contentsTargeting Fracture Permeability For Geothermal Developments In The North Perth Basin
Mark Ballesteros, Gary Meyer, Jim McDairimid, Adrian Larking, Green Rock Energy Ltd; Ralf Oppermann, OPPtimal Exploration and Development Pty Ltd
![Fig. 6. Comparison of reflectivity data with regional fault extraction results. Same time slice) through different volumes: (a) Reflectivity, (b) Fracture Network (red=high confidence, green=low confidence), (c) Fracture Network Reflectivity [Beharra Springs 3D seismic data set, time slices at 1,150 msec].](/images/article/1948/6.jpg)
Introduction
The northern Perth Basin (NPB) offers one of the most attractive areas for Hot Sedimentary Aquifer (HSA) geothermal projects in Australia. Key geological ingredients for a working geothermal system include a heat source and a heat trap in the form of thermally insulating rocks such as coals or low thermal conductivity shales with underlying naturally permeable, cavernous or fractured reservoir rocks (Cooper and Beardsmore, 2008). Available data suggests that the NPB has all of these attributes.
The technical evaluation of the NPB has identified a number of attractive target areas, with permeability associated with open natural fractures providing the most attractive potential reservoir. The Upper Rhine Graben in Europe provides a valuable analogue for the NPB geological model (Larking et al, 2010). Germany currently hosts two commercial HSA geothermal power plants, including the Unterhaching plant near Munich in southern Germany, and one in Landau in the Upper Rhine Graben. A third plant is now under construction at Insheim, close to Landau. An additional facility, the Soultz Research and Development plant, is located nearby in the Upper Rhine Graben in France (Schindler et. al, 2010). Importantly, all of these HSA plants produce exclusively from fractures.
Regional Setting
The NPB target area is located at the northern end of the Perth Basin, a north-south trending 1000 km-long rift basin which formed during the Permian to Early Cretaceous. The area has a complex tectonic history resulting in the development of extensive fault and fracture networks. The regional geology of the area is discussed by Mory and Iasky (1996) and the geological setting of the project area is described in more detail by Larking et al (2010, 2011). The structural setting in the project area is shown in Figure 1.
The primary source of geological and thermal information about the sedimentary sequences in the Perth Basin comes from petroleum wells and shallower water bores. Over 250 wells have been drilled in the NPB, reaching depths up to 4850 m, resulting in the discovery of more than 13 oil and gas fields. Most of these discoveries are located in and around the Green Rock permits and as a result the area has the highest density of petroleum wells and seismic data in the basin. Approximately 24,000 km of 2D seismic and over 2800 km2 of 3D seismic has been acquired to date and provides valuable assistance in understanding the detailed structural and stratigraphical history of the area (Figure 2).
Heat Flow and Temperatures
Temperature gradients exceeding 5°C/100 m and heat flows exceeding 100 mW/m2 have been measured from wells and deep bores in the study area (Larking, et al, 2011). Modelled surface heat flows in the Perth Basin ranged from 60–140 mW/m2, with a median value for all wells in the northern Perth Basin of 95 mW/m2. This compares to a median value of around 76 mW/m2 for the entire Perth Basin and 64.5 mW/m2 for Australia as a whole (WADME, 2008; Waining, 2009; Beardsmore, 2010).
Five wells within the Green Rock permit area have encountered bottom-hole temperatures greater than 150°C (Warradong-1, Redback-1 and -2, West Eregulla-1 and Mountain Bridge-1) at depths ranging from 3416 m to 4065 m. In addition, temperature and heat flow modelling for an additional 95 wells drilled in the area indicate locations where temperatures of 150°C should be reached at depths less than 3500 m deep. Figure 2 shows the area within the Green Rock permits where there is 3D data coverage and temperatures are expected to reach 150°C within the sedimentary section at depths less than 3500 m. This area represents our primary area of interest; west of this area basement is too shallow for HSA reservoirs and to the east deeper drilling would be required to reach the 150°C isotherm.
Reservoirs
HSA geothermal projects require relatively high water production rates to be commercially viable – generally in excess of 75 litres/sec (40,000 bpd) per well (Mortimer, 2010). Higher flow rates than these have been achieved from producing geothermal fields at Landau and Unterhaching in Germany and from shallow aquifers in the Perth Basin. Geothermal wells achieve this by producing from much thicker water saturated reservoir sections than typically encountered in petroleum wells.
A number of potential reservoirs are present in the permit area; the most prospective of which are all Permian in age, including the Upper Permian Wagina formation and the Caryngina formation, Irwin River coal measures and High Cliff Sandstone (Larking et al., 2011). Younger formations in the Triassic exhibit good reservoir qualities, but appear to have more limited potential as geothermal reservoirs due to their relatively shallow depths and consequent lower temperatures within the permit areas.
Nevertheless, despite examples of good reservoir quality at suitable depths locally, general trends show that reservoir quality deteriorates significantly below 3000 m (Laker, 2000), decreasing the probability of encountering a reservoir section with adequate matrix porosity and transmissivity to sustain the required flow rates. As a result, there is significant risk associated with relying solely on matrix permeability to supply adequate flow rates.
Natural Open Fractures
As noted previously, geothermal developments in the Rhine Graben produce sustainable high flow rates exclusively from open natural fractures due to lack of any significant matrix permeability in the sediments or in basement rocks. In the NPB, available well data proves that open natural fractures occur frequently in all of the target formations. For example, a fracture at 3334 m in Redback-1 had a measured in situ open thickness of 25 cm (King et al, 2008). These fractures are an attractive mechanism for recovering hot geothermal fluids from the sub-surface.
Review of borehole breakouts, rock densities and formation testing results, particularly leak-off tests has shown that the current stress regime in the NPB is transitional between the reverse and strike-slip regimes. This is due to the vertical and minimum horizontal principal stress components having similar magnitude. Break-out analysis has also shown that the maximum horizontal stress direction (σHmax) is oriented in an east-west direction (N84°E) (King et al, 2008).
Figure 3 illustrates the probability that a fault or fracture in this stress regime is critically stressed according to its dip and azimuth (plotted as a pole to plane). Critically stressed fractures are most likely to currently experience dilation and therefore be open to fluid movement (King et al, 2011).
A number of studies have been undertaken to establish the nature and orientation of fractures in the permit areas using borehole image logs (King et al, 2008, 2011; Bailey, in press). These efforts have catalogued the orientation of the fractures and classified them as open (conductive) and therefore likely to act as a conduit for fluid flow unless cemented with conductive minerals, or closed (resistive). Fractures generally tend to be oriented north, northwest and east depending on the local stress regime. There are examples of both resistive and conductive fractures in each orientation.
It is important to note that the dip of the fault or fracture is a critical variable. The most common azimuth for the fractures detected is approximately N10°W, which is essentially perpendicular to the σHmax. Figure 3 illustrates that vertical fractures with this azimuth have the lowest probability of reactivation, and therefore of being permeable. In contrast, when fractures with this azimuth are dipping 20° to 50° to the east or west, they are among the most likely to be reactivated.
Figure 4 illustrates the composite orientation of all the conductive fractures identified in the borehole image logs analysed. It shows that the most common orientation for conductive fractures in the study area is trending N10°W with dips of 30° to 60°. This, in turn, suggests that fractures with this orientation are likely to be critically stressed and therefore permeable. This suggests that the reservoir risk can be significantly reduced by targeting areas where seismic data shows a high density of appropriately oriented fractures is present.
Fracture Detection from Seismic Data
A number of 3D seismic surveys are available in the public domain over the project area. These surveys provide coverage over a number of the areas identified by Green Rock as having good geothermal prospectivity.
In addition to conventional structural mapping, a number of volume interpretation techniques have been applied to NPB seismic data, allowing 3D sub-surface features to be identified, filtered, classified and extracted quickly and accurately using an automated process. Automated fracture extraction is based on the physical measurement of spatial variation in amplitude, phase and/or frequency content of 3D seismic data, and is as such free of bias and interpretation. Fracture extraction allows to confidently identify small displacement faults in true 3D space that would often not be identified by manual interpretation (Oppermann 2010, 2012).
A key benefit of automated fracture extraction is that it facilitates the quantitative assessment of fracture properties such as distribution, orientation, size and intensity, as well as the confidence in the presence of a particular fracture. As the confidence value is often a proxy for fault throw or fracture aperture, it allows us to quickly visualise and separate larger (higher confidence) fractures from smaller (lower confidence) fractures (Oppermann, 2012). Results from unpublished studies show that fluid losses or fracture productivity are consistently linked with high confidence (larger) seismic fractures – though fracture orientation relative to present-day stress also plays a role.
Detailed correlation work in multiple study areas worldwide has demonstrated good matches between seismically identified fractures and fractures identified from well data (image logs, cores, correlation, well tests, productivity, fluid losses etc). Automated fracture extraction helps to narrow or close the scale gap between seismic and well data (Oppermann 2010, 2012).
Unpublished studies have also revealed that varying fracture densities encountered in wells (identified from image logs) are primarily related to the orientation of a well relative to a seismic fracture zone, or fracture corridor. Higher image log fracture densities are usually encountered when drilling parallel or sub-parallel to seismic fractures/fracture corridors, suggesting that a simple geometrical relationship exists between fracture density and well orientation. This highlights that detailed well planning using seismic fracture network data can allow us to increase fracture intersections and target productive sweet spots.
In the NPB, coherency processing was undertaken to help delineate major fault trends (Figure 5). However, this only shows the general trends of the faulting. This work is therefore being supplemented by both regional and high-resolution fracture processing that provide much greater detail regarding the orientation and density of the fracture networks (Figure 6).
Analysis of the Redback-1 image logs reveals two conductive fractures within the Kockatea Shale between 3334 and 3335 m (Figure 7). The orientation of the two fractures is azimuth 338° with a dip of 37° and 221° with a dip of 29°, respectively. The lower fracture displays a void of approximately 25 cm. Automated fracture detection results in the immediate vicinity of the well bore are shown in Figure 8 and illustrate two fractures were detected, one trending NW-SE, the other trending NE-SW; the features appear to intersect near the well bore (Figure 8a). Figure 8b show that both of these fractures have low dip angles consistent with the orientation observed in the image log.
While full evaluation of these results is still under way, they provide encouragement about the potential to correlate well bore and seismic data to accurately detect fractures. These results are being evaluated in conjunction with the heat flow and temperature data to identify the most favourable drilling location to directly target favourably oriented fractures, thus maximising the chance of being able to produce a sufficient volume of hot water to sustain a geothermal development.
Conclusions
The NPB displays all the necessary ingredients for successful development of a HSA geothermal development based on both matrix and fracture permeability type plays. Well data proves that temperatures in excess of 150°C can be reached at depths less than 3500 m. The most significant risk is identifying a zone with sufficient transmissivity to allow geothermal waters to be produced at commercially viable flow rates.
Work completed to date shows naturally occurring open fractures provide the best opportunity for intersecting a reservoir with the required transmissivity. Fault and fracture orientations most likely to be critically stressed, and therefore open and permeable, have been modelled based on the local stress regime in the NPB. Interpretation of borehole image logs provides data on the orientation of resistive and conductive fractures observed in the area.
These results are now being utilised to focus the seismic interpretation in order to identify a low risk drilling target. The application of new structural volume interpretation techniques in the NPB has shown that fault and fracture networks can be delineated at high resolution from seismic data. The integration of these seismic fracture networks with well and other data allows identification of areas with favourable fault orientations and densities that will significantly increase the chances of intersecting a suitable conduit for geothermal fluid production and ensure the potential for a successful HSA geothermal project.
References
Bailey, A.H.E., King, R. C. And Backe, G., in press, Integration of Structural, Stress and Seismic Data to Define Secondary Permeability Networks Through Deep Comented Sediments in the Northern Perth Basin, Proc of APPEA Conference 2012.
Beardsmore, G., 2010, Northern Perth Basin Project (NPBP) Geothermal Play, Statement of Geothermal Resources, unpublished report for Green Rock Energy Ltd.
Cooper, G.T and Beardsmore, G.M., 2008, Geothermal systems assessment: understanding risks in geothermal exploration in Australia, Proc. PESA Eastern Australian Basins Symposium III, pp. 411–419.
King, R. C., Hillis, R. R. and Reynolds, S. D., 2008,In situ stresses and natural fractures in the Northern Perth Basin, Australia, Australian Journal of Earth Sciences, 55:5,685–701.
King, R., Khair, A.A.,Bailey, A., Backe, G., Holford, S. and Hand, M., 2011, Integration of In-Situ Stress Analysis and Three-Dimensional Seismic Mapping to Understand Fracture Networks in Australian Basins, Australian Geothermal Energy Conference 2011, Melbourne.
Laker, N., 2000 Sedimentalogical and Petrographic Review of the Permo-Triassic Sandstones of the Northern Perth Basin, unpublished report by Oolithica Geoscience Ltd prepared for AWE.
Larking, A.L, 2010, Geothermal Energy in the Perth Basin, Australia: Comparisons with the Rhine Graben, proc of AGEA Conference, p. 30–34.
Larking, A.L., Meyer, Meyer, G., and Ballesteros, M., 2011, Mid West Geothermal Project, North Perth Basin, Australia, proc of AGEA Conference, p.139.
Mortimer, L., 2010, Geothermal Flow Rates in Australia - What Can We Expect and How We Can Mitigate Pre-Drill Risks, PESA News Resources April/May 2010, pp 66–67.
Mory, A.J. and Iasky, R.P, 1996, Stratigraphy and Structure of the Onshore Northern Perth Basin, Western Australia, Geol. Survey WA Report 46.
Oppermann, R., 2010, A new method for high-resolution fault imaging delivers groundbreaking insights into drilling and production of resources, ASEG 21st International Geophysical Conference and Exhibition, Extended Abstract, Sydney.
Oppermann, R., 2012, High-resolution fracture recognition from seismic data - applying oil and gas technology to mining, In: Deep Mining 2012, Y. Potvin (ed), Australian Centre for Geomechanics, Perth.
Schindler, Baumgartner, M. J., Gandy, T., Hauffe, P., Hettkamp, T., Menzel, H., Penzkofer, P., Teza, D., Tischner, T., and Wahl, G., 2010, Successful Hydraulic Stimulation Techniques for Electic Power Production in the Upper Rhine Graben, Central Europe, Proc World Geothermal Congress 2010, Bali, Indonesia.
Sagasco, 1993, Mountain Bridge-1 Well Completion Report.
Waining, B., 2009, Northern Perth Basin – extended heat flow analysis of wells for the GRK tenement areas. Unpublished report for Green Rock Energy Ltd.
