Reservoir Potential of Sands Formed in Glaciomarine Environments: an Analog Study Based on Cenozoic Examples from Mcmurdo Sound, Antarctica

University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln

Earth and Atmospheric Sciences, Department Papers in the Earth and Atmospheric Sciences of

4-2012

Reservoir potential of sands formed in glaciomarine environments: An analog study based on Cenozoic examples from McMurdo Sound, Antarctica

Christopher R. Fielding University of Nebraska-Lincoln, [email protected]

Brian A. Blackstone University of Nebraska-Lincoln

Tracy D. Frank University of Nebraska-Lincoln, [email protected]

Zi Gui University of Nebraska-Lincoln

Follow this and additional works at: https://digitalcommons.unl.edu/geosciencefacpub

Part of the Geochemistry Commons, Geology Commons, Geomorphology Commons, Geophysics and Seismology Commons, Glaciology Commons, Hydrology Commons, and the Oceanography and Atmospheric Sciences and Meteorology Commons

Fielding, Christopher R.; Blackstone, Brian A.; Frank, Tracy D.; and Gui, Zi, "Reservoir potential of sands formed in glaciomarine environments: An analog study based on Cenozoic examples from McMurdo Sound, Antarctica" (2012). Papers in the Earth and Atmospheric Sciences. 314. https://digitalcommons.unl.edu/geosciencefacpub/314

This Article is brought to you for free and open access by the Earth and Atmospheric Sciences, Department of at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Papers in the Earth and Atmospheric Sciences by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. Published in M. Huuse, J. Redfern, D. P. Le Heron, R. J. Dixon, A. Moscariello, & J. Craig, eds., Glaciogenic Reservoirs and Hydrocarbon Systems; Geological Society, London, Special Publications, 368 (2012); doi:10.1144/SP368.10 Online at http://dx.doi.org/10.1144/SP368.10 Copyright © 2012 The Geological Society of London. Used by permission. Published online April 5, 2012.

Reservoir potential of sands formed in glaciomarine environments: An analog study based on Cenozoic examples from McMurdo Sound, Antarctica

Christopher R. Fielding, Brian A. Blackstone, Tracy D. Frank, and Zi Gui

Department of Earth & Atmospheric Sciences, University of Nebraska–Lincoln 214 Bessey Hall, Lincoln, Nebraska 68588-0340, USA Corresponding author — C. Fielding, email [email protected]

Abstract This paper provides documentation of unexpectedly high-reservoir-quality glaciomarine sands found in the Cenozoic succession beneath McMurdo Sound, Antarctica, as an analogue study for evaluations of hydrocarbon prospectivity in basins elsewhere. The Oligocene to Lower Miocene succession of the Victoria Land Basin, an extant portion of the West Antarctic Rift System, comprises diamictites, mudrocks, and sandstones with minor conglomerates. These lithologies are arranged in repetitive stacking patterns (cycles), interpreted to record repeated advance and retreat of glaciers into and out of the basin, with attendant eustatic and isostatic effects. Phases of ice retreat within the cycles comprise an ar- ray of mudrocks, sandy mudrocks, and sandstones, deposited mainly during relative sea-level highstands. Clean, well- sorted, unconsolidated, and porous sands <25 m thick from such intervals, which are interpreted to be mainly deltaic in origin, were encountered. Some of these sands, which have visible porosity as high as 41%, flowed into the well bore together with significant volumes of cold formation water. Diagenetic modification of sands in these intervals is minimal, which can be attributed to the low-temperature nature of the subsurface environment. Accordingly, glaciomarine sands in near-field glaciogenic successions should be considered as potential reservoir facies in prospectivity assessments.

The fluid reservoir potential of most terrigenous clas- ity in Antarctica; rather, we use information from a well- tic depositional settings is well-documented in the geo- studied, largely unmodified and geologically extant basin logical literature (e.g. Scholle & Spearing 1982; Barwis in Antarctica to develop models that might be used as an- et al. 1990; Galloway & Hobday 1996; Shepherd 2009). alogues for potential exploration targets elsewhere. Conspicuously lacking from these reviews, however, Productive sandstone reservoirs of interpreted gla- and from the literature in general, is any evaluation of ciomarine origin have been documented from the Upper the reservoir potential of glaciogenic successions, partic- Carboniferous Itarare Group of the Paraña Basin in Bra- ularly those formed in glaciomarine settings. Nonethe- zil (Franca & Potter 1991; Potter et al. 1995; Vesely et al. less, glaciogenic successions both in the Ordovician and 2007). Similar sands formed in apparently non-marine late Paleozoic of Gondwana have proved to be effective water bodies (lakes) have also been documented from and producible petroleum systems, and so a fuller un- the Upper Carboniferous to Lower Permian Al Khlata derstanding of reservoir potential in glaciogenic settings Formation of Oman by Levell et al. (1988) and Osterl- would seem to be a desirable objective. off et al. (2004), and age-equivalent Unayzah Formation In this paper, we focus on an element of glaciogenic Members B and C in Saudi Arabia (Melvin & Sprague successions that is poorly documented in terms of reser- 2006; Melvin et al. 2010). There is, furthermore, consider- voir potential, that of glaciomarine facies. We detail the able glaciomarine reservoir hydrocarbon potential in the depositional environments, stratigraphic contexts, tex- Neoproterozoic and Early Paleozoic systems of North ture and porosity characteristics, as well as the diagene- Africa (e.g. Hirst et al. 2002; Le Heron et al. 2004, 2009; sis of Neogene sands from McMurdo Sound, Antarctica, Craig et al. 2009). Accordingly, there is reason to believe as an analogue for glaciomarine deposits elsewhere in the that analogue studies of glaciomarine reservoir poten- world and the geological record. We emphasize that in tial might be of value to future exploration efforts in gla- no way do we encourage or advocate exploration activ- cial petroleum systems.

1 2 Fielding et al. in Glaciogenic Reservoirs and Hydrocarbon Systems (2012)

Figure 1. Location map showing geological context of southern McMurdo Sound, Antarctica (inset map), the locations of fully cored drillholes mentioned in the text (black circles), Neogene, rift-related volcanic edifices (black triangles) and the principal research bases in the area (black squares).

Geological setting Main Rift, in which sediment accumulation was no lon- ger confined to individual sub-basins, but rather formed The data on which this analysis is based derive from a lens that thickens basinward to a central depocenter Cenozoic successions accumulated in the Victoria Land (29–24 Ma); (iv) thermal subsidence, in which sediments Basin (VLB), Antarctica (Figure 1). The VLB forms the formed a blanket over the earlier synrift section (24– 13 westernmost compartment of the West Antarctic Rift Ma); and (v) renewed rifting (Terror Rift: Cooper et al. system, a failed rift complex of Cenozoic age now un- 1987; Henrys et al. 2007), in which sedimentary section derlying the Ross Sea embayment (Cooper et al. 1987; again thickens markedly into a series of depocenters Davey & Brancolini 1995; Stagpoole 2004; Figure 1). and which was accompanied by extensive magmatic ac- As there is essentially no surface exposure of the VLB tivity (13–0 Ma). stratigraphic succession, all available information is The first drilling program to acquire substantial sub- sourced from extensive seismic reflection surveys (e.g. surface drillcore from the McMurdo Sound region was Cooper et al. 1987; Brancolini et al. 1995a, b; Bartek et the Dry Valleys Drilling Project of the 1970s (DVDP: Mc- al. 1996; Hamilton et al. 2001) and from a series of fully Ginnis 1981) in which 15 holes were drilled, all but one cored stratigraphic drillholes (Barrett 2007; Figure 1). onshore, to a maximum depth of 328 m. A parallel drill- Seismic reflection data show the VLB succession to be ing program was carried out by the Eastern Taylor Val- up to 5 km thick and to be divisible into a series of tec- ley Project in the early 1980s (Robinson 1983; Robinson tonostratigraphic units. Based on integration of drilling & Jaegers 1984). Thirteen holes were drilled to a maxi- results with seismic reflection data, Fielding et al. (2006, mum of 71 m below surface. Acquisition of cored strati- 2008a) divided the history of the VLB into five major graphic sections from offshore locations using sea-ice phases: (i) Exhumation and erosion of the Transantarc- platforms commenced with the drilling of DVDP-15 tic Mountains and formation of initial rift topography in 1975, in western McMurdo Sound (Barrett & Treves (older than 34 Ma); (ii) Early Rift, in which sediments ac- 1981; Figure 1). The hole penetrated to 65 m below sea cumulated in mainly fan environments in discrete, rap- floor (mbsf), and various drilling issues limited core re- idly subsiding extensional sub-basins (34–29 Ma); (iii) covery to 52%. A further attempt to penetrate the gla- Reservoir potential of sands formed in glaciomarine environments 3 cial succession of New Harbor was made in 1979 with posits during early rift development and preceding the the drilling of MSSTS-1 somewhat inboard of the site of initial growth of ice centers in the region (Fielding et DVDP-15, with 56% recovery (Barrett 1986). This hole al. 2001). These strata pass gradationally upwards into penetrated to 227 mbsf, but, like earlier offshore holes, increasingly lithologically diverse successions contain- did not intersect basement. The next operation was the ing progressively more abundant diamictite, which drilling of CIROS-2 further inshore in New Harbor (Fig- are interpreted to record the onset of large-scale ice ure 1), again from a sea-ice platform, in 1984. Core re- center development at c. 34 Ma, and later waxing and covery was improved to 67%, and a complete section to waning of ice centers (Fielding et al. 2000, 2001; Bar- crystalline basement was recovered from a presumed rett 2007). Although some of the numerous lithological glacially scoured trough in the mouth of the Ferrar Val- changes coincide with tectonostratigraphic boundaries ley (Barrett & Hambrey 1992). CIROS-1, drilled in 1986, as summarized above, most were probably controlled was a still more ambitious project, drilled considerably by changes in environmental conditions related to the further seaward from a sea-ice platform over a shallow long-term development and evolution of the Cenozoic sea floor ridge identified from seismic reflection surveys Icehouse. Changes in lithofacies assemblage upwards (Barrett 1989). The hole was drilled to 702 mbsf with through the Oligocene, Miocene and Pliocene section 98% recovery, representing a massive achievement, but in the various cores, particularly in the relative propor- did not intersect basement. tions of diamictites v. other terrigenous clastic facies, In the late 1990s, the international Cape Roberts Proj- has been interpreted as recording changes from one ect (CRP) executed the drilling and characterization of type of climatic and glacial regime to another over time a complete stratigraphic transect through a 1500 m Ce- (e.g. Barrett 2007; McKay et al. 2009; Naish et al. 2009). nozoic section unconformably overlying probable De- A summary of the principal lithofacies found in the Ce- vonian basement with an aggregate .95% recovery nozoic succession of the southern VLB is given in Ta- (Cape Roberts Science Team 1998, 1999, 2000; Barrett ble 1. A broadly glaciomarine environment of deposi- 2007). This was achieved by the drilling of three holes tion is inferred for the facies assemblage throughout all east to west through east-dipping strata in three succes- but the oldest sediments penetrated by CRP-3. Physi- sive drilling seasons (1997–1999) so as to capture a com- cal and biogenic sedimentary structures, preserved in- posite record of the basin fill. A detailed chronology of vertebrate fossils and sediment granulometry together this succession was determined from a variety of data, argue for deposition almost entirely within shelfal wa- and the cored stratigraphy was correlated to seismic re- ter depths (Fielding et al. 1998, 2000, 2001; Naish et al. flection data acquired over the same transect. More re- 2001; Dunbar et al. 2008). cently, the ANDRILL program has led to the recovery Seismic reflection data show a generally tabular, of two further, fully cored sections in the southern VLB. eastward-thickening cross-sectional geometry to se- AND-1B, drilled in 2006–2007 to 1284.87 mbsf, recov- quences in western McMurdo Sound (Fielding et al. ered an expanded section recording the interval 13–0 2000, 2008a). The bases of sequences (sequence bound- Ma from the deep-water flexural moat adjacent to Ross aries), where identifiable on seismic reflection records, Island (Figure 1, Naish et al. 2007, 2009; McKay et al. are essentially planar, with local development of ero- 2009). AND-2A, drilled in late 2007, recovered a core to sional channel forms interpreted to record sites of sub- 1138.54 mbsf, recording the time interval from 20–0 Ma, glacial tunnel valleys cut during glacial advances into comprising an expanded succession from 20–14 Ma and the basin (Fielding et al. 2008a). Clear evidence of a a more fragmented record from 14 Ma onward (Har- shelf break is generally lacking from seismic reflection wood et al. 2008). Both projects achieved 98% core recov- data except in the uppermost levels of the stratigra- ery over their total depth. phy (Red reflector, Bclino of Fielding et al. 2008a, inter- The collective stratigraphic archive achieved by the preted to be Pliocene in age), which led Fielding et al. recovery of these various cores provides an excellent ba- (1997) to propose a continuously sloping ramp geome- sis from which to address aspects of the Cenozoic stra- try to explain the pre-Miocene section across the west- tigraphy and sedimentology of the VLB. ern slope of the VLB. A cyclical arrangement of facies was noted in the CRP-1 core (Fielding et al. 1998) and in subsequent cores Stratigraphy and sedimentology (Fielding et al. 2000, 2001, 2011; Naish et al. 2001, 2009) and has been interpreted to record the formation of ge- The character of the stratigraphic succession in the netic sequences forced by cycles of glacial advance and southern VLB varies considerably from the basal, ?up- retreat into/out of the basin, with attendant eustatic and permost Eocene strata upward. Basal conglomerate, isostatic effects. A summary of the generalized vertical breccia and sandstone-dominated successions are in- distribution of facies in an idealized sequence, and its terpreted as subaerial to submarine fan and apron de- interpretation, are given in Figure 2. For an alternative 4 Fielding et al. in Glaciogenic Reservoirs and Hydrocarbon Systems (2012)

Table 1. List of lithofacies recognized in drillcores from McMurdo Sound, based on the facies scheme for AND-2A (Fielding et al. 2008b; Passchier et al. 2012) Facies Description Interpretation 1 Diatomite, opal-bearing or calcareous, fossil-rich mudstone, typ- Open marine shelfal conditions (60–100 m water ically intensively and diversely bioturbated, little if any dis- depth), minimal ice influence persed gravel 2 Siltstone to very fine-grained sandstone, intensively and di- Open marine shelfal conditions (60–100 m water versely bioturbated, dispersed small gravel and calcareous fos- depth), deposition from suspension settling, distal ice sils, soft-sediment deformation structures influence 3 Interstratified siltstone and very fine- to fine-grained sandstone, Open marine to restricted coastal settings (10–80 m wa- bioturbated and/or stratified with ripple-scale structures, typ- ter depth), deposition from suspension settling, and ically restricted and modest intensity bioturbation, dispersed gentle currents and waves, minor ice rafting calcareous fossils, soft-sediment deformation 4 Stratified, fine- to coarse-grained sandstone, dispersed to locally Open marine shoreface to upper offshore (10–40 m wa- concentrated small gravel, typically well sorted, with flat lami- ter depth), minor ice rafting of gravel nation, ripple cross-lamination, cross-bedding and hummocky cross-stratification, variable bioturbation, locally fossiliferous 5 Muddy fine-grained sandstone to sandy siltstone with dispersed Open marine conditions (10–60 m water depth) affected gravel, typically soft-sediment-deformed or unstratified, locally by hemipelagic fallout and supply of sediment from fossiliferous, rarely bioturbated ice-rafting 6 Delicately stratified, interlaminated fine- and coarse-grained silt- Distal glaciomarine conditions affected by hemipelagic stone, fine-grained sandstone and diamictite, dispersed gravel fallout, bottom currents and ice-rafting including diamictite clasts, some lamination rhythmic at sub- millimeter scale, soft-sediment and brittle deformation 7 Stratified diamictite, generally clast-poor, muddy to sandy ma- Open, proximal glaciomarine conditions, in distal por- trix, commonly fossiliferous tions of grounding-line fans 8 Massive diamictite, clast-poor to clast-rich and typically sandy Ice-proximal glaciomarine (grounding-line fan) to sub- matrix, some alignment of clast long-axes, shear structures in- glacial environments cluding boxwork fracturing and brecciation 9 Interbedded conglomerate and sandstone, thin units typically in- Ice-proximal environments in the presence of meltwa- terbedded with other coarse-grained facies, locally fossiliferous ter, various possible depositional settings 10 Vesicular, porphyritic, basaltic lava Primary sheet flows erupted from a proximal vent or fissure 11 Volcaniclastic breccias, monomict, composed of basaltic clasts in Autobrecciation of subaerially erupted lavas basaltic matrix 12 Pyroclastic lapilli tuff, composed of glassy juvenile clasts of Settling of pyroclasts through the water column coarse ash to lapilli size 13 Volcanic sedimentary breccias and sandstones, comprising tex- Reworking of volcaniclastic deposits in shallow marine turally unmodified to modified juvenile volcanic clasts and waters grains, local ripple cross-lamination

Estimates of formative water depths are based on analogy with calculations of Dunbar et al. (2008), and on contained fossil assemblages. cross-sectional representation of a complete glacial ad- coarsening upward succession or more complex interval vance–retreat cycle, see figure 8 of Dunbar et al. (2008). of interbedded mudrocks and sandstones formed dur- The internal facies geometry of sequences was summa- ing relative sea level highstand, and in many cases coin- rized by Dunbar et al. (2008) and Fielding et al. (2008a) as ciding with a clinoform set evident in seismic reflection showing a basal lowstand/transgressive (glacial maxi- data (Figures 2 & 3). The reservoir potential of the var- mum to early retreat) diamictite or distal equivalent de- ious lithofacies (Table 1) is limited by the mud-rich na- posit, overlain by a fining-upward succession formed ture of most lithofacies and their generally low porosity during transgression and glacial retreat, and in turn by a and (inferred) permeability. However, at various lev- Reservoir potential of sands formed in glaciomarine environments 5

Figure 2. An idealized genetic stratigraphic sequence from the Cenozoic succession of southern McMurdo Sound, showing vertical stacking of lithologies, and interpretation in terms of paleobathymetry, depositional environment and systems tracts. Facies are as in Table 1. See Figure 5 for key to symbols used. els within the succession are well-sorted, clean sands/ vals of Oligocene to Miocene age (Figure 4). They were sandstones that are internally massive, flat/low-an- interpreted as the deposits of coastal and nearshore ma- gle/hummocky cross-stratified, or cross-bedded. Some rine environments, possibly deltas in many cases. The show varying degrees of faunal bioturbation. These are association of physical sedimentary structures suggests common in CIROS-1 (Barrett 1989; Fielding et al. 1997), current dominance with a wave and storm influence on CRP-1 and 2/2A (Fielding et al. 1998, 2000), and AND- the depositional process. Accordingly, it is suggested 2A (Fielding et al. 2008b; Passchier et al. 2012) in inter- that deltas were probably lobate in planform.

Figure 3. Cross-sectional diagram illustrating the depositional dip-parallel geometry of lithofacies in a typical genetic sequence from the Cenozoic succession of McMurdo Sound, Antarctica. Vertical sequences in ice-proximal locations are dominated by diamictite (blue), which overlie the basal sequence boundary. Distally, basal diamictite units thin and pass ultimately into sandy conglomerate (orange). The basal unit is overlain by an interval of sandy mudstone (olive brown) with variable interbedded sandstone (yellow) that fines upward into a condensed interval of fine-grained mudrock (grey). This in turn coarsens upward into a section of interbedded mudrocks and sandy mudrocks (olive brown) and sandstones (yellow). Ice-proximal sequences are typically thin (<10 m), incomplete in terms of systems tracts, condensed and strongly top-truncated, whereas those in more ice-distal locations are thicker (25–50 m), more complete and less top-truncated. SB, sequence boundary; TST, transgressive systems tract; MFZ, maximum flooding zone; HST, highstand systems tract (little or no lowstand systems tract deposits are thought to be preserved above the basal sequence boundary). 6 Fielding et al. in Glaciogenic Reservoirs and Hydrocarbon Systems (2012) Reservoir potential of sands formed in glaciomarine environments 7

Tabular beds of texturally submature to mature Reservoir quality sands in the Oligocene and Miocene sand/sandstone are interpreted to occur primarily in succession the highstand systems tracts of genetic stratigraphic sequences (e.g. Fielding et al. 2000, 2011), where they are Several sandstone beds from AND-2A were sub- interbedded with more mud-rich sandstones, sandy jected to petrographic examination in order to obtain mudstones and mudstones formed in generally ice-dis- quantitative estimates of porosity, to determine porosity tal, nearshore marine and coastal settings (Figs 5–7). types (and hence likely permeability ranges), and from Lesser development of generally more mud-rich sand- the framework grain composition to assess the potential stones was also noted in the Transgressive Systems for secondary porosity development in the subsurface. Tract of some sequences (Figure 5). These muddy sands Point-counted determination of porosity was facilitated were interpreted as turbidity flow deposits by Howe et by having thin-section billets impregnated with blue- al. (1998). As a general observation, the greatest abun- dyed resin. Samples from the same depths were also dance of texturally mature, potential reservoir sands subjected to granulometric analysis using a laser dif- lies in genetic sequences and successions that show ev- fraction particle sizer (Malvern Mastersizer 2000E) fol- idence of only distal or minimal glacial influence. They lowing disaggregation (Blackstone 2009), and grain-size also occur interbedded with thick diamictites in places, spectra were plotted to illustrate the degree of textural however, and so were also formed during periods of maturity (Figure 9). more austere climate. Sands from AND-2A displayed a wide range in During the drilling of at least two cored holes (CRP- mud content and therefore in sorting (Table 2). Sands 2A and AND-2A: Cape Roberts Science Team 1999; Har- that formed during the post-glacial retreat and atten- wood et al. 2008), penetration of clean, porous sand beds dant relative sea level rise of cycles (Transgressive Sys- led to significant loss of drilling fluid circulation, sand tems Tract) are, in general, less well-sorted and con- production into the well bore, and influx of large vol- tain greater proportions of mud (Table 2). Those from umes of low-temperature groundwater. Downhole log- highstand systems tract contexts, are, in contrast, better ging operations indicated increased porosity through an sorted and contain generally modest amounts of mud. unconsolidated sand at 143–155 mbsf in CRP-2A (Cape Sands examined from CRP-1 and AND-2A showed Roberts Science Team 1999). During the drilling of AND- modest levels of compaction, consistent with the depths 2A in 2007, an interval of unconsolidated sand with some from which they were recovered. Dominant grain–grain gravel was encountered at 341.71 mbsf, which flowed contacts are floating and point contacts, with lesser long into the wellbore and migrated 27 m and later 48 m up contacts and virtually no interlocking or sutured con- inside the core barrel and drill pipe, requiring remedial tacts evident Figure 9). This typically minimal level of operations to stabilize the hole (Falconer et al. 2008). The compaction has facilitated the preservation of signifi- sand bed was found to be saturated with brine at high cant amounts of primary intergranular porosity in many pressure, which evidently facilitated sand production. samples (Figure 9). Most samples also showed signifi- The significance of these events for reservoir evaluation is cant proportions of microporosity between smaller (silt that they demonstrated that porous and permeable sand and clay grade) particles. The relative proportion of beds are indeed preserved within the glaciomarine suc- macroporosity and microporosity is evidently a func- cession of the VLB and are effectively sealed by enclos- tion of the mud content of the sample, sand-rich sam- ing lithologies. Figure 8 shows the thickness distribution ples showing a dominance of macroporosity, and vice for the texturally submature to mature (clean) sand facies versa. Overall, estimated porosity values are high, sug- in the Cape Roberts Project cores CRP-1 and CRP-2/2A, gesting considerable reservoir potential for these sands CIROS-1, and AND-2A. These sands typically form a mi- and for analogous sands from other basins. Highest po- nor proportion of the vertical section, and are isolated rosity values correlate with highest quartz contents and within intervals of mudstones and sandy mudstones (e.g. lowest mud contents, and with a dominance of macro- Figure 5). porosity (Table 2, Figure 9).

← Figure 4. Representative core photographs of the clean sand facies in the Cenozoic succession of McMurdo Sound. (a) The interval 632.15–628.13 mbsf in CIROS-1 (Barrett 1989), showing a flat- and cross-stratified sandstone interval passing abruptly upward into a dolerite clast conglomerate. (b) The interval 217.47–214.47 mbsf in CRP-3, showing an interval of soft-sediment- deformed sandstone and mudstone containing a large dolerite lonestone (left column), overlain abruptly by a clean, composite sandstone unit that fines upward from pebbly medium-grained flat-stratified sandstone to fine-grained, soft-sediment-deformed sandstone. (c) Detail of medium-grained sandstone passing abruptly upward into ripple cross-laminated and flat-stratified, fine- grained sandstone, with locally developed climbing ripple cross-lamination and wave-influenced ripple cross-lamination, from 178.76–179.00 mbsf in AND-2A. (d) Detail of the interval 114.00–114.25 mbsf in AND-2A, showing low-angle cross-stratified (probable hummocky cross-stratified), fine-grained sandstone.e ( ) Detail of the interval 1013.68–1013.85 mbsf in AND-2A, showing pervasively ripple cross-laminated, fine-grained sandstone. 8 Fielding et al. in Glaciogenic Reservoirs and Hydrocarbon Systems (2012)

Figure 5. Summary graphic log of the AND-2A core showing age ranges of intervals in mil- lions of years, based on the Ar– Ar geochronology of Di Vin- cenzo et al. (2010) and age model presented by Acton et al. (2008). Also shown are positions of the two component intervals shown in greater detail in Figures 6 and 7. Red star denotes tephra dated at 17.39+0.11 Ma (Di Vincenzo et al. 2010). Key to symbols used in graphic logs is given. Reservoir potential of sands formed in glaciomarine environments 9

Figure 6. Graphic sedimentological log of the interval 650–600 mbsf in AND-2A (17.5–17.0 Ma: Di Vincenzo et al. 2010), showing the stratigraphic context of clean sand/sandstone beds within an interval spanning two genetic sequences. Locations of samples examined petrographically in this study are also shown. Wavy horizontal lines denote interpreted sequence boundaries, and numbers denote genetic sequences of Fielding et al. (2011). TST, transgressive systems tract; MFZ, maximum flooding zone; HST, highstand systems tract. Red dots denote the positions of samples taken for petrographic examination. See Table 1 for explanation of lithofacies and Figure 5 for key to symbols. 10 Fielding et al. in Glaciogenic Reservoirs and Hydrocarbon Systems (2012)

Figure 7. Graphic sedimentological log of the interval 1044–994 mbsf in AND-2A (20– 19.5 Ma: Di Vincenzo et al. 2010), showing a substantial ice-distal succession spanning five genetic sequences formed in relatively ice-distal environments, and incorporat- ing coarsening-upward units interpreted as deltaic progradational cycles, and an interval of interpreted estuarine origin. Potential reservoir sands are developed at various levels within this succession. Wavy horizontal lines denote interpreted sequence boundaries, and numbers denote genetic sequences of Fielding et al. (2011). TST, transgressive systems tract; MFZ, maximum flooding zone; HST, highstand systems tract. See Ta- ble 1 for explanation of lithofacies, and Figure 5 for key to symbols. Reservoir potential of sands formed in glaciomarine environments 11

diagenetic modification in the subsurface. Downhole logging operations in these holes indicated evidence of significant (cold) fluid flow where the typically unconsolidated, permeable sand horizons were encountered (Cape Roberts Science Team 1999, 2000; Harwood et al. 2008).

Diagenetic insights from isotope geochemistry of sediment and pore water

ANDRILL is the first geological drilling program to include pore water analysis as part of routine coring efforts in shallow marine embayments around Ant- arctica. During recovery of the AND-2A core, 21 samples of pore water were extruded from whole-round core sections and subjected to a suite of isotopic and geochemical analyses (Panter et al. 2008). The pore water salinity profile revealed a fivefold increase downcore, from seawater values in the shallowest sample to 198 PSU (practical salinity units; c. 5.6 × sea water values) in the deepest sample. The convex-upward shape of the downcore trend indicates that the upper c. 200 m of the sedimentary succession contains connate seawater and that strata below that depth are saturated with brine (Figure 10a). Patterns in major element concen- trations suggest that the brine is a by-product of progressive freezing of seawater along the margins of ad- vancing ice sheets (Frank et al. 2010). Pore water δ18O values decrease rapidly downcore, from seawater-like values at the top to a value of –10.5‰ relative to the Figure 8. Cumulative thickness distribution for discrete, clean VSMOW (Vienna Standard Mean Ocean Water) scale sand intervals in McMurdo Sound drillcores. by 50 mbsf (Figure 10b). Given the high salinity of the pore water, low δ18O values cannot easily be attributed to the input of 18O-depleted glacial meltwaters. How- ever, removal of fresh water as ice and precipitation

The framework grain composition of Cenozoic of mirabilite (Na2SO4 ∙ 10H2O) during the progressive sands from the southern VLB shows an abundance of freezing of seawater are known to result in modest de- labile constituents that would in other situations be ex- creases in the δ18O value of the residual brine (Stewart pected to have been significantly diagenetically mod- 1974). Another factor at play in AND-2A is the alter- ified during burial. That such an abundance of volca- ation of volcanic glass, which is abundant throughout nic lithic grains, plagioclase and alkali feldspars, and the core (Panter et al. 2008). Petrographic observations in places pyroxene and other ferromagnesian min- reveal that the glass is undergoing devitrification and eral grains are preserved suggests a subsurface envi- zeolitization, processes known to cause strong 18O de- ronment in which the rate of diagenetic reactions is pletions in pore water (Garlick & Dymond 1970; Law- significantly slowed. This may in large part be due to rence et al. 1979 ). The δ13C values of dissolved inor- the low subsurface temperatures in the region facili- ganic carbon (DIC) in the pore water range from –4 to tated by active flow of cold groundwater, despite the –24‰ relative to the VPDB (Vienna Pee Dee Belemnite) magmatically active rift context of the VLB (Esser et al. standard and suggest that remineralization of organic 2004). Recorded bottomhole temperatures of 24.0 °C at matter has contributed 13C-depleted carbon to the DIC 672 mbsf in CIROS-1, 17.2 °C at 624 mbsf in CRP-2A, pool (Figure 10c). 23.0 °C at 870 mbsf in CRP-3 and 56.9 °C at 1138 mbsf Secondary calcite occurs throughout AND-2A, but is in AND-2A (Barrett 1989; Cape Roberts Science Team highly variable in abundance and distribution (Fielding 1999, 2000; Harwood et al. 2008) suggest that low tem- et al. 2008b). Most of the secondary calcite does not occur peratures are mainly responsible for slowing rates of as cement, but rather as a microcrystalline to microgran- 12 Fielding et al. in Glaciogenic Reservoirs and Hydrocarbon Systems (2012)

Figure 9. Representative photomicrographs of studied samples from AND-2A, together with the relevant grain-size spectrum derived from laser diffraction particle size analysis. Grain-size spectra show 5% error limits. Areas of blue correspond to porosity in the photomicrographs. See Figure 6 for location and context of samples. (a) Well-sorted highstand sandstone from 621.69 mbsf, showing a tightly constrained grain-size spectrum. A significant proportion of the visible porosity in this specimen is microporosity within and between volcanic lithic grains. (b) Moderately to poorly sorted highstand sandstone from 631.61 mbsf, showing an extensive fine tail to the grain-size spectrum. This sample shows limited macroporosity, extensive microporosity and internal heterogeneity. (c) Well-sorted highstand sandstone from 642.02 mbsf, showing tightly constrained grain-size spectrum. Note the quartz-rich character of this sample, interpreted as resulting from extensive working by marine currents and waves. (d) Poorly sorted transgressive sandstone from 648.49 mbsf, showing extensive fine tail to the grain-size spectrum, and extensive mud-grade material in part derived from volcanic lithic grains. ular replacement phase in the matrix of fine-grained li- sured on bulk samples (Gui 2009), are highly variable, thologies. Although most sandstone beds remain unlith- ranging from –3 to –14‰ relative to the VPDB scale ified and maintain high porosity (e.g. Figure 9), select (Figure 10b). Using the low-temperature calcite-water beds contain microgranular to blocky calcite in primary oxygen isotope fractionation equation of Kim & O’Neil pore space. The δ18O values of secondary calcite, mea- (1997) yields the range of possible δ18O values of diage- Reservoir potential of sands formed in glaciomarine environments 13

Table 2. Porosity and key textural data from studied samples of AND-2A drillcore (see Figure 5 for location of samples) Depth Depositional Porosity Porosity Quartz Mud Texture (mbsf) Setting (%) type (%) (%) 611.62 Highstand 27 Primary macro & 28 20 Moderate to poor sorting, delta microporosity heterogeneous, subangular to subrounded grains 614.71 Highstand 23 Predominantly 35 28 Moderate to poor sorting, delta microporosity in heterogeneous, subangular to the clay fraction subrounded grains 621.69 Highstand 29 Primary macro & 46 8 Well sorted, homogeneous, delta microporosity subangular to subrounded grains 627.48 Highstand 31 Predominantly 34 53 Moderate to poor sorting, delta microporosity in heterogeneous, subangular the clay fraction to rounded grains 631.61 Highstand 16 Primary macro & 30 33 Moderate to poor sorting, delta microporosity heterogeneous, subangular to subrounded grains 642.02 Highstand 41 Predominantly 63 8 Well sorted, homogeneous, delta primary subrounded to rounded macroporosity grains 648.49 Transgressive 24 Primary and 35 26 Poorly sorted, heterogeneous, reworking secondary macro subangular to subrounded & microporosity grains

Examples of both clean and more mud-rich sandstones were examined for comparative purposes.

netic calcite that would precipitate from Southern Mc- positions similar to those in AND-2A. In the absence of Murdo Sound (SMS) seawater and AND-2A pore water pore water data from these sites, the negative δ18O val- at temperatures of 0–25 °C (Figure 10b). These calcula- ues were interpreted to record the influence of glacial tions suggest that much of the calcite precipitated from meltwaters. The discovery of 18O-depleted brine dur- the evolved pore water (connate seawater and brine) ing recovery of AND-2A, however, may require revi- rather than from seawater or glacial meltwater. The δ13C sion of such interpretations. Documenting the regional values of secondary calcite tend to fall between values extent of brine in the deep subsurface of the Ross Sea of pore water DIC and SMS seawater (Figure 10c), sug- awaits the results of future drilling efforts. However, gesting carbon contributions from both seawater and models of cryogenic brine formation (Starinsky & Katz pore water DIC. Because carbon isotope fractionation is 2003) and groundwater flow beneath ice sheets advanc- negligible during the precipitation of calcite from water, ing atop unconsolidated sediment (Boulton et al. 1995) differences in composition between the pore water and suggest that although sites of brine formation are local- the secondary calcite suggests that the calcite formed be- ized to near ice sheet margins, subsurface flow can dis- fore the pore water evolved to its present carbon isoto- tribute brine over broad regions. pic composition, namely at some point after brine infil- tration but before organic matter remineralization had affected the dissolved carbon pool. Synthesis: Reservoir potential of VLB glaciomarine Observations from AND-2A suggest that diagenetic sands processes that have led to the lithification of fine-grained lithologies via calcitization/replacement have had lit- We have established that loosely consolidated to un- tle effect on coarser lithologies such as sandstone. Cal- consolidated, clean, porous and permeable sands are citization of the matrix in fine-grained lithologies is at- preserved within parts of the Cenozoic succession of the tributed to interaction with the cryogenic brine that now VLB. The stratigraphic context of these beds and their saturates the sediments. Previous work on CRP-1 (Baker properties can be used to derive some generalizations & Fielding 1998) and CRP-3 (Aghib et al. 2003) docu- about reservoir sand distribution in the VLB. From these mented the presence of secondary calcite with petro- conclusions, some useful insights may be gained into graphic characteristics, distributions and isotopic com- the reservoir potential of glaciomarine sands in general. 14 Fielding et al. in Glaciogenic Reservoirs and Hydrocarbon Systems (2012)

Figure 10. Geochemical and isotopic data from pore water and sediment in AND-2A. (a) Salinity (in PSU) calculated from chlo- rinity values. (b) Oxygen isotope data from pore water (VSMOW scale) and calcite cement (VPDB scale). Shaded fields show the range of calcite δ18O values in VPDB that would precipitate from pore water (grey field) and southern McMurdo Sound seawater (blue field) between 0° and 25 °C. Values were calculated using the low-temperature calcite-water oxygen isotope fractionation equation of Kim & O’Neil (1997). (c) Carbon isotope data from pore water (DIC, dissolved inorganic carbon), calcite cement and organic matter.

Sands with reservoir potential are interpreted to have gests that such sands were dispersed some tens of kilo- formed in coastal (probably deltaic) to nearshore marine meters offshore into the western VLB during the Early (delta front/shoreface) settings during ice minima asso- Miocene (Figure 3). The thickness distribution presented ciated with sea-level highstands. The greatest develop- in Figure 8 suggests that although generally <5 m thick, ment of clean sand facies appears from available data to potential exists for discrete reservoir intervals up to at have been during the Oligocene period (Main Rift phase least 25 m thick. of basin history), with less abundant development dur- The reservoir quality sands are mainly in discrete ing the Miocene, when the climatic regime is interpreted beds enclosed by more mud-rich lithologies, which in to have become more polar. The greatest abundance and the VLB at least appear to have some sealing capability. thickness of these sands has also been in holes drilled They are best-developed in the highstand systems tracts close to the present coastline, which suggests that sands of genetic sequences that are relatively complete and will thin, pinch out and become more mud-rich east- least top-truncated. Such sequences in the VLB are in- ward into the basin from the western (Transantarctic terpreted elsewhere (Fielding et al. 2000, 2011; Powell et Mountains) basin margin (Figure 3). Although no data al. 2000; Powell & Cooper 2002) to record time intervals on the plan geometry of these sands are available, lobate characterized by cold temperate to subpolar glaciations plan shapes are likely given the facies interpretation. involving significant volumes of meltwater runoff. This The location of the AND-2A drillhole (Figure 1) sug- situation contrasts with successions of diamictite-dom- Reservoir potential of sands formed in glaciomarine environments 15 inated, incomplete, strongly top-truncated sequences tative of climatic regimes less austere than today’s cold that are interpreted to record periods of more austere polar regime, in which substantial volumes of meltwater climate, when there was little or no meltwater runoff were released from glacier termini, allowing establish- into the VLB (e.g. McKay et al. 2009). The latter case typ- ment of substantial deltas and other sand-prone coastal ifies successions 14.5 to c. 5 and <3 million years in age, systems that dispersed texturally mature sands into the whereas the sections with greater reservoir potential in basin during times of glacial minima and sea-level high- the VLB are older than 14.5 million years old, and po- stand. Sands encountered in several wells have main- tentially also 5 to 3 million years in age. tained high porosity at depth in part due to relatively If the glaciomarine sands of the VLB are representa- modest diagenetic modification, despite the chemically tive of glaciomarine sands in general, then such sands reactive nature of their constituent framework grains. could hold considerable reservoir potential in analo- This is in turn influenced by the low temperatures of gous, hydrocarbon-prospective basins elsewhere in the groundwaters in the subsurface of the VLB. Some sands world. In addition to the potential for structural entrap- were so poorly consolidated that they flowed into the ment (antiforms, fault offsets), significant potential also well bore during coring operations at several hundred exists for stratigraphic traps formed by updip pinchout meters below sea floor. Accordingly, we submit that the of sand bodies in a direction that could be predicted data and observations presented herein may be useful from a knowledge of basin geometry. Data herein could as an analogue for glaciomarine successions elsewhere be used to predict reservoir quality in as yet unexplored that are targets for hydrocarbon exploration, and sug- basins. They could also be used to help evaluate addi- gest that the reservoir potential of glaciomarine suc- tional potential in currently active areas of petroleum cessions in paleo-ice-proximal settings should not be exploration such as the Paleozoic basins of North Af- dismissed. rica and the Middle East and the Late Paleozoic basins of South America (Potter et al. 1995; Le Heron et al. 2009; Acknowledgments — This paper is based on work sup- López-Gamundí 2010). ported by the National Science Foundation under Coopera- In the Upper Carboniferous Itarare Group of the tive Agreement No. 0342484 through sub-awards adminis- Paraña Basin in Brazil, high-quality glaciomarine sand- tered by the ANDRILL Science Management Office (SMO) stone reservoirs occur most abundantly in the highstand at University of Nebraska–Lincoln as part of the ANDRILL systems tract of a thick (100 m+) glacial–deglacial cycle US Science Support Program. We thank the SMS Drillsite within the Villa Velha Sandstone of the Campo Mourão and Science teams for their efforts in acquiring and char- Formation (Vesely et al. 2007). Although somewhat acterizing the AND-2A core, and SMO staff for adminis- trative support. The ANDRILL Program is a multinational thicker than most sequences described from the VLB, collaboration between the Antarctic programs of Germany, this unit strongly resembles the sequences of the VLB in Italy, New Zealand and the United States. Antarctica New terms of vertical facies stacking (Vesely et al. 2007, their Zealand is the project operator and developed the drilling Figure 4). Available porosity/permeability data from system in collaboration with A. Pyne at Victoria University subsurface expressions of this interval are also consis- of Wellington and Webster Drilling and Exploration Ltd. tent with the interpretations presented here. Levell et al. Antarctica New Zealand supported the drilling team at (1988), Melvin & Sprague (2006) and Melvin et al. (2010) Scott Base; Raytheon Polar Services Corporation supported describe hydrocarbon reservoirs of interpreted progra- the science team at McMurdo Station and the Crary Sci- dational delta origin from the Late Paleozoic succes- ence and Engineering Laboratory. The ANDRILL Science sions of the Arabian Peninsula, although the receiving Management Office at the University of Nebraska–Lincoln basins are there characterized as lakes rather than ma- provided science planning and operational support. Scien- rine water bodies. Nonetheless, the facies closely resem- tific studies are jointly supported by the US National Sci- ble those found in the McMurdo Sound cores and can be ence Foundation (NSF), NZ Foundation for Research, Sci- considered as broadly analogous. No doubt the future ence and Technology (FRST), the Italian Antarctic Research will reveal further instances of oil and gas production Program (PNRA), the German Research Foundation (DFG) from glaciomarine sandstone deposits. and the Alfred Wegener Institute for Polar and Marine Re- search (AWI). We thank the Special Publication Editors for the invitation to participate in this project, an anonymous Conclusions individual, and editor M. Huuse for constructive reviews of the manuscript. Porous and permeable sands were unexpectedly encountered during the recovery of continuous drillcores References through Cenozoic successions of the VLB, Antarctica.

These sands form parts of cyclically stacked genetic se- Acton, G., Crampton, J. et al. 2008. Preliminary integrated quences that bear a strong imprint of their glaciogenic chronostratigraphy of the AND-2A core, ANDRILL South- origin. They are best developed in sequences represen- ern McMurdo Sound Project Antarctica. In: Harwood, D. 16 Fielding et al. in Glaciogenic Reservoirs and Hydrocarbon Systems (2012)

M., Florindo, F., Talarico, F. & Levy, R. H. (eds.) Studies CAPE ROBERTS SCIENCE TEAM 1998. Initial Report on from the ANDRILL. Southern McMurdo Sound Project, Ant- CRP-1, Cape Roberts Project, Antarctica. Terra Antarctica, arctica. Terra Antarctica, 15, 211–220. 5, 1–187. Aghib, F. S., Fielding, C. R. & Frank, T. D. 2003. Carbonate CAPE ROBERTS SCIENCE TEAM 1999. Studies from the diagenesis of the Cenozoic sedimentary succession from Cape Roberts Project, Ross Sea, Antarctica: initial report on the CRP-3 core, Ross Sea, Antarctica. Terra Antarctica, 10, CRP-2/2A. Terra Antarctica, 6, 1–173. 27–37. CAPE ROBERTS SCIENCE TEAM 2000. Studies from the Cape Baker, J. C. & Fielding, C. R. 1998. Diagenesis of glacimarine Roberts Project, Ross Sea, Antarctica: initial report on CRP- Miocene strata in CRP-1, Antarctica. Terra Antarctica, 5, 3. Terra Antarctica, 7, 1–209. Cooper, A. K., Davey, F. J. & 647–653. Behrendt, J. C. 1987. Seismic stratigraphy and structure of Barrett, P. J. (ed.) 1986. Antarctic Cenozoic history from the the Victoria Land Basin, western Ross Sea, Antarctica. In: MSSTS-1 borehole, McMurdo Sound. New Zealand Depart- Cooper, A. K. & Davey, F. J. (eds.) The Antarctic Continental ment of Scientific and Industrial Research Bulletin, 237, 174 pp. Margin: Geology and Geophysics of the Western Ross Sea. Cir- Barrett, P. J. (ed.) 1989. Antarctic Cenozoic history from the cum-Pacific Council for Energy and Mineral Resources, CIROS-1 drillhole, McMurdo Sound. New Zealand Depart- Earth Science Series, Houston, TX, 5B, 27–65. ment of Scientific and Industrial Research Bulletin, 245, 254 Craig, J., Thurow, J. W., Thusu, B., Whitham, A.G. & Abutar- pp. ruma, Y. (eds.) 2009. Global Neoproterozoic Petroleum Sys- Barrett, P. J. 2007. Cenozoic climate and sea level history from tems: The Emerging Potential in North Africa. Geological So- glacimarine strata off the Victoria Land coast, Cape Rob- ciety, London, Special Publications, 326. erts Project, Antarctica. In: Hambrey, M. J., Christoffersen, Davey, F. J. & Brancolini, G. 1995. The Late Mesozoic and Ce- P., Glasser, N. F. & Hubbart, B. (eds.) Glacial Processes and nozoic structural setting of the Ross Sea region. In: Cooper, Products. International Association of Sedimentologists, A. K., Barker, P. F. & Brancolini, G. (eds.) Geology and Seis- Special Publications, 39, 259–287. mic Stratigraphy of the Antarctic Margin, Antarctic Research Barrett, P. J. & Hambrey, M. J. 1992. Plio-Pleistocene sedi- Series, 68. American Geophysical Union, Washington, DC, mentation in Ferrar Fiord, Antarctica. Sedimentology, 39, 167–182. 109–123. Di Vincenzo, G., Bracciali, L., Del Carlo, P., Panter, K. & Roc- 40 39 Barrett, P. J. & Treves, S. 1981. Sedimentology and petrology chi, S. 2010. Ar– Ar dating of volcanogenic products of core from DVDP-15, Western McMurdo Sound. In: Mc- from the AND-2A core (ANDRILL Southern McMurdo Ginnis, L. D. (ed.) Dry Valley Drilling Project, Antarctic Re- Sound Project, Antarctica): correlations with the Erebus search Series, 33. American Geophysical Union, Washing- Volcanic Province and implications for the age model of ton, DC, 281–314. the core. Bulletin of Volcanology, 72, 487–505. Bartek, L. R., Henrys, S. A., Anderson, J. B. & Barrett, P. J. 1996. Dunbar, G. D., Naish, T. R., Barrett, P. J., Fielding, C. R. & Seismic stratigraphy of McMurdo Sound, Antarctica: im- Powell, R. D. 2008. Constraining the amplitude of Late plications for glacially influenced early Cenozoic eustatic Oligocene bathymetric changes in western Ross Sea dur- change? Marine Geology, 130, 79–98. ing orbitally-induced oscillations in the East Antarctic Ice Sheet: 1. Implications for glacimarine sequence strati- Barwis, J. H., McPherson, J. G. & Studlick, J. R. J. (eds.) 1990. graphic models. Palaeogeography, Palaeoclimatology, Palaeo- Sandstone Petroleum Reservoirs. Springer-Verlag, New York. ecology, 260, 50–65. Blackstone, B. A. 2009. Mid-Miocene stratigraphic succession of Esser, R. E., Kyle, P. R. & MacIntosh, W. C. 2004. 40Ar/39Ar the Victoria Land Basin in the AND-2A drillcore from South- dating of the eruptive history of Mount Erebus, Antarctica: ern McMurdo Sound, Antarctica: sequence stratigraphy, palaeo- volcano evolution. Bulletin of Volcanology, 66, 671–686. bathymetry and reservoir development. MSc thesis, University of Nebraska– Lincoln. Falconer, T., Pyne, A., Wilson, D., Levy, R. H., Nielsen, S. & Petrushak, S. 2008. Operations overview for the ANDRILL Boulton, G. S., Caban, P. E. & van Gijssel, K. 1995. Ground- Southern McMurdo Sound Project, Antarctica. In: Har- water flow beneath ice sheets: Part I – large-scale patterns. wood, D. M., Florindo, F., Talarico, F. & Levy, R. (eds.) Quaternary Science Reviews, 14, 545–562. Studies from the ANDRILL Southern McMurdo Sound Project, Brancolini, G., Cooper, A. K. & Coren, F. 1995a. Seismic facies Antarctica. Terra Antarctica, 15, 41–48. and glacial history in the Western Ross Sea (Antarctica). In: Fielding, C. R., Woolfe, K. J., Purdon, R. G., Lavelle, M. & Cooper, A. K., Barker, P. F. & Brancolini, G. (eds.) Geology Howe, J. A. 1997. Sedimentological and stratigraphical re- and Seismic Stratigraphy of the Antarctic Margin. Antarctic evaluation of the CIROS-1 core, McMurdo Sound, Antarc- Research Series, 68. American Geophysical Union, Wash- tica. Terra Antarctica, 4, 149–160. ington, DC, 209–233. Fielding, C. R., Woolfe, K. J., Howe, J. A. & Lavelle, M. 1998. Brancolini, G., Busetti, M. et al. 1995b. Descriptive text for the Sequence stratigraphic analysis of CRP-1, Cape Roberts seismic stratigraphic atlas of the Ross Sea, Antarctica. In: Project, McMurdo Sound, Antarctica. Terra Antarctica, 5, Cooper, A. K., Barker, P. F. & 353–361. Brancolini, G. (eds.) Geology and Seismic Stratigraphy of the Ant- Fielding, C. R., Naish, T. R., Woolfe, K. J. & Lavelle, M. A. 2000. arctic Margin, Antarctic Research Series, 68. American Geo- Facies analysis and sequence stratigraphy of CRP-2/2A, physical Union, Washington, DC, A271–A286. McMurdo Sound, Antarctica. Terra Antarctica, 7, 323–338. Reservoir potential of sands formed in glaciomarine environments 17

Fielding, C. R., Naish, T. R. & Woolfe, K. J. 2001. Facies archi- Ordovician glacial succession, Illizi Basin, Algeria. Journal tecture of the CRP-3 drillhole, Victoria Land Basin, Antarc- of Petroleum Geology, 25, 297–324. tica. Terra Antarctica, 8, 217–224. Howe, J. A., Woolfe, K. J. & Fielding, C. R. 1998. Lower Mio- Fielding, C. R., Henrys, S. A. &Wilson, T. J. 2006. Rift history cene glacimarine gravity flows, Cape Roberts Drillhole-1, of the western Victoria Land Basin: a new perspective Ross Sea, Antarctica. Terra Antarctica, 5, 393–399. based on integration of cores with seismic reflection data. Kim, S.-T. & O’Neil, J. R. 1997. Equilibrium and nonequilib- In: Futterer, D. K., Damaske, D., Kleinschmidt, G., Miller, rium oxygen isotope effects in synthetic carbonates. Geochi- H. & Tessensohn, F. (eds.) Antarctica: Contributions to Global mica et Cosmochimica Acta, 61, 3461–3475. Earth Sciences. Springer-Verlag, Berlin, 309–318. Lawrence, J. R., Drever, J. I., Anderson, T. F. & Brueckner, H. Fielding, C. R., Whittaker, J., Henrys, S. A., Wilson, T. J. & 1979. Importance of volcanic alteration in the sediments Naish, T. R. 2008a. Seismic facies and stratigraphy of the of Site 323: Chemistry 18O/16O and 87Sr/86Sr. Geochimica et Cenozoic succession in McMurdo Sound, Antarctica: Cosmochimica Acta, 43, 537–588. implications for tectonic, climatic and glacial history. Pal- Le Heron, D. P., Sutcliffe, O., Bourgig, K., Craig, J., Visentin, aeogeography, Palaeoclimatology, Palaeoecology, 260, 8–29. C. & Whittington, R. 2004. Sedimentary architecture of Up- Fielding, C. R., Atkins, C. B. et al. ANDRILL-SMS SCIENCE per Ordovician tunnel valleys, Gargaf Arch, Libya: impli- TEAM 2008b. Sedimentology and stratigraphy of the cations for the genesis of a hydrocarbon reservoir. GeoAr- AND-2A core, ANDRILL Southern McMurdo Sound Proj- abia, 9, 137–160. ect, Antarctica. In: Harwood, D. M., Florindo, F., Talarico, Le Heron, D. P., Craig, J. & Etienne, J. L. 2009. Ancient glacia- F., Levy, R. H. (eds.) Studies from the ANDRILL, Southern tions and hydrocarbon accumulations in North Africa and McMurdo Sound Project, Antarctica. Terra Antarctica, 15, the Middle East. Earth-Science Reviews, 93, 47–76. 77–112. Levell, B. K., Braakman, J. H. & Rutten, K. W. 1988. Oil-bearing Fielding, C. R., Browne, G. H. et al. 2011. Stratigraphic cyclicity sediments of Gondwana glaciation in Oman. American As- from the ANDRILL AND-2A drillcore, Antarctica: a long- sociation of Petroleum Geologists Bulletin, 72, 775–796. term, ice-proximal record of Early to mid-Miocene climate, sea-level and glacial dynamism. Palaeogeography, Palaeocli- López-Gamundí, O. R. 2010. Transgressions related to the de- matology, Palaeoecology, 305, 337–351. mise of the Late Paleozoic Ice Age: their sequence stratigraphic context. In: López-Gamundí, O. R. & Buatois, L. A. Franca, A. B. & Potter, P. E. 1991. Stratigraphy and reservoir (eds.) Late Paleozoic Glacial Events and Postglacial Transgres- potential of glacial deposits of the Itarare´ Group (Carbon- sions in Gondwana. Geological Society of America, Special iferous–Permian), Paraná Basin, Brazil. American Associa- Paper, 468, 1–35. tion of Petroleum Geologists Bulletin, 75, 62–85. McGinnis, L. D. (ed.) 1981. Dry Valley Drilling Project, Antarctic Frank, T. D., Gui, Z. ANDRILL SMS SCIENCE TEAM 2010. Research Series, 33. American Geophysical Union, Wash- Cryogenic origin for brine in the subsurface of southern ington, DC. McMurdo Sound, Antarctica. Geology, 38, 587–590. McKay, R., Browne, G. et al. 2009. The stratigraphic signa- Galloway, W. E. & Hobday, D. K. 1996. Terrigenous Clastic Dep- ture of the late Cenozoic Antarctic Ice Sheets in the Ross ositional Systems, 2nd edn. Springer- Verlag, Berlin. Embayment. Geological Society of America Bulletin, 121, Garlick, G. D. & Dymond, J. R. 1970. Oxygen isotope exchange 1537–1561. between volcanic material and ocean water. Geological Soci- Melvin, J. & Sprague, R. A. 2006. Advances in Arabian stra- ety of America Bulletin, 81, 2137–2142. tigraphy: origin and stratigraphic architecture of glacio- Gui, Z. 2009. Origin of brines in Neogene sediments of the genic sediments in Permian–Carboniferous lower Unayzah Ross Sea, Antarctica: AND-2A core, ANDRILL South- sandstones, eastern central Saudi Arabia. GeoArabia, 11, ern McMurdo Sound Project. MSc thesis, University of 105–152. Nebraska–Lincoln. Melvin, J., Sprague, R. A. & Heine, C. J. 2010. From bergs to Hamilton, R. J., Luyendyk, B. P., Sorlien, C. C. & Bartek, L. R. ergs: the late Palaeozoic Gondwanan glaciations and its af- 2001. Cenozoic tectonics of the Cape Roberts Rift Basin and termath in Saudi Arabia. In: López-Gamundí, O. R. & Bua- Transantarctic Mountain Front, southwestern Ross Sea, tois, L. A. (eds.) Late Paleozoic Glacial Events and Postglacial Antarctica. Tectonics, 20, 325–342. Transgressions in Gondwana. Geological Society of America, Harwood, D. M., Florindo, F., Talarico, F. & Levy, R. (eds.) Special Paper, 468, 37–80. 2008. Studies from the ANDRILL Southern McMurdo Naish, T. R., Woolfe, K. J. et al. 2001. Orbitally induced oscilla- Sound Project, Antarctica. Terra Antarctica, 15, 1–235. tions in the East Antarctic Ice Sheet: direct evidence from Henrys, S. A., Wilson, T. J., Whittaker, J. M., Fielding, C. R., the Cape Roberts Drilling Project. Nature, 413, 719–723. Hall, J. & Naish, T. R. 2007. Tectonic history of mid-Mio- Naish, T. R., Powell, R. D. & Levy, R. (eds.) 2007. Studies from cene to present southern Victoria Land Basin, inferred the ANDRILL McMurdo Ice Shelf Project, Antarctica. Terra from seismic stratigraphy in McMurdo Sound, Antarctica. Antarctica, 14, 110–328. USGS and National Academies, USGS OF-2007-1047, short Naish, T. R., Powell, R.D. et al. 2009. Obliquity-paced Pliocene research paper 049; doi: 10.3133/of2007-1047.srp049. West Antarctic Ice Sheet oscillations. Nature, 458, 322–329. Hirst, J. P. P., Benbakir, A., Payne, D. F. & Westlake, I. R. 2002. Osterloff, P., Penney, R., Aitken, J., Clark, N. & Al-Husseini, Tunnel valleys and density flow processes in the Upper M. 2004. Depositional sequences of the Al Khlata Forma- 18 Fielding et al. in Glaciogenic Reservoirs and Hydrocarbon Systems (2012)

tion, subsurface Interior Oman. In: Al-Husseini, M. J. (ed.) logical and palaeoclimatic inferences. Terra Antarctica, 7, Carboniferous, Permian and Early Triassic Arabian Plate Stra- 313–322. tigraphy. Geo- Arabia (Special Publication), 3, 61–81. Robinson, P. H. 1983. Preliminary Report on the Lithostratigraphy Panter, K. S., Talarico, F. et al. ANDRILL-SMS SCIENCE and Sedimentary History of the Near Surface Glacigenic Drill- TEAM 2008. Petrologic and geochemical composition of Core Sediment of Eastern Taylor Valley, Antarctica. New Zea- the AND-2A core, ANDRILL Southern McMurdo Sound land Geological Survey Report G77. Project, Antarctica. In: Harwood, D. M., Florindo, F., Ta- Robinson, P. H. & Jaegers, A. 1984. The lithologic logs of East- larico, F., Levy, R. H. (eds.) Studies from the ANDRILL, ern Taylor Valley cores 2 to 13, South Victoria Land, Antarc- Southern McMurdo Sound Project, Antarctica. Terra Antarc- tica. New Zealand Geological Survey Report G89. tica, 15, 147–192. Scholle, P. A. & Spearing, D. (eds.) 1982. Sandstone Depositional Passchier, S., Browne, G., Field, B., Fielding, C. R., Krissek, Environments. American Association of Petroleum Geolo- L. A., Panter, K., Pekar, S. F. ANDRILL-SMS SCIENCE gists, Memoir 31. TEAM. 2012. Early and Middle Miocene Antarctic glacial Shepherd, M. 2009. Oil Field Production Geology. American As- history from the sedimentary facies distribution in AND- sociation of Petroleum Geologists, Memoir 91. 2A, Ross Sea, Antarctica. Geological Society of America Bulle- tin, 123, 2352–2365. Stagpoole, V. 2004. Bathymetry of the Ross Dependency and adjacent Southern Ocean. Institute of Geological and Nuclear Potter, P. E., Franca, A. B., Spencer, C. W. & Caputo, M. V. Sciences Geophysical Map 17, 1:5,000,000. 1995. Petroleum in glacially-related sandstones of Gond- wana: a review. Journal of Petroleum Geology, 18, 397–420. Starinsky, A. & Katz, A. 2003. The formation of natural cryogenic brines. Geochimica et Cosmochimica Acta, 67, Powell, R.D. & Cooper, J. M. 2002. A glacial sequence strati- 1475–1484. graphic model for temperate, glaciated continental shelves. In: Dowdeswell, J. A. & O’Cofaigh, C. (eds.) Glacier-Influ- Stewart, M. K. 1974. Hydrogen and oxygen isotope studies on enced Sedimentation on High- Latitude Continental margins. the McMurdo ice shelf, Antarctica. New Zealand Journal of Geological Society, London, Special Publications, 203, Geology and Geophysics, 18, 49–64. 215–244. Vesely, F. F., Rostirolla, S. P., Appi, C. J. & Kraft, R. P. 2007. Powell, R. D., Krissek, L. A. & Van der Meer, J. J. M. 2000. Late Paleozoic glacially related sandstone reservoirs Preliminary depositional environmental analysis of in the Paraná Basin, Brazil. American Associatio of Petroleum CRP2/2A, Victoria Land Basin, Antarctica: palaeoglacio- Geologists Bulletin, 91, 151–160.