal faulting (S hmin < S Hmax < S 0 v )stress regime cannot be ruled out, however, because of0the lack of data constraining the magnitude of S Hmax .Consequently, in our analysis of fault reactivation andseal breach risk, three cases (Table 1) have been considered:a strike-slip faulting (S hmin < S v < S Hmax ) stress0 0 000regime case, a normal faulting (S hmin < S Hmax < S 0 v )stress regime case, and a case on the boundary of strikeslip-normalfaulting stress regimes. The magnitude ofthe in-situ stress field for the three cases was determinedat a depth of 1000 m (3300 ft) below seabed.Pore PressurePore pressure measurements were only conductedin the Jerboa-1 and Greenly-1 wells. Repeat formationtests (RFTs) in the two wells indicate hydrostatic poreTABLE 1. Parameters used in the three cases to model fault reactivation and seal integrity in the0Bight Basin. The cases cover a range of possible values of S Hmax within the frictional limits. Themagnitude values have been calculated for a depth of 1000 m (3300 ft) below seabed.Case0S Hmax(MPa) S v0(MPa)0S hmin(MPa) Fault Regime S Hmax OrientationI 18.7 10.5 6.0 strike slip 1308NII 10.5 10.5 6.0 strike slip-normal 1308NIII 8.5 10.5 6.0 normal 1308N

66 Reynolds et al.pressures to a depth of approximately 3600 m (11,800 ft)(10.2 MPa/km). Below 3600 m (11,800 ft), the RFT measurementsin Greenly-1 indicate a moderate overpressureof 11.8 MPa/km. To obtain a better understanding of thepore-pressure distribution, the mud weights have beenconsidered as an indicator for pore pressure (Figure 7).In general, most of the region is normally pressured, withonly a small indication of overpressure below 3600 m(11,800 ft) in Greenly-1. Hydrostatic pressures are assumedin our analysis of fault reactivation risk.FAULT REACTIVATION ANDSEAL INTEGRITY IN THEBIGHT BASINIn case I (strike-slip stress regime), vertical faultsstriking between approximately 100 and 1608N are themost likely to be reactivated (Figure 8b). Hence, trapsrequiring such faults to be sealing are the most likelyto be breached in the in-situ stress field. Vertical faultsstriking 1308N are located between that conjugate shearpair and are also at high risk of reactivation and breach.Faults striking between 75 and 1808N show little reductionin their risk of reactivation with decreasing dip untilshallow dips (

Analysis of Fault Reactivation and Seal Integrity Based on Geomechanical Modeling 67FIGURE 8. Risk of fault reactivation. (a) Location in stress space of the three in-situ stress cases evaluated (Table 1).(b–d) Left-hand side is Mohr’s circle of stress and failure envelope (assumed) used to calculate the likelihood of reactivationfor each case. Right-hand side is the likelihood of fault and fracture plane reactivation, represented as polesto planes, for the three cases (Table 1). Numerical values on scale refer to increase in fluid pressure required to causereactivation (P). Equal-angle, lower hemisphere stereographic projection of poles to planes.accurate assessment of the fault reactivation risk is notpossible based on available information, some trendscan be established and discussed. Several precautionsare required when assessing the results from this study.First, hydrostatic pressures are assumed in our analysisof fault reactivation risk, and hence, in areas where overpressuresare anticipated, these predictions would needto be modified. Second, given the complex history ofchanges in the stress field that controls the structuralevolution of sedimentary basins, the in-situ stress field, asconstrained herein, cannot be extrapolated back in timeand applied to previous structural events. Knowledge ofthe in-situ stress field can only elucidate contemporarytectonic events.As shown in the previous section, faults that dip at258 (Figures 8, 9) have little risk of reactivation, irrespectiveof the fault orientation and the type of stressfield present. However, as the dips increase to 408, itbecomes apparent that the risk of fault seal failure becomesgreater in the northwest-trending fault arrays inthe Ceduna subbasin. In contrast, the more east- tonortheast-trending faults in the Eyre subbasin are atlow risk of reactivation. At dips of 558, the risk of faultseal failure appears to be low in the Eyre subbasin,where the faults have a generally northeast trend. However,in the more northwest-trending fault arrays, therisk of reactivation is much higher, especially in a strikeslipstress regime. An exception is the small, east–westtrending,intrabasinal Sea Lion faults that occur inthe overall, northwest-trending Sea Lion faults in theCeduna subbasin. These faults have a low risk of reactivationcompared to the northwest-trending faults thatdominate this part of the Bight Basin. At fault dips of708, a high risk of reactivation of the northwesttrendingfault sets of the Ceduna subbasin is presentfor all stress regimes. The exception is again the small,more east–west-trending fault arrays. The Eyre subbasinappears to have a low risk of reactivation, particularly

68 Reynolds et al.

Analysis of Fault Reactivation and Seal Integrity Based on Geomechanical Modeling 69in the strike-slip-normal and strike slip stress regimes. Itshould be noted, however, that stress trajectories calculatedfrom the Australian stress map database indicate amore east–west S Hmax orientation for the Eyre subbasinthan the 1308N S Hmax orientation used for our calculations.Therefore, the results for the Eyre subbasin shouldbe used with care.Overall, the results suggest that in a strike-slipnormalor normal stress regime, little risk of reactivationexists for fault systems that trend east–west ornortheast. For these fault trends, almost no sensitivityto fault dip is present. Clearly, the rift faults of the Eyresubbasin and the intrabasinal, east–west-trending faultsof the Ceduna subbasin all have a low risk of fault reactivation.In contrast, the results suggest that the riftand postrift faults of the Ceduna subbasin and, by inference,the Duntroon subbasin have a relatively highrisk of reactivation once dips exceed approximately 408for either a strike-slip-normal or normal stress regime.Traps with the lowest risk in the Ceduna and Duntroonsubbasins with respect to reactivation are those withlower (]408) dips on the bounding faults or those with amore east–west orientation.If we consider earthquake data, it appears thatthe most common earthquakes occur in the easternCeduna subbasin and in the Duntroon subbasin, broadlythrough the area with northwest-trending fault arrays.However,therestoftheBightappearstobelargelyaseismic, and this may suggest that the stress regime inthe central and western Bight is less conducive to reactivation.Thus, the trends of fault reactivation riskdetermined herein appear broadly consistent with theearthquake data.The distribution of oil slicks that have been mappedin the region (Struckmeyer et al., 2002) via the use ofsatellite-based synthetic aperture radar are generallymore common in the deep-water Ceduna subbasinand eastern parts of the Bight Basin, along the northwestfault arrays. Seismic gas chimneys are very commonin the eastern Ceduna and Duntroon subbasinsand correlate spatially with water column sniffer anomalies.These chimneys typically relate to the northwesttrendingfault arrays in the region, so this may supportsome loss of fault seal integrity in this area.Interestingly, the only confirmed paleo-oil columnin the region was located in the Jerboa-1 well in the Eyresubbasin. This area is predicted, under the present-daystress regime, to be of high fault seal integrity. However,this trap was actually breached in the Late Cretaceous(Ruble et al., 2001). This emphasizes the factthat the stress predictions only relate to the presentday and not to paleoreactivation events, when thestresses may have been quite different. It also emphasizesthe fact that the timing of hydrocarbon migration(and probably the nature of the hydrocarboncharge) is very important in relation to trap reactivation.Relatively low to moderate fault seal integritymay be beneficial in regions that are now experiencinga high gas charge because this may help to reducethe risk of gas flushing in traps in such areas, whichwere previously charged with oil (O’Brien and Woods,1995).These observations highlight the fact that theresults presented here should not be used in isolation.It does appear clear that relatively steeply dippingfaults with a generally northwest trend will be proneto reactivation, whereas more east–west- or northeasttrendingfaults are unlikely to become reactivated, irrespectiveof dip. To better determine how critical thepresent-day stress environment is to hydrocarbon prospectivityin the Bight Basin, a complete fault seal assessmentthat should take into consideration suchobservations should be integrated with other aspectsof the petroleum system, such as the generation history,remote-sensing results, and direct hydrocarbonindicator mapping.For individual traps, the assessment of fault breachingrisk requires detailed prospect studies, so that a geomechanicalanalysis can be applied to clearly defined,depth-converted fault planes interpreted from seismicdata. In addition, in-situ stress field characteristicsshould be based on fieldwide measurements and thefailure envelope constrained by properties specific tothe analyzed rocks.ACKNOWLEDGMENTSSpecial thanks to Jennie Totterdell and Barry Bradshawof Geoscience Australia for permission to usetheir fault interpretation and for providing the digitalfault files, to Peter Boult for valuable comments andsuggestions, and to Primary Industries and Resourcesof South Australia Publishing Services for assistancewith graphic files. Fugro Multi Client Services arethanked for permission to publish the seismic imagein Figure 3. We thank David Castillo, Isabelle Moretti,and Signe Ottesen for their constructive comments regardingthis manuscript.FIGURE 9. Fault reactivation risks calculated using FAST technique applied to Sea Lion and Tiger fault polygons shownin Figure 1, assuming constant dips of 25, 40, 55, and 708, respectively, for (a) case I (strike-slip stress regime) and(b) case III (normal stress regime). Results should be used with care in the western part of the Bight Basin and in areasof overpressure.

70 Reynolds et al.REFERENCES CITEDBlevin, J. E., J. M. Totterdell, G. A. Logan, J. M. Kennard,H. I. M. Struckmeyer, and J. B. Colwell, 2000, Hydrocarbonprospectivity of the Bight Basin— Petroleumsystems analysis in a frontier basin, in 2ndSprigg Symposium— Frontier Basins, Frontier Ideas,Adelaide, June 29–30, 2000: Geological Society ofAustralia, Abstracts no. 60, p. 24–29.Castillo, D. A., R. R. Hillis, K. Asquith, and M. Fisher,1998, State of stress in the Timor Sea area, based ondeep wellbore observations and frictional failure criteria:Application to fault-trap integrity, in P. G.Purcell and R. R. Purcell, eds., The sedimentary basinsof Western Australia: 2. Proceedings of PetroleumExploration Society of Australia Symposium, Perth:p. 326–341.Castillo, D. A., D. J. Bishop, I. Donaldson, D. Kuek, M. deRuig, M. Trupp, and M. W. Shuster, 2000, Trap integrityin the Laminaria high–Nancar trough region,Timor Sea: Prediction of fault seal failure using wellconstrainedstress tensors and fault surfaces interpretedfrom 3D seismic: Australian Petroleum Productionand Exploration Association Journal, v. 40,p. 151–173.de Ruig, M. J., M. Trupp, D. J. Bishop, D. Kuek, and D. A.Castillo, 2000, Fault architecture and the mechanicsof fault reactivation in the Nancar trough/Laminariaarea of the Timor Sea, Northern Australia: AustralianPetroleum Production and Exploration AssociationJournal, v. 40, p. 174–193.Edwards, D., D. M. McKirdy, and R. E. Summons, 1998,Enigmatic asphaltites from the southern Australianmargin: Molecular and carbon isotopic composition:Petroleum Exploration Society of Australia Journal,v. 26, p. 106–129.Finkbeiner, T., M. Zoback, P. Flemings, and B. Stump,2001, Stress, pore pressure, and dynamically constrainedhydrocarbon columns in the South EugeneIsland 330 field, northern Gulf of Mexico: AAPGBulletin, v. 85, p. 1007–1031.Gaarenstroom, L., R. A. J. Tromp, M. C. D. Jong, and A. M.Brandenburg, 1993, Overpressures in the CentralNorth Sea: Implications for trap integrity and drillingsafety, in J. Parker, ed., Petroleum geology of northwestEurope: Proceedings of the 4th Conference, GeologicalSociety (London): p. 1305–1313.Hillis, R. R., 1998, Mechanisms of dynamic seal failure inthe Timor Sea and central North Sea basins, in P. G.Purcell and R. R. Purcell, eds., The sedimentary basinsof Western Australia: 2. Proceedings of the PetroleumExploration Society of Australia Symposium, Perth,Western Australia: p. 313–324.Hillis, R. R., and S. D. Reynolds, 2000, The Australianstress map: Journal of the Geological Society (London),v. 157, p. 915–921.Jaeger, J. C., and N. G. W. Cook, 1979, Fundamentals ofrock mechanics: London, United Kingdom, Chapmanand Hall, 593 p.Jones,R.M.,P.Boult,R.R.Hillis,S.D.Mildren,andJ.Kaldi, 2000, Integrated hydrocarbon seal evaluationin the Penola trough, Otway Basin: Australian PetroleumProduction and Exploration AssociationJournal, v. 40, p. 194–211.Ludwig, W. J., J. E. Nafe, and C. L. Drake, 1970, Seismicrefraction, in A. E. Maxwell, ed., The sea: Ideas andobservations on progress in the study of the seas,New concepts of sea floor evolution: New York, Wiley-Interscience, v. 4, p. 53–84.Mildren, S. D., R. R. Hillis, and J. Kaldi, 2002, Calibratingpredictions of fault seal reactivation in the Timor Sea:Australian Petroleum Production and ExplorationAssociation Journal, v. 42, p. 187–202.Mildren S. D., R. R. Hillis, D. N. Dewhurst, P. J. Lyon, J. J.Meyer, and P. J. Boult, 2005, FAST: A new techniquefor geomechanical assessment of the risk of reactivation-relatedbreach of fault seals, in P. Boult and J.Kaldi, eds., Evaluating fault and cap rock seals: AAPGHedberg Series, no. 2, p. 73–85.O’Brien, G. W., and E. P. Woods, 1995, Hydrocarbonrelateddiagenetic zones (HRDZs) in the Vulcan subbasin,Timor Sea: Recognition and exploration implications:Australian Petroleum and ExplorationAssociation Journal, v. 35, p. 220–252.Reynolds, S. D., and R. R. Hillis, 2000, The in situ stressfield of the Perth Basin, Australia: Geophysical ResearchLetters, v. 27, p. 3421–3424.Reynolds, S., R. Hillis, and E. Paraschivoiu, 2003, In situstress field, fault reactivation and seal integrity in theBight Basin, South Australia: Exploration Geophysics,v. 34, p. 174–181.Ruble, T. E., G. A. Logan, J. E. Blevin, H. I. M. Struckmeyer,K. Liu, M. Ahmed, P. J. Eadington, and R. A.Quezada, 2001, Geochemistry and charge history of apaleo-oil column: Jerboa-1, Eyre sub-basin, GreatAustralian Bight, in K. C. Hill and T. Bernecker, eds.,Eastern Australasian Basins Symposium. A refocusedenergy perspective for the future: Petroleum ExplorationSociety of Australia Special Publication,p. 521–529.Sibson, R. H., 1974, Frictional constraints on thrust,wrench and normal faults: Nature, v. 249, p. 542–544.Sprigg, R. C., and J. B. Wooley, 1963, Coastal bitumen inSouth Australia with special reference to observationsat Geltwood beach, south-eastern South Australia:Transactions of the Royal Society of South Australia,v. 86, p. 67–103.Struckmeyer, H. I. M., J. M. Totterdell, J. E. Blevin, G. A.Logan, C. J. Boreham, I. Deighton, A. A. Krassay, andM. T. Bradshaw, 2001, Character, maturity and distributionof potential Cretaceous oil source rocksin the Ceduna sub-basin, Bight Basin, Great AustralianBight, in K. C. Hill and T. Bernecker, eds., EasternAustralasian Basins Symposium: A refocusedenergy perspective for the future: Petroleum ExplorationSociety of Australia Special Publication, p. 543–552.Struckmeyer, H. I. M., A. K. Williams, R. Cowley, J. M.Totterdell, G. Lawrence, and G. W. O’Brien, 2002,Evaluation of hydrocarbon migration and seepage in

Analysis of Fault Reactivation and Seal Integrity Based on Geomechanical Modeling 71the Great Australian Bight: Australian PetroleumProduction and Exploration Association Journal,v. 42, p. 371–384.Totterdell, J. M., J. E. Blevin, H. I. M. Struckmeyer, B. E.Bradshaw, J. B. Colwell, and J. M. Kennard, 2000, Anew sequence framework for the Great AustralianBight: Starting with a clean slate: Australian PetroleumProduction and Exploration Association Journal,v. 40, p. 95–117.Totterdell, J. M., B. E. Bradshaw, J. B. Willcox, 2003,Structural and tectonic setting, in G. W. O’Brien, E.Paraschivoiu, and J. E. Hibburt, eds., Petroleum geologyof South Australia, v. 5: Great Australian Bight,South Australia, Department of Primary Industriesand Energy Resources: Petroleum Geology of SouthAustralia Series 5, ch. 4: www.pir.sa.gov.au/dhtml/ss/section.php?sectID=1775&tempID=8 (accessed September23, 2003).Watts, N. L., 1987, Theoretical aspects of cap-rock andfault seals for single and two phase hydrocarbon columns:Marine and Petroleum Geology, v. 4, p. 274–307.Wiprut, D., and M. D. Zoback, 2000, Fault reactivation andfluid flow along a previously dormant normal fault inthe northern North Sea: Geology, v. 28, p. 595–598.Zoback, M. D., and J. H. Healy, 1984, Friction, faultingand in situ stress: Annales Geophysicae, v. 2, p. 689–698.