Tanami 3D model

• About the Tanami 3D model
• Model metadata

• View the Tanami 3D model online
• Snapshots of this model

About the Tanami 3D Model

This 4th version of the Tanami 3D model covers a 200 km x 200 km area of the Tanami Region and Arunta Inlier (Aileron Province). This model consists of a series of 3D surfaces of major geological elements ranging from relatively shallow cover sequences, down to the Moho. The 3D surfaces were constructed using 3D GeoModeller and combines the interpreted seismic data from the 2005 Tanami Seismic Collaborative Research Project, with geological mapping, interpreted solid geology maps, and gravity and magnetic data. Elements of the model have been tested by gravity forward modelling.

Geographic extent

Tanami Region, Northern Territory-Western Australia.

Contents

Digital elevation model; images of magnetics, gravity, radiometrics, satellite imagery, mapped geology and regolith; roads; images of solid geology interpretation maps; modelled cross-sections; 3D fault surfaces; gravity and magnetic inversion surfaces; gravity and magnetics strings; mineral occurrences; SHRIMP geochronology; seismic images; and depth to basement points, surfaces and contours.

Size

Approximately 27 MB. Startup download is 850 KB - the remaining datasets download when selected.

[back to top]

Metadata

• Introduction
• Reference layers
• Images
• Geology
• Sites
• Seismic Images
• Seismic Interpretation
• Modelled Sections
• 3D Model
• Pre-Seismic Sections
• Pre-Seismic Faults Surfaces
• Inversion surfaces
• Gravity and magnetics strings
• Depth to basement

Introduction

The aim of the X3D (extensible 3D) model is to assist visualisation of the three-dimensional distribution of the principal geological elements of the crust. Two 3D models of the crust are included and consist of: a series of fault surfaces constructed from geological sections, tested by potential field modelling; and, gravity and magnetic inversion surfaces generated by inverting gridded gravity and magnetic data into a 3D distribution of density and magnetic susceptibilities. The 3D models are viewable with a number of geological and geophysical datasets supplied by Geoscience Australia, the Northern Territory Geological Survey and the Geological Survey of Western Australia.

[back to top]

Reference layers

• Roads: Image of the tracks and roads in the region. The data were obtained from the Northern Territory Geological Survey, Geoscience Australia, and Barrick Gold of Australia Ltd. No attempt has been made to discriminate road types or the accuracy of the data.

[back to top]

Images

• Elevation: The digital elevation model (DEM) is a raster image of the land surface or topography based on the Geoscience Australia 9-second DEM (2nd edition). White and red shades correspond to high elevations, blue shades correspond to low elevations. Due to the moderate relief of the topography in the region, a 25x vertical exaggeration is recommended in order to illustrate the height variation.
• Gravity: Colour image of the Bouguer gravity field, draped on the DEM with a north-easterly illumination. Colours range from pink, for high field values to blue, for low field values. Typical spacing of the gravity stations range from 4 to 11 km.
• Magnetics (RTP): Pseudocolour image of the total magnetic intensity, reduced to pole magnetic (RTP) grid, draped on the DEM with a north-easterly illumination. The reduction to the pole process was applied to correct for the non-vertical inclination of the Earths magnetic field in the region. Colours range from pink, for high magnetic intensities to blue, for low magnetic intensities. A number of airborne surveys were merged together to form the complete dataset including those of GA/AGSO/BMR, the State or Territory Geological Surveys and open file company surveys.
• Magnetics (1VD): Grey scale image of the first vertical derivative (1VD) applied to the total magnetic intensity, reduced to pole magnetic grid, draped on the DEM with a north-easterly illumination.
• Gravity and Magnetics: Composite image of the Bouguer gravity field in colour and the total magnetic intensity field, with a north-easterly illuminated shading, draped on the DEM.
• Radiometrics: Ternary image (Potassium - red: Thorium - green: Uranium - blue) of the gamma-ray spectrometric data, draped on the DEM. A number of airborne surveys were merged together to form the complete dataset.
• Satellite Image: Landsat 7 (bands 1, 2, 3) image, draped on the DEM.

[back to top]

Geology

Geoscience Australia is the custodian of surface geological mapping at 1:250 000 scale for the Australian continent. The geology can be ordered as either digital maps or images. Scanned images of all 1:250 000 geology maps may be downloaded from the Geoscience Portal.

• Outcrops: Image showing the outlines of mapped Palaeoproterozoic basement, Proterozoic and Palaeozoic cover sequences. 1:250 000 scale digital geology maps were used to compile this information.
• Major Faults: The fault traces consist of the intercepts of the modelled fault surfaces with the topographic surface, as well as selected faults from the preliminary version of the Northern Territory Geological Survey's Northern Arunta Region integrated interpretation of geology and geophysics (First Edition) 1:500 000-scale map.
• Arunta geology: Image of a preliminary version of the North Arunta Region integrated interpretation of geophysics and geology, Northern Territory (First Edition) 1:500 000-scale map (Vandenberg et. al., 2003), produced by the Northern Territory Geological Survey, Alice Springs. Further information on the Arunta North Project is available on the NTGS website.
• Tanami geology: Image of a preliminary version of the Tanami SF 52-15 1:250 000 integrated interpretation of geophysics and geology. Edition 1. Northern Territory Geological Survey (Slater, 2000). Further information on the Arunta North Project is available on the NTGS website.
• West Tanami geology: Image of a preliminary version of the solid geology 1:250 000 scale interpretation of the Tanami region in Western Australia, produced by Geoscience Australia and the Geological Survey of Western Australia. Further information is available from the Geological Survey of Western Australia website.
• Regolith: Image of regolith-landform data taken from a regolith study conducted in the Tanami region. This study was completed in 1999-2000 by the Cooperative Research Centre for Landscape Evolution and Mineral Exploration (CRC LEME), of which Geoscience Australia is a member. The study was originally commissioned on behalf of Acacia Resources, Normandy NFM and Otter Gold NL and provides a regional framework for regolith and associated landforms over the Tanami region. A number of datasets are contained in a Geographic Information System (GIS) provided as a CDROM package by GA which includes: regolith-landform units; enhanced Landsat TM imagery; site descriptions and photo links; regolith profile descriptions (geochemistry and PIMA); ex-BMR drill hole geochemistry; gamma-ray spectrometry imagery; palaeochannels; geochemical sampling strategy maps; surface flow vector maps; enhanced DEMs; erosional scarps; and maps showing depth of transported cover. The entire regolith study (10 disks) can be ordered online and individual disks containing either MapInfo or ArcView data and images can be ordered through the Sales Centre.

References:

Slater, K.R., 2000. Tanami 1:250 000 integrated interpretation of geophysics and geology. Edition 1. Northern Territory Geological Survey.

Vandenberg, L.C., Johnstone, A., Donnellan, N., Green, M.G. & Crispe, A., 2004. Northern Arunta Region Integrated interpretation of geophysics and geology, Northern Territory (First Edition) 1:500 000 scale map. Northern Territory Geological Survey, Alice Springs.

[back to top]

Sites

• Deposits: This layer contains point locations of mineral occurrences from the Northern Territory Geological Survey's MODAT database, and from the Western Australian Geological Survey's WAMEX and MINEDEX databases. The points are coloured yellow (gold), red (uranium) and green (copper). Hovering over a location with the cursor will reveal its name and commodity.
• Geochronology: The data in this layer is derived from Geoscience Australia's (GA) Oracle GEOCHRON database of pooled age, Sensitive High-Resolution Ion Microprobe (SHRIMP), U-Pb analyses. Hovering over a location with the cursor will reveal a pooled age and a lithology name, if supplied. Hovering over a location and selecting with the left hand mouse button will open a link to that locations' sample record in GA's OZCHRON database. The database record contains the complete SHRIMP results for that sample including a link to a concordia diagram.

[back to top]

Seismic Images

The seismic data was acquired as part of the 2005 Tanami Seismic Collaborative Research Project (Huston et al., 2006), a collaborative project between Geoscience Australia, the Geological Survey of Western Australia, the Northern Territory Geological Survey, Newmont Exploration and Tanami Gold NL, using seismic facilities of ANSIR, the National Research Facility of Earth Sounding. The aims of the project were to understand both the crustal architecture and mineral systems of the Tanami Region using deep seismic reflection techniques.

A total of 720 line-km of 60-fold seismic reflection data were acquired to 20 s TWT along four lines using 60 000 lb peak force vibrators. The data were processed by the Seismic Acquisition and Processing Group within Geoscience Australia and included the following key processing steps: calculation of refraction and automatic residual statics, spectral equalisation, stacking velocity analysis, DMO, and migration. These images consist of the final migrated seismic sections.

Reference:

Huston, H., Goleby, B., Vandenberg, L., Bagas, L., Jones, L., Korsch, R., Lyons, P., Johnston, W., Smith, T., Barton, T., and Johnstone, D., 2006. The Tanami Deep Seismic Reflection Experiment: a look under the surface of a major gold province in northern Australia. Australian Earth Sciences Conference, Extended Abstract.

[back to top]

Seismic Interpretation

The interpretation was conducted on the migrated sections. All sections were plotted at a V/H ratio equal to 1, assuming an average crustal velocity of 6 kms^-1. Detailed interpretations of the 6 s sections were done at 1:50 000 scale; interpretations of 20 s sections were done at both 1:100 000 and 1:200 000 scales.

The seismic data were interpreted in conjunction with regional geophysical data (gravity and magnetics), integrated geological and geophysical interpretations (Slater, 2000 a,b; Vandenberg et al., 2004) and integrated geological-geophysical modelling (Vandenberg et al., 2004; Meixner and Vandenberg, 2004).

References:

Meixner, A.J., Vandenberg, L.C., and Nicoll, M.G., 2004. 3-D modelling of the Tanami Gold Region. Bulletin of the Society of Economic Geologists - Perth conference 2004, extended abstracts. 304-308.

Slater, K.R., 2000a. Tanami SE 52-15 1:250 000 Integrated interpretation of geophysics and geology. Northern Territory Geological Survey.

Slater, K.R., 2000b. The Granites SE 52-15 1:250 000 Integrated interpretation of geophysics and geology. Northern Territory Geological Survey.

Vandenberg, L.C., Johnstone, A., Donellan, N., Green, M.G., and Crispe, A.J., 2004. Northern Arunta Region Geophysical Interpretation. Northern Territory Geological Survey.

Vandenberg, L.C., Meixner, A.J., and Nicoll, M.G., 2004. The Tanami 3-D geological model - integrating geology and potential field data. ASEG 17th Geophysical Conference and exhibition, Sydney 2004, Extended abstracts.

[back to top]

Modelled Sections

Gravity modelling was conducted in order to constrain the architecture delineated by interpretation of the seismic data.

The Willowra 1 section approximates the southeast portion of the back-bone seismic line 05GA-T1 where it crosses the east-west-trending Willowra Gravity Ridge. The modelling was conducted on a straight section that approximates the line of the seismic transect as it follows the roads and tracks of the region. The seismic data show a thickening of the crust across the gravity ridge, with a marked difference in the character of reflectivity: moderate apparent southeast-dipping reflectors to the northwest of the ridge, and strong apparent northwest-dipping reflectors to the southeast of the ridge. A region of strong reflectivity separated by zones of strong reflectors is coincident with the gravity ridge. This region of strong reflectivity is interpreted as a wedge of upthrust lower crustal material, which now separates the Tanami basement to the northwest and the Arunta basement to the southeast. The apparent southeast and northwest-dipping zones of strong reflectivity separating the Tanami basement from both the crustal wedge and the Arunta basement is interpreted as a "crocodile" structure typical of terrain collision (Meissner, 1989).

The seismic interpretation formed the geometrical constraints to the Willowra 1 modelled section. A density of 3.3 g/cm3 was assigned to the mantle and 2.7 g/cm3 to the undivided Tanami Group and Lander Rock Beds. The remaining densities were altered until a reasonable fit was achieved between the modelled gravity and the long wavelength variation of the observed gravity.

The geometry of the Willowra Gravity Ridge is somewhat complex in the region where it is crossed by the seismic line, compared to portions east and west of the transect, where it is relatively narrow and trends east-west. Where the seismic line crosses the ridge, it is broader, 'dog-legs' to the north and there is a drop in the magnitude of the gravity anomaly. The Willowra 2 section is located to the west of Willowra 1, where the crust is interpreted to be structurally simpler. The longer wavelength of the observed gravity is successfully modelled in the Willowra 2 section using the same densities and similar geometries to the Willowra 1 section, but without the need for the dense body (3.0 g/cm3) at the base of the Arunta crust employed in the Willowra 1 section. The interpretation that the seismic line (05GA-T1) crosses a structurally complex portion of the Willowra gravity ridge is supported by a number of major east-west trending faults identified from magnetic and gravity data in the region of the seismic line (Slater, 2000; Vandenberg et al., 2004).

References:

Meissner, R., 1989. Rupture, creep, lamellae and crocodiles: happenings in the continental crust. Terra Nova, 1(1). 17-28.

Slater, K.R., 2000. The Granites SF 52-3 1:250 000 integrated interpretation of geophysics and geology. Edition 1. Northern Territory Geological Survey.

Vandenberg, L.C., Johnstone, A., Donellan, N., Green, M.G., and Crispe, A.J., 2004. Northern Arunta Region Geophysical Interpretation. Northern Territory Geological Survey.

[back to top]

3D Model

The Tanami 3D model covers a 200 km x 200 km area of the Tanami Region and Arunta Inlier (Aileron Province) and extends to a depth of 60 km. The model consists of a series of 3D surfaces defining major geological elements ranging from relatively shallow cover sequences, down to the Moho. The 3D surfaces were constructed using 3D GeoModeller, combining the interpreted seismic data from the 2005 Tanami Seismic Collaborative Research Project (Huston et al., 2006) with geological mapping, interpreted solid geology maps, and gravity and magnetic data. Elements of the model have been tested by gravity forward modelling.

3D GeoModeller is a mathematically-based surface modelling package that allows construction of 3D volumetric models based on a range of geological information ( http://www.geomodeller.com/geo/index.php). Meixner et al. (2006) describe the use of 3D GeoModeller in constructing 3D models. The geological input consists of (1) defining a stratigraphic pile with the geological units related by conformable, onlapping, or erosional contacts; (2) geological contact points; (3) geological orientation data, and; (4) faults. Geological boundaries within the models are then computed as mathematical functions that take into account the lithological contacts and orientation measurements supplied by the user. The advantage 3D GeoModeller has over the more traditional CAD-based packages for surface construction is that it takes less time to build complex 3D surfaces, which can easily be modified by altering the initial input parameters.

The 3D model consists of the following surfaces:

• Redcliff Pound Group: This surface represents the unconformity at the base of the Redcliff Pound Group and the base of an overlying unnamed basin, consisting of the Pedestal Beds and Lucas Formation. This unconformity is imaged in the seismic data on the south-western ends of lines 05GA-T3 and 05GA-T4. The surface was generated using the seismic data to provide depth control, while geological mapping and geophysical interpretation (Slater, 2000a; Vandenberg et al., 2004) provided the spatial extents of the unconformity. Clastic rocks in the Redcliff Pound Group have the potential to host sandstone uranium deposits. These rocks and the reduced rocks of the Tanami Group, below the unconformity, have the potential for hosting unconformity-related uranium deposits.
• Birrindudu Basin: This 3D surface represents the unconformity at the base of the Birrindudu Basin. The surface is a generalised representation of the unconformity, as it is beyond the scope of this study to incorporate all of the highly complex structure in the region. The unconformity is imaged in the seismic data at relatively shallow depths (< 150 m) on the north-eastern end of 05GA-T2 and to the northwest of the Coomarie Granite on 05GA-T1. On the north-eastern end of 05GA-T3 the unconformity is at a depth of approximately 1900 m. The 3D surface was constructed using the seismic data, as well as the results of the 2D 'pre seismic' modelled sections, which were based on geological mapping and potential field modelling, as constraints. Spatial extents of the unconformity were defined by geological mapping and geophysical interpretation (Slater, 2000 a,b). The digenetic history of the Birrindudu Group sediments shows that uranium bearing fluids were active (Huston et al., 2007). These fluids have the potential to form unconformity-realted uranium deposits in the reduced rocks of the Tanami Group below the unconformity and in the overlying rocks of the Birrindudu Group.
• Callie Fault: The Callie fault consists of two surfaces which are offset by the Tanami Fault and were constructed using faults identified in the seismic and magnetic data as constraints.
  Line 05GA-T4 crosses a series of apparent north-dipping thrusts that reach depths of approximately 8 km. These faults correspond to northwest-trending structures on magnetic images of the region (Vandenberg et al., 2004), including the principal fault associated with the Callie deposit. Magnetic data show the fault continues to the northwest, where it coincides with an apparent north-dipping structure on 05GA-T3.
  Approximately 2 km to the east of this north-dipping structure is the east-southeast-trending strike-slip Tanami Fault which has had horizontal dextral displacements of at least 10 km (Vandenberg et al., 2001). It is younger than the northwest-trending faults. The Tanami fault is not imaged in the seismic data and is, therefore, inferred to be vertical and shallow, less than 1 km deep. To the east of the Tanami Fault on 05GA-T3 are three apparent north-dipping thrusts extending to approximately 1 km, as well as an apparent north-dipping thrust that offsets the top of the crystalline basement. Of the three shallow thrusts, the eastern most thrust corresponds to a northwest-trending fault identified in magnetic images that is interpreted by Vandenberg (pers comm.) as an extension of the Callie fault.
• Frankenia Granite: This surface represents the base of the Frankenia Granite as interpreted from the seismic data, geological mapping and geophysical interpretation (Slater, 2000b). There is a distinct change in seismic character from incoherent low reflectivity to high reflectivity of folded layers at approximately 1 km depth on 05GA-T1. This change of seismic character also occurs at the north-eastern end of 05GA-T3, although on this line the change in seismic character is not as pronounced.
• Coomarie Granite: This surface represents the base of the Coomarie Granite as interpreted from the seismic data, geological mapping and interpretations of magnetic and gravity data (Slater, 2000 a,b). There is a change in seismic character from low to high reflectivity at approximately 3000 m depth, although the change is not as distinct as the change in seismic character for the Frankenia Granite on 05GA-T1.
• Top of crystalline basement: This surface represents the unconformity between the top of crystalline basement and the base of the Bald Hills Sequence (northwest 05GA-T1) and the Tanami Group (Mt Charles, Killi Killi and Dead Bullock Formations) and was delineated on seismic character. The top of crystalline basement is generally between 6 km and 9 km depth across most of the region, except for a area towards the northeast of line 05GA-T1 which is at approximately 11 km) and to the southeast of 05GA-T1 where it shallows to approximately 3 km.
• Suture and back thrust: These surfaces represent the structural zone separating the Tanami basement, Arunta basement and an interpreted wedge of upthrust lower crustal material. The suture and back thrust are characterised by zones of dipping reflectors separating regions of differing reflectivity (Tanami basement - overall apparent south-easterly-dipping reflectors: Arunta basement - apparent north-westerly-dipping high amplitude reflectors; upthrust lower crustal material - strong reflectivity) in the mid to lower crust on the south-eastern portion of line 05GA-T1 (Huston et al., 2006). The suture and back thrust, therefore, define the northern and southern limit of the relatively dense upthrust lower crust, which correspond to the northern and southern limits of the Willowra Gravity Ridge (see Willowra 1 and 2 modelled sections). The upthrust wedge of lower crustal material is, therefore, interpreted as the source of the Willowra Gravity Ridge. The northern and southern extents of the Willowra Gravity Ridge was used to constrain the suture and backthrust surfaces to the east and west of the Willowra 1 and 2 modelled sections.
• Moho: The Moho is clearly imaged in the four seismic transacts and is at approximately 35 km to 40 km depth except on line 05GA-T1 where it increases in depth to approximately 60 km below the Willowra Gravity Ridge.

References:

Huston, D.L., Vandenberg, L., Wygralak, S., Mernagh, T.P., Bagas, L., Crispe, A., Lambeck, A., Cross, A., Fraser, G., Williams, N., Worden, K., Meixner, T., Goleby, B., Jones, L., Lyons, P., and Maidment, D., 2007. Lode-gold mineralization in the Tanami region, northern Australia. Miner Deposita 42. 175-204.

Huston, D.L., Goleby, B., Vandenberg, L., Bagas, L., Jones, L., Korsch, R., Lyons, P., Johnston, W., Smith, T., Barton, T., and Johnstone, D., 2006. The Tanami Deep Seismic Reflection Experiment: a look under the surface of a major gold province in northern Australia. Australian Earth Sciences Conference, Extended Abstract.

Meixner, A.J., Lane, R., Czarnota, K., and Cassidy, K., 2006. Constructing geologically-constrained 3D models using 3D GeoModeller: an example from the Paterson Orogen. Geoscience Australia, Record, 2006/20.

Slater, K.R., 2000a. The Granites SF 52-3 1:250 000 integrated interpretation of geophysics and geology. Edition 1. Northern Territory Geological Survey.

Slater, K.R., 2000b. Tanami SE 52-15 1:250 000 integrated interpretation of geophysics and geology. Edition 1. Northern Territory Geological Survey.

Vandenberg, L.C., Johnstone, A., Donnellan, N., Green, M.G., and Crispe, A., 2004. Northern Arunta Region integrated interpretation of geophysics and geology, Northern Territory (First Edition) 1:500 000 scale map. Northern Territory Geological Survey, Alice Springs.

Vandenberg, L.C., Hendrickx, M.A., and Crispe, A.J., 2001. Structural geology of the Tanami Region. Northern Territory Geological Survey, Record 2001-004.

[back to top]

Pre-Seismic Sections

The modelled sections are based on geological interpretations that are consistent with known geological and structural data in the region. The geological cross-sections were tested by potential field modelling using ModelVision, a 2D interactive program. Subsequent geological and geophysical revisions resulted in the final modelled sections. Each section has been modelled to a depth of 15 km.

Cross-section positions were chosen such that many previously identified or interpreted faults, contacts and structures related to mineralisation were intersected at right angles. This effectively simulated a 2D environment perpendicular to the section lines in order to simplify the forward modelling process.

Gravity data consisted mostly of 4 to 11 km station spacing. The sparse station spacing results in an under-sampled gravity field which precludes the modelling of near surface features. Gravity modelling, however, was useful for modelling the deep crustal structure. In contrast, magnetic data, acquired at 500 m or less flight line spacing, were suitable for modelling near-surface structure.

For section construction, basic stratigraphic sub-divisions for the Tanami were adopted (Hendrickx et al, 2000). Gravity modelling incorporated standard standard density values, taking into account metamorphic grade and proportions of various rock types within a stratigraphic unit (eg. siliclastic sedimentary rocks vs. mafic rocks). Magnetic modelling incorporated magnetic susceptibility values that are within acceptable ranges for the given rock compositions. A regional baseline is included in the magnetic modelling for a number of sections where a low order polynomial (order 2, for section 1) or a tilt is interpreted to exist in the regional magnetic field. Modelling of this anomalous field is not possible using variations of magnetic material within the top 15 km of the crust. For further information refer to Vandenberg & Meixner (2004) and Meixner & Vandenburg (2004).

There are two sets of check boxes associated with each cross-section in the model. The right-hand column "c" box will display an image of the cross-section with profiles of the modelled and observed gravity and magnetic data. The left-hand column check box will display the cross-section in 3D space.

References

Vandenberg, L.C., & Meixner, A.J., 2004, The Tanami 3-D geological model - integrating geology and Potential field data. In, ASEG 17th Geophysical Conference and Exhibition, Sydney 2004, extended abstracts.

Meixner, A.J. & Vandenberg, L.C., 2004. 3-D modelling of the Tanami Gold Region. Bulletin of the society of economic geologists - Perth conference 2004, extended abstracts. 304-308.

[back to top]

Pre-Seismic Fault Surfaces

Yellow - D1

Green - D4

Blue - D5

Crimson - D6

Red - D6a

Brown - D6b

The fault surfaces were created in GoCad from fault traces located on the solid geology interpretation maps and the modelled cross-sections. The faults were constructed to a depth of 15 km to be consistent with the modelled sections. The faults were assigned different colours according to the deformation event that they were interpreted to be most active. Some faults, for example the Supplejack Fault, formed in an earlier deformation event (D1) and were reactivated in subsequent deformation events.

Tanami Region structural summary

A thorough description of the characteristics, relationships and ages of tectono-stratigraphic units and granitoids in the Tanami region is given in Crispe et al. (2005). The oldest rocks in the Tanami region are thought to be the high-grade metasedimentary rocks and leucogranites of Neoarchaean age within the Billabong Complex (~2510 Ma; Page et al., 1995), although these rocks are being redated. The most extensive unit is the Tanami Group (~1840 Ma), subdivided into the basal Dead Bullock and overlying Killi Killi formations. Dolerite sills intrude the Tanami Group.

D1-M1

Folding and regional greenschist-to amphibolite-facies metamorphism of the Tanami Group during the Tanami Event (~1830 Ma; Vandenberg et al., 2001). The D1-M1 Tanami Event was followed by deposition of sandstone, siltstone and felsic volcanic rocks of the Ware Group and high level intrusion of the Winnecke Granophyre (~1820 Ma).

D2 and D3

Two periods of compressional deformation with concomitant intrusion of the Grimwade and Frederick magmatic suites (1825-1795 Ma, Cross et al., 2005). D2 is characterised by NW trending fold structures, D3 is characterised by NE trending fold structures.

D4

Regional faulting and the formation of the rift-related Mount Charles Formation. The D4 Supplejack Fault may be associated with earlier rifting accompanying the deposition of the Ware Group. D4 structures may also have been relatively long lived structures, undergoing reactivation during subsequent events. Forward modelling of potential field data indicate that D4 faults can be traced to a depth of at least 10-15 km.

D5

Shear zones and faults associated with many of the gold deposits in the Tanami region. D5 may include two or more discrete deformational events (D5a/5b Crispe et al., 2005). D5a structures are regional strike slip and thrust faults that were active during late syn-to post-regional granitoid emplacement (1825-1795 Ma, Cross et al., 2005). Forward modelling of potential field data indicate that D5 faults can be traced to a depth of at least 10-15 km. D5b structures are second order thrusts and strike slip faults that are locally associated with gold mineralisation. D5 structures probably reflect the effects of WNW-trending regional transpressional shearing (SW directed shortening). Post-D5 deposition of the Pargee Sandstone (~1765 Ma, Cross et al., 2005). The Pargee Sandstone is unconformably overlain by arenite, siltstone, shale and carbonate of the Birrindudu Group.

D6

Late post-mineralisation brittle faults that cut all units in Tanami region. D6 structures have caused fault-propagated folding within the Birrindudu Group cover sequence (post 1758 Ma) and displaced Antrim Plateau Volcanics (513 +/- 12 Ma; Hanley and Wingate 2000). D6 structures probably reflect the affects of the Strangways Orogeny (1730-1690 Ma), Liebig Orogeny (1640 Ma), Chewings Orogeny (1590-1570 Ma), Musgrave Orogeny (1200-1160 Ma), Petermann Orogeny (570-530 Ma) and the Alice Springs Orogeny (400-330 Ma). Two broad phases have been distinguished (D6a/6b).

D6a

Early D6 faults, generally striking WNW.

D6b

Latest faults that cut D6a faults and generally strike NW or NE.

References

Crispe, A. Scrimgeour, I. and Vandenberg, L., (2005). Geological framework of the Palaeoproterozoic Tanami region, Northern Territory. Mineralium Deposita, in review.

Cross, A., Crispe, A., and Williams, I., (2005) Mineralium Deposita, in review.

Hanley, L.M., and Wingate, M.T.D., 2000. SHRIMP zircon age for an Early Cambrian dolerite dyke: an intrusive phase of the Antrim Plateau Volcanics of Northern Australia. Australian Journal of Earth Sciences, 47/6, 1029-1040.

Page, R.W., Sun, S-s., Blake, D.H., Edgecombe, D.R., and Pearcey, D.P., 1995. Geochronology of an exposed late Archaean basement terrane in The Granites-Tanami region. AGSO Res. Newsl. 22: 19-20.

Vandenberg, L.C., Hendrickx, M.H., and Crispe, A.J., 2001. Structural geology of the Tanami region. Northern Territory Geological Survey Record 2001-004.

[back to top]

Inversion surfaces

3D inversion surfaces

The low density (red) and highly magnetic (green) inversion surfaces were generated using recently developed 3D inversion modelling software. This software, which utilises gravity and magnetic data, was developed by the University of British Columbia – Geophysical Inversion Facility (Li & Oldenburg, 1996; 1998) and is an advance over the more traditional 2D forward modelling method, allowing the generation of full 3D models in an automated environment.

This study is the first application of 3D inversion modelling to the central Australian region (Meixner & Lane, 2005). 3D inversion is a means of transforming observed gravity or magnetic data into a 3D model, populated by a mesh of cells consisting of density or magnetic susceptibility values. The process is iterative, whereby adjustments are made to a starting model in order to minimise the misfit between the observed and the computed data. The final models, containing the density and magnetic susceptibility values, reproduce the observed gravity or magnetic field to within a small degree of error.

3D inversion surfaces were generated to enclose regions within the models of anomalous physical property values. The red surfaces enclose regions of low density within the gravity model of less than 2.655 g/cm3, or 0.015 g/cm3 below the mean density for the model of 2.67 g/cm3. These surfaces correspond mostly to mapped and interpreted granites and are interpreted to simulate the 3D distributions of the granites. The green surfaces enclose regions of high magnetic susceptibilities within the magnetic inversion model with susceptibility values of greater than 0.01 SI. These surfaces correspond mostly to the magnetic stratigraphy (BIFs and mafic units) within the Tanami group sediments and are also interpreted to simulate the 3D distribution of these units. The inversion models, therefore, may be used as a regional-scale guide for determining where the magnetic units, which are considered to be potential traps for gold mineralisation, may appear at depth, as well as where these units might occur close to the surface, beneath younger cover.

References:

Li, Y. & Oldenburg, D.W., 1996. 3-D inversion of magnetic data. Geophysics, VOL. 61(2), 394-408.

Li, Y. & Oldenburg, D.W., 1998. 3-D inversion of gravity data. Geophysics, VOL. 63(1), 109-119.

Meixner, A.J.,& Lane, R., 2005. 3D inversion of gravity and magnetic data for the Tanami Region. Annual Geoscience Exploration Seminar (AGES) 2005: Record of Abstracts. Northern Territory Geological Survey. Record 2005-001.

[back to top]

Gravity and Magnetics Strings

Multi-scale analysis of potential field (gravity and magnetic) data, or "strings", is based on the property that data collected at or near the Earth's surface contain a spectrum of wavelengths. Short wavelengths usually result from near-surface property distributions (or "bodies"), while longer wavelengths usually result from deeper bodies. Such generalisations must be used with caution because, due to the non-uniqueness of potential field data, an infinite number of property distributions are possible which give rise to the same measured field at the surface. This concept of depth as a function of wavelength led to the development of a multi-scale approach to the detection of property boundaries. From potential field theory, data collected at one level above the Earth's surface can be transformed to data that could be acquired at any other level. This process is termed "continuation", and while there are practical considerations concerning noise degradation in the downward continuation process, upward continuation, which is essentially a smoothing process, can be successfully achieved. Thus, if data are upward continued to a number of levels, which represent successive removal of short-wavelength information as height increases, the maximum gradients (edges) selected from such data provide information concerning the depth and attitude of source discontinuities (e.g. Archibald et al., 1999; Hornby et al., 1999; Hobbs et al., 2000). For an explanation of the method used to create the strings, contact Tony Meixner of Geoscience Australia.

References

Archibald, N., Gow, P. & Boschetti, F., 1999. Multiscale edge analysis of potential field data. Exploration Geophysics, 30, 38-44.

Canny, J., 1986. A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 8, 679-698.

Hobbs, B.E., Ord, A., Archibald, N.J., Walshe, J.L., Zhang, Y., Brown, M. & Zhao, C., 2000. Geodynamic modelling as an exploration tool: Published in: After 2000: the future of mining. The impact of new technology and changing demands on the mining industry, Sydney, 10-12 April, 2000. Proceedings. AusIMM Publication Series.

Hornby, P., Boschetti, F. & Horowity, F.G., 1999. Analysis of potential field data in the wavelet domain. Geophysical Journal International, 137, 175-196.

Milligan, P.R., Lyons, P. & Direen, N., 2003. Spatial and directional analysis of potential field gradients - new methods to help solve and display three-dimensional crustal architecture. Extended Abstracts. ASEG 16th Geophysical Conference and Exhibition, February 2003, Adelaide.

[back to top]

Depth to basement

• Euler depths: The depth solutions shown in this layer were generated using the Euler deconvolution method. The values in this study represent the depth to the top of magnetic bodies beneath the topographic surface and are interpreted to estimate the depth to the Tanami basement situated beneath all platform cover. Cover material in the region consists of the Proterozoic Birrindudu Basin, the Palaeozoic Wiso and Canning basins up to more recent cover sequences including the most recent quaternary sands and gravels. Only depth solutions to magnetic sources interpreted to be at or near the top of basement were included. Magnetic sources interpreted to exist within the cover were ignored.

  The process of generating a final Euler depth solution requires some interpretation by the user in order to: filter spurious solutions; and select the most valid structural index (SI) for each observed magnetic anomaly. The value of each depth solution may, therefore, not be reproducible by different users. As such, each Euler depth solution must be considered as a depth value plus or minus a depth error. Experience suggests that a depth error of +/- 15% is a reasonable estimate.

  Tanami basement rocks are highly weathered, with weathering profiles in places extending up to 100 m and more below the outcropping surface. The magnetic properties of rocks tend to decrease with increased weathering due to magnetite destruction. This will result in an overestimation of the depth solution with respect to the basement to cover contact. The calculated Euler solution will lie somewhere between the basement to cover contact and the base of the weathering profile.

  The quality of the depth solution is dependent on the data resolution. The majority of the Tanami region is covered by airborne surveys with line spacings of 400 and 500 m. The line data for each survey were gridded to a 100 m cell size, with the resulting grids merged together to form a seamless composite grid. Unfortunately high frequency magnetic anomalies, sourced by near surface magnetic bodies, are not adequately sampled by the 100 m gridded data resulting in inaccuracies of estimating the depth to these bodies. Experience suggests that depth estimates of less than 200 m are not accurate and may be overestimations of the true depth to magnetic basement. Only substandard 1500 m line spaced aeromagnetic data were available for the Stansmore 1:250 000 sheet area. It was considered that the inaccuracies involved in generating depth estimates, in this region, were too high to produce any meaningful results.

• Modelled depths: This layer shows the depths to the top of forward modelled bodies below the topographic surface. Forward modelling was conducted on selected anomalies where the Euler process failed to generate valid depth solutions. Only discrete anomalies were chosen that were interpreted to be sourced by a dipping dyke source body. Forward modelling involves a degree of interpretation by the user, resulting in depth estimates that may not be reproducible by different users. Each modelled depth solution must therefore be considered as depth value plus or minus a depth error. Experience suggests that a depth error of +/- 10% is a reasonable estimate.

• Basement 3D surface: This layer is a 3D surface generated using the Euler, the modelled depth solutions and the locations of outcropping basement as constraints. Caution should be applied when viewing the surface, particularly in the central-north of the model area. In this region, Tanami Basement lie beneath thick sequences of Birrindudu cover sediments and are not constrained due to the lack of magnetic bodies within the basement and subsequent lack of depth solutions.

• Basement contours: This layer consists of contours generated from the 3D surface layer, projected onto the topographic surface. The contours are at 200 m depth intervals and coloured with a red-blue colour palette with red contours defining shallow depth and blue contours defining greater depths. Caution should be taken when viewing the near surface contours (200 m) due to the inherent errors of the near surface depth estimates.

Euler deconvolution

The Euler solutions were generated using the Intrepid geophysical software, applied to gridded magnetic data. The solutions generated used the "extended" Euler deconvolution method as implemented by FitzGerald et al. (2003) following the methods of Mushayandebvu et al. (2001) and Nabighian and Hansen (2001). Depth estimates were generated using the two Hilbert equations which calculate solutions for the spatial position of the source, the structural index (SI) and associated errors (Milligan, et al., 2004). Two window sizes (7 and 15 cells) were used in order to resolve depth solutions for near surface, as well as deeply buried magnetic bodies.

A large number of solutions were often generated for each magnetic anomaly, while solutions for more than one SI were often also generated for an individual anomaly. Filtering of spurious solutions was achieved by applying cut-off values on a number of parameters as suggested by FitzGerald, et al. (2004) and rejecting solutions where the values for a given parameter fell outside excepted values.

The SI of a solution was used as a way of associating the "shape" of the magnetic anomaly with a particular source body. For magnetic data the SI values are: 0 for a contact; 0.5 for a thick dyke; 1 for a fault (small step) and a thin dyke; 2 for a horizontal cylinder; and, 3 for a sphere. The SI values should be integer values and are theoretical model-derived estimates of field fall-off rates. In practice, however, the extended Euler method generates non-integer values which are then rounded to the nearest integer for subsequent analysis. The most accurate depth estimate will be given by the appropriate SI for a particular source body. A discrete elongated anomaly, for example, will be produced by either a thin (0.5 SI) or thick (1 SI) dyke. Some interpretation by the user is required in order to select the appropriate SI for each magnetic anomaly.

Selected forward modelling

The Euler method failed to generate solutions for a number of deeply sourced elongate anomalies interpreted to be sourced from dyke like bodies (SI 0.5 or 1). The larger window size of 15 cells, which is the maximum number of cells allowable in the Intrepid application, should have been large enough to generate valid solutions. Solutions were also not generated for a series of subtle north-west trending anomalies in the north-western portion of the model area. Again, these bodies, interpreted to be dykes, should have generated solutions (SI 0.5 or 1). The anomalies, though discrete, are subtle (5-10 nT), and may have been below a threshold value set within the Intrepid application.

Forward modelling, using ModelVision, was conducted on selected anomalies where the Euler process failed to generate valid depth solutions. Only discrete anomalies were chosen that were interpreted as sourced by a dipping dyke body. The widths of the "dykes" were set as the width of the anomaly at half maximum amplitude and are therefore simulating a thick dyke source or the theoretical equivalent of an SI of 0.5. The depth estimates to the top of the source body using the thick-dyke model will return a minimum depth to the source body. A thin dyke model (SI 1), simulating the same magnetic anomaly, will tend to return a larger depth estimate to the top of the body. The modelled depths should therefore be considered a minimum depth to basement.

[back to top]

References

FitzGerald, D., Reid, A., and McInerney, P., 2004, New discrimination techniques for Euler deconvolution: Computers & Geosciences, 30, 461-469.

FitzGerald, D., Reid, A., and McInerney, P., 2003, New discrimination techniques for Euler deconvolution: 8th SAGA Biennial Technical Meeting and Exhibition, October 2003.

Milligan, P.R., Reed, G., Meixner, T., and FitzGerald, D., 2004, Towards automated mapping of depth to magnetic basement – examples using new extensions to an old method: Extended Abstract, ASEG Conference, Sydney, 2004.

Mushayandebvu, M.F., van Driel, P., Reid, A.B., and Fairhead, J.D., 2001, Magnetic source parameters of two-dimensional structures using extended Euler deconvolution: Geophysics, 66, 814-823.

Nabghian, M.N., and Hansen, R.O., 2001, Unification of Euler and Werner deconvolution in three dimensions via the generalised Hilbert transform: Geophysics, 66, 1805-1810.

[back to top]

Product Information

For a complete metadata reference, please see the ANZLIC metadata record 11115.

For product information, please see Geocat record 65367.

[back to top]