Structural and geomechanical analysis of naturally fractured hydrocarbon provinces of the Bowen and Amadeus Basins: Onshore Australia
Gillam, Daniel J.
Doctor of Philosophy 2004
University of Adelaide
Significant hydrocarbon resources onshore Australia occur within reservoirs with low matrix permeability. Such low permeability reservoirs include those of the mature Palm Valley and Mereenie Fields of the Amadeus Basin, the emergent coal seam methane (CSM) play in eastern Australia and stranded tight gas within existing fields of the Bowen and Cooper Basins. This thesis investigates structural and geomechanical aspects of the Bowen and Amadeus Basins, with particular reference to the role of natural fractures in enhancing hydrocarbon recovery in the low permeability reservoirs of onshore Australia.
This thesis first investigates natural fracture development and in-situ stress in the Scotia coal seam methane (CSM) field of the eastern Bowen Basin (Chapter 2). The key factors responsible for good well performance in the field are identified as: 1) relatively low stress to facilitate hydraulic fracture treatments; and, 2) a high degree of natural fracturing. However, the Scotia study is restricted by data limitations typical of most hydrocarbon development projects including: limited core and image log data, an absence of nearby outcrop similar to the reservoir, and ambiguous stress data.
Chapters 3–6 present an analogue study from the Goonyella Riverside coal mine of the northern Bowen Basin aimed at thorough characterisation of natural fracture development and in-situ stress within a coal measure succession similar to that of the Scotia Field. The study uses 3D seismic data, image log data and outcrop observations to identify the key controls on natural fracture development: namely the mechanical stratigraphy and proximity to seismically resolvable faults. Different fracture sets at Goonyella Riverside are linked to the tectonic evolution of the area: 1) coal cleats and compaction-related normal faults are interpreted to have formed during Late Permian burial and diagenesis; 2) thrust faults and subsidiary low-angle shear fractures are most likely to have formed during the Early to Mid Triassic Hunter Bowen Orogeny; and, 3) tensile joints are interpreted to have formed during Cretaceous uplift and unloading. The
influence of natural fracture permeability in the tight coal measure sequence is highlighted by groundwater observations that show a single thrust fault and associated damage zone account for half the water entering the underground mine adit.
The Goonyella Riverside study also uses in-situ stress measurements taken during the course of the underground mine development to illustrate the partitioning of stress within a background reverse stress regime. The sandstone intervals are characterised by a reverse-stress regime with maximum horizontal stress (SHmax) approximately twice the magnitude of the vertical stress (Sv) while the coal seams are characterised by a normal-stress regime with SHmax approximately half the magnitude of Sv. Drilling strategies and predictive pre-drill techniques have been developed from the Scotia and Goonyella Riverside studies for similar CSM fields in eastern Australia. Fracture swarms that cluster around seismic-scale intraformational faults may be targeted for high-rate wells and cleat orientations may be predicted pre-drill by the orientation of seismic scale structures. The study also highlights potential problems with conventional hydraulic fracture stimulation in formations with highly anisotropic rock properties such as coal measure sequences where there is high background stress typical of onshore Australia.
Chapters 7–10 investigate the structural evolution and natural fracture development in the Palm Valley and Mereenie Fields of the northern Amadeus Basin. Regional balanced cross-sections demonstrate the interaction of a basement-cored wedge with the sedimentary cover and a weak basal detachment in an intracratonic setting. Shortening across the northern margin of the Amadeus Basin during the Alice Springs Orogeny (ASO) is ~ 30 km, significantly less than the ~110 km implied by previous thin-skinned models. Deformation within the foreland is relatively mild compared to other fold and thrust belts. The Mereenie and Palm Valley anticlines developed as thin-skinned, salt-cored detachment folds above the basal detachment. Isostatic modelling shows that the effective elastic thickness of the lithosphere was significantly less than its contemporary thickness of ~88 km during the ASO. The modelled effective elastic thickness of 10 km is similar to contemporary intracratonic extensional provinces such as east Africa and
the Basin and Range Province, implying significant heat flow prior to the ASO.
Natural fractures are vital for gas production in the Palm Valley Field and naturally fractured sweet spots could also provide high-rate wells in the Mereenie Field. The study builds on previous work and identifies two generations of natural fracturing that influence reservoir performance in the fields: 1) bed-bound fold-related fractures that provide access to the low permeability matrix; and, 2) pervasive, orogen-related regional fractures that plumb the entire reservoir by connecting the fold-related fractures and different reservoir intervals. The foldrelated fractures may be predicted by modelling the fold geometry with Gaussian curvature. A strong correlation exists between areas of high fracturing, good well performance and areas of high Gaussian curvature in the Palm Valley Field. The study shows that curvature analysis is
field-specific and needs to be calibrated with fracture or production data from within the field and cannot necessarily be calibrated to data from other fields. Regional fracture spacing varies systematically across the basin and can be correlated to background orogenic intensity defined by the balanced 2D sections. The most productive wells in both fields are expected to occur in locations where the pervasive regional fractures intersect zones of dense fold-related fractures or where there are large reservoir-effective crestal fractures (as currently targeted).
The Amadeus Basin study also uses the structural modelling to assess the untested Mereenie sub-thrust play as the structural evolution is critical to trap integrity. A balanced reservoir-scale 2D section constrained by seismic data shows the anticline developed as a salt-cored detachment fold. Faulting was late in the structural development, but prior to maximum burial and final hydrocarbon charge. Footwall closure and fault seal are identified as the critical risks to the sub-thrust play. Much of the Mereenie seismic data does not extend over the footwall of the fault but the structural modelling indicates footwall closure is likely. Fault seal risk is assessed as moderate based on juxtaposition relationships, likely fault rock properties and reactivation potential. The study identifies the Mereenie sub-thrust play as a medium to high-risk near-field exploration opportunity due mainly to poor seismic imaging of the footwall.