Geological Survey

Darling Basin - Geological Overview

The Darling Basin lies in western New South Wales, across an area of approximately 90,000 square kilometres. The Darling Basin contains over 8000 metres of mainly Palaeozoic sediments. It stretches from Broken Hill in the west, to near Cobar in the east. To the north, the basin extends as far as White Cliffs and Louth at the surface but probably underlies the Eromanga Basin and continues into the state of Queensland along basement structural lows. Southward its surface limits extend to Ivanhoe and Roto, continuing under the Murray Basin as the Wentworth Trough into Victoria.

With approximately 20 petroleum wells drilled in the basin, mostly during the 1960s and 1970s, and 1550 km of modern multi-fold seismic coverage, the Darling Basin represents one of the major frontier basinal regions of onshore Australia. New aeromagnetic data, acquired by the Department as part of the Discovery 2000 initiative, have greatly supplemented the existing seismic coverage and allow for the first time, a coherent data base from which a consistent structural framework for the Darling Basin can be established.

The initial phase of petroleum exploration was discouraged by the lack of shows, the likelihood of a gas-prone source and presence of a thick, red-bed dominated, organically lean, Late Devonian sequence. Renewed interest in the Darling Basin's prospectivity followed from favourable, albeit superficial, comparisons between the Darling Basin and the Adavale Basin in the state of Queensland, where commercial gas is produced at the Gilmore Gas Field. Additionally, new geochemical analyses of extracts collected from core holes and outcrop suggest the presence of at least one active petroleum system, responsible for generating oil in some regions and possibly the substantial quantities of gas found dissolved within artesian waters in the overlying, shallow Mesozoic sequences.

Geology

The Darling Basin contains over 8000 m of Late Silurian to Early Carboniferous (but predominantly Devonian) sediments contained in a number of distinct structural depressions. The nature of basement beneath the Darling Basin is largely unknown, but it is underlain in part by Lower Palaeozoics of the Kanmantoo Fold Belt and perhaps rifted slivers of Proterozoic. The majority of the basin sequences are covered with 100-200 metres of Cainozoic alluvium and minor thickness of Cretaceous Eromanga Basin sediments. Outcrop is limited, confined mainly to the east and west basin fringes or to sporadic occurrences coincident with regional basin structural highs. Because redbed related lithologies dominate the Late Devonian sequences, most source rock potential is confined to the Early Devonian and possibly older sequences beneath the Darling Basin. Earlier notions of marine-dominated Early Devonian and terrestrial dominated red-bed, Late Devonian, stratigraphic subdivisions have proven to be too simplistic. Red-bed deposition was contemporaneous with marine deposition, at least during a portion of the Early Devonian.

Regional magnetic and gravity coverage (coupled with available seismic data and outcrop geological mapping) shows that there is good correlation between the broad sedimentary depressions and Bouguer gravity lows. The principal depressions are the Blantyre Trough, Pondie Range Depression, Poopelloe Lake Troughs, Nelyambo, Neckarboo, Bancannia and Menindee Troughs. Intervening structural highs tend to be less well defined and include the Mt Jack High, Lake Wintlow High, Wilcannia High and the Wonaminta Block (Koonenberry High). Based on gravimetric response, depth-to-basement estimates indicate maximum sediment thicknesses of 8-12 km are attained within the Blantyre Trough with at least 3.5 seconds (8 km) of sediment interpreted on seismic lines across the area.

The present day architecture of the Darling Basin appears to be principally the result of structural deformation and erosion. Time-structure maps and other depth-to-basement maps are not necessarily indicative of palaeostructural relief at the time of Devonian deposition. Recent drilling, the revision of wire-line log correlations and addition of new biostratigraphic control, suggest that at least some of these depressions shared a common sequence stratigraphic framework and may have shared common depositional histories. The seismic also suggests some troughs are of remnant structural or depositional origins. For these reasons, the sediment deeps are referred to as troughs or depressions rather than depocentres, because there is evidence of considerable basin inversion. Seismic data, coupled with gravity and magnetic data, suggest strong structural control along virtually all the boundaries separating the regional structural elements. Regional seismic lines imply considerable, post-depositional (i.e. Mid Carboniferous Kanimblan Orogeny), deformation involving either wrenching or compression. There is a conspicuous absence of significant basin-forming tensional faulting. Rather, tensional faulting is restricted to crestal faults associated with major anticlines.

Prominent regional basement highs, such as the Lake Wintlow, Wilcannia and Mt Jack High, are wrench or thrust-bound. The poor quality of even modern, high (240) fold, seismic data recorded across these highs, including a lack of lateral continuity of reflectors, implies that these may be intensely deformed and unlikely to have retained significant trap integrity. Moreover, wells such as Mt Jack-1 and Mt Emu-1, sited on these highs, penetrated very indurated lithologies that further downgrade the prospectivity. In contrast, intervening depressions, such as the Blantyre, Pondie Range, Lake Poopelloe, Neckarboo and Bancannia Troughs, contain relatively undeformed sediments. Tied to available seismic control, computer modelling of the aeromagnetic data and downward continuation, enabled smaller scale structuring within these depressions to be mapped. Reverse faulting, thrusting and basin inversion are amongst the variety of structural styles recognised as characterising an overall compressional regime. Several broad structures, reminiscent of the Gilmore Gas Field in Queensland and capable of individually containing several hundred BCF of gas, have been delineated along a number of structural fairways.

Hydrocarbon shows

There is little in the way of direct hydrocarbon indications in the Darling Basin. The best indications have been found in the Bancannia Trough and in Scopes Range. Gas shows were reported in Bancannia South 1 well together with bitumen traces. Further south, at Little Topar, a stratigraphic bore encountered a small interval logged as 'oil stained'. Minor gas shows were also recorded in non-marine Devonian sandstones in Jupiter 1 and Bancannia North 1. Ignitable quantities of methane have been recorded from flowing water bores on the flanks of the Bancannia Trough.

A study by the Department has revealed that most water bores in the artesian basin in NSW are flowing gas to the surface. Gas seeps have also been recorded in mines at Cobar. As we understand it, gas has been reported from fractures and from bore holes within the underground workings. Most interestingly, they report the gas as being wet, with significant amounts of ethane, propane and butane, although a thorough analysis has yet to be made.

Gas is now produced from the Gilmore Gas Field in the Adavale Basin some 400 km to the north, a basin with which the Darling Basin shares many common features. The Department, as part of its water bore sampling program, has had the Gilmore Gas isotopes analysed. It has been found that several of the water bores in NSW have a similar isotope signature to that of the Gilmore gas. Gas analyses undertaken by the Australian Geological Survey Organisation (AGSO) indicate that the Gilmore Gas Field produces a dry gas (methane 76-88%, ethane less than 3% and non-hydrocarbon gases including nitrogen up to 21%) most likely sourced from the Middle Devonian Log Creek and Bury Limestone units, although some deep, crustal, contribution of methane may be involved.

As part of Discovery 2000, in 1996 the Department initiated a study of putative oil shows recorded within the basin. On the Kookenberry High where oil occurrences had been reported in the Cymbric Vale area, surface samples of the Cambrian aged First Discovery Limestone and Funeral Creek Limestone were analysed. Small amounts of extract were obtained from each of three samples from the First Discovery Limestone and whole extract and saturate fraction were analysed by Gas Chromatography (GC) for the Funeral Creek Limestone. In addition to the surface samples, extracts from samples retrieved from seven wells or mineral bores within the Darling Basin, were analysed. All yielded very low extracts and most suggested possible contamination with crude oil products; however, samples from Berangabah and the Kiri wells also may represent a mixture of both refined and natural hydrocarbons.

Three samples from Berangabah 1 were solvent extracted. The overall picture of the three GC saturate traces obtained are virtually identical and characterise the three extracts as genetically related. A baseline hump characterising an unresolved complex mixture of compounds is present in all three samples. Overall these GC traces might suggest two phases of hydrocarbons. The results suggest strongly anoxic conditions during deposition of the organic source matter. One sample had indications of marine organic matter. The presence of C30 norhopanes is a marker for input from a carbonate source. Nevertheless the presence of cadalene and the relatively high Higher Plant Index (HPI) of 3.42 are markers for input from a terrestrial source. Branched/cyclic and aromatic maturity parameters characterise the hydrocarbons as moderately mature to mature, between approximately 0.85 and 0.95% Vr. In the Kiri wells, the presence of isopimarane and retene, as well as cadalene, reflects input from higher plant derived organic matter which indicate a contribution from Devonian or younger source rocks.

The extract from Kewell East 1 seems basically similar to the ones from the area of the Bancannia Trough. The biomarkers in the samples from Bancannia South-1 and Little Topar wells are very similar to the extract from the Cymbric Vale sample.

The Cymbric Vale sample is from an outcrop to the east of the Bancannia Trough. The carbonate markers found in the Bancannia South extracts could thus be real rather than derived from a refined product. This does not mean that all the hydrocarbons in this sample are natural; refined products could be responsible for part of the extract. The similarity of the Little Topar extracts to those from Bancannia South is also interesting, as the Little Topar wells occur on the southern margin of the Bancannia Trough. Taken together, this evidence suggests that the Bancannia Trough is underlain by Cambrian carbonate facies that may provide a viable source rock for economic accumulations in younger reservoirs.

In March/April 1999 the Department undertook a shallow drilling program designed to investigate potential source rocks within fine grained sequences across the Koonenberry High, on the eastern margin of the Bancannia Trough. Early results suggest TOC values are low (0.02 - 0.53%). However, a petroliferous odour during drilling of the Gnalta Group in Koonenberry-9, led to an extract characterisation indicating mature hydrocarbons believed to be sourced from mixed organic matter with possibly a carbonate association. Some evidence of contamination of the extract was noted.

Source rocks within the Darling Basin remain elusive, possibly hidden at depths un-penetrated in the middle of the deeper troughs or simply not encountered at shallower depths due to the small number of wells and lack of outcrop in the basin. However there is ample evidence of hydrocarbon (gas and oil) generation within the basin indicating that source rocks exist somewhere within (or below ?) the basin, possibly in pre-Devonian sequences.

Reservoirs

Industry's perception is that the reservoir potential of the Early Devonian is poor because of diagenesis, and by implication, so too is that of the Late Devonian sequences contained within the troughs. To some extent this perception has been recently rebutted, or at least partially revised, following the drilling of three stratigraphic boreholes. At Kewell East 1, quite good porosities and permeabilities were encountered in a sandy interval between 800 and 1 200 m, within the Pragian. At Booligal Creek 1 and 2 wells, excellent porosities and permeabilities were encountered. These wells were drilled on the flank of the Wilcannia High, bordering the Blantyre Trough where seismic data indicated that the penetrated section had been previously much more deeply buried, having subsequently been uplift and eroded.

However, and unlike earlier well results at Mt Jack 1 and Mt Emu 1, drilled on major highs, excellent porosities and permeabilities were encountered over the 450m interval sampled. These results extend the potential distribution of reservoirs both stratigraphically and geographically to include not only the Late Devonian Ravensdale and Snake Cave intervals within the troughs and across the flanks of the adjacent highs, but also portions of the Early Devonian Winduck interval. Quantitative data are not numerous, most porosity and permeability data consisting of electric log interpretations. In general the porosities are good to excellent - ranging from 10% to nearly 30% in some cases. Commonly the permeabilities range from a few hundred millidarcies up to 8000 millidarcies.

Where quantitative poroperm data are available, the log interpretations suggest that the sand units were commonly deposited in high energy fluvial environments. These probably largely consist of channel sands from braided stream environments with some input from meandering stream conditions.

Structural History and Trap Timing

The structural history of the Darling Basin remains relatively unknown. Most wells either penetrate structurally complex areas where seismic imaging of geological horizons is poor, or alternatively, wells in the less deformed thicker depocentres only penetrate through the Late Devonian sequence. At least three discrete periods of structuring impacted upon the present day geometry of the Darling Basin, following the onset of deposition. These are the Bindian (Early Devonan), Tabberabberan (Middle Devonian) and Kanimblan (Middle Carboniferous) Orogenies. It is speculated that the Kanimblan Orogeny, associated with the Alice Springs Orogeny, brought about the cessation of Darling Basin deposition and imposed much of the conspicuous structuring now recognised.

One of the potential exploration risks in the Darling Basin is the relative timing of trap formation and hydrocarbon generation. Ideally, trapping mechanisms with the least risk are associated with palaeo-closures, highs formed during the earlier periods of deformation. Significant basin inversion, uplift and reactivation during the Kanimblan Orogeny occurred during the period when it is predicted Early Devonian sediments would have passed into the oil generation window. Therefore there is a question of relative timing between trap formation and generation, expulsion and migration. Moreover, some of the early formed traps may well have been breached by faulting or vertical seal integrity breached by a reduction in overburden pressure following basin inversion.

Play Concepts

With only 1 550 km of modern, high-fold seismic data acquired within the Darling Basin, its structural framework cannot be considered well delineated. However, tied to aeromagnetic and gravity data, a wide range of structural styles are already recognised involving virtually all of the major basinal boundaries. Shear zones are identified with high angle, basement involved reverse faults, cored anticlinal folds, most with faulted flanks and basement detached thrust complexes are represented. Most of the conspicuous structuring is post-Devonian, interpreted to be Middle Carboniferous, related to the Kanimblan Orogeny whose deformation is widely recognised at this time throughout much of eastern Australia, including the Adavale Basin.

Potential play concepts include:

Structural targets associated with closures formed across thrust, wrench and normal fault blocks, forming the Nelyambo Ridge with Snake Cave and Ravensdale interval reservoirs, sealed regionally and intraformationally; sourced primarily by the Amphitheatre Group sequences.
Structural targets associated with the upthrust margins of the Mt Jack High, Wilcannia High, Lake Wintlow High, and Scopes Range High, with Middle Devonian Snake Cave and Winduck Group reservoirs sourced by Amphitheatre Group marine shales and siltstones.
Traps associated with wrench, thrust and reverse fault zones which were subjected to significant vertical uplift and subsequent erosional truncation. Targets within the restricted "Middle" or Late Devonian Snake Cave or Winduck Group and underlying Amphitheatre Group are sourced by marine shales and siltstones of the latter sequence as well as marine shales and siltstones. An example of this play type could be tested by drilling the previously identified Snake Flat prospect.
Traps formed by compressional folds and associated reverse fault trends within the deeper portions of the basins where potential traps involve closure of more complete Ravendale Interval sediments, low angle thrust fault structures often having a detachment surface near the top of the Amphitheatre Group. Reservoir targets in the Late Devonian are sealed by thick regionally extensive shallow shale units, whereas thick shale and siltstone units provide potential intra-formational seal deeper within the Middle Devonian. This play type could be tested by drilling the previously identified Avon or North Avon prospects. The greatest exploration risk is considered to be adequate seal and lack of certainty as to source, although the latter is to some extent ameliorated as recent seismic reprocessing has confirmed the high amplitude nature of the reflectors and preliminary amplitude-versus-offset analysis has supported some amplitude variations consistent with those anticipated in a gas charged reservoir. Estimates of potential gas reserves range from 300 to 600 BCF.

The presence of these fairways, together with encouraging results from geochemistry and petrophysical studies, proximity to the Sydney to Moomba gas pipeline and gas market opportunities in the Sydney region, highlight the need for an ongoing review of the Darling Basin.

Please refer to the following publications for further details.

New South Wales Petroleum Potential

Structural Diagram

Structural elements and interpreted sedimentary thickness

68 Kb

Stratigraphic Table

Darling Basin Stratigraphic Table To Download (Image)

39 Kb

Exploration Data

Seismic Shot Points (October 2001) - Data		312 Kb
NSW well locations		44 Kb

For further information, please contact petroleum@dpi.nsw.gov.au