Reservoir characterisation of the Patchawarra Formation ... · the development and sedimentary fill of the basin (Stuart et al, 1988; Apak et al, 1993, 1997). Normal, strike-slip

APPEA Journal 2014—45

T. Stephens, B. Richards and J. LimSenex Energy LtdLevel 14, 144 Edward Street Brisbane, Queensland [email protected]

ABSTRACT

An exploration program to assess the basin-centred gas (BCG) and stratigraphic trap potential of the Mettika Embay-ment in the southern Cooper Basin resulted in the discovery of gas at Hornet–1 and Kingston Rule–1. The embayment is a confined fluvial sedimentary depocentre surrounded by pro-lific gas fields producing from structurally closed anticlines. Gas pay was identified and both wells produced sustained gas flows to surface of between 1.2 and 2.2 MMscf/d after fracture stimulation. Core collected from the Patchawarra Formation sandstone reservoir was analysed to constrain the depositional environment and establish petrophysical properties by routine and special core analysis. An integrated reservoir study was undertaken to understand depositional setting, reservoir architecture, trapping mechanisms, perme-ability, and saturation controls on productivity. Gas identified in the embayment appears to have accumulated in subtle stratigraphic and combination structural traps against the flanks of existing fields and does not display the geological and physical characteristics of a BCG play. The impact and analysis of hydrocarbon migration and reservoir trapping influences in this basin-margin gas accumulation may be applicable to other under-explored flank and trough plays of the Cooper Basin.

KEYWORDS

Mettika Embayment, Cooper Basin, basin-centred gas, Hor-net discovery, Patchawarra Formation, stratigraphic trap, core analysis, relative permeability, capillary pressure, log analysis, reservoir characterisation, appraisal concepts.

INTRODUCTION

Recent exploration and appraisal by several operators has identified a potentially significant unconventional and basin-centred gas (BCG) resource in the Nappamerri Trough, Cooper Basin (Hillis et al, 2001; Trembath et al, 2012). An exploration and appraisal drilling program was initiated by Senex Energy Ltd to assess the potential for unconventional, BCG and off-structure conventional stratigraphic traps in the Mettika Embay-ment, southern Cooper Basin (Fig. 1). The two discovery wells, Hornet–1 and Kingston Rule–1, are sited in the embayment on the eastern flank of the Toolachee Field High and flowed gas after fracture stimulation from tight Patchawarra Formation sandstone. Hornet–1 was originally drilled by Victoria Petro-leum to test a Jurassic oil play in the shallow Eromanga Basin. After the well failed to intersect hydrocarbon saturation in the primary oil target, the well was deepened to evaluate the gas

potential of the Patchawarra Formation. Strong gas shows while drilling and a drill stem test (DST) that recovered gas to surface at 50 Mscf/d led the well to be cased and suspended for future evaluation. The Kingston Rule–1 and Talaq–1 wells were drilled by Senex Energy to evaluate the unconventional gas potential of the Roseneath-Epsilon-Murteree (REM) sequence and the nature of the off-structure Patchawarra Formation gas identified in Hornet–1. Kingston Rule–1 intersected several gas-charged sandstone packages outside of structural closure while Talaq–1 observed minor gas shows on drilling. Hornet–1 and Kingston Rule–1 recorded post-stimulation flow rates of 2.2 MMscf/d and 1.2 MMscf/d, respectively (Scott et al, 2013). Talaq–1 is located in the structurally deepest section of the embayment and pet-rophysical log analysis suggests that water saturation is high throughout most of the intersected Patchawarra Formation.

The Kingston Rule–1 and Talaq–1 wells were cored through the REM and Patchawarra formations as part of an uncon-ventional reservoir characterisation effort. An extensive tight sandstone core analysis program was conducted to evaluate petrophysical properties including porosity, permeability, cap-illary pressure, relative permeability and electrical parameters at in situ (laboratory) conditions. The recovered core improved understanding of stratigraphic complexity, reservoir and re-source potential, and helped to establish a regional deposi-tional model and the nature of likely trapping and play styles present in the embayment. It is proposed that the Patchawarra Formation in the Mettika Embayment was deposited as a north to south axial fluvial system influenced by basin subsidence and syn-depositional faulting. While gas appears to be pervasive in the three off-structure appraisal wells, saturations were not suf-ficient to justify completion of many sandstone packages. It is likely the Patchawarra Formation in the Mettika Embayment is not a true BCG system as was initially envisaged. The sandstone reservoirs are normally pressured and exhibit petrophysical properties characteristic of a conventional system dominated by stratigraphic, structural and combination hydrocarbon traps. In this study, the impact of reservoir architecture, rock physics and concepts of conventional and BCG trapping mechanisms is discussed to define the potential lower and upper limits of gas productivity along the western flank of the Mettika Embayment.

REGIONAL GEOLOGY OF THE COOPER BASIN

The Cooper Basin is Australia’s most significant onshore oil and gas province. It spans a present-day areal extent of 130,000 km2 striking northeast from northeast SA into south-west Queensland. The basin represents an intra-cratonic Car-boniferous-Triassic siliciclastic depocentre that unconformably overlies the Cambrian-Devonian sediments of the Warburton Basin and Carboniferous granitoids of the Big Lake Suite. The Cooper Basin is again unconformably overlain by Jurassic-Cretaceous sediments of the Eromanga Basin (Gravestock et al, 1998). Regional unconformities mark significant compres-sional events within the Cooper Basin (Apak et al, 1993, 1997), however background subsidence has been hypothesised to be driven by exponential thermal decay during the waning heat

Reservoir characterisation of the Patchawarra Formation within a deep, basin-margin gas accumulation

Lead authorTimStephens

46—APPEA Journal 2014

T. Stephens, B. Richards and J. Lim

flow period post-Carboniferous granitiod emplacement (Grave-stock et al, 1998). The interplay of compressional tectonics and subsidence resulted in fault reactivation playing a large part in the development and sedimentary fill of the basin (Stuart et al, 1988; Apak et al, 1993, 1997). Normal, strike-slip and reverse faulting have been invoked to explain the deformational fabric of the Cooper Basin. The differing views on the structural style

of the Cooper Basin reveal an anisotropic stress field and that no single mechanism of deformation has predominated (Stuart et al, 1988; Apak et al, 1997; Gravestock et al, 1998).

Early deposition within the Cooper Basin was strongly in-fluenced by glacial geomorphology. Glacial retreat initiated in the Carboniferous is represented by the glaciofluvial deposits of the Late Carboniferous-Early Permian Merrimelia Formation and Tirrawarra Sandstone (Williams and Wild, 1984). Waning glacial influence in the late Early-Permian gave rise to a flu-vial, flood plain to peat swamp, lacustrine, and deltaic domi-nated depositional system spanning 25 Ma during which the Patchawarra Formation accumulated (Lang et al, 2002; Apak et al, 1997). The Patchawarra is conformably overlain by the la-custrine-deltaic transgressive and regressive cycles of the REM and Daralingie formations. The Early Permian-Late Permian boundary is marked by the regional Daralingie Unconformity generated during compressional uplift. This resulted in erosion of the reactivated basement highs and structural flanks (Apak et al, 1993, 1997). The Toolachee Formation marks the top of the Gidgealpa Group and was deposited on a surface of low relief above the Daralingie Unconformity and consists of fluvial, lacustrine and peat swamp deposits (Gravestock et al, 1998). The remainder of the Late Permian-Triassic is interpreted to be tectonically quiescent with a slight down to the northeast structural tilting as a flexural response to Australian Eastern Margin loading (Gravestock et al, 1998). Triassic deposition is represented by the Nappamerri Group, which conformably overlies the Toolachee Formation and consists of a fluvial-la-custrine sand and shale sequence. The top of the Nappamerri Group is marked by a regional unconformity related to regional compression and Australia Eastern Margin tectonics (Apak et al, 1997). The unconformity marks the end of the Cooper Basin sedimentation, and the commencement of deposition in the Eromanga Basin.

Geology of the Mettika Embayment

The Mettika Embayment is located along the southeastern flank of the Cooper Basin (Fig. 1) and is a north–south striking depression, which continues along a regionally dipping base-ment trend towards the Nappamerri Trough. Some consider the embayment to be an extension of the Tenappera Trough (Gravestock et al, 1998), however, this study indicates the Mettika Embayment was a separate depocentre separated by the Toolachee Field High during the Early Permian. The sedi-mentary sequence in the Mettika Embayment consists of Early Permian to Triassic Cooper Basin strata unconformably depos-ited on metasediments of the Cambrian-Devonian Warburton Basin and unconformably overlain by the Eromanga Basin sedimentary sequence (Fig. 2). Williams and Wild (1984) noted ambiguity surrounding the lithostratigraphic identification of basement lithologies owing to spatial heterogeneity in the ba-sin floor. Sidewall and conventional core recovered from Pre-Permian basement within the Mettika Embayment are barren of the spore pollens used in the palynological subdivision of the Copper Basin (Price et al, 1985). The basement was conven-tionally cored in Toolachee East–2 on the western flank of the Mettika Embayment and consists of sucrosic, steeply dipping, metamorphosed siltstone that supports the interpretation of the basin floor lithology within the embayment.

The absence of the Late Carboniferous to Early Permian pro-glacial deposits of the Merrimelia Formation and Tirrawarra Sandstone has been noted in the existing wells within the em-bayment. This may suggest either non-deposition or deposition with subsequent erosion. The absence of the Merrimelia For-mation and Tirrawarra Sandstone resulted in the Patchawarra Formation deposited unconformably on basement.

Figure 1. Cooper Basin and Mettika Embayment location map with gas fields and Senex Energy land position.



Syn-depositional faulting and its effect on the accumulation of the Patchawarra Formation have been extensively document-ed throughout the Cooper Basin (Apak et al, 1997; Lang et al, 2001; Strong et al, 2001). The vertical stacking patterns of the Patchawarra Sandstones observed within the Mettika Embay-ment are strongly controlled by the ratio of sediment supply to changes in accommodation, which was primarily driven by tectonic alteration of base level. Regional scale basement structuring reflects a broad north to northeast tilting across the Mettika Embayment with further enhancement during Tertiary subsidence. A west–east traverse across the embayment high-lights the slightly asymmetric profile of the depocentre, with the main trough located towards the western flank (Figs 3 and 4). The southeastern margin is a shelf-like, regional basement high with subordinate basement four-way closures (Fig. 3). Steep, near vertical north to northeast striking normal faults occur on the western and eastern flanks of the embayment formed synchronous with Patchawarra Formation deposition (Fig. 4). Post Patchawarra deformational events are predominantly re-verse reactivation of existing features. Figure 4 shows a steeply dipping fault with reverse offset flanking the eastern side of the Toolachee Field. The lineament exhibits decreasing displace-ment to the south implying a component of strike-slip move-ment. The absence of significant thickening of the Patchawarra Formation in the footwall block of this fault combined with offset at the Cadna-owie horizon suggests this is a late-stage structural reactivation of a pre-existing basement weakness. The steeply dipping nature of the faults within the Mettika Em-bayment makes them difficult to image on seismic.

The Toolachee Field is located on the largest structure flank-ing the western side of the Mettika Embayment (Fig. 3). The Toolachee Field High reflects exposed basement that influ-enced sedimentation until accommodation space was filled within the embayment. The result is a regionally extensive up-per Patchawarra and Murteree Shale deposited over structural highs. The Patchawarra Formation thickens off-structure simi-lar to that observed within the Nappamerri Trough, flanking

the Gidgealpa-Merrimelia-Innamincka (GMI) Ridge (Apak et al, 1997). Thickening in Permian sediments off structure reflects the interplay of tectonic driven subsidence with the prevailing climatic conditions (Lang et al, 2002; Strong et al, 2002; Hamlin et al, 1996; and Taylor et al, 1991). Rates of subsidence within local depocentres of the Cooper Basin have been recognised as being non-uniform leading to localised variations in the fluvial geometry of the Patchawarra Formation (Apak et al, 1997; Stuart et al, 1988).

Intra-Patchawarra unconformities present on the crests and flanks of the GMI Ridge were generated during periods of negative accommodation. The erosional pulses were localised to the ridge system while the Nappamerri Trough was under-going near-continuous deposition. These observations along with palynological and seismic evidence led Apak et al (1993, 1997) to the interpretation that fault related movement was predominantly reverse, with accommodation present in the footwall blocks of this fault system. Conversely, off structure thickening of the Patchawarra Formation within the embay-ment is interpreted to be a combination of synchronous normal fault driven subsidence and paleo-topographic infilling (Figs 4 and 5). These conclusions are based on thickening in the hang-ing wall of normal faults, seismic reflectors onlapping base-ment, and the absence of significant hiatal breaks in the paly-nology. The onlapping geometry of the Patchawarra Formation on the Toolachee Field High was also noted by Apak et al (1997).

Devine and Gatehouse (1977), in a review of the Toolachee Field, described localised thick sandstone development within the Patchawarra Formation overlying basement in the Toolachee East–1 appraisal well. The sandstone was intersect-ed within the hanging wall of a normal fault and interpreted to be indicative of syn-depositional tectonic influence. The Toolachee Field High is composed of a series of fault related sub-culminations with fluvial erosion of exposed basement enhancing these features (Devine and Gatehouse, 1977). The presence of sub-aerially exposed basement during Patchawarra Formation deposition has also been noted on paleohighs else-where within the Copper Basin (Apak et al, 1997; Lang et al, 2002).

The Early to Late Permian boundary within the Mettika Embayment is marked by the regionally extensive Daralingie Unconformity, which is accepted to have been generated in response to tectonic uplift (Apak et al, 1993, 1997). The absence of the Daralingie Formation and thinned Roseneath Shale on the Toolachee Field High indicates positive relief was present on the flanks of the embayment resulting in erosion (Figure 6). Late Permian deposition of the Toolachee Formation and Post-Permian deposition of the Nappamerri Group within the em-bayment is isochronally consistent, suggesting a reduction in the influence of basement topography on deposition. The ces-sation of the Nappamerri Group deposition marks the transi-tion from sedimentation within the Mettika Embayment to the deposition of the Eromanga Basin.

Figure 2. Idealised stratigraphic section of the Cooper basin (after Apak et al, 1993).

Continued next page.



Figure 4. Toolachee 3D west to east seismic traverse of the western Mettika Embayment.

Figure 3. Regional Patchawarra Formation basement depth structure and isochore maps with developed gas fields delineated in red. Basement topography and isochores suggest the Patchawarra Formation was deposited in paleolows as an infilling sequence. Lines of section are referenced in Figures 4, 6, 8 and 9.

Continued from previous page.



Figure 5. Model of the stratigraphic evolution of the Mettika Embayment. The north–south fluvial axis present during deposition of the Patchawarra Formation is structurally controlled by existing paleotopography and local tectonics. Stage one infill reflects initial sediment bypass and subsequent deposition of the Patchawarra on basement. Stage two infill is dominated by increased accommodation. Stage three shows infilling of the paleotopography and regional deposition of the up-permost Patchawarra units.

Continued next page.

Figure 6. West to east cross-section showing the absence of the Daralingie Formation over the Toolachee Field High and the preserved remnants within the central Mettika Embayment. The unconformity marks the end of the Early-Permian deposition of the Gidgealpa Group.



DEPOSITIONAL ENVIRONMENTS OF THE PATCHAWARRA FORMATION

The vertical succession cored and recovered from the Patcha-warra Formation in both the Kingston Rule–1 (113 mTVT) and Talaq–1 (98 mTVT) wells provided good control on the fluvial elements comprising the Patchawarra Formation within the Mettika Embayment. The main depositional elements recog-nised and discussed in this study include channels, crevasse splays, and floodplain deposits (Figs 7–9).

Channel facies (CF)

The channel facies consists of a basal channel lag deposit comprising of a scoured contact with the underlying lithology and a conglomerate lag. The conglomerate lag consists pre-dominantly of coarse to pebble sized, rounded to sub-elongate quartz and metasedimentary clasts. The channel fill contains predominantly medium to very coarse grained, poorly sorted, rounded-subangular, quartzarenite-sublitharentie sandstones. Observed bed forms include planar, tabular, and trough cross bedding representative of in-channel bars and dunes (Fig. 7). Individual channel cycles range from 3–5 m in thickness but commonly occur as stacked sequences between 10–15 m in thickness. An absence of well-developed point bar deposits over cored intervals, rapid vertical grainsize transition into heterolithic fine-grained floodplain sediments, and the pre-dominantly sandy bed load suggests that channel sinuosity was relatively low. The channel facies are observed to be in vertical and lateral contact with both the crevasse splay facies and the flood plain facies. The pore system consists of primary inter-granular porosity, a poorly interconnected secondary porosity system formed by the leaching of framework grains, and a microporosity system associated with authigenic, pore filling kaolinite booklets. Destruction of primary porosity oc-curs mainly through mechanical compaction and ductile de-formation of mica rich framework grains and significant quartz overgrowths.

Crevasse splay facies (CSF)

The crevasse splay facies consists of very fine–medium grained, rounded, well sorted, texturally and composition-ally mature, quartzarenite sandstones. Observed bed forms include massive, ripple and climbing ripple cross-laminations and planar laminations (Fig. 7). The observed bed forms are consistent with rapid deposition of sediment from suspen-sion during decelerating flow on an unconfined flood plain. The sandstones commonly have sharp basal contacts and contain rip up clasts consisting of organically rich flood plain siltstones and mudstones. The crevasse splay facies are in ver-tical and lateral contact with the floodplain sediments and channel facies. Individual crevasse splay sandstones range from 0.1–3 m in thickness and can stack to between 5–10 m. The pore system consists of primary inter-granular porosity, minor secondary leached grain porosity and microporosity associated with authigenic, pore-filling kaolinite. Pyrite and siderite are also an abundant pore-filling component in these facies. Destruction of primary porosity is mainly by quartz overgrowths, which are likely enhanced by smaller pore throats and increased grain to grain contact area. Mechanical compaction and deformation of ductile grains also contributes to a reduction in porosity.

Floodplain facies (FPF)

The floodplain facies is dominated by heterolithic interbeds of siltstone, carbonaceous siltstone, mudstone and sandstone ranging from 1–50 cm. Siltstone and mudstone represent the primary lithologies comprising the facies and are interpreted to have been deposited during ephemeral sheet floods and from suspension settling in flood plain lakes. Coarser lithologies are interpreted to have been sourced from crevasse splays and irregular flood events. The observed bed forms include both planar and ripple laminations as well as convoluted bedding (Fig. 7). Secondary structures include faulting and bioturbation. Bioturbation occurs primarily in the organic rich mudstones and siltstones and is dominated by sand filled, horizontal bur-rows and escape structures. The presence of convoluted bed-ding and faulting within very fine grained sandstones indicates instability on the flood plain and is likely to have a tectonic origin. Coals have been included in this facies as they com-monly overlay the carbonaceous siltstones and mudstones of the flood plain and contain carbonaceous siltstone bandings, which suggest periodic clastic influx. The coals may have root-lets developed within their basal strata (Fig. 7) and range in thickness from 0.2–7 m. Maceral analysis of Patchawarra coals from the Talaq–1 well suggests that the coals consist primarily of Vitrinite with a significant portion of Inertinite. The high In-ertinite composition of the Patchawarra coals may have a num-ber of potential origins including partial combustion in forest fires, oxidation occurring either in situ or during transport, bio-chemical alteration, or may be primary in origin (Ward, 1984).

DEPOSITIONAL ARCHITECTURE OF THE PATCHAWARRA FORMATION

Sequence stratigraphic concepts have been previously ap-plied to the Patchawarra Formation to predict reservoir oc-currence and connectivity, generate prospects, appraise fields and assess geological risk in exploration drilling (Strong et al, 2002; Nakanishi and Lang, 2001, 2002; Lang et al, 2002). Lang et al (2001) presented a review on the application of sequence stratigraphy to Eastern Australian non-marine basins. The re-view demonstrates, with context and caveats, the successful application of sequence stratigraphy to the interpretation of continental successions.

Fluvial stacking pattern recognition is critical to the appli-cation of sequence stratigraphy in continental sequences. The stratal geometry of fluvial sequences reflects perturbations in the fluvial equilibrium profile under the prevailing climate, which produces changes in the rate of accommodation gen-eration versus sediment supply (Lang et al, 2001). Regional re-view of the Mettika Embayment demonstrates localised normal faulting, regional tilting and outer-basinal reverse fault move-ment to be occurring synchronously with Patchawarra Forma-tion deposition. The resulting proximal and distal influence of these factors on relative changes in base level and sediment supply is reflected in fluvial stacking patterns of the Patcha-warra Formation. The stratal geometries expected within each of the systems tracts is outlined for context below.

Alluvial transgressive systems tract (TST)

During rising base level, accommodation creation out-com-petes sediment supply and allows fluvial reservoirs to stack with increasing isolation up-section. The generation of accommoda-tion generally results in thick peat development on the flood plain.

Continued from previous page.



Figure 7. Red annotations represent key bedding features. Kingston Rule–1 core, 10 cm diameter (4”); Talaq–1 core, 8 cm diameter (3¼”). Petrographic annotation: IG is inter-granular pore space, LG is leached grain, RF is rock fragment, QZ is quartz grain, PY is pyrite, SO is siderite, QO is quartz overgrowth, KA is kaolinite. All depths refer to driller’s depth.



Alluvial highstand systems tract (HST)

The maximum rate of accommodation creation is marked by a lacustrine maximum flooding surface in the distal regions and thin peat or floodplain lakes in the proximal areas. These peat derived coals form regional correlation markers (Hamilton and Trados, 1994). As the rate of accommodation creation declines relative to the sediment supply, lakes are infilled and fluvial chan-nel deposits can become more amalgamated. Lang et al (2001) document that in parts of the Cooper Basin, this systems tract is dominated by the development of widespread flood plains and splay complexes with tortuous main channel pathways.

Alluvial lowstand systems tract (LST)

This was initiated during basin wide tectonic uplift or tilt-ing. Falling base level leads to widespread erosion on the basin margins and over basinal highs, generating an unconformity equivalent to a sequence boundary. If base level fall is pro-ceeded by low accommodation, fluvial systems rework the flood plain, resulting in a high net to gross sheet-like amalgamation of predominantly sandy bed load deposits.

Sequences identified within the Mettika Embayment are reflective of a relatively proximal-to-source fluvial system. The

vertical stacking patterns of the depositional elements iden-tified within the embayment (Fig. 7) and their interpretation within the sequence framework outlined by Lang et al (2001) are presented in Figures 8 and 9, and are discussed subse-quently.

Transgressive systems tracts

Isolated channel systems and the preservation of thick coal seams, flood plain deposits, and crevasse splay complexes in Talaq–1, Kingston Rule–1 and Hornet–1 reflect relatively stable conditions in which accommodation generation was close to the rate of sediment supply, resulting in vertical aggradation. TSTs within the Mettika Embayment commonly result in increasing sandstone isolation up-section, consistent with a rising base level. Individual channel fill cycles are commonly on the order of 10 m in thickness as opposed to the 5 m cycles developed within the LST system. Sandstone reservoirs deposited during the TST are likely to be orientated along structurally controlled depositional axes and potentially of limited lateral extent.

Highstand systems tracts

Marked by a flooding surface (FS), the HST consists pre-dominantly of siltstones and mudstones deposited within the flood plain environment. HST deposits reflect the maximum accommodation generation relative to sediment supply, which ultimately leads to poor reservoir development and vertical stratigraphic compartmentalisation. The absence of an under-lying TST is interpreted to indicate rapid positive increases in accommodation generated by localised tectonic modification of base level. Spectral gamma ray logs reflect an increase in thori-um and uranium associated with increased clay sized sediment deposition and distinct peaks mark potential flooding surfaces. Coals developed within the HST are interpreted to represent near-chronostratigraphic equivalent horizons (Hamilton and Trados, 1994). The interpreted HST deposits were not cored in the Kingston Rule–1 and Talaq–1 wells.

Lowstand systems tracts

The LST is dominated by fluvial channels and crevasse splay complexes stacked in vertical succession when sediment supply increases relative to accommodation creation. Dropping base level results in the vertical juxtaposition of amalgamated, proxi-mal-to-channel sandstones with distal, fine-grained flood plain facies. Subordinate development of coals and thin flood plain sediments throughout the LST deposits suggest the presence of accommodation outside the channel axes and that falling base level was not followed by a period of low accommodation and flood plain reworking. The absence of sheet sands char-acteristic of LST deposits, combined with an uncharacteristic sandstone amalgamation, is interpreted to reflect an increase in sandy sediment supply into a dynamically subsiding depocen-tre. The lower rate of accommodation generation combined with the high sediment load produces regionally significant reservoir systems.

Summary

The application of sequence stratigraphic principles to the Patchawarra Formation within the Mettika Embayment has as-sisted in defining a correlation framework that captures both al-locyclic and autocyclic processes within a third to fourth order sequence scale. At a second-order sequence scale, the Patcha-warra Formation represents part of a basinal transgressive fill. Initially confined by paleotopography, sediment carried within

Figure 8. Mettika Embayment, west to east traverse highlights the onlap geometry from the axis of the embayment at Hornet–1 onto the Toolachee Field High tested by Witchetty–1.



the fluvial drainage network bypasses the Mettika Embayment and is deposited in the distal portions of the catchment. As the base level rises (accompanying continued deposition), the drain-age network becomes retrogradational, infilling the previously bypassed, topographically higher portions of the Permain Basin (Fig. 5). This fill cycle is in agreement with the progressive loss of stratigraphically older Patchawarra Formation sediments to-wards the basin edge, the vertical stacking patterns within the Patchawarra Formation, and the onlapping geometries seen within the Mettika Embayment. It is interpreted that both tec-tonics affecting the distal portions of the fluvial drainage network and localised tectonics have influenced the internal stacking ge-ometry of the Patchawarra Formation within the embayment.

Legacy lithostratigraphic correlations of the embayment overestimate the continuity of sandstone bodies. The complexi-ty of the Patchawarra Formation captured using sequence strati-graphic principles suggests potentially good connectivity in a north–south orientation for sandstones deposited within the TST and LST. Conversely, east–west connectivity is considered unlikely for sands within the TST and is potentially tortuous for sands within the LST owing to continued accommodation generation during periods of increased sandy sediment load. The results from this ongoing analysis have yielded a number

of potential stratigraphic trapping geometries including strati-graphically isolated sandstones, subtle onlap or pinch outs in the north–south orientation, and onlap plays in an east–west orientation. At the time of publication, multiples present within the existing seismic data impeded further stratigraphic inter-pretation and identification of potential appraisal targets.

REGIONAL PRESSURE DATA

BCG plays are broadly characterised by regionally per-vasive gas saturation below identified conventional water contacts, and abnormal pressure regimes. Pressures can be significantly higher or lower than the regional aquifer trend and may either imply a single large continuous gas column or compartmentalisation into multiple isolated reservoirs (Fig. 10). Examples are well documented in the literature (Yurewicz et al, 2008; Burnie et al, 2008; Shanley et al, 2004; Spencer, 1989; Masters, 1984; Law, 2001). Pressure and pro-duction data from recent exploration and appraisal wells in the Nappamerri Trough, Cooper Basin have identified a BCG accumulation (Hillis et al, 2001; Trembath et al, 2012). The system is much higher in pressure relative to the regional Eromanga hydrostatic gradient and gas saturation is found

Figure 9. Mettika Embayment, south to north traverse. The absence of the lowermost LST and TST at Azolla–1 provides evidence for the existence of a north–south orientated pinch out or onlap play.



within many deep tight-sandstone intervals. Pressure data ob-tained from wells surrounding the Mettika Embayment were analysed to establish pressure-depth trends and assess the potential for an extant BCG play. A regional hydrostatic trend was established from Permian discovery well DST pressures producing water and wireline pressure data from shallow Ero-manga Basin sandstones (Fig. 11). DST data was interpreted using standard pressure transient analysis techniques and extrapolated to obtain an infinite acting pressure (P*) while FMT data with mobility less than 10 mD.ft/cP was excluded due to possible super-charging effects. Care was taken to avoid pressure depletion by only interpreting wells drilled before substantial production had commenced.

The Patchawarra Formation aquifer in the southern Cooper Basin is likely recharged from the Eromanga Basin sub-crop edge to the south and west of the Mettika Embayment (Dunlop et al, 1992). Water chemistry mapping indicates fresh water ingress into the Tenappera Trough west of the Toolachee Field (not shown). Salinity increases substantially east and north

of Toolachee Field and the Mettika Embayment appears iso-lated from active hydrodynamic flow. Regional Patchawarra Formation pressure data west of the Toolachee Field implies a potentiometric head and recharge datum similar to the re-gional Eromanga Basin, at about 55 m above sea level. The equivalent recharge datum appears to be slightly lower in the selected Mettika Embayment data but may be accounted for by pressure gauge variability and inappropriate pressure extrapolation. Two lines are fitted to the data in Figure 11 to capture the upper and lower range of hydrostatic trends. As-suming the regional hydrostatic line is the reference pressure and the small sampling of wells are representative, then it would appear that the Patchawarra Formation aquifer is not abnormally pressured. Potential hydrocarbon column heights were examined by plotting pressures obtained from embay-ment gas discovery well DSTs and referenced to the regional and local hydrostatic trends (Fig. 12). The wells plotted in this figure do not show depletion as large-scale gas production from neighbouring fields was yet to commence. Accounting for variability and analysis inaccuracy, a selection of pressure points fall somewhat below the regional trends. Extrapolating a gas gradient to a down-dip water leg to estimate column height is inappropriate as the connected aquifer pressure has not been recorded. The slight under pressure may have devel-oped by a process similar to that described in the literature and will be discussed subsequently (Masters, 1984; Burnie et al, 2008; Law, 2001). Individual sands appear to be separate accumulations, which is to be expected from lenticular flu-vial sandstone bodies with variable hydraulic connectivity and charge history. Sandstones developed within the LST are likely to have better connectivity than the TST, implying lower strati-graphic trapping potential. While it appears the geological criteria to accumulate a significant BCG resource is present, it is not supported by the presently available pressure data.

METTIKA EMBAYMENT PRODUCTION ANALOGUES

Very few wells surrounding the Mettika Embayment can be considered production analogues to Hornet–1 and Kingston Rule–1. Several off-structure wells drilled by the neighbouring operator targeted the Patchawarra Formation and discovered gas in subtle anticline and combination traps with varying degrees of success. Historic production data suggests moderate to good recovery and high initial rates without stimulation (Fig. 13). Many

Figure 10. Extant pressure regimes observed in BCG type accumulations.

Figure 11. Pressure versus depth data for discovery wells drilled surrounding the Mettika Embayment prior to significant production. All plotted DST pressures were obtained from tests returning a full string of water with minimal to no gas produced. Dashed and solid lines are the equivalent gradient for water at 0.433 psi/ft and are fitted through trends.

Figure 12. Pressure versus depth data for discovery wells drilled surrounding the Mettika Embayment that flowed significant volumes of gas to surface. Pressures lie both above and below the established regional and embayment aquifer trends. Approximate gas trend lines are fitted for reference through each point.



nearby fields discovered on four-way closed anticline structures appear to be impacted by early and increasing water production. Analysis of the few publically available production logs suggests gas is mostly produced from one or two perforations while the remainder do not flow efficiently or produce water. These pro-duction logs were generally performed to identify the source of increasing water and liquid loading late in the well life and are not useful for diagnosing early water breakthrough and petro-physical log calibration. FMT and DST pressure data collected from off-structure gas discoveries suggests many of the gas ac-cumulations are relatively modest in size (Fig. 14). DST pressure from Allambi–1 points to a significant gas accumulation in the upper Patchawarra, while the lower Patchawarra lies below hy-drostatic. It is possible that pressure depletion from the nearby Kidman Field extends to Allambi–1 and Hornet–1 accumulations in some sandstone sequences. Some surrounding wells drilled on structural highs in the Toolachee, Kidman, Kerna and Brumby fields also display high early water cut (not shown). It is possible that many sandstone packages have distinct saturation and pres-sure systems and a common free water level (FWL) is not shared among the stacked sequences.

CORE AND PETROPHYSICAL ANALYSIS

An extensive tight sandstone core analysis program was per-formed on core recovered from Kingston Rule–1 and Talaq–1. The primary objective was to acquire reliable grain density, slippage corrected permeability at overburden stress (Kinf ) and porosity, and to identify sandstone qualities thought to be characteristic of BCG such as:• low and variable permeability and porosity trends;• abnormally high capillary entry pressure; • high irreducible water saturation; and,• evidence of significant diagenesis, and unfavourable relative

permeability curves.Both a biased and traditional regularly spaced sampling

approach was used to select plugs (Worthington, 1991). Up to four plugs were obtained in each sandstone sequence with one sampled at the top and base of the section. Fractures and lami-nations were avoided to minimise permeability heterogeneity effects. By broadening the plug selection process to capture as many ranges of sandstone types and qualities as possible, it was envisaged that a suitable porosity and permeability rela-tionship could be developed for simple modelling purposes. In parallel, a single sequence was sampled at close regular spacing for log interpretation calibration. Where possible, the same set of routine plugs was used in subsequent special core analysis.

It is well established that interpretation of permeability in tight sandstone requires stress and gas slippage effects to be accounted for during testing (e.g. Spencer, 1989; Thomas and Ward, 1972; Byrnes, 1979). Much of the publically avail-able Patchawarra Formation core data from local analogue wells was deemed unrepresentative as neither gas slippage nor overburden stress effects were included in plug testing. A small subset of plug porosity and permeability data is re-ported to be corrected for overburden stress and is plotted in Figure 15. A wide range of porosities were captured with the plug sampling regime and overall, the porosity-permeability trend from both Talaq–1 and Kingston Rule–1 appears favour-able when compared with selected analogue data (Fig. 15). To simplify the analysis in this study, three exponential trends capturing the median, 90th and 10th percentile were adopted for permeability prediction.

Relative permeability

Characteristics of two-phase relative permeability can be useful to define possible trapping and productivity in tight gas plays. Several previous studies have described the permeability

0.0001

0.001

0.01

0.1

1

10

100

1000

0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20%

Perm

ea

bil

ity (

mD

)

Porosity

Analogue Core Data

Kingston Rule and Talaq Core

Figure 15. Klinkenburg corrected air permeability and helium porosity data from the Patchawarra Formation at overburden conditions. Analogue data are sourced from overburden and slippage corrected Patchawarra plugs throughout the Cooper Basin.

Figure 13. Gas rate and normalised time for five off-structure or deep analogue wells in the Mettika Embayment—Allambi–1, Marsilea–1, Amyema–1, Kerna–2A and Plantago–1.

Figure 14. FMT data from Allambi–1, Marsilea–1, Amyema–1, and Hornet–1. Data suggests potentially moderate to small gas accumulations. Identification of the aquifer trend, however, is problematic as data is unreliable. Approximate gas trend lines are fitted for reference through each point.



jail concept, whereby a severe reduction in relative permeabil-ity occurs at certain saturation ranges in tight gas sandstones (Masters, 1984; Shanley et al, 2004; Blasingame, 2008; Cluff and Byrnes, 2010). Secondary gas migration can be reduced dra-matically when the absolute permeability to both phases is low, and in such cases the reservoir itself is considered the trap. If combined with large and ongoing local generation, substantial volumes of gas may accumulate and be producible at economic rates after stimulation.

Steady-state gas and water drainage and imbibition relative permeability curves were obtained from four Kingston Rule–1 plugs at overburden stressed conditions. A wide range of plug permeabilities was selected to capture curves from various rock qualities. Despite more than four orders of magnitude of differ-ence between sample permeability to air (Kinf), relative perme-ability curves show mobile gas and water phases at a wide range of water saturations (Table 1 and Fig. 16). If this observation is regionally consistent, it is possible that capillary entry pressure, subtle structuring, and stratigraphy control the accumulation and distribution of hydrocarbon saturation. This will be dis-cussed in detail in a subsequent section.

Imbibition relative permeability data was collected on two samples to demonstrate the effect of introducing water into the system, likely to occur during well workovers and fracture stimulation (Fig. 17). The curves show a severe reduction in gas relative permeability when wetting phase saturation is in-

creased from irreducible. At the crossover point, the absolute permeability to water and gas has reduced and the curves begin to resemble jail-like behaviour. Quantifying the effect of water injection during fracture stimulation is difficult given the lack of representative in situ data that is not confounded by other po-tential damage mechanisms. The most important characteristic is clean-up efficiency, which will follow a secondary drainage curve and it is possible that some degree of hysteresis occurs and full permeability to gas is never regained. Further work is required to assess imbibition damage mechanisms in the Patchawarra Formation.

Petrophysical log interpretation

A deterministic petrophysical analysis was conducted to support reservoir characterisation in conjunction with the core analysis and production test analysis results. Poor-hole conditions were commonly observed in the reservoir section and thus, bulk density, PE (photoelectric factor), and micro-resistivity logs were deemed inappropriate for log analysis. Po-rosity was calibrated from NMR (nuclear magnetic resonance) logs and core plug porosity measurement. Clay volume (Qv) was determined from gamma ray logs, which were not affected by hole conditions. With the exception of pad type tools, log qualities were generally reliable for interpretation purposes. Cementation and saturation exponent data were acquired from Kingston Rule–1 and Talaq–1 preserved core plugs.

Petrophysical analysis was performed using the Interactive Petrophysics (IP) software package (Senergy). Clay volume (V

cl)

determination is integral to the interpretation workflow, how-ever, due to the poor quality of the bulk density log data, V

cl was

determined from the linear GR method (Eq. 1).

Vcl =

(GR − GR0)

GR100

− GR0

(1)

In Equation 1:• GR is the recorded gamma ray log reading;• GR

0 is the gamma ray log reading in 100% clean zone;

• GR100

is the gamma ray log reading 100% clay; and,• V

cl is the clay volume from gamma ray log (fractional).

A variable matrix sonic model was used to determine poros-

Sample 1 2 3 4

Helium porosity 6.7% 10.2% 13.1% 5.7%

Klinkenburg perme-ability at OVB (mD) 0.0522 0.551 30.2 0.00421

Effective permeabil-ity to gas at Swi (mD) 0.0194 0.411 29.0 0.00198

Relative permeability to water at 100% Sw

0.48 0.36 0.31 0.42

Irreducible Sw 10% 15.6% 3% 10%

Table 1. Selection of relative permeability data from Mettika Embayment core. OVB is overburden stress conditions, Swi is ir-reducible water saturation, and Sw is water saturation.

Figure 16. Wetting and non-wetting relative permeability by drainage from four grouped Patchawarra Formation sandstone samples at stressed overburden conditions.

Figure 17. Wetting and non-wetting relative permeability by imbibition from two grouped Patchawarra Formation sandstone samples at overburden stressed conditions. Primary drainage data overlain for reference.



ity as core derived reservoir descriptions indicated existence of minerals other than quartz. Porosity was calculated for each mineral (sandstone/limestone/dolomite) from the neutron and sonic logs. After correction for clay, salinity and hydrocarbon influence, neutron porosity was derived from the recorded neu-tron log while sonic porosity was computed using the Wyllie equation (Wyllie et al, 1956, 1958). The cross-plot porosity was calculated and calibrated to core and NMR porosities whenever available (Eq. 2 and Fig. 18).

ϕ = ϕS1

+(ϕ

N1 − ϕ

S1)

1 −(ϕ

N1 − ϕ

N2)

(ϕS1

− ϕS2

) (2)

In Equation 2:• ϕ

N1 is the neutron corrected porosity for matrix 1;

• ϕN2

is the neutron corrected porosity for matrix 2;• ϕ

S1 is the sonic corrected porosity for matrix 1; and,

• ϕS2

is the sonic corrected porosity for matrix 2.The Juhasz equation was used to calculate water saturation.

Cation exchange capacity per unit total pore volume (nor-malised Qv) was calibrated to available core Qv data (Juhasz, 1979; 1981). The Qv parameter quantified the clay effect on the recorded resistivity log and thus was used to correct the calcu-lated water saturation for clay influence. Patchawarra Forma-

tion water analysed from production testing in Kingston Rule–1 and nearby wells was estimated at 12,000 ppm NaCl equiva-lent or 0.119 ohm-m at a reservoir temperature of 155°C. This resistivity is comparable to that derived from the Pickett plot generated for the 2,800–2,866 m interval in Talaq–1 (Fig. 19). Talaq–1 is the structurally deepest well sited in the embayment and many sandstone sections appear to be water saturated, in-cluding the aforementioned section used as a calibration point.

Accurate and consistent porosity measurement from pad-type tools was complicated by rugose borehole surfaces throughout the logged intervals in all wells. Sonic logs were primarily used to determine porosity as they were generally unaffected by these irregularities (Porter, 1976; Morton 1989). While the sonic log has the advantage of not being affected by borehole rugosity, low tool resolution can bias porosity esti-mates. This bias is especially evident in laminated sections with multiple thin cemented beds, resulting in pessimistic porosity measurements, while carbonaceous inclusions can have the opposite effect. The neutron porosity log was used cautiously as the tool can overestimate porosity in shaly intervals and is again affected by irregular borehole walls. Care was taken to appro-priately determine and calibrate the neutron-sonic cross-plot porosity to the available core plug and NMR porosities (Fig. 18).

The main uncertainty in the log interpretation was the in situ water saturation and efforts were made to constrain water resis-tivity values and uncertainty in key electrical parameters and calculated porosity. Considerable time was expended to moni-

Figure 18. Petrophysical log interpretation of Hornet–1, Kingston Rule–1 and Talaq–1.



tor the salinity of the production water from Kingston Rule–1. It is believed that the salinity of 12,000 ppm NaCl equivalent at reservoir conditions is representative of the Patchawarra Formation in the area. Uncertainties in water saturation can be limited to either the application of a constant cementation exponent, overestimation of porosity, or the effects of both.

Initial gas saturations determined using a constant ce-mentation exponent were deemed unreasonably high given production test characteristics and results were not in line with expectations. Early time post clean-up production data and logging from Hornet–1 and Kingston Rule–1 has identi-fied mobile off-structure gas and water in varying degrees. In total, 11 Patchawarra and one Epsilon Formation zones over both wells were stimulated successfully and produced for three weeks after clean up. Kingston Rule–1 and Hornet–1 produced at approximately 1.2 MMscf/d and 2.2 MMscf/d, respectively at a tubing head pressure of between 600 and 700 psi(g). Water-gas-ratio (WGR) averaged 200 bbl/MMscf in Hornet–1 and stabilised at 600 bbl/MMscf in the structurally deeper Kingston Rule–1. Production logging identified water co-production from several gas bearing intervals. Two mobile phases implies the reservoir is likely to be within a transition zone and water saturations are above irreducible. This was not unexpected given that many wells on surrounding structural highs also co-produce significant volumes of water early in production life. A preliminary wireline log interpretation (not shown) used to pick perforation and stimulation intervals sug-gested water saturations were close to irreducible but did not match observed water and gas rate production characteristics. This early interpretation assumed a constant cementation exponent of 1.96 while the interpretation shown in Figure 18 employs a variable cementation exponent, applied when V

cl

was greater than 30% (Eq. 3). The updated water saturations are more in-line with expectations and observed water and gas production rates.

m = m × 10Vcl−VclCutoff (3)

To assess the impact of water saturation on productivity, a simple single-well numerical model (Schlumberger Eclipse) was used to match the observed WGR of each stimulated zone. The relative permeability curves presented in Figure 16 were input and water saturation adjusted to match the measured gas and water rates. Individual zone water saturation needed

to be raised by between 15% and 25% (absolute) above the pre-liminary petrophysical interpretation to obtain approximately similar WGR to production log measurements. This appears to support the revised interpretation incorporating a variable cementation exponent. While every effort was made to ensure accurate measurement, interpreted zone water and gas produc-tion rates from production logging can be in error and emphasis was not placed on an exact match. Given that all perforated horizons were fracture stimulated, it is conceivable that the well is accessing an area of high water saturation beyond the observable wellbore and the log interpretation does not give an accurate impression of post-stimulation performance. It is dif-ficult to draw firm conclusions in this situation. A broad obser-vation can be made by analysing the apparent saturations and production characteristics in the Hornet–1, Kingston Rule–1 and Talaq–1 wells; the occurrence of gas charged sands with mobile hydrocarbons decreases with depth. In both BCG and conventional, buoyancy dominated reservoirs, relative perme-ability and structural elevation have important implications for well productivity and resource distribution.

CONCEPTS AND IMPLICATIONS OF RELATIVE PERMEABILITY IN DEEP GAS ACCUMULATIONS

Masters (1984) described a theoretical multi-step process by which gas generation and secondary migration in low perme-ability sandstone can lead to large abnormally pressured gas columns (Fig. 10). This process relies on:1. sufficient and ongoing gas generation and percolation to

remove water to up-dip permeable aquifers;2. a commensurate reduction in the permeability to water;

and,3. a phase trap by which water cannot flow counter-currently

to replace gas.If relative permeability characteristics do not limit buoyan-

cy-driven counter-current flow, or gas generation and migra-tion rates are insufficient to overcome the ingress of water, then the result is a conventional normally pressured reservoir. Basins with poor or limited regional aquifer contact, large generative potential, low porosity and unfavourable relative permeability characteristics are more likely to contain a pervasive gas ac-cumulation. Previous studies have described derived relative permeability curves that are purported to act as an effective trap (Hillis et al, 2001; Masters, 1984; Burnie et al, 2008, Cluff and Byrnes, 2010), however, few studies have quantified trap effec-tiveness using measured reservoir derived data and a numerical simulator that combines the many physical processes at play.

A simple vertical, one dimensional, two phase black oil simulation model (Schlumberger Eclipse) was constructed to examine three effects: trapping of gas against a conventional stratigraphic seal; a relative permeability and phase block trap; and, the impact of aquifers on gas column height. To approxi-mate basin processes, gas was injected for a finite period at the base of the model at 200 psi over an input reservoir pressure of 3,200 psi(a). Two aquifers were added at the top and base of the model to control water ingress and gas escape during and after gas generation and migration. Where possible, measured res-ervoir data from the Mettika Embayment was used, including capillary pressure (to be discussed subsequently), relative per-meability curves (Fig. 16), porosity to permeability relationship and fluid PVT (pressure, volume, temperature) properties. The experimental cases and results are summarised in Figure 20. Reservoir permeability and porosity were constant across the model except for a single horizontal model cell acting as the seal. A control case lacking any facies seal demonstrates that only low saturation immobile gas remained trapped and the derived relative permeability curves do not act as a phase trap.

Figure 19. Talaq–1 Pickett plot used to estimate in situ water resistivity in the Patchawarra Formation. Water saturation quality lines are added for reference.



Cases 1 and 2 examine gas trapped by capillary entry pressure alone. As expected, the hydrocarbon column height will be pro-portional to the entry pressure of the seal facies and the rela-tive density of the gas and water phases. In both cases 1 and 2, water remains connected and mobile across the capillary trap and can flow by gravity to recharge the bottom aquifer. Hence, the gas-water pressure profile resembles a normally pressured conventional gas on water system and the absence of the bot-tom aquifer has minimal effect on the gas column height and pressure regime (Fig. 10). The situation may be different if gas saturation across the trap is at irreducible but if some relative permeability to water remains, the bottom aquifer will gradu-ally approach the regional hydrostatic pressure after charge.

Cases 3 and 4 examine the effect of a severe jail-like relative permeability trap. To approximate the jail effect, the Corey ex-ponents for gas and water were increased to create a minimum permeability to both phases of less than 10 nano-Darcies, or approximately three orders of magnitude less than that de-termined from the Mettika Embayment core study (Fig. 16).

Case 3 lacks a bottom aquifer and significant quantities of gas are trapped as a result of the phase block effect. In contrast to cases 1 and 2, the pressure profile for case 3 is sub-normally pressured in both the gas and down dip aquifer and is analo-gous to the experiment presented by Gies (1984). The system was still equilibrating when gas generation ceased and gas was continually escaping from the trap into the shallow aquifer. The jail-like relative permeability curves only slowed counter-cur-rent flow and did not arrest it completely. Hence, the phase trap is not effective across long time periods unless combined with a capillary trap. If a normally pressured aquifer recharges from below, gas is forced out of the system as per case 5 and no gas remains in the reservoir above residual saturation. In the pa-per presented by Gies (1984), it was explicitly stated that water could not flow down-dip to the lower aquifer. Using the relative permeability curves measured in this study, some degree of water permeability and phase continuity is retained through the trap even as the gas saturation approaches near-irreducible. It appears that a permeability jail is not an effective barrier to

Figure 20. Schematic and results of various trapping experiments. Reservoir parameters are input from obtained routine and special core analysis results. Cases 1 and 2 examine the effect of capillary pressure on trapping while cases 3 and 4 investigate the concept of permeability jail. Cases 1, 2 and 4 will resemble normally pressured gas/water systems while case 3 will be sub-normally pressured.



migration in this simulation experiment. In situ, such a phase trap needs to be accompanied by a diagenetic reduction in po-rosity and permeability to slow gas migration significantly and potentially add a component of capillary trapping. A balance is ultimately struck between continued low level gas generation, which maintains the gas column and ingress of water recharg-ing the base water leg. (Burnie, et al 2008).

If gas generation continues or accelerates after diagenesis, then regional abnormal-pressure can evolve across an extended period of time. In lenticular, isolated sandstone bodies such as that found in the Mettika Embayment, overpressure can result if the conditions are appropriate. It is reasonable to anticipate that regionally low porosity and high capillary entry pressure, combined with continued gas generation from proximal source rock, and unfavourable relative permeability characteristics will have a greater propensity to trap gas and form a continuous gas accumulation. One or more of these tenets appears not to be regionally active in the Mettika Embayment.

Complex concepts of phase trapping or diagenesis do not need to be invoked to create an off-structure, sub-normally pressured hydrocarbon accumulation. For demonstration purposes, it is assumed that the two-dimensional depocentre shown in Figure 21 is a full syncline and water is unable to by-pass the anticline high to replenish the lower water leg near the source rock. Gas charges the reservoir and expels water up dip and to an infinite regional aquifer. Ultimately, generation and charge ceases and the up-dip aquifer water is allowed to par-tially ingress back into the reservoir. In this scenario, there are two spill points—at the top and base of the reservoir. If there is sufficient structural relief, water will only replace gas down the lower spill point and the lower water leg will not be recharged. The pressure-depth profile close to the regional aquifer appears as a conventional gas-water system while the gas and water leg near the source rock will be sub-normally pressured (Fig. 10). This simple reservoir configuration and charge model does not invoke a stratigraphic seal or jail-like relative permeability curves. It is likely that various trapping and secondary migra-tion concepts manifest in any particular off structure gas play, the degree of which controls the amount, distribution and pro-ductivity of the hydrocarbon bearing formations.

Routine and special core analysis combined with production and pressure data suggest that the Patchawarra Formation in the Mettika Embayment is not likely to contain a pervasive BCG system, but a series of conventional stratigraphic and subtle structural accumulations. Exploration of such systems relies on discovering reservoirs where rock property degradation is limited and sandstone packages are of sufficient quality and structural relief to avoid early water production. In stacked flu-vial reservoir environments, charge and migration behaviour can be subtle and complex. It can be difficult to identify sand-stone bodies in pressure and saturation continuity or trapping features without the benefit of high density drilling or adequate seismic data. The reality is that gas saturation encountered in a complex depositional environment may defy rigorous expla-nation. At the very least, production data and log analysis of Kingston Rule–1 and Hornet–1 suggest that water saturation in many of the sandstone packages are likely above irreducible and within a transition zone. A generalised approach has been adopted to assess the potential for the reservoir to accumulate commercially viable gas saturation.

USE OF CAPILLARY PRESSURE DATA TO ASSIST EXPLORATION AND APPRAISAL

For exploration and appraisal purposes in complex poorly de-fined reservoirs, standard capillary pressure curve dimensions of water saturation and pressure or height are not particularly useful for mapping and assessment of regional saturation trends. Transformations can be made to these curves to solve for poros-ity, height above free water level (hAFWL) and capillary entry pressure. Ambient air-brine centrifuge and air-mercury injec-tion capillary pressure was obtained from representative plugs sampled from Kingston Rule–1 and Talaq–1 and combined with analogous Patchawarra core data obtained from Sasanof–1 and Skipton–1. Data normalised using the Leverett-J function re-sulted in a somewhat consistent clustering of curves (Fig. 22). This observation may imply that the pore network structure and distribution is broadly similar across various pore scales. A single Skelt-Harrison function was fitted to the J data to simplify subse-quent analysis (Skelt and Harrison, 1995) (Eq. 4). Classification of J-curve by rock type or facies was not attempted as a full core lith-ological description was not available at the time this study was completed. This model assumes the pore network connectivity 1

2

3

Regional Aquifer

Limited Recharge

Water Recharge

GasLoss

WaterGasInitial

Condition

Generation &Migration

GenerationCeases

Under-PressuredOff Structure Gas

Upper Spill

Figure 21. Simplistic hydrocarbon charge and migration behaviour in a two-dimensional anticline in contact with a regional infinite aquifer.

Figure 22. Capillary data acquired from Kingston Rule–1, Talaq–1, Skipton–1 and Sasanof–1 core plug normalised to dimensionless J. A single Skelt-Harrison J-function was fitted to the data and used for subsequent analysis.



is similar across all porosity ranges and capillary pressure, poros-ity and permeability are related in all sampled rock types. Using the fitted function, a series of theoretical curves were derived to describe a continuum of porosities in dimensions of pressure and hAFWL with wetting phase saturation (Fig. 23).

Sw

= = 1 − × 1 − +

1 − e−

a0

J − JD = 1 −

× 1 − +

= 1 − × 1 − +

a1 = 1 −

× 1 − +

× (1 − Sirr

) + Sirr

(4)

In Equation 4:• J is the J-function parameter.• JD, a

0 and a

1 are the Skelt-Harrison fitting parameters;

• Sw

is the water saturation; and,• S

irr is the irreducible water saturation.

Capillary entry pressure is a useful metric to examine con-cepts of seal capacity and saturation potential across a reservoir unit of varying quality. To enable a non-wetting gas to enter the pore space and create a continuous phase through the plug, a threshold pressure must be reached. This entry pressure is re-lated to the largest connected pore system available in the plug and can be estimated from capillary data. To establish the qual-ity of a potential up-dip stratigraphic seal facies, a correlation was created between average plug porosity and capillary entry pressure. The theoretical capillary entry pressure was calculated by solving for J from the fitted Skelt-Harrison curve (Eq. 5). Con-verting J to capillary pressure using estimates of in situ gas-water interfacial tension and contact angle, porosity and permeability, the entry pressure at slightly under 100% water saturation is es-timated. This approximates the pressure difference required to overcome capillarity and ideal entry of the non-wetting phase into the largest continuous pores in the plug at reservoir con-ditions. The required equivalent gas column height required to enter a plug of given porosity is easily calculated and can be con-sidered as the maximum supportable column before seal breach occurs. (Fig. 24). The relationship was validated by plotting entry pressure from both centrifuge and mercury capillary pressure curves. Care was taken to correct for dead volume and entry ef-fects at high water saturation in the measured laboratory data and broadly speaking, the actual estimated entry pressures fit within the range of theoretical derived values.

Several approximations have been made when developing this technique including: • The average plug porosity has been assumed to correlate to

the average diameter of the largest continuous pore network in the plug. This is acceptable if the distributions of pore

radii are consistent and the J-function normalised capillary pressure curves are similar.

• An estimate of permeability is required to relate porosity to capillary pressure. The three fitted permeability and poros-ity relationships (high, mid and low) presented in Figure 15, attempt to capture the range of entry pressures and column heights. For the purposes of this basic study, the mid-case porosity-permeability relationship and one J-function curve was used for subsequent analysis.

• Mercury and centrifuge capillary pressure curves were ob-tained at laboratory conditions under a minimal confining stress of 800 psi. Entry pressure, saturation and irreducible saturation may be affected by confining stress. At the time of publication, this data was unavailable.

S = S

w − S

irr

1 − Sirr

= + − ln1 − × − ln1 −

= − 1 −

= + − ln1 − × − ln1 −

= − 1 −

J = [ a0 + J

D (–ln(1−S))

(1/a1)] × [–ln(1−S)

(–1/a1)]

(5)

In Equation 5:• J is the J-function parameter.• JD, a

0 and a

1 are the Skelt-Harrison fitting parameters;

• S is the reduced water saturation;• S

w is the water saturation; and,

• Sirr

is the irreducible water saturation.The chart presented in Figure 24 can be thought of as the

maximum porosity before entry of gas into the pore network for a given hAFWL and porosity-permeability relationship. This is useful for understanding hydrocarbon migration. Gas is generated from source rock and migrates preferentially along the highest porosity connected carrier beds until stratigraphic or structural barriers are encountered (Fig. 25). Low porosity intervals will generally be avoided at early stage secondary migration (Wardlaw, 1980). As gas accumulates against a trap and the buoyancy force increases, the low porosity previously bypassed will begin accumulating hydrocarbon commensurate with the capillary entry pressure and position in the column. It can be inferred that if a particular reservoir was completely charged and the largest gas column possible was trapped, a porosity measured below the poro-perm trend lines (Fig. 25D) is likely not contain hydrocarbon. This might be thought of as a

Figure 24. Relationship between supportable column height and porosity derived from capillary pressure curves. Dashed lines represent modelled entry pressures from the fitted J-function for the high, mid and low permeability trends.

Figure 23. Wetting phase saturation versus hAFWL and capillary pressure for various porosities derived from the fitted J-function.

0

100

200

300

400

500

600

700

800

0

100

200

300

400

500

600

700

800

900

1000

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

He

igh

t A

bo

ve F

ree W

ate

r L

eve

l (m

)

Cap

illa

ry P

res

su

re (

ps

i)

Wetting Phase Saturation

ϕ = 14%

10%

8%

6%

4%

2%



theoretical porosity cut-off for a net-to-gross ratio and it varies depending on the height of the gas column, the buoyant force and capillary pressure.

This information can be used to exclude reservoirs that are likely to contain residual hydrocarbon saturation and affect deep resistivity readings (Figure 25D). At high water saturations, gas is essentially immobile (Fig. 16) and hence, zones produc-ing only water may still contain both water and gas in the pore space. Mud log and petrophysical analysis suggests that many sequences in the Mettika Embayment contain small volumes of residual gas and may have acted as secondary migration con-duits. If it is assumed that gas has filled the structure to a hypo-thetical spill, porosity that is unlikely to have had gas charge can be identified. Fully water bearing zones are most likely to occur low in the column in low porosity. The lower the input FWL parameter, the more stringent the selection of porosity will be for wireline log and Pickett plot interpretation. Given the appar-ent complexity of the individual traps, however, this approach needs to be combined with knowledge of FWL and facies in each sandstone package to be effective. By re-dimensioning the curves developed in Figure 23, a useful relationship has been created between the height of the hydrocarbon column and porosity, which can be measured and mapped relatively easily.

Reservoir productivity and appraisal concept

To achieve high gas saturation within a trap and avoid water production, sufficient porosity contrast must exist between the reservoir and the trap (Fig. 26). In this example derived from Patchawarra capillary pressure data, reservoir 1 will be close to irreducible water and produce only gas while reservoir 2 will be in transition and produce both gas and water. For reservoir 2 to produce water-free, the seal porosity must diminish substan-tially, thereby raising the capillary entry pressure and potential

height of hydrocarbon column. The most prospective reservoirs will have good capillary contrast between lateral seal facies or a large trapped column to lower water saturation close to ir-reducible.

Some channel sequences of the Patchawarra Formation within the Mettika Embayment exhibit fining up sequences bounded by shale and coal facies. Care must be taken when perforating and stimulating such packages as lower porosity does not necessarily equate to higher irreducible water satura-tion as shown by the relative permeability curves in Figure 16. A fining up sequence with increasing water saturation will have a greater propensity to produce free water, compared to a block-like sequence with consistent high porosity and low water saturation (Fig. 27) (Shanley et al, 2004). Limited perforation or fracture height growth control may limit water production. Res-ervoir development at the well, however, can only imply facies change aerially and stimulation may connect low-quality wet reservoirs unpredictably. In the absence of good well control and pressure data, it is sometimes problematic to model well performance post stimulation.

Given this inherent complexity and lack of well data, a pragmatic method has been used to map potential production fairways for appraisal and exploration. In charged, buoyant hydrocarbon traps at equilibrium, saturation is dependent on several elements including the height of the hydrocarbon col-umn, capillary pressure characteristics, pore geometry and pore connectivity. Generally, the higher the reservoir permeability, porosity or the larger the gas column, the greater the propensity will be to achieve low water saturation and economically viable hydrocarbon flow rates. This simple and well-established tenet of reservoir physics does not directly account for important pro-duction metrics such as WGR and relative permeability.

To convert a simple capillary pressure curve into a tool for exploration purposes, production metrics may be included to

Figure 25. Conceptual hydrocarbon migration in a stratigraphic trap through a semi-continuous sandstone body. (A) Initial conditions with the reservoir completely saturated with water. (B) Migration from source rock bypasses low porosity sandstones with high capillary entry pressure. (C) Post charge and migration. Low porosity interval near the trap may contain low hydrocarbon saturations near residual. (D) Porosity and hAFWL chart for the given reservoir. Points 1 to 4 describe various reservoir levels and qualities.



establish rate limits. Figure 28 shows the relationship between WGR and water saturation for the first month of production and is based on a simple single-well model presented earlier. A 70% water saturation results in an early WGR of 200 bbl/d per MMscf/d and efficient production becomes problematic. The well has a high chance of liquid loading early in life should the water cut increase through aquifer influx. For the purposes of this study, a slightly conservative limit of 60% water saturation was adopted, which translates to approximately 25 bbl/d per MMscf/d of gas. Gas productivity is also severely impaired by this high water saturation as the effective permeability to gas is less than 10% of that at irreducible.

The relative permeability curves presented in Figure 16 are unfortunately not able to provide an estimate of in situ per-meability alone. When acquiring relative permeability data in low porosity samples, it is sometimes found that measured plug permeability at S

wi is significantly lower than dried plug

permeability. Figure 29 is derived from Patchawarra Forma-tion relative permeability data and demonstrates that samples lower than 12% porosity require normalisation to the effective gas permeability at S

wi rather than the dried plug permeability.

A simple linear regression between porosity and reduction in effective permeability to gas at S

wi was used to correct dried

plug permeability (Fig. 29) and was applied to porosity less than 12%. Taking the corrected permeability at S

wi and the relative

permeability curves, the permeability to gas at a given water saturation can be estimated. To demonstrate our method, a somewhat arbitrary gas permeability limit of 1 µD was adopted independent of net pay or economic considerations. For more rigorous applications, net thickness, the permeability-thickness product, and completion strategy may be modelled to assess more case-specific productivity limits.

Figure 30 describes a theoretical relationship between reser-voir hAFWL and porosity for a given water saturation. The curves presented here are based on the single derived J-function depict-ed in Figure 22 and are simply the measured capillary pressure data in dimensions of height above free water and porosity. To intersect a reservoir with low water saturation, either the porosity must increase or there must be a larger trapped gas column. The shaded area below the 60% water saturation quality line is the upper limit of water saturation required to avoid potential early liquid loading. The curved dashed line is the absolute perme-ability limit trend of 1 µD and is based on the median porosity-permeability relationship, bounded by the high (90th) and low (10th) percentile defined in Figure 15. An intersected sandstone

0

50

100

150

200

250

300

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

He

igh

t A

bo

ve

Fre

e W

ate

r L

eve

l(m

)

Wetting Phase Saturation

Reservoir 1ϕ = 14%

Seal Capillary Pressure

Reservoir 2ϕ = 6%

Water Free

Production

Supportable Column Height

Co-Produced

Water

Figure 26. Relationship between water and hydrocarbon production, capillary pressure and the height of column.

Potential Frac

Height Growth

PerforationGas

ProductionLow

High

Sw

ϕ Sw

Potential Frac

Height Growth

Perforation

Water and Gas

Production

GasProduction

Case 1 - Fining Up

Case 2 - Massive

Rel-Perm

Rel-Perm

Figure 27. Comparison of production characteristics between fining up and block-like sand packages. Early production from case 1 will likely co-produce water while case 2 will predominantly produce gas due to varying water saturations and relative permeability to gas and water phases.

0.01

0.1

1

10

100

1000

10000

0% 20% 40% 60% 80% 100%

Wate

r-G

as

Ra

tio

(b

bl/

MM

sc

f)

Water Saturation

Figure 28. Relationship between simulated WGR and water saturation.

0%

2%

4%

6%

8%

10%

12%

14%

0% 20% 40% 60% 80% 100%

Plu

g P

oro

sit

y (

%)

Reduction in Permeability (Kinf to Krg @ Swi)

Figure 29. Reduction in gas permeability as a result of irreducible water.



package below the curved permeability cut-off line and below the 60% S

w quality line will either have sufficient permeability but too

high WGR (blue shaded area) or may be too impermeable (yel-low shaded area) to provide sufficient unstimulated productivity. The green shaded area includes reservoirs with sufficient perme-ability and gas saturation to produce unstimulated at high rate. Net pay can be defined here by several inputs including elevation above the FWL, porosity, relative permeability, capillary pressure and production modelling.

To define an appraisal fairway in the Mettika Embayment using this technique, Patchawarra Formation arithmetically averaged porosity was contour mapped between well control and co-krigged with the observed porosity and depth trend (Fig. 31). The mid-point of the Patchawarra at the deepest part

of the embayment was input as the hypothetical lowest FWL and the mid-Patchawarra depth structure map used to calcu-late a height above this reference (Fig. 32). Grid control points were classified by location on the chart presented in Figure 30, and mapped. The appraisal areas defined in Figure 33 repre-sent the most optimistic case where gas has charged the trap completely down to a hypothetical spill point. It implies that even if the entire reservoir was charged with gas, reservoir in-tersected in the blue shaded area down dip of Kingston Rule–1 and Hornet–1 is not likely to be productive for gas. The yellow area has sufficient structural relief to produce with moderate to low water rates but the porosity is too low to flow unstimu-lated. New appraisal well locations located in the green area are constrained by the lower and middle-Patchawarra onlap edges. The map implies that further appraisal drilling should focus on the green and yellow shaded areas away from high water saturation and low porosity. The reservoir architecture model predicts good north to south reservoir development and connectivity and intermittent connectivity in an east to west orientation. Stratigraphic trap development potential is likely on the flanks of the Toolachee Field High.

There are several important caveats when using this technique:• Given the apparent complexity of the reservoir, it is possible

that each sequence has a unique FWL and saturation profile. The map is only useful for defining the lowest limit of pro-ductivity should the entire trap be charged with gas. As the lowest likely FWL was used, a saturation map overlay may be required to define areas of high water saturation up-dip.

• It assumes a deterministic case rather than capturing the range of input trends. It is somewhat sensitive to porosity given the log-linear relationship (Figs 24 and 30).

• More well control is needed to define trends appropriately. Detailed intra-formation porosity maps may be needed to improve reliability.

• Vertical and lateral sandstone heterogeneity associated with changes in depositional environments is not captured.

Figure 30. Relationship between reservoir hAFWL and porosity derived from capillary pressure data. Water saturation, WGR and productivity cut-offs are overlaid for reference.

Kidman-8

Kidman-3

Kidman-9

Kidman-1

Coochilara-2Boongala-2

Kingston Rule-1

Hornet-1

Allambi-1

Toolachee-37

Kelbrook-1

Verbena-1

Toolachee-50

Toolachee East-2

Toolachee-40

Toolachee-21

Toolachee-41

6%

7%

8%9%

10%

11%12%

Porosity

11%

10%

9%

8%

7%

6%

5%

North

12%

Porosity

11%

10%

9%

8%

7%

6%

5%

0 1 2

Km

Middle PatchawarraZero Edge

Lower PatchawarraZero Edge

Figure 31. Average Patchawarra Formation porosity map within the study area.

Kidman-8

Kidman-3

Kidman-9

Kidman-1

Coochilara-2Boongala-2

Kingston Rule-1

Hornet-1

Allambi-1

Toolachee-37

Kelbrook-1

Verbena-1

Toolachee-50

Toolachee East-2

Toolachee-40

Toolachee-21

Toolachee-41

2340m

2240m

2140m

2100

Sub-Sea Depth (m)

2150

2200

2250

2300

2350

2400

2450

North

0 1 2

Km

Middle PatchawarraZero Edge

Lower PatchawarraZero Edge

Figure 32. Top Patchawarra Formation depth map.



Treating sands as a homogenous package will potentially lead to production problems.

• Permeability was used to define the lower productive limit of the reservoir. The permeability-thickness product, however, is more useful to define production rate. This will need to be considered in future work.

Migration, charge pathways and trapping

Thick coals distributed throughout the Permian Patchawarra Formation are widely regarded as the most-likely source for oil and gas accumulations within the Cooper Basin (Gravestock et al, 1998). Available vitrinite reflectance data from wells within the Mettika Embayment suggests the Patchawarra Formation is mature for gas expulsion. Diagenetic destruction of Perm-ian sandstone porosity and permeability is documented to have occurred both prior-to and, in some cases, synchronous with charge (Hamlin et al, 1996; Seggie, 1997). The lack of suf-ficient porosity and permeability reduction in the Patchawarra Formation within the embayment may have prevented the de-velopment of a self-sourced BCG play. As gas is expelled from source rocks during primary migration, capillary data suggests it may be difficult to accumulate and trap this gas locally and develop abnormal pressures on a regional scale. This leads to a conventional stratigraphic and structural trap system as the gas undergoes secondary migration. At the very least, mobile gas production at Kingston Rule–1 and Horent–1 sufficiently de-risks charge into the embayment flanks in a conventional sense. The presence of low gas saturations in many sandstone packages in wells outside of structural closure, coupled with water production may suggest one or more of the following:• Charge adequacy—insufficient gas has been generated to fill

all sandstone bodies during primary migration and much of the reservoir is a transition zone or gas saturation is close to irreducible.

• Charge bypass—channel axes are preferentially orientated north–south, coupled with broad north–south tilting of the

embayment. If the western and eastern flanks of the embay-ment are not steep enough to create a sufficient drainage divide, hydrocarbon migration will be focused to the south resulting in bypass. Pre-charge faulting may also compart-mentalise the reservoir preventing access to charge.

• Charge preservation—post Cadna-owie fault reactivation focused on the flanks of the embayment may have resulted in leaked columns.

• Timing of expulsion—the trapping geometries seen present day did not exist at the onset of expulsion.

• Trap adequacy—structural relief is not sufficient to achieve high gas saturation in the flanks at the given porosity.

• Loss of gas through diffusive processes.

SUMMARY AND CONCLUSIONS

Appraisal drilling targets have been better defined by the improved petrophysical and engineering concepts and appli-cation within the proposed geological framework. Reservoir characterisation undertaken as part of this integrated study has provided a number of tools that can be applied early to high-grade appraisal targets within the geological framework. The Mettika Embayment is not likely to contain pervasive gas saturation and hydrocarbons appear to be trapped in struc-tural, stratigraphic and combination traps. Given the lack of 3D seismic and well control in the Mettika Embayment, the geological model can be refined through additional seismic acquisition and further appraisal drilling.

Senex Energy Ltd presently holds a significant land posi-tion in many major trough regions of the SA Cooper Basin (Fig. 34). These regions are generally unexplored and have the potential to host stratigraphic and pervasive BCG plays similar

Figure 33. Appraisal area prospectivity map. Reservoir quality is delineated by water saturation and porosity risk.

PatchawarraDepth (m)

-1050

-2000

-2950

-1530

-2500

North

-3450

0 10 20 40

0 10 20 40

km

Senex Permit

Figure 34. Senex Energy Cooper Basin land position with highlighted troughs at the Patchawarra depth horizon.



to recent discoveries in the Nappamerri Trough. The insight gained from studying the stratigraphic and structural trapping in the Mettika Embayment has created an opportunity for the company to leverage in-house expertise and knowledge to ef-ficiently explore other areas of the basin.

ACKNOWLEDGMENTS

The authors wish to thank the Senex Energy Ltd manage-ment team and the APPEA Technical Program Committee for the opportunity to publish this paper. The authors wish to thank Alan Sansome of DMITRE for assistance in the reconnaissance of data, James Crowley for technical reviews and support dur-ing the writing of this paper, Michaela Farrow of Schlumberger for her ongoing assistance, David Warner and Bob Frears for technical review, Tony Kennaird for assistance with core analy-sis, and Mark Riley and Ian Moffat of Precipice Training who provided insightful opinions and facilitated numerous interest-ing discussions during interpretation. The authors also wish to thank previous investigators for documenting the stratigraphic complexity of the Cooper Basin.

The views presented in this paper are those of the authors at the time of publication and do not necessarily reflect those of Senex Energy Ltd management.

REFERENCES

APAK, S.N., STUART, W.J. AND LEMON, N.M., 1993—Struc-tural-stratigraphic development of the Gidgealpa-Merrimelia-Innamincka Trend with implications for petroleum trap styles, Cooper Basin, Australia. The APEA Journal, 33 (1), 94–104.

APAK, S.N., STUART, W.J., LEMON, N.M. AND WOOD, G., 1997—Structural evolution of the Permian - Triassic Cooper Basin, Australia: Relation to hydrocarbon trap styles. AAPG Bulletin, 81 (4), 533–55.

BLASINGAME, T.A., 2008—The Characteristic Flow Behaviour of Low-Permeability Reservoir Systems. SPE Unconventional Reservoirs Conference, Keystone, Colorado, 10–12 February, SPE 114168.

BURNIE, W.S., MAINI, B., PALMER, B.R. AND RAKHIT, K., 2008—Experimental and empirical observations support-ing a capillary model involving gas generation, migration, and seal leakage for the origin and occurrence of regional gasifers. In: Cumella, S.P., Shanley, K.W. and Camp, W.K. (eds) Understanding, Exploring, and Developing Tight-gas Sands. 2005 Vail Hedberg Conference, AAPG Hedberg Series, no. 3, 29–48.

BYRNES, A.P., SAMPATH, P.K. AND RANDOLPH, P.L., 1979—Effect of Pressure and Water Saturation on Permeability of Western Tight Sandstones. Fifth Annual Department of Energy Symposium on Enhanced Oil and Gas Recovery and Improved Drilling Technology, Tulsa, Oklahoma, 1–15.

CLUFF, R.M. AND BYRNES, A.P., 2010—Relative Permeability in Tight Gas Sandstone Reservoirs – The ‘Permeability Jail’ Model. Society of Petrophysicists and Well Log Analysts. SPWLA 51st Annual Logging Symposium. Perth, Western Australia, 19–23 June.

DEVINE, S.B. AND GATEHOUSE, C.G., 1977—Sandstone reser-voir geometry of non-marine sediments in the Toolachee Gas Field. The APEA Journal, 17 (1), 50–7.

DUNLOP, E.C., BROWNE, M.V. AND TADIAR, E.F., 1992—De-plation of gas reservoirs by molecular diffusion – a case study. The APEA Journal, 32 (1), 369–89.

GIES, R.M., 1984—Case history for a major Alberta deep basin gas trap: the Cadomin Formation. In: Masters, J.A. (ed.) Case Study of a Deep Basin Gas Field. Tulsa, Oklahoma: AAPG.

GRAVESTOCK, D.I., HIBBURT, J.E. AND DREXEL, J.F. (eds), 1998—The petroleum geology of South Australia. Volume 4: Cooper Basin. Report Book, 98/9. Adeladie: South Australian Department of Primary Industries and Resources.

HAMILTON, D.S. AND TADROS, N.Z., 1994—Utility of coal seama as genetic stratigraphic sequence boundaries in nonma-rine basins: an example from the Gunnedah Basin, Australia. AAPG Bulletin, 78 (2), 267–86.

HAMLIN, H.S., DUTTON, S.P., SEGGIE, R.J. AND TYLER, N., 1996—Depositional controls on reservoir properties in a braid-delta sandstone, Tirrawarra Oil Field, South Australia. AAPG Bulletin, 80 (2), 139–56.

HILLIS, R.R., MORTON, J.G.G., WARNER, D.S. AND PENNEY, R.K., 2001—Deep basin gas: a new exploration paradigm in the Nappamerri Trough, Cooper Basin, South Australia. The APPEA Journal, 41 (1), 185–200.

JUHASZ, I. 1979—The central role of Qv and formation water salinity in the evaluation of shaly formations. In: SPWLA Sym-posium Transactions, June 1979. Houston, Texas: Society of Petrophysicists and Well Log Analysts.

JUHASZ, I., 1981—Normalised Qv. The Key to Shaly Sand Evalu-ation Using the Waxman-Smits Equation in the Absence of Core Data. SPWLA 22nd Annual Logging Symposium, Mexico City, Mexico, 23–26 June.

LANG, S.C., CEGLAR, N., FORDER, S., SPENCER, G. AND KASSAN, J., 2002—High resolution sequence stratigraphy, reservoir analogues and 3D seismic interpretation – applica-tion to exploration and reservoir development in the Baryulah Complex, Cooper Basin, Southwest Queensland. The APPEA Journal, 42, 511–22.

LANG, S.C., GRECH, P., ROOT, R., HILL, A. AND HARRISON, D., 2001—The application of sequence stratigraphy to exploration and reservoir development in the Cooper-Eromanga-Bowen-Surat Basin System. The APPEA Journal, 41 (1), 223–50.

LAW, B.E., 2001—Basin-centered gas systems. AAPG Bulletin, 86 (11), 1,891–919.

MASTERS, J.A., 1984—Elmworth, case study of a deep basin gas field. AAPG Memoir 38. Tulsa, Oklahoma: The American Association of Petroleum Geologists.

MORTON, J.G.G., 1989—Petrophysics of Cooper Basin reservoirs in South Australia. In: O’Neil, B.J. (ed.) The Cooper and Eroman-ga Basins Australia. Proceedings of the Cooper and Eromanga Basins Conference, PESA, Adelaide, South Australia, 153–63.

NAKANISHI, T. AND LANG, S.C., 2001—The search for strati-graphic traps goes on – visualisation of fluvial-lacustrine suc-cessions in the Moorari 3D Survey, Cooper-Eromanga Basin. The APPEA Journal, 41 (1), 115–37.



NAKANISHI, T. AND LANG, S.C., 2002—Towards an efficient exploration frontier: vonstructing a portfolio of stratigraphic traps in fluvial-lacustrine successions, Cooper-Eromanga Basin. The APPEA Journal, 42 (1), 131–50.

PORTER, C.R., 1976—An empirical approach to the determina-tion of porosity, shale percentage and permeability of Permian sandstones in the Cooper Basin South Australia. The APEA Journal, 16 (1), 111–5.

PRICE, P.L., FILATOFF, J., WILLIAMS, A.J., PICKERING, S.A. AND WOOD, G.R., 1985—Late Palaeozoic and Mesozoic paly-nostratigraphical units. CSR Oil and Gas Division, palynology facility report, 274/25. South Australia. Department of Primary Industries and Resources. Open file Envelope, 6846:41-90, un-published.

SCOTT, M.P., STEPHENS, T., DURANT, R.H., MCGOWEN, J.M., THOM, W.W. AND WOODROOF, R.A., 2013—Investigating Hydraulic Fracturing in Tight Gas Sand and Shale Gas Res-ervoirs in the Cooper Basin. SPE Unconventional Resources Conference and Exhibition-Asia Pacific, Brisbane, Queensland, 11–13 November, SPE 167073.

SEGGIE, R.J., 1997—Reservoir characterisation of the Moorari/Woolkina Field Complex, Cooper Basin. The APPEA Journal, 37 (1), 70–86.

SHANLEY, K.W., CLUFF, R.M. AND ROBINSON, J.W., 2004—Factors controlling prolific gas production from low-permeability sandstone reservoirs: implications for resource assessment, prospect and risk analysis. AAPG Bulletin, 88 (8), 1,083–121.

SKELT, C. AND HARRISON, B., 1995—An Integrated Approach to Saturation Height Analysis. SPWLA 36th Annual Logging Symposium, Paris, France, 26–29 June.

SPENCER, C.W., 1989—Review of Characteristics of Low-Per-meability Gas Reservoirs in the Western United States. AAPG Bulletin, 73 (5), 613–29.

STRONG, P.C., WOOD, G.R., LANG, S.C., JOLLANDS, A., KARALAUS, E. AND KASSAN, J., 2002—High resolution pa-leogeographic mapping of the fluvial-lacustrine Patchawarra Formation in the Cooper Basin, South Australia. The APPEA Journal, 42 (1), 65–81.

STUART, W.J., KENNEDY, S. AND THOMAS, A.D., 1988—The influence of structural growth and other factors on the con-figuration of fluviatile sandstones, Permian Cooper Basin. The APEA Journal, 28 (1), 255–66.

TAYLOR, S., SOLOMON, G., TUPPER, N., EVANOCHKO, J., HORTON, G., WALDECK, R. AND PHILLIPS, S., 1991—Flank plays and faulted basement: new directions for the Cooper Basin. The APEA Journal, 31 (1), 56–73.

THOMAS, R.D. AND WARD, D.C., 1972—Effect of overburden pressure and water saturation on gas permeability of tight sand-stone cores. Journal of Petroleum Technology, 24 (2), 120–4.

TREMBATH, C., ELLIOTT, L. AND PITKIN, M., 2012—The Nap¬pamerri Trough, Cooper Basin unconventional plays: proving a hypothesis. APPEA Conference Proceedings, 52, extended abstract.

WARD, C.R. (ed.), 1984—Coal geology and coal technology. Melbourne: Blackwell Scientific Publications.

WARDLAW, N.C., 1980—The effects of pore structure on displacement efficiency in reservoir rocks and in glass micro-models. SPE/DOE Enhanced Oil Recovery Symposium, Tulsa, Oklahoma, 20–23 April, SPE 8843.

WILLIAMS, B.P.J. AND WILD, E.K., 1984—The Tirrawarra Sandstone and Merrimelia Formation of the Southern Cooper Basin, South Australia – the sedimentation and evolution of a glaciofluvial system. The APEA Journal, 24 (1), 377–92.

WORTHINGTON, P., 1991—Effective Integration of Core and Log Data. The Society of Core Analysts International Sympo-sium, London, England, conference paper 9102.

WYLLIE, M.R.J., GREGORY, A.R. AND GARDNER, G.H.F., 1956—Elastic wave velocities in heterogeneous and porous media. Geophysics, 21 (1), 41–70.

WYLLIE, M.R.J., GREGORY, A.R. AND GARDNER, G.H.F., 1958—An experimental investigation of factors affecting elastic wave velocities in porous media. Geophysics, 23 (3), 459–93.

YUREWICZ, D.A., KRONMUELLER, K., BOHACS, K.M., KLI-MENTIDIS, R.E., MEURER, M.E., YEAKEL, J.D., RYAN, T.C. AND KENDALL, J. 2008—Controls on Gas and Water Distribution, Mesaverde Basin-Centred Gas Play, Piceance Basin, Colorado. In: Cumella, S.P., Shanley, K.W. and Camp, W.K. (eds) Under-standing, Exploring, and Developing Tight-gas Sands. 2005 Vail Hedberg Conference, AAPG Hedberg Series, no. 3, 105–36.

Authors' biographies next page.



Tim Stephens received a Bachelor of Biotechnology with first-class Honours from the University of Queensland (UQ), and a Master of Science in Petroleum Engineering from Heriot-Watt University, Edinburgh. Tim has previously held bio-medical research positions at the Insti-tute for Molecular Bioscience (UQ), and the Diamantina Institute at the Princess

Alexandria Hospital, investigating the role of protein structure and function on disease phenotypes. Tim has worked in vari-ous subsurface engineering roles in the upstream oil and gas industry during the past five years, including conventional gas, tight gas, tight oil, coal seam and shale gas, and is presently a reservoir engineer at Senex Energy Ltd.

[email protected]

Brenton Richards graduated from the Queensland University of Technology in 2008 with a BSc in geoscience. He joined Origin Energy in 2009 where he worked on oil and gas assets within the Taranaki Basin, New Zealand. Brenton joined Senex Energy Ltd in 2013 as a develop-ment geologist working on oil and tight gas assets within the Cooper Basin.

[email protected]

THE AUTHORS

Joseph Lim completed a BSc (honours) in geology with the National University of Malaysia, and a Master of Business Administration with Edith Cowan University of Australia. He joined DownUnder GeoSolutions Pty Ltd in 2007 as junior petrophysicist/rock physicist where he worked on projects from different geological settings globally. Joseph held the role of geo-pressure analyst from 2012–13. In 2013 he joined Senex Energy Ltd as petrophysicist.

[email protected]

Documents

Reservoir characterisation of the Patchawarra Formation ... · the development and sedimentary fill of the basin (Stuart et al, 1988; Apak et al, 1993, 1997). Normal, strike-slip