2.1.2.2 Statistical analysis and interpolation


Observed datasets (Section 2.1.2.1) were analysed and processed to form derived datasets for use in the bioregional assessment of the Hunter subregion. These datasets were analysed and interpolated to develop a three-dimensional geological model of the Carboniferous to Triassic stratigraphic sequences of the Hunter subregion. The three-dimensional geological model with 2 km2 cell size was built using Roxar Reservoir Management Software (Roxar RMS). Developed by Emerson (2016), this product is traditionally used in the oil and gas industry to make hydrocarbon reservoir or basin-scale models. Other packages are available, such as GOCAD® Mining Suite or Leapfrog® three-dimensional geological mining software. Roxar RMS was adopted because it has the option of computing 3D facies grids based on a stratigraphic analysis approach within the grid cells (see Emerson (2016) for more details).

2.1.2.2.1 Workflow

Having established the need to build a regional geological model for bioregional assessment purposes, a workflow was defined to construct the first-order subsurface structural and stratigraphic architecture of the Hunter subregion. The approach is based on classical three-dimensional geological modelling approaches (Ross et al., 2004) to produce a simple regional model from poorly constrained datasets. The aim of this workflow is to model the large-scale stratigraphic units without introducing structural complexity into the grid geometry that is not supported by the available hard data (Wellmann et al., 2010). The workflow comprised:

  1. selection and processing of the observed data to form derived datasets and implementing the model numerical database (see Section 2.1.2.2.2), including:
    1. defining regional horizons
    2. determining horizon tops and lithological datasets from the deep well dataset and geological maps
    3. mapping the topography and bathymetry of the subregion from DEM data
  2. three-dimensional non-faulted and non-eroded geological modelling (see Section 2.1.2.2.3):
    1. selecting reference horizons and creating a horizon depth map
    2. isopach mapping
    3. building a preliminary (non-faulted and non-eroded) geological model
    4. extracting depth structure maps from the geological model
  3. fold and fault analysis to refine the geological model (see Section 2.1.2.2.4).

2.1.2.2.2 Data selection and processing

Spatial distribution of the deep wells data in the Hunter subregion is poor: only heterogeneous scattered well data are available as shown by Figure 6. These types of data are point source and do not provide much insight into the three-dimensional structure of the geological units at depth.

Generalised stratigraphic columns for the Hunter, Newcastle, Western and Central coalfields (shown in the geology section of the companion product 1.1 for the Hunter subregion (McVicar et al., 2015)) and correlations proposed by NSW Trade and Investment (Oliveira et al., 2014) were used to define ‘regional horizons’ for the purpose of this Assessment. Nine regional horizons were determined. They are named according to the nomenclature used by NSW Trade and Investment, with ‘M’ referring to Mesozoic and ‘P’ to Paleozoic. The relationship between these regional horizons and the stratigraphic units in each of the four coalfields are shown in Table 4.

Well completion reports (WCR) were analysed to determine the top depth of regional horizons. Of the 105 wells in the original dataset, 44 wells had information that could be used in the model. The markers of the horizon top depths in the wells are called ‘well picks’. Table 5 summarises for each well in the derived dataset the pick depths and the top of regional horizon to which it corresponds. The uncertainties in the depths to the tops of the horizons are not known. They are a function of the original well stratigraphic interpretation; a new integrated interpretation of all the well logs could remove some uncertainty at this stage. However, due to operational constraints, this type of integrated interpretation could not be achieved for this geological model.

The well top data and relevant structural contours from the outcrop limits and mapped isopachs in the Newcastle Coalfield (see Figure 35 in companion product 1.1) and Western Coalfield (see Figure 36 in companion product 1.1) were mapped for each of the regional horizons (Figure 7). P500 and P100 are the best constrained horizons in terms of number of well tops. Figure 7 shows the distribution of the nine horizon tops in nine separate maps.

Table 4 Relationship between regional horizons and stratigraphic units in the coalfields of the Hunter subregion


Regional horizon name

Age (geological stage)

Newcastle Coalfield

Hunter Coalfield

Western Coalfield

Central or Southern

coalfields

M600

Top Anisian

Top Hawkesbury Sandstone

Top Hawkesbury Sandstone

Top Hawkesbury Sandstone

Base Wianamatta Group

M700

Top Olenekian

Base Hawkesbury Sandstone

Base Hawkesbury Sandstone

Base Hawkesbury Sandstone

Base Hawkesbury Sandstone

P000

Top Changhsingian

Base Narrabeen Group

Base Narrabeen Group

Base Narrabeen Group

Base Narrabeen Group

P100

Upper Wuchiapingian

Base Newcastle Coal Measures

Base Newcastle Coal Measures

Top Watts Sandstone

Top Bargo Claystone

P500

Mid Capitanian

Base Tomago Coal Measures

Base Wittingham Coal Measures

Base Illawarra Coal Measures

Base Illawarra Coal Measures

P550

Top Wordian

Base Mulbring Siltstone

Base Mulbring Siltstone

Base Berry Siltstone

Base Berry Siltstone

P600

Mid Roadian

Base Maitland Group

Base Maitland Group

Base Shoalhaven Group

Base Shoalhaven Group

P700

Upper Kungurian

Base Greta Coal Measures

Base Greta Coal Measures

P900

Base Serpukhovian

Base Seaham Formation

Base Seaham Formation

Refer to the Sydney Basin stratigraphic column in Hodgkinson et al. (2016) for more details

Table 5 Depth to top of regional horizons from deep wells across the Hunter subregion


Well name

Regional horizon top

Pick depth

(m TVD ssa)

Well name

Regional horizon top

Pick depth

(m TVD ssa)

Allambi_1C

M700

–367.84

Howes_Swamp_1

P600

1545.87

Allambi_1C

P000

–258.84

Howes_Swamp_1

P900

2256.96

Allambi_1C

P500

–154.24

Hunter_Bulga_1

P000

24.53

Allambi_1C

P550

–138.04

Hunter_Bulga_1

P100

30.59

Allambi_1C

P900

–103.54

Hunter_Bulga_1

P500

463

Baulkham_Hills_1

M600

–89.4

Hunter_Bulga_2

P000

34.46

Baulkham_Hills_1

M700

154.44

Hunter_Bulga_2

P100

50.28

Baulkham_Hills_1

P500

977.4

Hunter_Bulga_2

P500

449.44

Berkshire_Park_1

M600

396.3

Hunter_Coricudgy_1

M700

–947.2

Berkshire_Park_1

M700

1062.3

Hunter_Coricudgy_1

P000

–462.2

Berkshire_Park_1

P000

3281.6

Hunter_Coricudgy_1

P500

403.15

Berkshire_Park_1

P500

3490.6

Jilliby_13

M700

5.3

Big_Adder_Hill_1

P500

550.32

Jilliby_13

P000

389.41

Black_Springs_1

P000

24.88

Jilliby_13

P100

550.69

Brawboy_1

M700

–356.18

Jilliby_13

P500

601.3

Brawboy_1

P000

238.92

Kenthurst_1

M600

0

Brawboy_1

P100

550.8

Kenthurst_1

M700

259.08

Brawboy_1

P500

667.12

Kenthurst_1

P000

856.18

Brawboy_2

M700

–171.5

Kenthurst_1

P100

1067.1

Brawboy_2

P000

225

Kulnura_1

M600

0

Brawboy_2

P100

534

Kulnura_1

M700

94.48

Brawboy_2

P500

638

Kulnura_1

P000

836.67

Catherine_Hill_Bay_1

P000

–42.3

Kulnura_1

P100

981.76

Catherine_Hill_Bay_1

P100

369.1

Kulnura_1

P500

1446.27

Catherine_Hill_Bay_1

P500

1003.7

Kulnura_1

P550

1904.39

Coolahville_1C

M700

–430.84

Kulnura_1

P600

2472.53

Coolahville_1C

P000

–280.94

Longley_1

M700

–180.7

Coolahville_1C

P500

–156.34

Longley_1

P000

627

Coolahville_1C

P550

–129.14

Longley_1

P100

1031.1

Coolahville_1C

P900

–125.34

Meads_Crossing_1

M700

–59.39

Cuan_1

M700

–297.5

Meads_Crossing_1

P000

79.9

Cuan_1

P000

54.5

Meads_Crossing_1

P500

168.09

Cuan_1

P100

448.1

Mellong_1

M600

0

Cuan_1

P500

978

Mellong_1

M700

112.77

Dartbrook_1

P100

–205.83

Mellong_1

P000

755.9

Dartbrook_1

P500

618.65

Mellong_1

P100

905.25

Dural_1

M700

265.17

Monkey_Place_1

P000

–101

Dural_1

P000

844.29

Monkey_Place_1

P100

–33.67

Dural_1

P100

1150.62

Monkey_Place_1

P500

452

Dural_1

P500

1150.62

Monkey_Place_2

P000

–88.43

Dural_1

P600

1368.55

Monkey_Place_2

P100

–60.39

Dural_1

P700

1422.5

Monkey_Place_2

P500

479.57

Dural_South_1

M600

–161.23

Monkey_Place_3

P000

–95.3

Dural_South_1

M700

76.52

Monkey_Place_3

P100

–16.11

Dural_South_1

P000

634.3

Monkey_Place_3

P500

472.7

Dural_South_1

P100

787.61

Monkey_Place_4

P000

–105.2

Dural_South_1

P500

1416.41

Monkey_Place_4

P100

–24.9

Dural_South_1

P550

2217.12

North_Colah_1

M700

–5.79

Dural_South_1

P600

2851.72

North_Colah_1

P000

681.95

Dural_South_1

P900

2851.72

North_Colah_1

P500

745.84

Elizabeth_Macarthur_1H

M700

137.82

Oakdale_1

M700

–262.5

Elizabeth_Macarthur_1H

P000

507.7

Oakdale_1

P000

30.9

Fullerton_2

P500

1140.23

Oakdale_1

P100

105

Fullerton_4

P500

157.5

Oakdale_1

P500

167.3

Hawkesbury_Bunnerong_1

M600

–25

Oakdale_1

P550

231.5

Hawkesbury_Bunnerong_1

M700

211.78

Oakdale_1

P900

250.8

Hawkesbury_Bunnerong_1

P000

776.02

Riverstone_1

M600

68

Hawkesbury_Bunnerong_1

P500

1226.05

Riverstone_1

M700

287

Hawkesbury_Eveleigh_1

M600

25

Riverstone_1

P000

96

Hawkesbury_Eveleigh_1

M700

290

Rouchel_Rouchel_1

P000

–289.9

Hawkesbury_Eveleigh_1

P000

912

Rouchel_Rouchel_2

P100

329.2

Hawkesbury_Eveleigh_1

P500

1317

Tangorin1

P700

–18.66

Higher_Macdonald_1

M700

3.38

Turnermans_1

P100

–21.8

Higher_Macdonald_1

P000

567.87

Wappinguy_1

P000

166.86

Higher_Macdonald_1

P500

606.88

Wappinguy_1

P500

669.42

Howes_Swamp_1

M600

–300.61

Windermere_1

P500

479.69

Howes_Swamp_1

M700

–216.79

Windermere_3

P500

512.64

Howes_Swamp_1

P000

429.69

Windermere_4

P500

–14.15

Howes_Swamp_1

P500

1288.92

aTVD ss = total vertical depth subsea reported to the Australian Height Datum

Data: see the listing of well completion reports within the list of references following Section 2.1.2.3

Figure 7

Figure 7 Distribution of regional horizon tops from the wells derived dataset for the Hunter subregion

(a) P900 shows all 105 petroleum wells. The 45 well picks that contained sufficient information to define a formation top are coloured differently for each horizon: (a) P900, (b) P700, (c) P600, (d) P550, (e) P500, (f) P100, (g) P000, (h) M700 and (i) M600.

‘Well picks’ are the markers of the formation top depths in the wells.

Outcrop limits are shown for the (c) P600 and (e) P500 horizons.

Isodepth points are shown for (g) P000

Data: DTIRIS NSW (Dataset 1), CSIRO (Dataset 2, Dataset 4), Bioregional Assessment Programme (Dataset 3)

The 3-second DEM and 9-second bathymetry data (see Section 2.1.2.1.1) were extrapolated with a local B-spline algorithm (Roxar package) to produce a topographic and bathymetric map of the Hunter subregion (Figure 8). The local B-spline algorithm calculates the amplitude of a family of bell-shaped functions (B-splines) using a local heuristic approach. The sum of these functions defines a function in (x, y) that approaches the input data: this method generates stable and functional results for all types of mapping (Emerson, 2016).

Figure 8

Figure 8 Surface topography and offshore bathymetry of the Hunter subregion

TVD ss = total vertical depth subsea reported to the Australian Height Datum; negative values represent elevation above sea level

Data: Bioregional Assessment Programme (Dataset 3), Geoscience Australia (Dataset 5, Dataset 6)

2.1.2.2.3 Three-dimensional non-faulted and non-eroded geological model

The stratigraphic reference horizons selected for the Sydney Basin geological model of the Hunter subregion were the P900, P500, P100 and P000 horizons (Figure 9). These were chosen because they have the highest density of well picks or are important regional stratigraphic markers in the Sydney Basin (Oliveira et al., 2014).

Horizon P900 is the base of the Sydney Basin sedimentary sequence (i.e. the surface of the geological basement). The horizon depth map for P900 (Figure 9a) was obtained using a global B‑spline algorithm with a 500 m lateral step (x and y). The modelled depth varies between +5780 and –730 m total vertical depth subsea (TVD ss). This reference is used in the well database (Table 5) and is kept in the geological model for consistency purposes. Modelled depths have been compared with the OZ SEEBASE dataset and the calibration with well picks. Basement highs occur onshore along the western and north-east borders of the Sydney Basin; and offshore along a north-east to south-west trend. This basement structure is located deeper than 3000 m below sea level in the central and eastern parts of the basin where thick Permian and Mesozoic sediment layers are observed (purple colours, Figure 9a). This map is an intermediary modelling result that only considers the well picks of P900, without correlation with other structural data and the present-day surface topography.

The P500, P100 and P000 horizons are all important levels within the geological model as they are the stratigraphic boundaries of the main upper Permian coal-bearing units: the Newcastle, Wittingham, Illawarra and Tomago coal measures (Table 4). The initial modelling step maps each horizon independently and does not account for the present-day erosional level (as explained in the workflow, Section 2.1.2.2.1). The horizons were mapped with a 500 m increment in x and y based on the well picks and available mapping data. P500 depth varies between +1830 and –730 m TVD ss, P100 between 1050 and –733 m, P000 between 918 and –796 m. At this stage of the model development, the resulting horizon depth maps are non-deterministic, and the level of geological uncertainty increases proportionally with decreased number of data points. The horizons only conform to the available well data and regional maps, and are not further constrained by other sources of stratigraphic information. This means that there may be a poor level of consistency between different horizons, and in some cases they may even overlap at depth. To build a more coherent stratigraphic model, the next step is to generate isopach maps (thickness maps) of each geological unit, which can be used to constrain the three-dimensional structure of each horizon away from the wells (Figure 9 and Figure 10).

Figure 9

Figure 9 Non-eroded and non-folded regional horizon maps for Hunter subregion reference horizons: (a) P900, (b) P500, (c) P100 and (d) P000

(a) P900 shows all 105 petroleum wells. Well picks used to define formation tops are coloured differently for each horizon.

‘Well picks’ are the markers of the formation top depths in the wells.

Outcrop limits are shown in (b) for the P500 horizon.

Isodepth points are shown for (d) P000

Data: DTIRIS NSW (Dataset 1), CSIRO (Dataset 2, Dataset 4), Bioregional Assessment Programme (Dataset 3)

Six isopach maps were developed by extrapolation away from the calibration points defined by the 44 input wells and with reference to existing mapping (Table 4, Figure 7, Figure 9, Figure 10). The calibration points are called isodepths. An isodepth is the difference between two well picks for a given stratigraphic unit at a given well location. To obtain an isodepth measurement, two consecutive stratigraphic well tops must be available within the well dataset:

  1. top of P700 to top of P900 – lower Permian volcanic-bearing conglomerate, siltstone and sandstone units of the Dalwood Group
  2. top of P600 to top of P700 – Greta Coal Measures
  3. top of P500 to top of P550 – siltstone-dominated units, including Mulbring Siltstone in Newcastle and Hunter coalfields, and Berry Siltstone in Western Coalfield
  4. top of P000 to top of P100 – upper Permian coal-bearing units including the Newcastle Coal Measures in Hunter and Newcastle coalfields, and Illawarra Coal Measures in the Western Coalfield (base of Narrabeen Group to top of Watts Sandstone)
  5. top of M700 to top of P000 –Triassic Narrabeen Group
  6. top of M600 to top of M700 –Hawkesbury Sandstone.

These maps provide a two-dimensional representation of geological unit thickness. For the isopach maps that show the top of P000 to top of P100 (Figure 10d) and top of M700 to top of P000 (Figure 10e), thickness varies between 200 to 790 m and 154 to 789 m, respectively. Thickening occurs from the central Hunter subregion towards the north and south-east of the subregion.

For the other isopach maps, there are no obvious thickness trend variations and a constant thickness has been adopted. Average value of these thicknesses have been determined based on the well pick depths (Table 5) and thickness of formations given in Section 1.1.3.2 of companion product 1.1 (McVicar et al., 2015):

  • 365 m for the Dalwood Group (top of P700 to top of P900)
  • 88 m for Greta Coal Measures (top of P600 to top of P700)
  • 500 m for Mulbring Siltstone in Newcastle and Hunter coalfields and the Berry Siltstone in the Western Coalfield (top of P500 to top of P550)
  • 290 m for Hawkesbury Sandstone (top of M600 to top of M700).

The resulting isopach maps are non-deterministic and the uncertainty increases as the isopach data concentration decreases. These isopach maps are another intermediary step in the development of the final geological model and do not represent the final isopach data in the Hunter subregion geological model.

Figure 10

Figure 10 Isopach maps showing distribution of formation thicknesses from the wells derived dataset for the Hunter subregion

(a) Top of P700 to top of P900, (b) Top of P600 to top of P700, (c) Top of P500 to top of P550, (d) Top of P000 to top of P100, (e) Top of M700 to top of P100 and (f) Top of M600 to top of M700

An isodepth is the difference between two well picks for a given stratigraphic unit at a given well location

Data: DTIRIS NSW (Dataset 1), CSIRO (Dataset 2, Dataset 4), Bioregional Assessment Programme (Dataset 3)

The reference structural maps of P900, P500, P100, P000 (Figure 9) and the isopach maps (Figure 10) were used as input to build a non-faulted and non-eroded geological model. The geological modelling option of the Roxar RMS software takes all this information into account and can generate different simulations depending on the weight given to each dataset. A scenario giving a weight of ‘1’ to the reference horizons and ‘0.5’ to the isopachs was selected. The weighting adjustment is a classical approach in geological modelling, used to minimise the differences between the model results and the input data (Caumon et al., 2009). In this case, the reference horizons are known to be more constrained than the isopachs (Figure 9 and Figure 10) and can thus be used to build the model. The geological modelling option corrects the potential errors in each horizon interpolation, taking into account the interaction between all horizons to respect a minimum thickness between each horizon as well as the geological chronology (for example, P500 cannot occur above P100). This model has a horizontal resolution of 2 km × 2 km (x, y), with 109 layers along the vertical (z) for a total of 511,118 cells. The layers along the vertical can have variable size: 109 layers were adopted to allow a maximum thickness of 200 m to each cell and enable the generation of a facies grid with a non-homogeneous lithology within each unit. The depth ranges between 1185 m above sea level (the highest elevation known in the subregion) and 5062 m below sea level (the deepest part of the sedimentary infill sequence of the Sydney Basin).

Figure 11 shows depth structure maps of the non-faulted and non-eroded horizons extracted from the geological model.

Figure 11

Figure 11 Depth structure maps of folded and non-eroded horizons extracted from the geological model for the Hunter subregion

Outcrop limits are shown on (e) and (c) and the isodepth points on (g).

‘Well picks’ are the markers of the formation top depths in the wells.

The depths are in TVD ss AHD = total vertical subsea reported to Australian Height Datum; negative values represent elevation above sea level

Data: DTIRIS NSW (Dataset 1), CSIRO (Dataset 2, Dataset 4), Bioregional Assessment Programme (Dataset 3)

2.1.2.2.4 Preliminary three-dimensional geological model

The extracted non-eroded horizons (Figure 11) were compared with the well data in the geomodelling software (Roxar RMS) and the difference between the datasets was minimised as a function of the main fold trends observed within the Hunter subregion. Then, the resulting horizons were eroded by the topographic level to obtain a preliminary three-dimensional present-day geological model (CSIRO, Dataset 9). Folding seems to be the dominant first-order structure influencing the regional horizon structures: anticline and synclines are indeed the most represented structure within the subregion (Stewart and Alder, 1995).

Fault and fractures are present within the basin but there is insufficient information about their three-dimensional structure, dip, throw and displacement to inform the model in the time frame of the Assessment. Therefore, faults have not been represented in the model (see Section 2.1.2.3).

The three-dimensional structure of the preliminary geological model and the facies distribution within the structural grid are illustrated in Figure 12 and Figure 13. The facies were simply defined at regional scale from the regional generalised stratigraphic column of the Permian and Triassic units in the main coalfields of the Sydney geological basin (Figure 33 of companion product 1.1. for the Hunter subregion (McVicar et al., 2015)). Figure 13 provides an overview of the facies architecture, illustrating the mixed facies of the Permian coal-bearing intervals capped in the central and western parts of the basin by more uniform Triassic sandstone, siltstone and mudstone formations. Figure 14 shows the regional horizon tops extracted from this folded and eroded geological model. The main fold trends affect the entire stratigraphic sequence and the white areas reflect the zones eroded at present day. The model had been folded through several numerical iterations of Roxar RMS stratigraphic modelling plugging to minimise the difference between the primary maps (Figure 11) and the well picks (Table 4). While Figure 14 also shows the surface expressions of faults from previous geological mapping, they are not included in the Hunter geological model.

Figure 12

Figure 12 Three-dimensional perspective view of the eroded and folded geological model for the Hunter subregion

View from the south, showing the surface expression of model horizons across the subregion. Upland areas are dominated by horizon 1; in the more eroded areas, the Newcastle, Tomago, Wittingham and Illawarra Coal Measures (horizons 4 and 5) are at or closer to the surface, and are the main mining areas of the Hunter and Western coalfields where a mix of open-cut and underground mining occur; only underground mining occurs in the Newcastle Coalfield along the coast where the Greta Coal Measures (horizon 8) are deeper.

Data: CSIRO (Dataset 9)

Figure 13

Figure 13 Three-dimensional perspective view of the regional lithology pattern in the geological model for the Hunter subregion

View from the south, showing a change in surface lithology from siltstone and sandstone outcrops in the upland areas of the subregion to shale and sand bed lithologies in more eroded areas. Mining is undertaken in areas where the siltstone and sandstone outcrops have been eroded.

Data: CSIRO (Dataset 9)

Figure 14

Figure 14 Depth structure maps extracted from the folded and eroded geological model for the Hunter subregion

Depth in m TVD ss = total vertical depth to the Australian Height Datum; negative values represent elevation above sea level. Faults from previous geological mapping are shown, but are not included in the Hunter geological model.

Data: CSIRO (Dataset 9)

Last updated:
18 January 2019
Thumbnail of the Hunter subregion

Product Finalisation date

2018

ASSESSMENT