Characterization of Channels in the Mungaroo Formation, Exmouth Plateau, North Carnarvon Basin, Australia
Published:December 01, 2012
- PDF LinkChapter PDF
Minken Jon, Miller Josh, 2012. "Characterization of Channels in the Mungaroo Formation, Exmouth Plateau, North Carnarvon Basin, Australia", New Understanding of the Petroleum Systems of Continental Margins of the World, Norman C. Rosen, Paul Weimer, Sylvia Maria Coutes dos Anjos, Sverre Henrickson, Edmundo Marques, Mike Mayall, Richard Fillon, Tony D’Agostino, Art Saller, Kurt Campion, Tim Huang, Rick Sarg, Fred Schroeder
Download citation file:
The late Triassic Mungaroo Formation is a prolific gas-condensate reservoir along the landward margins of the Exmouth Plateau (i.e., Gorgon and Rankin trends). Recent exploration drilling has stepped outboard of these trends into the Exmouth Plateau adding calibration to an area of sparse well control. Reservoir units were primarily channelized complexes of variable widths and orientations within a large fluvial-deltaic complex that formed a thick (>2.5 kilometer) succession. Channel complexes were attractive exploration targets and have been the reservoirs of recent discoveries.
Mungaroo channelized complexes were thick >10 meters, multistorey, and greater than 1 kilometer in width. Their depositional origins were considered either distributaries or channels within an incised valley. Reservoir characterization studies has provided ranges in the properties of the channels, internal reservoir architecture and expected facies distribution that allowed for gas initially in place (GIIP) and estimated ultimate recovery (EUR) calculations.
Recent exploration activity in WA-390-P, Exmouth Plateau, has discovered gas-condensate in the late Triassic Mungaroo Formation (Figs. 1 and 2). Initial exploration of the Exmouth Plateau occurred from 1979-1980 with the drilling of eleven exploration wells leading to several non-commercial gas discoveries and the discovery of the Scarborough gas field (Barber, 1988). In recent years, exploration in the Exmouth Plateau has experienced a renaissance, prompting the acquisition of large 3D seismic data sets and the drilling of numerous exploration wells. Recent activity has added new control points to an area of previous sparse well control, allowing for the refinement of the depositional model. The Mungaroo Formation has been well explored along the Rankin Trend, the landward margin of the Exmouth Plateau, as the primary reservoir in the Wheatstone, Gorgon, Pluto, and Goodwyn fields (Fig. 2) (Palmer et al., 2005; Seggie et al., 2007; Tilbury et al., 2009; and Swinburn, et al., 2011).
The Exmouth Plateau formed a broad continental platform in the northwest margin of Australia (Fig. 2). It is bounded by abyssal plains from the north to west and the Exmouth subbasin and the Rankin Trend to the south and east, respectively. Jurassic to Early Cretaceous rifting of Australia and greater India defined the Exmouth Plateau. Extension created northeast to southwest tending horsts, tilted fault blocks, and grabens that defined the main structural elements. The horsts and tilted fault blocks created three-way dip closures that formed important trapping geometries for hydrocarbons. Hydrocarbons were gas prone, and sourced from the deeply buried coals and carbonaceous shales of the Mungaroo Formation (Bussell et al., 2001).
The Mungaroo Formation was deposited during the Ladinian to Norian (Middle to Late Jurassic), formed a succession of strata in excess of 2.5 kilometers, and exploration discovered gas in the Carnian and the Norian (Fig. 1). Preservation of a significant stratigraphic succession was facilitated by gentle subsidence of the Exmouth Plateau by late Permian intracratonic crustal thinning (Exon and Buffler, 1992). Silliciclastic deposition ceased during the Rhaetian (Late Triassic) as the Lhasa block rifted off the northern Indian margin, causing a major flooding event that drowned the Mungaroo Delta and deposited a carbonate over the Exmouth Plateau (Longley et al., 2002).
In the late Triassic, the present day Exmouth Plateau formed the northern margin of the Gondwana landmass (Exon and Buffler, 1992). Deltaic complexes dominated this margin (Jablonski, 1997, Bradshaw et al., 1988, and Longley et al., 2002) (Fig. 3). The approximate position of the paleo-shoreline of the Tethys Ocean was towards the north to northwest. In general, at a specific stratigraphic level more marine conditions were expected towards the north and northwest and continental conditions towards the south. The position of the coastline should have varied drastically depending on regional allogenic and localized autogenic conditions (i.e., relative sea-level fluctuations, delta lobe avulsion, increased sediment supply, etc.).
Palynology data indicated occasional flooding surfaces within the succession, which served as key surfaces for well log correlations and paleogeographic reconstructions. Increases in the abundances of Dino-flagellate cysts were related to near-shore marine depositional environments and marine flooding. Near the top of the Mungaroo, where the deltaic complex was being transgressed and the sedimentological information indicated marine to marginal marine conditions the Dinoflagellate cysts became more prevalent (Fig. 4).
Within the subsurface, the fluviodeltaic deposits were preserved as a succession of interbedded thick-(>10 meters) and thin- (<10 meters) bedded sands, silts, shales, and coals. The seismically resolvable bodies were characterized by channel morphologies identified on well-log and seismic data (Fig. 5 and Fig. 6). The extensive channel networks were the targets of several exploration wells and formed the primary reservoirs for several key discoveries, such as Gorgon, Goodwyn, Wheatstone, and Pluto fields (Palmer et al.; 2005, Seggie et al., 2007; Tilbury et al., 2009; and Swinburn et al., 2011). For reservoir characterization studies, constraining the channels within a stratigraphic framework was essential for the development of depositional models, reservoir architecture, and ranges in reservoir properties for the input into geo-cellular models to accurately estimate gas initially in place (GIIP) and estimated ultimate recovery (EUR).
Channels were identified utilizing seismic amplitude extractions and well data (Fig. 5 and Fig. 6). Well logs, conventional, and sidewall core data demonstrated that channels were sand-prone, sharp-based, and fine upward. The internal architecture was multistorey and single-storey as interpreted on standard wireline logs and dipmeter data (stories sensu Friend et al., 1977). In amplitude extractions, the channel morphologies varied in width, sinuosity, and orientations ranged from northeast to southwest (Fig. 6). Only rarely, did the seismic data identify internal morphology within the channel or evidence of lateral migration. This was attributed to seismic resolution and paleogeographic position of the channels. Low gradient environments such as a delta plain were consistent with rectilinear channels and a paucity of evidence for lateral accretion (Miall, 2002; Payenberg and Lang, 2003).
In general, multistorey channels were thicker than single-storey channels and considered more economically significant; therefore they formed the primary focus of this study (Fig. 7). Individual channel stories ranged from 1-12 meters, with a mode of 2 meters. They can either coarsen or fine upwards. In some instances, the stories were organized into higher order coarsening and fining upward packages within the overall fining upward succession (Fig. 8). The higher order stacking patterns represented either autogenic or allogenic changes to the depositional system (Fig. 9).
The channel morphologies were commonly wider than 1 kilometer and had variable orientation and sinuosity (Fig. 9). They formed single, rectilinear channel bodies and rarely showed evidence of lateral accretion. In Figure 10, a sinuous channel having meander belts and a rectilinear channel were found in the same seismic amplitude extraction. These channels were unlikely to have been coeval but illustrate the difference between the morphologies. Channels having meander belts were estimated to account for less than 10 percent of the channels identified. Occasionally, in association with larger multistorey channels were small, sinuous channels and had orientations that ranged from parallel to perpendicular to the main channels. They have been interpreted as tidal channels, small flood plain channels, and/or tributaries depending on the depositional environment (Fig. 10).
Multistorey channels were found either clustered at a stratigraphic level or as individual sand bodies within the delta plain succession (Fig. 11). When the channels were found as isolated bodies, they were interpreted as distributary channels. The frequency of occurrence of these channels was attributed to avulsion from delta lobe switching. Delta lobe switching was a regularly process that was facilitated by channel avulsion for shorter, more efficient paths to the shoreline (Allen and Chambers, 1998). Following avulsion, an incisional channel formed across the delta top (Hoyal and Sheets, 2009) to a lowest surface of erosion, which was controlled by relative sea-level (Fig. 12). The depth of incision from the delta plain surface increased landward on the delta plain, where aggradation elevated the channel and delta plain (Fig. 12). Multistorey channels amassed regionally at a stratigraphic level were interpreted as incised valleys. Incised valleys formed as a result of a decrease in relative sea-level that was identifiable by a sequence boundary at the valley base (Van Wagoner, et al., 1990; and Zaitlin et al., 1994). Sequence boundaries were identified regionally by stratigraphic levels showing channel clustering, based on well-log correlations and seismic amplitude extractions (Shanley and McCabe, 1991).
Distributary channels were thick (>10 meters) isolated sand-bodies intercalated with a succession of thin interbedded delta plain sands, silts, shales, and coals (Fig. 11 and Fig. 13). The isolated nature of these channels made them difficult to correlate subregionally on well logs. Their morphologies were low sinuosity, 12 kilometers in width, and had no internal morphology (Fig. 13). Their significant thickness was associated with their paleogeographic position within the delta plain following avulsion. Near the apex of the delta, the incision associated with avulsion is expected to be deeper. The laterally extensive nature of the Mungaroo delta plain provided the potential for significant incision as the channel returns to the lowest surface of erosion (Fig. 3). Channel fill occurred over multiple stages and reflected a complex history of scouring and deposition. The stories may have recorded repeated avulsion cycles in more distal, higher-order channels as the channel base-level profile evolved throughout the avulsion cycle (Hoyal and Sheets, 2009).
Incised valleys were identified by stratigraphic levels of channel clustering associated with a decrease in relative sea-level and the formation of a sequence boundary (Van Wagoner, et al., 1990; Shanley and McCabe, 1991; and Zaitlin et al., 1994). Figure 14 is an example of channel clustering on well data and seismic amplitude extractions. Numerous valleys were recognizable having widths that varied from 300 m to almost 10 km. The morphologies and orientations were variable and younger channels erode into older channel deposits. These valleys fill were typically thicker than 10 meters, sand-prone, sharp-based, and in an overall fining-upward succession. Variation in thickness in the same channel may indicate terraces or local scouring within the valley (Well 4 and Well 5 in Fig. 14).
The magnitude of incision that was constrained within channel morphology in some instances required a decrease in relative sea-level (Figs. 15). In Figure 15, the incised valley fill was in excess of 100 meters (not de-compacted), and valley widths vary between 2 to 5 km. Internal heterogeneities appear to be recognized throughout the valley. Well A missed the valley margin by less than 1 km and showed no evidence of the nearby incision.
Multistorey fill in the incised valleys reflected both autogenic and allogenic controls. Differentiating the two mechanisms was problematic. During greenhouse conditions like the Triassic, autogenic climatic conditions varied sediment supply, potentially creating the observed high frequency stacking patterns (Blum, 2009). Alternatively, allogenic changes can create a multistorey fill by incision and channel widening through multiple higher order relative sea-level cycles (Blum and Garvin, 2010). Incision occurred during periods of accelerated relative sea-level fall, causing the channels to narrow, and during decelerated relative sea-level fall and rise the valley widened (Strong and Paola, 2008). Allogenic controls created and resulted in the abandonment of the valley, but the internal stories were considered to reflect a combination of both allogeneic and autogenic processes. In Figure 9, channel stories were organized into lower-order staking patterns identifiable by changes in paleoflow direction. This internal hierarchy may represent an autogenic change in the valley. For example, channel abandonment within the valley (Fig. 9, Fig. 14, and Fig. 15). In Figure 15, the recognition of an internal shale bed within all three wells may indicate an allogenic increase in base-level causing a flood-back into the valley.
Distributary channels were expected to be prevalent throughout the succession as a result of delta avulsion and their size dependent on their position relative to their paleo-coastline. Near the apex of the delta, where the trunk fluvial channel initially bifurcates, the lower order distributaries were expected to be their greatest thickness. Their widths varied from 1 - 2 km and reflected incision without appreciable widening (Fig. 13). Towards the delta front, after successive bifurcations, the distributaries narrowed and shallowed, reflecting the dispersal of the fluvial discharge through several channels. The reservoir potential of the distributary channels was expected to decrease after each successive bifurcation. Distributaries, lower down on the delta plain may have sand bodies isolated by shale abandonment plugs (Payenberg and Lang, 2003) and were subject to more marine and tidal conditions potentially creating heterogeneous reservoirs.
Incised valleys were found within the succession in association with other channel complexes. They were variable in geometry, as some formed broad valleys and others did not (Fig. 14 and Fig. 15). A complex history of erosion and widening in the incised valleys created valleys that that ranged from 300 meters to under 10 km (Fig. 14) in width. The wider and more deeply entrenched valleys formed significant reservoir containers for hydrocarbons. In the shown examples, the incised valleys were commonly wider than the recognized distributaries and have the potential to become more deeply incised. Incised valley deposits formed large potential reservoir containers, tend to be sand-prone, and deposited under high energy conditions, thus making them of favorable reservoir quality.
Multistorey channels formed important reservoirs in the Exmouth Plateau and Rankin Trend and were considered to be distributary channels and incised valleys. Distributary channels were commonly found as individual, thick channel bodies (>10 m) within the delta plain succession. Incised valleys were thick (>10 m), potentially wide (> 1 km), and associated with other channel complexes.
Distributary channels frequency of occurrences were associated with avulsion from delta lobe switching. With avulsion, the channels entrenched into the delta plain and were progressively back-filled during delta plain aggradation. These channels formed important reservoirs owing to sand-prone fill that was deposited under high depositional energy.
Incised valleys formed important reservoirs owing to their large reservoir size. The multistorey fill was attributed to a complex history of incision and widening controlled by autogenic and allogenic factors. High depositional energies were associated with the valley filling facies making them favorable reservoirs. In general, incised valleys were expected to form larger reservoirs than the entrenched distributaries owing to a complex history of erosion and widening of the valley.
The author thanks Hess Australia Corporation for permission to publish the work on the depositional systems of the Mungaroo Formation. This work includes substantial contributions from numerous geoscientists that have worked WA-390-P from exploration, technology, and development during the identification and characterization of the gas-condensate fields. I thank Phil Seligmann, Steve Massie, Dean Griffin, Yohan, Josh Miller, Phil Cox, Woody Prescott, Pat Boss, Mark Whelan, Tim Kirst, Richard Harmer and Toni Munckton for the discussions and insights on the Mungaroo Formation.
Figures & Tables
New Understanding of the Petroleum Systems of Continental Margins of the World
- Carnarvon Basin
- clastic rocks
- continental shelf
- deltaic environment
- elastic waves
- Exmouth Plateau
- fluvial environment
- incised valleys
- Indian Ocean
- Middle Triassic
- natural gas
- petroleum exploration
- reservoir rocks
- sedimentary rocks
- seismic attributes
- source rocks
- Upper Triassic
- well logs
- Western Australia
- Mungaroo Formation
- northern Carnarvon Basin