2.1.3.2.1 Hydrochemistry
In descending stratigraphic order the hydrostratigraphic units used in this section are:
- Winton-Mackunda partial aquifer
- Rolling Downs Group aquitard (includes Allaru Mudstone, Toolebuc Formation, Wallumbilla Formation)
- Cadna-owie – Hooray Sandstone aquifer
- Injune Creek Group aquitard (includes Westbourne Formation, Adori Sandstone, Birkhead Formation)
- Hutton Sandstone aquifer
- Moolayember Formation aquitard
- Clematis Group aquifer (includes Warang Sandstone)
- upper Permian coal measures partial aquifer
- Joe Joe Group partial aquifer.
As expected the majority of samples are from major regional aquifers, with some samples from local aquifers situated within regional aquitards (i.e. the Injune Creek Group aquitard and the Moolayember Formation aquitard).
Hydrochemical trends in the hydrogeologic units were investigated using a variety of techniques discussed below.
2.1.3.2.1.1 Ion:chloride ratios
Chloride (Cl) is often assumed to be a conservative ion in solution. Once it is introduced into a groundwater system it remains dissolved because there is no water–rock interaction that can remove chloride from groundwater (Appelo and Postma, 2006), although halite dissolution processes may increase its concentration. In contrast, the other major ions can enter or leave a groundwater system through adsorption and desorption, or precipitation and dissolution.
The chemical processes which are not related to surface processes such as evapotranspiration can be identified by the changes in ion:Cl ratios in groundwater. The HCO3:Cl, (Na+K):Cl and (Ca+Mg):Cl ratios can be used to compare concentrations of ions (units meq/L) in excess of their respective recharge ratios. The concentrations are a measure of the input of ions to groundwater from water–rock interactions or mixing of different water bodies, relative to the concentration of ocean-derived salts in rainfall. Ion:Cl ratios were used to examine the sources of major ions in solution for the different hydrogeologic units.
2.1.3.2.1.2 Sodium Absorption Ratio
Sodium, magnesium and calcium are the dominant cations in natural waters. The sodium absorption ratio (SAR) is the fraction of exchangeable sodium over the square root of half of the sum of the calcium and magnesium concentrations (Fetter, 2001). High sodic groundwater, when used in surface environments (e.g. irrigation cropping), can affect soil quality and result in soil dispersibility. Calcium and magnesium ions are preferentially held over sodium ions in cation exchange sites in clays. Where sodium concentrations are high relative to calcium and magnesium, sodium may displace them from the cation exchange sites. Sodium has a large hydrated ionic radius and it tends to push the layered lattices of clay minerals apart, ultimately causing disaggregation and loss of soil structure.
SAR is expressed as:
|
|
(2) |
where Na+, Ca2+ and Mg2+ are expressed as milliequivalents per litre (meq/L) (Appelo and Postma, 2006). SAR was used in conjunction with ion:Cl ratios to examine which aquifer properties may be responsible for changes in groundwater chemistry.
2.1.3.2.1.3 Hydrochemical characterisation
In this section hydrogeologic units are assigned to hydrochemical systems based on major ion abundances. Hydrochemical trends in each hydrogeologic unit are identified using ion-TDS and ion-chloride relationships which are summarised in Table 8. It should be noted that the hydrogeological units used conform to the hydrostratigraphic units used for the Galilee groundwater model (see companion product 2.6.2 for the Galilee subregion (Peeters et al., 2018)).
The data used in this study are composited from analyses accumulated over a long timescale, from 1938 to 2013. Approximately 70% of samples were collected between 1970 and 2013. It is assumed that given the long residence time of water in the GAB there would be little short-term variation in the data, and, given the absence of significant resource development in the region, only minor changes in the water chemistry over the time period are represented by the samples. This view is supported by Moya et al. (2015) who performed hierarchical cluster analysis on samples from the Galilee subregion and observed that samples collected from the same bore at different times were consistently assigned to the same hierarchical cluster group. The data discussed below therefore represent a generalised picture of the hydrochemistry of the subregion over several decades. Such data should be appropriate for investigating regional trends and processes operating over a large part of the basin, but may not necessarily be suited to a detailed study of localised processes.
Major ions
Major ion abundances were used to classify the hydrogeologic units into hydrochemical systems. Figure 35 shows the abundances of major ions relative to TDS for the different hydrogeologic units in the subregion. Table 8 shows R2 values for regressions undertaken on major ion relationships to TDS. On the basis of the relative abundances of major ions, three distinct groundwater types can be differentiated:
- Na-Cl dominated groundwater type with minor SO4 (Winton-Mackunda partial aquifer and Rolling Downs Group aquitard)
- pH dependent, Na-HCO3-Cl groundwater type (the Cadna-owie – Hooray Sandstone aquifer, Injune Creek Group aquitard, and Hutton Sandstone aquifer)
- Na-Cl type with minor HCO3 (the Clematis Group aquifer, upper Permian coal measures partial aquifer, and Joe Joe Group partial aquifer).
Figure 35 Relative average abundances of elements in hydrogeologic units for the Galilee subregion
TDS = total dissolved solids
Data: Bioregional Assessment Programme (Dataset 1)
Table 8 Ion-TDS R2 values for hydrogeologic units in the Galilee subregion
aregression fitted to samples with TDS-EC relationships between 0.50 and 0.90, after eliminating EC data points with zero values and obvious errors
bMoolayember Formation data has virtually same Cl-TDS slope
cregression limited to groundwater samples with pH
TDS = total dissolved solids, EC = electrical conductivity
Data: Bioregional Assessment Programme (Dataset 1)
The reason for these different water types is likely due to a combination of mineralogical and depositional characteristics of the host material of the aquifers, and differences in hydrochemical evolution along flow paths. The hydrochemical evolution may be largely an indication of the age of the groundwater. The clustering of groundwater types closely mirrors the stratigraphy of the hydrogeologic units; each hydrochemical system in the subregion consists of a number of hydrogeologic units that overlie one another.
This suggests that there is limited flow across the regional aquitards identified in the subregion. These are the Rolling Downs Group aquitard, which separates the Winton-Mackunda partial aquifer from the Cadna-owie – Hooray Sandstone aquifer, and the Moolayember Formation, which separates the Hutton Sandstone aquifer from the Clematis Group aquifer. Between these regional aquitards there may be hydraulic continuity between overlying hydrogeologic units, causing similarities in groundwater chemistry.
It is worth noting that despite the relatively high average abundance of HCO3 in the Clematis Group aquifer, upper Permian coal measures and Joe Joe Group partial aquifers, the R2 values for HCO3 and TDS in these units are 0.02, 0.15 and 0.02 respectively. This suggests that HCO3 reaches high concentrations locally, but may not be a significant component of TDS through the full extent of these hydrogeologic units.
These hydrochemical groupings are slightly different from a revised hydrochemical stratigraphy proposed by Moya et al. (2015). Using a combination of hierarchical clustering, principal component analysis (PCA) and factor analysis, they identified three hydrochemical groups:
- a Na-Cl dominated group containing brackish waters, seen in the Winton-Mackunda formations, Allaru Mudstone and Toolebuc Formation. Similar to the Na-Cl dominated group identified with the exception that they exclude the Wallumbilla Formation based on a lower mean TDS
- a Na-HCO3-Cl group containing slightly brackish waters, seen in the Wallumbilla Formation, Cadna-owie Formation, Hooray Sandstone and Westbourne Formation. These aquifers are recognised as containing dissolved gas which is believed to be primarily CO2. This again is similar to the Na-HCO3-Cl group identified by the BA programme, though Moya et al. (2015) exclude the Hutton Sandstone from this group and include the Wallumbilla Formation
- a Na-HCO3 dominated group containing largely fresh waters and more dissolved gas than overlying units, occurring within the Adori Sandstone, Birkhead Formation, Hutton Sandstone and Clematis Sandstone.
The Na-HCO3 group differs significantly from the hydrochemical system for the Galilee Basin units identified by the BA programme, which is a Na-Cl dominated system seen in the Clematis Group, upper Permian coal measures and Joe Joe Group.
Key differences between the hydrochemical classification of Moya et al. (2015) and this work are the inclusion by Moya et al. (2015) of the Wallumbilla Formation (Rolling Downs Group aquitard) in the Na-HCO3-Cl system, and the grouping of the Hutton Sandstone with the Clematis Group in a Na-HCO3 type system. No hierarchical clustering was undertaken in the Galilee subregion BA, and hydrochemical systems were defined on the basis of average ionic abundance as a fraction of TDS, and the ions with which TDS was highly correlated. Significant data were included in the Galilee subregion BA for the upper Permian coal measures and Joe Joe Group, which were not analysed by Moya et al. (2015). The data indicates to these authors that the units underlying the Moolayember Formation comprise a generally more brackish hydrochemical system than the units overlying the Moolayember Formation, with a predominance of Na and Cl and a weaker relationship to HCO3 than in the Eromanga Basin aquifers. The possibility of a hydrological disconnect between the Hutton Sandstone and Clematis Group aquifers is discussed by Moya et al. (2015) who conclude that it seems probable that the Moolayember Formation forms a tight aquitard between these units, but there are insufficient data to discount the possibility of water being exchanged between these units. In this product the possibility of leakage across the Moolayember Formation is explored through a number of different methods.
Detailed hydrochemistry of each hydrogeologic unit is summarised below. These data have been treated as a regionally representative dataset to examine the processes controlling water quality in each hydrogeologic unit at a regional scale. However, differences in hydrochemical processes present at smaller scales have not been examined for each hydrogeologic unit. The complexity of the data suggests that there may be different processes acting along different flow paths for each hydrogeologic unit, rather than groundwater evolving over a single chemical pathway in the subregion.
Winton-Mackunda partial aquifer
The Winton-Mackunda partial aquifer overlies the Rolling Downs Group aquitard. A total of 611 samples passed the QC process. Of these, 440 had pH data available for the multivariate analysis. Salinity has a very broad range, with TDS from 79 to 20,400 mg/L. The principal ions are Na and Cl, with minor SO4. Their relative abundances are 32%, 47% and 9% respectively (Table 9, Figure 36).
Table 9 Hydrochemistry of the Winton-Mackunda partial aquifer
NA = data not available, SAR = sodium absorption ratio, TDS = total dissolved solids, std dev = standard deviation
Data: Bioregional Assessment Programme (Dataset 1)
Data: Bioregional Assessment Programme (Dataset 1)
Distribution of TDS is variable in the Winton-Mackunda partial aquifer, with a number of areas of higher salinity occurring through the subregion. The Winton-Mackunda partial aquifer shows little change in ion:Cl ratios with TDS (Figure 37).
TDS = total dissolved solids
Data: Bioregional Assessment Programme (Dataset 1)
Rolling Downs Group aquitard
The Rolling Downs Group aquitard overlies the Cadna-owie – Hooray Sandstone aquifer. A total of 159 samples passed the QC process. Of these, 127 had pH values available for the multivariate analysis. Salinity shows a broad range, with TDS from 170 to 12,735 mg/L. The principal ions are Na and Cl, with minor SO4. Their respective abundances are 29%, 43% and 12% (Table 10, Figure 38). It is apparent that some samples on Figure 38 have anomalous ionic concentrations, for example low Cl or HCO3 relative to TDS. Further investigation may be warranted to determine whether the anomalies are due to hydrogeological processes or is a sample artefact.
Table 10 Hydrochemistry of the Rolling Downs Group aquitard
NA = data not available, SAR = sodium absorption ratio, TDS = total dissolved solids, std dev = standard deviation
Data: Bioregional Assessment Programme (Dataset 1)
Data: Bioregional Assessment Programme (Dataset 1)
Water in the Rolling Downs Group aquitard is generally fresh with an area of higher salinity in the south of the subregion. Ion:Cl ratios show little variability with TDS (Figure 39). Other than some high Na+K:Cl and HCO3:Cl ratios in the fresher samples, ion:Cl ratios occupy a very narrow band.
TDS = total dissolved solids
Data: Bioregional Assessment Programme (Dataset 1)
Cadna-owie – Hooray Sandstone aquifer
There were 1302 samples for the Cadna-owie – Hooray sandstone aquifer after QA/QC filtering, of which 1269 samples had pH values. Samples were collected in the formation at a depth of up to 2920 m. The Cadna-owie – Hooray Sandstone aquifer shows a broad range in salinity as measured by TDS (180–7136 mg/L). The major ions are Cl, Na and HCO3. The mean concentrations of these ions as a percentage of TDS are 17%, 28% and 47% respectively (Table 11, Figure 40). It is apparent that some samples shown on Figure 40 have anomalous ionic concentrations, for example low Cl or HCO3 relative to TDS. Further investigation in the future may determine whether the anomalies are due to a hydrogeological process or are an artefact of the sample.
Table 11 Hydrochemistry of the Cadna-owie – Hooray Sandstone aquifer
NA = data not available, SAR = sodium absorption ratio, TDS = total dissolved solids, std dev = standard deviation
Data: Bioregional Assessment Programme (Dataset 1)
Data: Bioregional Assessment Programme (Dataset 1)
Like the Injune Creek Group and Hutton Sandstone aquifer, the Cadna-owie – Hooray Sandstone aquifer is generally fresh (TDS < 1000 mg/L) with higher salinities occurring on the Maneroo Platform. Ion:Cl ratios are variable where TDS is below about 2000 mg/L, but has a lower variance where TDS is greater than 2000 mg/L (Figure 41).
TDS = total dissolved solids
Data: Bioregional Assessment Programme (Dataset 1)
Injune Creek Group aquitard
The Injune Creek Group aquitard overlies the Hutton Sandstone aquifer. A summary of the hydrochemical data for the Injune Creek Group aquitard is presented in Table 12. There were 146 samples available for analysis after QA/QC filtering, of which 134 samples had pH values.
Table 12 Hydrochemistry of the Injune Creek Group aquitard
NA = data not available, SAR = sodium absorption ratio, TDS = total dissolved solids, std dev = standard deviation
Data: Bioregional Assessment Programme (Dataset 1)
Injune Creek Group aquitard shows a broad range in salinity (286–9541 mg/L TDS). The major ions are Cl, Na and HCO3. The mean concentrations of these ions as a percentage of TDS are 20%, 27% and 45% respectively (Figure 42).
Data: Bioregional Assessment Programme (Dataset 1)
Like the Hutton Sandstone aquifer, the Injune Creek Group aquitard is fresh in much of the subregion, with higher salinities occurring on the Maneroo Platform. This may be an indication of hydraulic connection between these two hydrogeologic units, or may reflect similarities in the composition of aquifer material. Ion:Cl ratios are variable in the Injune Creek Group aquitard, indicating a variety of processes acting on solute concentrations (Figure 43). There is a tendency for low Ca+Mg:Cl ratios where TDS is high.
TDS = total dissolved solids
Data: Bioregional Assessment Programme (Dataset 1)
Hutton Sandstone aquifer
The Hutton Sandstone aquifer overlies the Clematis Group aquifer, separated by the Moolayember Formation which acts as a regional aquitard (see companion product 1.1 for the Galilee subregion (Evans et al., 2014)). There were 1302 samples for the Hutton Sandstone after QA/QC filtering, of which 1269 samples had pH values. The Hutton Sandstone shows a significant range in salinity (55–3579 mg/L TDS). The major ions are Cl, Na and HCO3. The mean concentrations of these ions as a percentage of TDS are 14%, 23% and 54% respectively (Table 13, Figure 44). It is apparent that some samples shown on Figure 44 have anomalous ionic concentrations, for example low or high HCO3 relative to TDS. Further investigation may be warranted in the future to determine whether the anomalies are due to a hydrogeological process or is an artefact of the sample.
Table 13 Hydrochemistry of the Hutton Sandstone aquifer
NA = data not available, SAR = sodium absorption ratio, TDS = total dissolved solids, std dev = standard deviation
Data: Bioregional Assessment Programme (Dataset 1)
Data: Bioregional Assessment Programme (Dataset 1)
Hutton Sandstone aquifer is predominantly fresh (TDS < 1000 mg/L), with higher salinities on the Maneroo Platform. Ion:Cl ratios are highly variable in the Hutton Sandstone aquifer, indicating a variety of processes affecting solute loads (Figure 45). These may include evapotranspiration, mixing of waters from different hydrogeologic units, and water–rock interactions.
Data: Bioregional Assessment Programme (Dataset 1)
Clematis Group aquifer
The Clematis Group aquifer overlies the upper Permian coal measures and underlies the Moolayember Formation, which is thought to keep it hydraulically separate from the overlying Hutton Sandstone aquifer (see companion product 1.1 for the Galilee subregion (Evans et al., 2014)). A summary of the hydrochemical data is presented in Table 14. There were 98 samples available for analysis after QA/QC filtering, of which 88 samples had pH values for multivariate analysis.
Table 14 Hydrochemistry of the Clematis Group aquifer
NA = data not available, SAR = sodium absorption ratio, TDS = total dissolved solids, std dev = standard deviation
Data: Bioregional Assessment Programme (Dataset 1)
Salinity in the Clematis Group aquifer shows a broad range (103–3290 mg/L TDS). The major ions are Cl, Na and HCO3. The mean concentrations of these ions as a percentage of TDS are 31%, 26% and 33% respectively, with minor Ca (3%) and sulfate (3%) (Figure 46). The groundwater in the Clematis Group aquifer can be generally described as Na-Cl-HCO3.
Data: Bioregional Assessment Programme (Dataset 1)
It is apparent that some samples shown on Figure 46 have anomalous ionic concentrations, for example low HCO3 relative to TDS. Further investigation in the future may determine whether these anomalies are due to hydrogeological processes or are an artefact of the sample.
Groundwaters in the Clematis Group aquifer are fresh (TDS 100–200 mg/L) close to the Clematis Group outcrop in the east of the subregion. Further west salinities are higher (up to 3000 mg/L TDS). Close to the western extent of the Clematis Group aquifer, in the central part of the Galilee subregion between the Maneroo Platform and Aramac, groundwater in the Clematis Group has lower salinity (300–500 mg/L TDS).
Ion:Cl ratios in the Clematis Group aquifer (Figure 47) show little variation with TDS, much like the upper Permian coal measures (Figure 49) and Joe Joe Group, except for a region of elevated Na+K:Cl ratios and HCO3:Cl ratios at relatively low TDS.
Figure 47 Ion:chloride (Cl) ratios versus total dissolved solids (TDS) in the Clematis Group
TDS = total dissolved solids
Data: Bioregional Assessment Programme (Dataset 1)
Upper Permian coal measures partial aquifer
The upper Permian coal measures overlies the Joe Joe Group and underlies the Clematis Group. A summary of the hydrochemical data for the upper Permian coal measures is presented in Table 15. There were 132 samples available for analysis after QA/QC filtering, of which 46 samples had pH values for the multivariate analysis.
Salinity in the upper Permian coal measures shows a broad range. The major ions are Cl, Na and HCO3. The mean concentrations of these ions as a percentage of TDS are 34%, 28% and 26% respectively, with minor Ca (3%) and Mg (2%) and SO4 (5%) (Figure 48). The groundwater can be generally described as Na-Cl-HCO3 with possible calcium-magnesium carbonate species. The possibility of secondary gypsum or dolomite forming in the aquifer material should be investigated further.
Table 15 Hydrochemistry of the upper Permian coal measures partial aquifer
NA = data not available, SAR = sodium absorption ratio, TDS = total dissolved solids, std dev = standard deviation
Data: Bioregional Assessment Programme (Dataset 1)
Data: Bioregional Assessment Programme (Dataset 1)
Bores screened in the upper Permian coal measures partial aquifer have a limited distribution. The freshest water is located around areas of outcrop where recharge occurs (see companion product 1.5 for the Galilee subregion (Evans et al., 2015)). Ion:Cl ratios in the upper Permian coal measures partial aquifer show little variation with TDS (Figure 49). It is apparent that some samples shown on Figure 48 have anomalous ionic concentrations, for example low HCO3 relative to TDS. Further investigation may be warranted to determine whether these anomalies are due to hydrogeological processes or are a sample artefact.
TDS = total dissolved solids
Data: Bioregional Assessment Programme (Dataset 1)
Joe Joe Group partial aquifer
The Joe Joe Group partial aquifer is the lowermost hydrogeologic unit in the Galilee Basin sequence. A summary of the hydrochemical data are presented in Table 16. There were 106 samples available for analysis after QA/QC filtering, of which 99 samples had pH values for the multivariate analysis.
Salinity in the Joe Joe Group aquifer shows a broad range (175–11,060 mg/L TDS). The major ions are Cl, Na and HCO3. The mean concentrations of these ions as a percentage of TDS are 36%, 23% and 24% respectively with minor Ca (4%) and SO4 (6%). The groundwater can be generally described as Na-Cl with minor bicarbonate.
Table 16 Hydrochemistry of the Joe Joe Group partial aquifer
NA = data not available, SAR = sodium absorption ratio, TDS = total dissolved solids, std dev = standard deviation
Data: Bioregional Assessment Programme (Dataset 1)
Data: Bioregional Assessment Programme (Dataset 1)
The strong correlation between TDS and sodium and chloride (Figure 50) indicates that these two elements account for the majority of salinity changes in the Joe Joe Group. In the south of the Galilee subregion a small area of high bicarbonate concentration is at odds with this trend. Anomalously, TDS for the Joe Joe Group is highest in the north-east of the Galilee subregion, close to where the Permian and Carboniferous rocks outcrop. This may be an indication that recharge does not occur in this region, or that potential for flow is from the deep central parts of the basin to the margins. Ion:Cl ratios in the Joe Joe Group show very little variation with TDS (Figure 51).
TDS = total dissolved solids
Data: Bioregional Assessment Programme (Dataset 1)
2.1.3.2.1.4 Discussion
Solute sources
Defining the sources of solutes to groundwater is important for understanding how groundwater will evolve along a flow path, and to understand the potential for connectivity between different hydrogeologic units. Moya et al. (2015) concluded that in the recharge area in the east of the basin, evaporative concentration of cyclic salts is an important process and leads to a dominance of Na and Cl ions. This is a trend which is common to many of the hydrogeologic units studied here, with the exception of samples collected from the central and western parts of the Galilee subregion. As stated above, the complexity of the dataset indicates that more than one process operates in each hydrogeologic unit in different parts of the subregion. The following discussion attempts to define the major processes operating in each hydrogeologic unit but may not identify local processes operating only in a small area of the subregion.
Na-Cl system (Winton-Mackunda partial aquifer, Rolling Downs Group aquitard)
Both the Winton-Mackunda partial aquifer and the Rolling Downs Group aquitard show very high TDS values in some samples (> 10,000 mg/L). Both these units show ion:Cl ratios to be unaffected by increases in TDS at higher salinities; beyond 5,000 mg/L in the Winton-Mackunda partial aquifer, and beyond 2,000 mg/L in the Rolling Downs Group aquitard. This may be indicative of evaporative concentration, which increases TDS without altering the ion:Cl ratio. Radke et al. (2000) suggested that diffusion of original marine salts contained within the marine units of the Rolling Downs Group aquitard and the Mackunda Formation were a controlling factor on high solute loads.
It seems likely that marine salts are the dominant control on very high TDS values. The greater variance in ion:Cl ratios where TDS values are low may represent variable ion:Cl ratios in recharge water, or a mixing signal between marine salts and the cyclic salts contained in recharge water, which becomes overprinted by the marine salt signal as water flows through the system.
Na-HCO3-Cl system (Cadna-owie – Hooray Sandstone aquifer, Injune Creek Group aquitard, Hutton Sandstone aquifer)
All three units in this hydrochemical system show greater variability in ion:Cl ratios than the Na-Cl system, and do not show the same restriction of ion:Cl ratios where TDS is high. The variability in ion:Cl ratios may reflect an increased importance of water–rock interaction, such as carbonate dissolution, in this system.
The bicarbonate concentrations of the Hutton Sandstone aquifer and Injune Creek Group aquitard are up to twice that observed in the Joe Joe Group and upper Permian coal measures partial aquifers, or Clematis Group aquifer. This is consistent with observations of carbonate cements in the Jurassic and Cretaceous sequence of the Eromanga Basin (Draper, 2002).
All three units in this hydrochemical system display a wide range of Na+K:Cl, Ca+Mg:Cl, and HCO3:Cl ratios where TDS is low (below 1000–2000 mg/L). Where TDS values are higher, Na+K and HCO3:Cl values may also be high, but Ca+Mg:Cl ratios are consistently low. High Na+K:Cl ratios where TDS is high coupled with low Ca+Mg:Cl ratios may be an indication of cation exchange processes in the aquifer. Herczeg et al. (1991) observed a similar trend and suggested that a combination of carbonate dissolution and cation exchange was the dominant influence on solute concentrations in the Hutton Sandstone aquifer. Based on similarities in the hydrochemistry, this process may also be operating in the Cadna-owie – Hooray Sandstone aquifer and Injune Creek Group aquitard. This process is further explored using the sodium absorption ratio (SAR).
SAR is a measure of the ratio of sodium to calcium and magnesium in solution. In the Galilee subregion, SAR shows a positive correlation with TDS in most units. This is most pronounced in the Hutton Sandstone aquifer, Injune Creek Group aquitard, and Cadna-owie – Hooray Sandstone aquifer. Carbonate dissolution would create high HCO3:Cl ratios, but would be expected to be accompanied by a rise in Ca+Mg:Cl ratios (or reduction in SAR) as well. Instead, elevated SAR levels occur where HCO3:Cl ratios are high. Herczeg et al. (1991) observed a similar trend in samples from the western region of the Eromanga Basin, and suggested that carbonate dissolution may be followed by cation exchange processes, in which Na in clay minerals is exchanged for Ca and Mg in solution, reducing the ratio of Ca+Mg:Cl and allowing SAR to increase where HCO3 concentrations are high.
Another process which could lead to high HCO3:Cl and Na:Cl ratios with low Ca:Cl ratios is albite dissolution, which contributes sodium, bicarbonate and dissolved silica to solution. However, Si:Cl ratios are lowest where SAR is high (Figure 52), which is inconsistent with silicate mineral dissolution contributing solutes to groundwater. The process of carbonate dissolution followed by cation exchange, outlined by Herczeg et al. (1991), adequately explains the trends seen in the data and is assumed to be the dominant control on solute loads in the Na-HCO3-Cl hydrochemical system.
Data: Bioregional Assessment Programme (Dataset 1)
Carbonate dissolution is also considered to be a dominant process in controlling groundwater chemistry by Moya et al. (2015), who observed degassing of groundwater samples, probably associated with dissolved CO2, from the Jurassic, Triassic and Cretaceous units when pumped to the surface.
Na-Cl system with minor HCO3 (Clematis Group, upper Permian coal measures, Joe Joe Group)
This hydrochemical group shows greater variability in the relative average abundance of ions than the Na-Cl or Na-HCO3-Cl systems, however, they are grouped as a single hydrochemical system based on the similarity in ion-TDS relationships; all three units show high R2 values for Cl and Na with respect to TDS. While HCO3 is high in some Clematis Group aquifer, and upper Permian coal measures partial aquifer samples, there is poor correlation between HCO3 and TDS in these units (R2 < 0.2), indicating that high HCO3 values are a local phenomenon rather than a regional trend.
The distribution of solutes in the Clematis Group aquifer is difficult to explain given that low salinity (TDS < 500 mg/L) groundwater is found to the west of much higher salinity (TDS up to 3000 mg/L TDS) groundwater, at depths of 600 m where no recent recharge is expected. These low salinity groundwaters, in the area between the Maneroo Platform and Aramac, also tend to have higher HCO3:Cl ratios than samples further east. A possible explanation for low salinity and elevated HCO3 concentrations in the Clematis Group is the interactions with fresher, HCO3-rich water from the Hutton Sandstone aquifer. This hypothesis is discussed further in the ‘Inter-aquifer mixing’ section.
Another possible reason for elevated HCO3 concentrations in the Clematis Group aquifer is upward migration of CO2 from the underlying upper Permian coal measures, though this would not account for the low salinity of these samples when groundwater in the Clematis Group is of higher salinity further east. Potentiometric head difference maps in Section 2.1.3.2.4 show the upper Permian coal measures, which are known to have potentially economic gas contents, has a higher potentiometric surface than the Clematis Group aquifer between the Maneroo Platform and Aramac, meaning upward migration of gas containing CO2 is possible given the hydrologic pressure regime in this area.
Finally, it should be noted that there is a groundwater divide in the Clematis Group aquifer in the central part of the subregion, which may separate the high TDS groundwater in the east, and low TDS groundwater in the west (Section 2.1.3.2.2). The implication here is that low salinity samples close to the Maneroo Platform may not be connected to possible recharge areas of the Clematis Group aquifer, that are thought to occur in areas of outcrop along the eastern margin. This compartmentalisation of the Clematis Group aquifer makes it a complex and unusual hydrogeologic system, and further work is needed to fully understand its hydrogeologic processes.
The region of high TDS in the Clematis Group aquifer, with low hydraulic gradients and high hydraulic heads, may represent an area of stagnant groundwater where solute concentrations are high due to long residence times allowing for high levels of water–rock interaction.
In the Joe Joe Group the independence of ion:Cl ratios with respect to TDS suggests that evaporative concentration dominates the solute budget. Higher Ca+Mg:Cl and HCO3:Cl ratios in some fresh samples may indicate carbonate dissolution.
Inter-aquifer mixing
A primary objective of this hydrochemical analysis is to help confirm whether inter-aquifer mixing may be occurring between the Galilee Basin hydrogeologic units and the main basal hydrogeologic unit of the Eromanga Basin, the Hutton Sandstone aquifer. The Hutton Sandstone is generally separated from the Clematis Group by the Moolayember Formation. However, there are areas adjacent to the Maneroo Platform margin where the Moolayember Formation is absent, allowing the Hutton Sandstone to be in direct contact with Clematis Sandstone and other Galilee Basin stratigraphic units (Figure 55). Leakage of this sort would require either flow through the Moolayember Formation aquitard, or flow in areas where the Moolayember Formation pinches out.
The distribution of TDS and HCO3 concentrations in the Clematis Group aquifer are consistent with possible interactions occurring between the Hutton Sandstone aquifer and the Clematis Group aquifer. High HCO3:Cl ratios only occur in the Clematis Group in comparatively low TDS range samples. This is consistent with mixing with water from the Hutton Sandstone aquifer, as increasing the bicarbonate concentration of the Clematis Group enough to raise the HCO3:Cl ratio would require the addition of significant amounts of fresher water from the Hutton Sandstone aquifer.
As stated above, the high HCO3:Cl and Na+K:Cl ratios in the Clematis Group are consistent with carbonate dissolution followed by cation exchange. Primary carbonates or carbonate cements are not commonly reported in the upper Permian coal measures and Clematis Group, meaning this process seems unlikely in these hydrogeologic units. Biological activity, such as acetate fermentation may account for the high HCO3 concentrations in some samples (Herczeg et al., 1991; Burra et al., 2014), but cannot explain the high Na+K:Cl and low Ca+Mg:Cl ratios observed. Figure 53 and Figure 54 show the chemistry of Clematis Group and Hutton Sandstone aquifer groundwater samples in a Piper diagram. Two distinct paths of groundwater evolution can be seen in the Clematis Group aquifer: an increasingly Na-Cl dominated groundwater type with minor SO4, which mirrors the groundwaters of the upper Permian coal measures, and a groundwater type with significant bicarbonate, which seems to follow the trend of increasing bicarbonate dominance seen in the Hutton Sandstone aquifer. Hydraulic head differences presented in Section 2.1.3.2.4 show the Hutton Sandstone aquifer to have generally higher hydraulic head than the Clematis Group aquifer, indicating that the proposed direction of mixing is possible based on relative pressure in the aquifers.
Moya et al. (2015) noted that for Hutton Sandstone aquifer that the majority of hydrochemistry samples clustered into one hierarchical cluster group, however, there was also some clustering of samples in other hierarchical groups. This indicates that some groundwater exchange may be occurring with other aquifers. In the Moya et al. (2015) study area this was attributed to connectivity created by the Hulton-Rand structure (the location of the structure is outlined in Figure 55).
In addition to the possibility of water in the Hutton Sandstone aquifer mixing with waters of the Clematis Group aquifer, it is possible that near the margin of the Maneroo Platform the Hutton Sandstone aquifer receives water from deeper units such as the upper Permian coal measures and Joe Joe Group. In the western parts of the subregion the Moolayember Formation is absent, and the Hutton Sandstone aquifer directly overlies the upper Permian coal measures, Clematis Group, or Joe Joe Group. With only a few exceptions, bores with TDS above the 90th percentile of samples in the Hutton Sandstone aquifer are located in areas where the Moolayember Formation is absent (Figure 55). Figure 54 shows the chemistry of samples from the Hutton Sandstone aquifer with anomalously high TDS, as well as groundwaters in the upper Permian coal measures and Clematis Group, on a Piper diagram. The samples from the Hutton Sandstone aquifer with anomalous TDS values are consistent with hydrochemical trends for groundwater from aquifers in the Galilee Basin.
Samples from the Hutton Sandstone with anomalous TDS also show a marked difference in some major ion R2 values with respect to TDS. Samples with TDS below the 90th percentile (underlain by the Moolayember Formation) have R2 values for Cl and alkalinity with respect to TDS of 0.60 and 0.72. In samples with TDS above the 90th percentile (mostly not underlain by the Moolayember Formation), the R2 value for Cl rises to 0.83, and the value for alkalinity is reduced to 0.37. In the Joe Joe group, R2 values for Cl and alkalinity with respect to TDS are 0.96 and 0.02 respectively. This is a strong indication that the high TDS values in the Hutton Sandstone aquifer to the west of the subregion are the result of mixing with Na-Cl dominated water from underlying aquifers in the Galilee Basin.
Overall there is potential for some leakage to occur between aquifers in the Galilee Basin and the Hutton Sandstone aquifer in areas where the Moolayember Formation is either absent, thin (acts as a leaky aquitard) or in areas where faulting has significantly offset the aquifers.
Due to the large number of samples available, this plot is limited to samples collected after 1990.
Data: Bioregional Assessment Programme (Dataset 11)
Data: Bioregional Assessment Programme (Dataset 11)
High TDS samples in the Hutton Sandstone aquifer tend to occur where the Moolayember Formation is thin (less than 100 m thick) or absent. High TDS samples can occur in areas where aquifers in the Galilee Basin are in direct contact with the Hutton Sandstone aquifer.
Data: Bioregional Assessment Programme (Dataset 12, Dataset 13), Geoscience Australia (Dataset 14)
2.1.3.2.1.5 Summary
Groundwaters of all the hydrogeologic units in the subregion show high variability in solute concentrations, ion:Cl ratios, and sample depth. Based on major ion chemistry, hydrogeologic units in the subregion can be grouped into three hydrochemical systems: two are recognised in the Eromanga Basin and one in the Galilee Basin. In descending hydrostratigraphic order these are: a strongly Na-Cl dominated system with minor SO4, recognised in the Winton-Mackunda partial aquifer and the Rolling Downs Group aquitard; a Na-HCO3-Cl dominated system, evident in the Cadna-owie – Hooray Sandstone aquifer, the Injune Creek Group aquitard, and the Hutton Sandstone aquifer; and a Na-Cl system with minor to significant HCO3, within the Clematis Group aquifer, the upper Permian coal measures partial aquifer and the Joe Joe Group partial aquifer.
These hydrochemical systems appear to be hydraulically separated at a regional scale by aquitards: siltstones and mudstones in the Rolling Downs Group aquitard separating the Winton-Mackunda partial aquifer from the Cadna-owie – Hooray Sandstone aquifer, and the Moolayember Formation separating the Hutton Sandstone aquifer from the Clematis Group aquifer.
Very high salinities are observed in the Na-Cl hydrochemical system in the upper Eromanga sequence. These are attributed to a combination of evaporative concentration of cyclic salts and diffusion of connate marine salts held in the Mackunda Formation and Rolling Downs Group aquitard.
Bicarbonate concentrations in the Na-HCO3-Cl hydrochemical system of the Jurassic and Lower Cretaceous Eromanga Basin are up to twice those in the Na-Cl and minor HCO3 hydrochemical system in the Galilee Basin. The high HCO3 concentrations may be the result of dissolution of secondary carbonate cement, however Ca+Mg:Cl ratios do not tend to increase where HCO3:Cl ratios do, as would be expected during carbonate dissolution. Herczeg et al. (1991) observed a similar pattern in groundwater from the GAB, and suggested that Ca and Mg:Cl values were reduced through cation exchange for Na. This process accounts for the high Na:Cl values also observed in samples with high HCO3:Cl.
Two possible pathways for vertical flow between the Hutton Sandstone aquifer and a number of units in the Galilee Basin sequence are outlined:
- At a local scale there may be the potential for some downward vertical flow from the Hutton Sandstone aquifer into the Clematis Group aquifer, resulting in locally high HCO3:Cl ratios and low TDS in the Clematis Group aquifer. This process may occur in a small area east of the Maneroo Platform and west of Aramac. Vertical flow from the Hutton Sandstone into the Clematis Group would require the Moolayember Formation to act as a leaky aquitard, which may be caused by small-scale faults, erosional holes, or areas where the Moolayember Formation is anomalously thin. Investigations of inter-aquifer mixing would benefit from study at a finer scale than the regional work reported here.
- In the west of the subregion, particularly around and on the Maneroo Platform, there is potential for upward vertical flow to occur from the upper Permian coal measures and Joe Joe Group partial aquifers into the Hutton Sandstone aquifer where the Moolayember Formation is absent, and where these units directly underlie the Hutton Sandstone aquifer. Salinity values are highest in the Hutton Sandstone aquifer where the Moolayember Formation and Clematis Group pinch out, and TDS in the samples from the Hutton Sandstone aquifer in this area show an anomalously high correlation with Cl, similar to that seen in the upper Permian coal measures and Joe Joe Group partial aquifers. It seems likely that the higher salinities on the Maneroo Platform are caused by more saline water from units in the Galilee Basin sequence entering the Hutton Sandstone aquifer where they pinch out.
2.1.3.2.2 Water levels
In descending stratigraphic order the hydrostratigraphic units used in this section are:
- Cenozoic aquifers
- Wallumbilla and Winton-Mackunda partial aquifer
- Cadna-owie – Hooray Sandstone aquifer
- Hutton Sandstone aquifer
- Moolayember Formation aquitard
- Clematis Group aquifer (includes Warang Sandstone)
- upper Permian coal measures partial aquifer split into BC1, BC2 and BC3 units
- Joe Joe Group partial aquifer.
2.1.3.2.2.1 Cenozoic groundwater system
Cenozoic aquifers
Figure 56 shows the potentiometric surface of the Cenozoic unconsolidated sediments in the eastern zone of the Galilee subregion. The pressure surface was constructed from recent water levels in bores screened in both the Quaternary and Cenozoic aquifers and assumes the two are hydraulically connected. Figure 56 can be considered to be a plot of the watertable in the eastern zone. The Cenozoic aquifers obviously comprise a local flow system having a strong relationship with the surface drainage system. Figure 56 indicates from the curvature of potentials around certain streams (Jordan Creek, the Alice River and the Belyando River) that these are potentially gaining streams. Conversely, Dunda Creek, Tallarenha Creek and Lagoon Creek are potentially losing streams. This aspect will be further explored in Section 2.1.5 on surface water – groundwater interaction.
Figure 56 Watertable in the eastern zone of the Galilee subregion
Contours are constructed from recent water levels in bores screened in both the Quaternary and Cenozoic aquifers and assume vertical hydraulic connectivity between the two systems.
Data: Bioregional Assessment Programme (Dataset 15, Dataset 16)
2.1.3.2.2.2 Eromanga Basin (GAB) groundwater system
Wallumbilla Formation and Winton-Mackunda formations partial aquifer
Many location descriptions referenced in this section refer to 1:250,000 map sheets that cover the Galilee subregion. Refer to Figure 69 and Table 3 in Section 2.1.2 for further information on these maps sheet locations. The Winton Formation includes some brown coal resource developments (for further detail see Section 2.3.4 in companion product 2.3 for the Galilee subregion (Evans et al., 2018)).
Figure 57 and Figure 58 show the potentiometric surface in the Wallumbilla Formation and Winton-Mackunda formations in the central eastern, central western and western zones of the Galilee subregion. These pressure surfaces are considered to represent the Eromanga Basin regional watertable, as the groundwater head in these units equilibrates with the first water cut. The contours are well constrained by water levels from over a thousand bores. In most places, the watertable appears to transition smoothly from the Wallumbilla Formation to the Mackunda Formation except on the Muttaburra and Tangorin 1:250,000 geological map sheets. In the case of the latter, the watertable in the Wallumbilla Formation appears to be higher than that in the Mackunda Formation, but the reverse is true in the former sheet. Although normally regarded as an aquitard, the Wallumbilla Formation supplies over 200 water bores in the eastern zone, some with remarkably fresh water. The majority of these bores are located in the south on the Augathella sheet (n = 82), Charleville sheet (n = 61) and Mitchell sheet (n = 30). Most of the bores that take from the Wallumbilla Formation are screened in sand beds near the base of the Doncaster Member – the basal member of the Wallumbilla Formation.
Figure 57 and Figure 58 indicate that the groundwater flow directions in the Winton-Mackunda formations approximately follow the regional dip and topography to the west and south-west. Although the flow system is a regional one, it nevertheless displays some groundwater mounding in topographically higher areas more typical of a local groundwater flow system. Such areas include the Forsythe Range on the southern Winton 1:250,000 sheet (Figure 57), Yellow Mountain and Opal Hill on the south-eastern Jundah 1:250,000 sheet (Figure 58) and the Grey and Gowan Ranges on the southern Blackall and northern Adavale 1:250,000 sheets.
The Winton-Mackunda formations are classified as a partial aquifer because it is by no means certain that it will supply sufficient quantities of good quality groundwater to a waterbore. There have been a significant number of dry wells drilled in this formation and there have been others where the supply was described as ‘insufficient’ (<0.1 L/sec). There have also been reports of equipped bores in the Winton-Mackunda formations which have been abandoned because the groundwater became saline with continued pumping. Most of the acceptable supplies, in terms of both well yield and salinity, have been obtained from sand beds in the lower part of the Winton Formation or in the underlying Mackunda Formation. It is generally true that all groundwater intersections throughout the Winton-Mackunda formations rise to equilibrate with the level of the first water cut, indicating good vertical hydraulic connection.
Water bores in the Winton-Mackunda formations are all sub-artesian. The watertable in the Winton-Mackunda formations generally lies about 50 to 60 m below ground surface in the eastern and central eastern zones, however, the depth to water can extend to 80 to 100 m in some topographically higher areas (e.g. the aforementioned Grey and Gowan Ranges). The watertable dips more shallowly than the regional topographic slope and depth to water in the central western and western zones is between 20 and 40 m below ground surface. In the north-west, the watertable in the Winton-Mackunda formations lies 10 m or less below ground surface over most of the Julia Creek and Richmond sheets (see Figure 10 in Section 2.1.2).
Data: Bioregional Assessment Programme (Dataset 16, Dataset 17)
Data: Bioregional Assessment Programme (Dataset 15, Dataset 16, Dataset 17)
Cadna-owie – Hooray Sandstone aquifer
Figure 59 and Figure 60 show the potentiometric surface for the Cadna-owie – Hooray Sandstone aquifer. The potentiometric contours are generally smoother and spaced further apart than those in the overlying Winton-Mackunda partial aquifer and do not show any relationship with the topography except where the aquifer is unconfined near the intake beds in the eastern zone. The steep hydraulic gradients near the intake beds are probably due to weathering of the rock outcrop which has produced a lower hydraulic conductivity in the rock outcrop. The groundwater flow potential is to the west and south-west, except around the Thomson River channel on the Muttaburra and Longreach 1:250,000 sheets. Here the configuration of the potentials suggests upward leakage from the aquifer into the bed of the Thomson River.
The Hooray Sandstone aquifer is a GAB icon with extensive areas of flowing artesian bores in the central eastern, central western and western zones of the Galilee subregion. Before development in the late nineteenth century, the extent of artesian conditions was even larger than it is today. The water yielded by the Hooray Sandstone is fresh (mean TDS 834 mg/L) but the salinity range is high (180–7136 mg/L, n = 1302). Most of the fresh waters are produced either from the highly permeable and high yielding Wyandra Sandstone at the top of the sequence, or from coarse-grained sand beds near the base of the Hooray Sandstone. Groundwater salinity increases to the west of the Galilee subregion, where the aquifer overlies the Maneroo Platform (see companion product 1.5 for the Galilee subregion (Evans et al., 2015)). Here, most of the bores are drawing groundwater from the Cadna-owie Formation, which immediately underlies the Wyandra Sandstone member. In marked contrast to the overlying Allaru Mudstone / Wallumbilla Formation, and Winton-Mackunda formations waters, the hydrochemical water type of Hooray Sandstone groundwater is Na-HCO3-Cl dominant.
In compiling the potentiometric surface of the Hooray Sandstone in the eastern zone it was necessary to partition the Ronlow beds into the three Jurassic members – Hooray Sandstone, Injune Creek Group and Hutton Sandstone. The Ronlow beds are mapped on three geological sheets – Buchanan, Galilee and Jericho (Figure 59). Vine and Doutch (1972) describe the Ronlow beds as a marginal sandstone facies of the complete Early Jurassic to Early Cretaceous terrestrial sequence from the Hutton to the Hooray Sandstone. These authors were of the opinion that most of the eastern Ronlow beds in the south were equivalent to the Hutton Sandstone but that the sequence younged northward. With these criteria in mind, the Ronlow beds were partitioned into the three Jurassic members shown in Figure 61.
Data points are adjusted to a datum of 25 °C and equivalent fresh water head
Data: Bioregional Assessment Programme (Dataset 15, Dataset 16, Dataset 17)
Data points are adjusted to a datum of 25 °C and equivalent fresh water head
Data: Bioregional Assessment Programme (Dataset 15, Dataset 16, Dataset 17)
Data: Bioregional Assessment Programme (Dataset 12, Dataset 18, Dataset 19)
Hutton Sandstone aquifer
Figure 62 shows the potentiometric surface (corrected for temperature and salinity) of the Hutton Sandstone. This pressure surface looks very much like that of the Hooray Sandstone and the same patterns are common to both. Groundwater flow is to the west or south-west, except on the Julia Creek sheet which indicates an additional minor northerly component of flow across the Euroka Arch into the Carpentaria Basin. Like the Hooray Sandstone, hydraulic gradients are steepest in the area of the intake beds of the Hutton Sandstone, and the same causal factor of rock weathering is advanced here.
The area of flowing artesian bores in the Hutton Sandstone is slightly larger than that of the Hooray Sandstone and throughout most of the central eastern zone, and over much of the central western and western zones, heads in the Hutton Sandstone are higher than heads in the Hooray Sandstone. In some places this head difference is up to 50 m. The Hutton Sandstone aquifer yields the best quality groundwater of any of the Galilee subregion aquifers.
Figure 62 Potentiometric surface of the Hutton Sandstone, Galilee subregion
Data points are adjusted to a datum of 25 °C and equivalent fresh water head
DST = drill stem tests
Data: Bioregional Assessment Programme (Dataset 15, Dataset 16, Dataset 17, Dataset 19)
2.1.3.2.2.3 Galilee Basin groundwater system
Clematis Group aquifer
Figure 63 shows the potentiometric surface of the Clematis Group aquifer (corrected for temperature and salinity). Groundwater flow directions are slightly different to those of the overlying Eromanga Basin aquifers. For the Clematis Group in the south, the potential groundwater flow direction is to the north-west; in the central area the potential groundwater flow direction is westward, and in the north the groundwater flow direction in the Warang Sandstone is to the south-west. A minor but significant exception to this general pattern occurs in the Carmichael area. Here, a component of potential groundwater flow is to the east, and focuses towards the Carmichael River and the Doongmabulla Springs complex. This means that there must be a groundwater divide separating the easterly and westerly flow regimes. Its inferred location is shown in Figure 63. The groundwater divide occurs to the west of the Carmichael river basin and approximates a topographically elevated area found between the Great Dividing Range and prominent ridges along the margin of the Eromanga Basin.
The Clematis Group aquifer subcrops near surface in the vicinity of the Doongmabulla Springs complex, in the headwaters of the Carmichael river basin. The focusing of potential groundwater flow in the Clematis Group aquifer eastwards towards these areas suggests there is potential for discharge from the Clematis Group aquifer to provide baseflow to the Carmichael River and to be a source aquifer for the Doongmabulla Springs complex. Further discussion on the origin of the groundwater divide and the source aquifer for Doongmabulla Springs complex is provided in Section 2.3.2 of companion product 2.3 (Evans et al., 2018) as well as Section 3.4 and Section 3.5 of companion product 3-4 (Lewis et al., 2018).
Most bores in the Clematis Group are sub-artesian but there are some minor artesian areas near the limits of the formation in the west. Bore yields are generally significantly lower than those of the overlying Hutton Sandstone. Nevertheless this aquifer produces good quality water.
Figure 63 Potentiometric surface of the Clematis Group aquifer, Galilee subregion
Data points are adjusted to a datum of 25 °C and equivalent fresh water head
DST = drill stem tests
Data: Bioregional Assessment Programme (Dataset 15, Dataset 16, Dataset 17, Dataset 19)
Upper Permian coal measures partial aquifer
Coal seams in the upper Permian coal measures are the primary target for CSG exploration in the Galilee subregion as well as coal mining proposals located along the eastern margin of the Galilee subregion (for further detail see Section 2.3.4 in companion product 2.3 for the Galilee subregion (Evans et al., 2018)). The formal stratigraphic units included in the upper Permian coal measures are outlined in Section 2.1.2. In general, six major seams separated by interburden sandstones, ranging from the A seam in the Bandanna Formation at the top of the upper Permian coal measures, to the F seam in the upper Permian coal measures at the base, with the upper split in the D seam referred to as DU or D1, and the lower split called DL or D2. The coal seams are highly variable in thickness, ranging from 0.1 m for the E seam at Kevin’s Corner to 18 m for the A seam at Carmichael (data from Bleakley et al., 2014). The mean thickness of the coal seams is about 3 m. The thickest total accumulation of coal occurs at Carmichael with a total thickness of 39 m; the thinnest accumulation is at South Galilee where the total coal thickness is considerably less – 14.5 m. The interburden sandstones are thicker than the coal seams, with the thickest being the BC interburden. This unit ranges in thickness from 60 m at Carmichael to 90 m at Kevin’s Corner, Galilee and South Galilee (Bleakley et al., 2014).
The mining proponents have established extensive bore monitoring networks, measuring water levels in all coal seams and interburden sandstones. The company water level data show significant vertical hydraulic gradients exist through the upper Permian coal measures. For this reason, and to more adequately model the complexity of the upper Permian coal measures, the upper Permian coal measures have been split into three hydrogeological sub-units. From top to bottom these sub-units are informally designated BC1, BC2 and BC3. The partitioning was done on the basis of approximately similar groundwater pressures existing in each sub-unit (Table 17).
Table 17 Subdivision of the upper Permian coal measures
The upper Permian coal measures are designated as a partial aquifer because of their low hydraulic conductivity (Kh < 0.2 m/day). Figure 64 shows the Kh profiles at China First, Kevin’s Corner and Carmichael coal projects. The data points are from Bleakley et al. (2014). A few aspects of these plots are notable:
- The plots show a trend (albeit with only three sites) of Kh being highest in the south at China First (Galilee) and decreasing northwards to a minimum at Carmichael. The implication is that there are broad trends in hydraulic conductivity, with it decreasing (getting tighter) to the north. Detailed numerical modelling may need to take these trends into consideration.
- There is no consistent trend of Kh with depth. One interpretation is that a trend of Kh increasing with depth in BC2 at Kevin’s Corner and China First (Galilee) down to the C seam with Kh then decreasing in the DU and DL seams near the base of BC2, but the reverse trend is displayed at Carmichael. Here Kh is maximised in the DU and DL seams, and in the interburden sandstone.
- In coal basins like the Bowen Basin there is anecdotal evidence to support the theory of preferential flow of groundwater through cleats in the coal and minimal groundwater flow through the interburden layers. This appears not to be so in the Galilee Basin. The Kh plots in Figure 64 show that, in general, the hydraulic conductivity in the interburden sandstones is higher than in the coal seams.
These coal mine developments are located along the eastern margin of the Galilee subregion. Carmichael coal project is the northern most of the three mentioned here, while China First (Galilee) is the southernmost of the three.
Data: Bleakley et al. (2014), Queensland Department of Employment, Economic Development and Innovation (Dataset 4)
The data points for Figure 65, Figure 66 and Figure 67 come from two distinctly different sources. Along the eastern margin the data points are water level measurements in boreholes drilled by the mining proponents. However, in the western areas the pressures were calculated from drill stem tests (DST) done in petroleum exploration wells. This was necessary because there are no bores deep enough to measure water levels of these units in the west. Although erroneous DSTs were culled during the rigorous QA/QC process carried out by the Assessment team, there still remains an element of uncertainty in the formation pressure estimates, even in the better tests; consequently this uncertainty is propagated to the interpreted potentiometric surfaces presented in Figure 65, Figure 66 and Figure 67.
The potentiometric surfaces (corrected for temperature and salinity) for BC1, BC2 and BC3 are shown in Figure 65, Figure 66 and Figure 67 respectively. The patterns are similar for all three. All show an easterly component of flow towards the Belyando River valley on the Galilee and northern part of the Jericho 1:250,000 sheets (Figure 10 in Section 2.1.2). Elsewhere the flow direction is largely westward. Based on the available data a major groundwater divide must exist for all three sub-units. Groundwater flow is directed away from this divide towards the eastern (against the regional west dip of bedding) and western margins of the Galilee Basin. Figure 65, Figure 66 and Figure 67 show that prominent north-trending groundwater mounding approximates a topographically elevated area situated between the Great Dividing Range and prominent ridge that defines the margin of the Eromanga Basin.
The genesis of the groundwater divides warrants investigation as it is not immediately obvious why such divides should exist. The upper Permian coal measures are confined by overlying sedimentary sequences (Rewan Group, Clematis Group, Moolayember Formation) in the vicinity of the groundwater divide; hence the groundwater divide cannot be due to direct recharge from the surface. One possible explanation is that the north-trending groundwater divide represents a northward extension of the west-south-west-trending recharge mound that is apparent to the east of Blackall. This explanation may be plausible for BC1 (Figure 65) but not really so for BC2 (Figure 66) and BC3 (Figure 67) or the underlying Joe Joe Group (see next subsection). Further discussion on the origin of the groundwater divide is provided in Section 2.3.2 of companion product 2.3 (Evans et al., 2018).
Figure 65 Potentiometric surface of BC1, Galilee subregion
Data points are adjusted to a datum of 25 °C and equivalent fresh water head
DST = drill stem tests
Data: Bioregional Assessment Programme (Dataset 15, Dataset 16, Dataset 17, Dataset 19)
Figure 66 Potentiometric surface of BC2, Galilee subregion
Data points are adjusted to a datum of 25 °C and equivalent fresh water head
DST = drill stem tests
Data: Bioregional Assessment Programme (Dataset 15, Dataset 16, Dataset 17, Dataset 19)
Figure 67 Potentiometric surface of BC3, Galilee subregion
Data points are adjusted to a datum of 25 °C and equivalent fresh water head
DST = drill stem tests
Data: Bioregional Assessment Programme (Dataset 15, Dataset 16, Dataset 17, Dataset 19)
Joe Joe Group aquitard
Although the Joe Joe Group is regarded as a regional aquitard, the unit nevertheless has the capacity to store and transmit groundwater and supplies water to a few dozen stock and domestic bores in the southern areas. The (corrected) potentiometric surface for the Joe Joe Group is shown in Figure 68. Unfortunately data are sparse for this formation. Like the upper Permian coal measures data points, the eastern points are water level measurements from boreholes and the western points are pressures calculated from DSTs. Like the overlying upper Permian coal measures partial aquifer, the Joe Joe Group exhibits an easterly flow component towards the Belyando River valley (northern half of Jericho sheet) but elsewhere the flow direction is westward. In common with the upper Permian coal measures partial aquifer and the Clematis Group aquifer, a major groundwater divide must occur that segregates eastern- and western-directed flow components of the groundwater system. Further discussion on the origin of the groundwater divide is provided in Section 2.3.2 of companion product 2.3 (Evans et al., 2018). In general, heads in the Joe Joe Group are higher than the heads in BC3.
Figure 68 Potentiometric surface of the Joe Joe Group aquitard, Galilee subregion
Data points are adjusted to a datum of 25 °C and equivalent fresh water head
DST = drill stem tests
Data: Bioregional Assessment Programme (Dataset 15, Dataset 16, Dataset 17, Dataset 19)
2.1.3.2.3 Water level trends
Time series groundwater level data for 202 observation wells were obtained from the Queensland groundwater database (Queensland Department of Natural Resources and Mines, Dataset 2). However, not all available time series data were applicable for the purposes of the bioregional assessment. For instance, for some wells it was not possible to assign a hydrogeological unit to the time series water level data.
In descending stratigraphic order the stratigraphic units used in this section are:
- Cenozoic (Alluvium)
- Cenozoic (Paleogene–Neogene)
- Hooray Sandstone
- Adori Sandstone
- Hutton Sandstone
- Clematis Group
- Rewan Group
- upper Permian coal measures.
Time series data from all bores in the dataset (Dataset 20) were analysed using the Theil-Sen regression method (Singh and Singh, 2013) to identify whether there were statistically significant trends in the time series records of the observation bores. The Thiel-Sen method is preferable to a simple linear regression as it is insensitive to outliers and more likely to identify meaningful trends in datasets with a high level of variability.
Many bores had time series record lengths of less than two years. The GAB is a groundwater system where water can have residence times of tens or hundreds of thousands of years, and water levels at a given time are the product of numerous different influences, some of which may operate on very long time spans. It seems unlikely then that a record of only two years would be representative of the long-term trends in the system. Therefore bores with record lengths of less than two years will be discussed separately to bores with long records of water level. Similarly, a number of bores had only one or two observations of water level throughout their recording history. Obviously these were not sufficient data on which to draw conclusions about trends over time.
Figure 69 shows the distribution of observation bores in the Galilee subregion with time series water level monitoring records for a period of time greater than two years. Most of these bores are non-artesian. However, a number of bores monitoring Great Artesian Basin (GAB) aquifers under artesian conditions occur in the western part of the subregion. Most observation wells are in the east of the subregion, with clusters of bores around proposed coal resource developments. A smaller number of bores occur in the north-west and south-west.
Refer to Figure 71, Figure 72 and Figure 73 for hydrographs of the selected observation bores.
Data: Bioregional Assessment Programme (Dataset 20, Dataset 21)
2.1.3.2.3.1 Bores with long recording periods
Of the 129 bores with records longer than two years, 24 had records which finished before 1997 and cannot be assumed to represent the current state of the aquifer they are screened in. These bores were mostly screened in the Cenozoic or Alluvium, or had unknown hydrogeology for the screened interval so were of limited use for investigating the hydrogeology of the subregion even if they had modern measurements.
The number of observation bores screened in each hydrogeologic unit in the subregion can be seen in Table 19.
This left 105 bores with a record longer than two years, with the most recent water level measurement taken within the last 20 years. Of these, 50 bores did not show a statistically significant trend in the time series data, and 55 showed a statistically significant increase or decrease in water level over the period of measurement. Table 18 summarises the results of statistical analysis of the observation data.
Table 18 Statistically significant trends in observation bores in the Galilee subregion
Data: Bioregional Assessment Programme (Dataset 20)
Of the 50 bores with records longer than two years with no statistically significant trend, 16 were screened in alluvial deposits or Paleogene-Neogene rocks, which typically show high variability in water level data (see Figure 70). It seems likely that the absence of a statistically significant trend is the result of rapid responses to rainfall in these bores due to their shallow screened interval.
On the “cumulative years” axis, year 0 = 1976, year 25 = 2001
Data: Bioregional Assessment Programme (Dataset 20)
The majority of non-artesian bores with a statistically significant trend and records longer than two years showed a decreasing trend in water levels over the recording period. The difference in water level recorded in bores with a decreasing trend ranged from 0.2 to 14 m over the recorded interval. The mean change was 2.5 m with a standard deviation of 3.03 m. The magnitude of decline in water level is strongly influenced by the length of the record. Some bores have measurements dating back to the 1950s and 1960s when it seems water levels were up to ten metres higher than observed today. Bores with recent (post 1990) records of only 10 to 20 years duration typically show declines in water level on the order of 10 cm to 3 m.
Non-artesian bores showing an increase in hydraulic head over the recording period showed changes in water level of 0.2 to 41 m with a mean change of 3.1 m and standard deviation of 9.2 m.
The majority of artesian bores did not show any statistically significant trend in the data. For some bores this is likely due to the small number of observations available for statistical analysis (only three or four points for some bores), but for many bores a change in the hydrological regime caused by the Great Artesian Basin Sustainability Initiative (GABSI) program may be the reason. The hydrographs of several bores in which a statistically significant trend could not be identified show stable water levels, or a decline in water levels, until sometime in the 1990s, after which water levels begin to rise sharply. Figure 71, Figure 72, and Figure 73 show examples of this hydrograph trend for three aquifers in the subregion. All three show a sharp increase in water levels in the early 1990s, when GABSI began. By 1995, 51 bores had been rehabilitated in just these three aquifers.
The change in trend caused by GABSI’s influence makes a trend in water levels for these bores impossible to detect with linear analysis methods without pre-treatment of the data to separate observations from before and after the sealing of flowing artesian bores.
Figure 71 Hydrograph data for artesian observation bore RN 389 in the Hooray Sandstone
Refer to Figure 69 to see the location of RN 389.
Data: Bioregional Assessment Programme (Dataset 20)
Figure 72 Hydrograph data for artesian observation bore RN 3887 in the Hutton Sandstone
Refer to Figure 69 to see the location of RN 3887.
Data: Bioregional Assessment Programme (Dataset 20)
Figure 73 Hydrograph data for artesian observation bore RN 3274 in the Cadna-owie - Hooray Sandstone
Refer to Figure 69 to see the location of RN 3274.
Data: Bioregional Assessment Programme (Dataset 20)
Of the artesian bores which did show a statistically significant trend, all but one showed an increase in water levels over the recording period, also consistent with an expected increase in pressure resulting from capping and piping of flowing artesian bores.
Both increasing and decreasing trends in water level were seen in most hydrostratigraphic units. There were, however, some units which showed trends in only one direction. The trends observed in each hydrostratigraphic unit are summarised in Table 19 and discussed in detail below.
Table 19 Statistically significant trends in water levels for bores with records longer than two years in the Galilee subregion
Data: Bioregional Assessment Programme (Dataset 20)
Data: Bioregional Assessment Programme (Dataset 22)
Cenozoic (Alluvium)
All bores in the alluvium are non-artesian and monitor unconfined alluvial aquifers in the Belyando river basin. Many are clustered in the north of the subregion in the vicinity of Charters Towers (Figure 74a). Six of these bores show a decrease in water level with time, and the remaining two show no statistically significant trend. The remaining six bores in the alluvium are spread broadly across the subregion, and show both increasing and decreasing trends (Figure 74a). The variety of trends seen in bores screened in alluvial material is consistent with expectations that the alluvium is not a highly connected system, but contains a number of local flow systems which are unlikely to influence one another.
Cenozoic (Paleogene–Neogene)
Like the alluvium, bores screened in Cenozoic rocks and sediments are non-artesian and located in the eastern part of the subregion (Figure 74b), mostly in the Belyando river basin. These bores show both increasing and decreasing trends, as well as a large number of records in which there is no statistically significant trend. Water levels can be highly variable in these records, possibly due to rapid responses to rainfall due to the shallow screened intervals of the bores. Similar to bores in alluvial material, the different trends seen in bores screened in Cenozoic rocks may be an indication that flow systems in these rocks are relatively local and have little impact on one another.
Hooray Sandstone
There are 18 monitoring bores in the Hooray Sandstone in the filtered dataset. Four of these are sub-artesian and 14 are artesian.
Of the sub-artesian bores, one has no statistically significant trend, another has an increasing trend (greater than 40 m over 6 years), whilst the remaining two display a decreasing trend (of 10 m and 14 m). The magnitude of water level changes is greater in the Hooray Sandstone than in any other hydrogeological unit in any other GAB aquifer in the Galilee subregion. The increase of 40 m occurred in the early 2000s. Natural hydrogeologic processes would not cause such a large change over such a short time, and it is assumed that the rise in head is a result of the GABSI bore rehabilitation programme.
The two sub-artesian bores showing large decreases in water level are located in the north-west of the subregion and are within 30 km of each other.
Of the artesian bores, one shows a decrease in water levels whilst another three show a statistically significant increase. The other ten bores had no statistically significant trend. The increase in water levels (pressures) in some bores may again be due to the GABSI programme, with the capping and piping of a large number of uncontrolled flowing artesian bores in the Hooray Sandstone that commenced in the mid-1990s. Figure 74c shows the locations of observations bores screened in the Hooray Sandstone.
Adori Sandstone
Five bores in the filtered dataset were tapping into the Adori Sandstone aquifer. Two are sub-artesian and three are artesian.
Of the two sub-artesian bores, one was found to have a statistically significant increasing trend, while the other was found to have a decreasing trend. This is notable because the two bores are within 2 km of each other. Water levels in these bores are highly variable, which may be related to both bores being located in the recharge area of the Adori Sandstone; water levels observed here may be influenced by infiltrating recharge water. Both records show a sharp rise in water level at the end of the record.
Of the three artesian bores, two show a statistically significant increase in water levels. One of which shows a very large increase of greater than 11 m over the recording period.
All five monitoring bores screened in the Adori Sandstone show an increase in water levels from 1996. Figure 74d shows the locations of observation bores screened in the Adori Sandstone.
Hutton Sandstone
There were 22 bores in the filtered dataset in the Hutton Sandstone aquifer. Of these six are sub-artesian and 16 are artesian.
Of the six sub-artesian bores, five show a decrease in water level over the recording time, and one shows no statistically significant trend. The decrease occurs in all bores between 1993 and 2014, and is of a magnitude in the range 0.2 to 0.6 m. The distribution of these bores in the subregion is a roughly north–south transect covering most of the subregion (Figure 75c). This suggests that the reduction in water levels is occurring in the Hutton Sandstone aquifer throughout the Galilee Basin. The records of decreasing water levels in the Hutton Sandstone aquifer begin in the early 1990s and may be related to the abstraction of water from these units, as the Hutton Sandstone is an important water source in the region.
Of the artesian bores, four show an increase in water levels over the recording period, and eight show no statistically significant trend. Increases in water level are from 1 to 3 m over the recording period, and tend to be steepest after 1990. This may be related to the capping and piping of uncontrolled flowing artesian bores in the Hutton Sandstone during the GABSI project.
Clematis Group
There were five bores in the Clematis Group in the filtered dataset, clustered in two locations. All of these were sub-artesian. Only one showed a statistically significant trend, which was of increasing water levels. It is difficult to base any interpretation of water level trends in an aquifer on a single bore. The absence of a statistically significant trend in other bores screened in the Clematis Group may be due to their location in or close to the recharge area for the aquifer. This means the water level can be highly variable in response to rainwater entering the aquifer as recharge. Figure 75a shows the location of bores screened in the Clematis Group relative to the recharge zone.
Rewan Group
There are two bores in the Rewan Group in the filtered dataset. Both are sub-artesian. They both show an increasing trend in water levels. Both bores are located in or close to the recharge area for the Rewan Group.
Upper Permian coal measures
Two bores in the filtered dataset were present in the upper Permian coal measures (Figure 75b). Both of these are sub-artesian. One shows an increase in water levels over the recording period, while the other shows a decrease. Both of these bores are located in the outcrop area of the upper Permian coal measures. The bore showing an increase in water levels is located in the south-eastern part of the subregion, while the bore showing a decrease is located in the north.
Data: Bioregional Assessment Programme (Dataset 22)
2.1.3.2.3.2 Bores with short recording periods
All bores with recording periods of less than two years had data collected in 2012 or later. Therefore no further filtering of this dataset was required to ensure water levels representative of the modern system. Forty bores show a decreasing trend in water level. Nine bores show an increasing trend and 11 do not show any statistically significant trend. Most (45 of 67) are screened in the upper Permian coal measures. The majority of these bores were installed in association with proposed mining development in the Galilee Basin, in particular the Carmichael and Alpha coal projects. While piezometers have been installed, no mining operations have commenced, and the trends observed in these bores are a feature of the system pre-development rather than the result of any resource development.
While the declines in head show a statistically significant trend, the change in head is typically in the order of 5 or 10 cm and it is difficult to consider this trend as representative of the long-term behaviour of the aquifer due to the short duration of the record. Results of statistical analysis for each hydrogeologic unit are shown in Table 20.
Table 20 Statistically significant trends in water levels for bores with records shorter than two years in the Galilee subregion
Data: Bioregional Assessment Programme (Dataset 20)
2.1.3.2.3.3 Nested bores
A small number of bores in the dataset were nested allowing comparison of water level changes in different aquifers at the same location. Time series data from nested bores indicates that deep and shallow aquifers are connected in some parts of the subregion, and disconnected in others. A pair of bores screened in the upper Permian coal measures and alluvium close to the proposed site for the Kevin’s Corner development show concurrent increases in water level in both aquifers (Figure 76). Bores screened in the upper Permian coal measures and Cenozoic sediments close to the proposed Alpha coal development site show a similar relationship between the upper Permian coal measures and the shallow aquifer (Figure 77).
In both cases the water level of the upper Permian coal measures is greater than in the alluvium, meaning a confining layer must separate the two aquifers. The similarity in water level trends could be due to minor leakage across a confining layer, allowing the upper Permian coal measures to influence the water level of overlying alluvium, or may be the result of mechanical loading of water in the alluvium causing water levels in the upper Permian coal measures to rise in response to increased downward pressure. Similar trends were observed in the Condamine Alluvium and Walloon Coal Measures, where mechanical loading was considered the most likely explanation due to the close correlation between rainfall events and water level rise in the alluvium (DNRM, 2015).
Examination of time series precipitation data in conjunction with water levels, and/or pump test data from both aquifers may be necessary to distinguish between mechanical loading and leakage across a confining layer as the cause of the trends observed.
On the “cumulative years” axis, year 0 = 2006, year 10 = 2016
Data: Bioregional Assessment Programme (Dataset 20)
On the “cumulative years” axis, year 0 = 2011, year 1 = 2012
Data: Bioregional Assessment Programme (Dataset 20)
2.1.3.2.4 Head differences between aquifers and seal characteristics of aquitards
It is instructive to examine head differences between aquifers to get a sense of potential direction of leakage (if pathways exist) and also to assess the seal characteristics of the intervening aquitard. In the following section the potentiometric surfaces of two aquifers are overlain, and one subtracted from the other. In areas where the head difference residual is less than ±10 m, the heads are assumed to be approximately equal (the 10 m buffer about zero is to allow for possible errors in contouring and data). This is a necessary, but not the only condition for an intervening aquitard to be considered to be leaking or to indicate the occurrence of inter-aquifer leakage. Conversely, where the head difference residual is larger than 10 m (i.e. a value less than –10 m or a value greater than +10 m), this indicates areas where significant pressure differences exist between the two aquifers. Such a condition is interpreted as indicating areas where the intervening aquitard may be acting as a tight seal or where inter-aquifer leakage is negligible.
2.1.3.2.4.1 Hutton Sandstone – Hooray Sandstone head difference
Figure 78a shows the (corrected) head difference between the Hutton and Hooray sandstones. This can only be calculated where the Hooray Sandstone aquifer overlies the Hutton Sandstone aquifer and where data exists. Positive values are areas where the head in the Hutton Sandstone is higher than the head in the Hooray Sandstone. These positive areas occur near the intake beds in the eastern zone (where the outcrop of Hutton Sandstone is topographically higher than the Hooray Sandstone) and in the western artesian areas on the Winton and Mackunda 1:250,000 sheets. These are areas where it appears that the intervening aquitard, the Injune Creek Group, forms a tight seal. Elsewhere, in the central eastern, and in parts of the central western and western zones, the Injune Creek Group aquitard appears to be leaky (Figure 78b).
The Injune Creek Group aquitard comprises three units. At the top of this aquitard lies the Late Jurassic Westbourne Formation (carbonaceous siltstone and mudstone) overlying the Middle to Late Jurassic Adori Sandstone (labile sandstone). At the base of the Injune Creek Group lies the Middle Jurassic Birkhead Formation (labile sandstone, siltstone, minor coal). The Adori Sandstone is a GAB aquifer in its own right, though it is not as extensively utilised as the Hutton and Hooray sandstones. The Birkhead Formation is thought to be leakier and more permeable than the Westbourne Formation. In fact, in the Western Eromanga Basin oil and gas fields, the Birkhead Formation is regarded as a reservoir rock, being second only to the Hutton Sandstone in hydrocarbon production (Gravestock et al., 1998). Thus, in areas where it appears the Injune Creek Group forms a tight aquitard, the tightness is largely due to the impermeable nature of the Westbourne Formation.
According to Figure 78b, about 40% of the Injune Creek Group aquitard forms a tight seal and the remaining 60% is leaky. The notion of a significant component of the Injune Creek Group being leaky is supported by the hydrochemistry data. Like the Hutton and Hooray sandstones, groundwater in the Injune Creek Group is a Na-HCO3-Cl type water. Table 21 shows the mean and ranges of TDS in the Hutton and Hooray sandstones and the Injune Creek Group.
Table 21 Mean and range of TDS values in Hooray Sandstone, Injune Creek Group and Hutton Sandstone
|
Formation |
Number of samples |
Mean TDS (mg/L) |
Range of TDS (mg/L) |
|---|---|---|---|
|
Hooray Sandstone |
1302 |
834 |
180–7136 |
|
Injune Creek Group |
146 |
597 |
133–2041 |
|
Hutton Sandstone |
1269 |
482 |
55–3579 |
TDS = total dissolved solids
Data: Bioregional Assessment Programme (Dataset 1)
The mean TDS and TDS range in Table 21 indicates that chemically the Injune Creek Group is intermediate but with a TDS range that is closer to that of the Hutton Sandstone than the Hooray Sandstone. Therefore the dominant vertical flow direction through the leaky aquitard would be upwards from the Hutton Sandstone.
Data: Bioregional Assessment Programme (Dataset 17)
2.1.3.2.4.2 Hutton Sandstone – Clematis Group head difference
Figure 79a shows the (corrected) head difference between the Hutton Sandstone and Clematis Group. This can only be calculated where Hutton Sandstone overlies the Clematis Group (central and western portions of the subregion) and where data exists. Positive values occur where the head in the Hutton Sandstone aquifer is higher than the head in the Clematis Group aquifer. These areas occur near the intake beds of the Hutton Sandstone and are also propagated down gradient into some parts of the central eastern zone. Such areas are interpreted as those where the intervening aquitard, the Moolayember Formation, forms a tight seal.
Heads in the Hutton Sandstone aquifer are higher than the heads in the Clematis Group aquifer everywhere except for a small area just south of Hughenden. Figure 79b indicates the Moolayember Formation forms a tight seal in the east but becomes leaky westward over the majority of the central eastern zone and for all of the Manuka 1:250,000 sheet in the central western zone.
Data: Bioregional Assessment Programme (Dataset 17)
2.1.3.2.4.3 Clematis Group aquifer – BC1 partial aquifer head difference
Figure 80a shows the (corrected) head difference between the Clematis Group aquifer and the BC1 partial aquifer. This can only be calculated where the Clematis Group aquifer overlies the BC1 partial aquifer and where data exists. Negative values indicate areas where the head in BC1 is higher than the head in the Clematis Group, and such values populate the majority of the mapped area (a small but important exception occurs on the Tambo 1:250,000 sheet where the heads in the Clematis Group aquifer are higher than those in the BC1 partial aquifer). These are areas where it appears that the intervening aquitard, the Rewan Group, forms a tight seal (Figure 80b).
Data: Bioregional Assessment Programme (Dataset 17)
2.1.3.2.4.4 BC1 – BC2 partial aquifer head difference
Unlike the examples in previous subsections of Section 2.1.3.2.4, the BC1 and BC2 partial aquifers all occur within the one hydrostratigraphic unit (upper Permian coal measures). Lithologies within the upper Permian coal measures which can have aquitard properties include shale-rich sequences, coal or tight sandstone.
Figure 81a shows the (corrected) head difference between the BC1 and BC2 partial aquifers. This can only be calculated where the BC1 partial aquifer overlies the BC2 partial aquifer and where data exists. Negative values are areas where the head in BC2 is higher than the head in BC1, and such areas are to be found on the Muttaburra and Jericho 1:250,000 sheets. These are places where it appears that the aquitard at the top of BC2 (the BC interburden sandstone) forms a tight seal to exclude vertical hydraulic connection between BC1 and BC2 and is approximately collinear with the groundwater divide. The tight aquitard occurs across about 40% of the mapped area. A small but significant area where heads in BC1 are higher than BC2 (positive head difference values) occurs in the south-east corner of the Jericho 1:250,000 sheet. This area was mentioned previously as being the major recharge zone for the upper Permian coal measures and Figure 81 indicates the potential is for downward leakage from BC1 to BC2. The BC interburden sandstone aquitard is thought to be leaky over about 60% of the mapped area (Figure 81b).
Data: Bioregional Assessment Programme (Dataset 17)
2.1.3.2.4.5 BC2 – BC3 partial aquifer head difference
The BC2 and BC3 partial aquifers comprise parts of the upper Permian coal measures. Figure 82a shows the (corrected) head difference between the BC2 and BC3 partial aquifers. This can only be calculated where the BC2 partial aquifer overlies the BC3 partial aquifer and where data exists. Positive values are areas where the head in BC2 is higher than the head in BC3. Such areas occur on the Tangorin, Muttaburra, Galilee, Jericho and Springsure 1:250,000 sheets. These are places where it appears that the intervening aquitard (DE interburden sandstone) forms a tight seal. Elsewhere, on the eastern and western margins, the DE interburden sandstone aquitard appears to be leaky (Figure 82b).
Data: Bioregional Assessment Programme (Dataset 17)
2.1.3.2.4.6 BC3 partial aquifer – Joe Joe Group aquitard head difference
Figure 83a shows the (corrected) head difference between the BC3 partial aquifer of the upper Permian coal measures and the Joe Joe Group aquitard. This can only be calculated where the BC3 partial aquifer overlies the Joe Joe Group aquitard and where data exists. Negative values are areas where the head in the Joe Joe Group is higher than the head in BC3, and this condition applies over most of the mapped area. While there is a potential for vertical upwards flow, leakage is excluded by the tight seal afforded by the Joe Joe Group aquitard (Jochmus Formation). Figure 83b shows an area on the western margin, and a smaller one on the eastern margin, where the Joe Joe Group aquitard appears to be leaky.
Data: Bioregional Assessment Programme (Dataset 17)
2.1.3.2.5 Groundwater recharge and discharge
2.1.3.2.5.1 Recharge
Cenozoic groundwater system
Limited information is available to estimate recharge to the Cenozoic system. Bleakley et al. (2014) documents information available in environmental impact statements, which form the basis for the estimated Kh and recharge values, shown in Figure 84. The recharge rate in the Quaternary alluvium is thought to be an order of magnitude higher than that of the Cenozoic sediments and is likely to be in the order of 5 mm/year.
The low recharge rate of 0.2 mm/year in the Cenozoic unconsolidated sediments has been assigned based on the impedance to vertical infiltration by a dense, plastic, clay layer, 10 to 20 m thick, which is likely to be present throughout much of the area north of Barcaldine (A Bleakley, 2015, pers. comm.). This clay is well exposed in the Alpha test pit where its top lies 10 m below ground surface. At Alpha this layer has been mapped as a green clay, but its colour depends on whether the clay underwent pedogenesis under an oxidising or reducing environment. The clay texture is the critical hydrological property, not its colour. In places where the clay layer is absent, recharge rates for the Cenozoic unconsolidated sediments are considered to be similar to that of the underlying Eromanga Basin and Galilee Basin units (Figure 85). The saturated zone in the Cenozoic sediments may not be laterally continuous throughout the entire eastern zone, indicating poor intra-formational connectivity.
The same low recharge rate in the Cenozoic consolidated sediments is due to the tough silica cement which occurs as secondary infills in voids of the conglomerate and sandstone of the Glendower Formation (the brown areas apart from that on the Mackunda sheet in Figure 84). The latter (which has been incised by the Diamantina River) comprises sandstone of the Old Cork Beds and Mueller Sandstone. Though not of the same degree of silicification as the Glendower Formation, these rocks nonetheless are virtually impervious to water and have very low porosity, hence the very low recharge rate.
Areas where clay layer absent - horizontal hydraulic conductivity and recharge have not been estimated.
Data: Bioregional Assessment Programme (Dataset 23)
Eromanga and Galilee groundwater systems
Kellett et al. (2003) estimated total groundwater recharge of 21,360 ML/year in the Hooray Sandstone and 25,710 ML/year in the Hutton Sandstone intake beds in the eastern Galilee Basin. These authors used chloride mass balance to derive their recharge rates.
Importantly, the recharge areas used in the recharge flux calculations by Kellet et al. (2003) included both rock outcrop and sub-crop of the aquifers at the intake beds. In the BA programme only the rock outcrop areas of the aquifers have been used in the recharge estimates. The sub-crop areas are assumed to be blanketed by the Cenozoic clay described earlier in Section 2.1.3.2 .5. This clay greatly impedes recharge and wherever it occurs, a recharge rate of 0.2 mm/year was applied irrespective of the substrate. Amended diffuse recharge rates in mm/year are shown for all formations in Figure 85. These recharge rates have been calculated using rainfall chloride accession rates outlined in Leaney et al. (2011) and groundwater chloride concentrations from Dataset 1.
Table 22 shows recharge fluxes by hydrogeological unit. As mentioned previously these recharge rates were calculated by chloride mass balance. These recharge estimates do not take into account episodic recharge from point sources such as leakage from river channels into underlying aquifers.
Table 22 Estimated recharge fluxes by geologic formation for the Galilee subregion
na = not applicable
Data: Bioregional Assessment Programme (Dataset 26)
The surprising and comparatively high recharge flux for the Wallumbilla Formation is due to the large area of outcrop (97,624 km2) and to the low chloride concentrations in bores on the Muttaburra, Longreach, Tangorin, Richmond and Julia Creek 1:250,000 sheets.
Data: Bioregional Assessment Programme (Dataset 23, Dataset 24)
Recharge to the Winton-Mackunda formations occurs over its entire area of outcrop, but the recharge rate is not uniform. In many places, the Winton-Mackunda formations are blanketed by a thick layer of saprolite and the lower horizon of the weathered profile greatly impedes downward infiltration of the wetting front and therefore also recharge. The Winton-Mackunda formations exhibit groundwater mounding in some areas, a characteristic more typical of a local flow system rather than a regional one. These mounds occur in those places where the saprolite has been eroded exposing relatively unweathered (fresh) rock. Recharge rates in such areas are about 1 mm/year, but in places where the saprolite cover has been preserved, recharge rates are considerably lower, of the order of 0.1 mm/year. Although the Winton-Mackunda formations have the largest area of occurrence (109,911 km2), its recharge flux is only the second highest because of its low recharge rate.
The estimated recharge fluxes presented in Table 22 are lower than those estimated by Kellett et al. (2003) – only 73% of that estimated for the Hutton Sandstone and 57% for the Hooray Sandstone – for reasons explained earlier.
The recharge fluxes for the Rewan Group, upper Permian coal measures, and the Joe Joe Group, are particularly low.
Discharge
Artificial discharge includes groundwater pumping from bores, or discharge from free flowing artesian wells. For the Hooray and Hutton sandstones, flow from controlled or uncontrolled artesian water wells is by far the largest proportion of discharge from these aquifers (approximately 17,000 ML/year from the Hooray Sandstone aquifer and approximately 23,000 ML/year from the Hutton Sandstone (see companion product 1.5 for the Galilee subregion (Evans et al., 2015)). The remainder of the groundwater flux in these two aquifers, except for a minor component of flow to rejected recharge springs in the Barcaldine Springs complex (Figure 86), is ultimately naturally discharged in springs, salt lakes or vertical leakage in the south-west Eromanga Basin.
Natural groundwater discharge occurs from several spring supergroups in the Galilee subregion. Fensham et al. (2016) provides detail on spring supergroups and spring complexes that occur within the Galilee subregion (spring complexes are clumped together to define spring supergroups). The Barcaldine Springs complex (Figure 86) consists of over 300 individual springs (Fensham et al., 2016). The springs are concentrated in two lines along the flanks of the Eromanga Basin margin, which is where outcrop occurs of the Ronlow beds, Hooray Sandstone and Hutton Sandstone. The source aquifers for these springs are primarily the Hooray Sandstone aquifer, Hutton Sandstone aquifer and the Ronlow beds aquifer (which is mainly a Hooray Sandstone equivalent).
The Doongmabulla Springs complex (Figure 86) lies about 10 km west of the proposed Carmichael Coal Mine. The Doongmabulla Springs complex includes over 180 vents that feed some relatively large wetlands (Fensham et al., 2016). Some of the more well-known springs are called Joshua, Moses and Little Moses (the Moses springs includes more than 65 individual vents). The source aquifer is likely to be the Clematis Group aquifer. The Carmichael River receives outflow from the Doongmabulla Springs complex and baseflow from the Clematis Group aquifer.
Fensham et al. (2016) considers the upper Permian coal measures (specifically the Colinlea Sandstone) to be the likely source aquifer for the Mellaluka Springs complex, Lignum Springs and Albro Springs (Figure 86). Whilst the Dunda beds (part of the Rewan Group) is likely to be the source aquifer for Hector Springs, Greentree Springs and Hunter Springs (Fensham et al., 2016).
With the notable exception of the Clematis Group aquifer with groundwater pumping of about 1400 ML/year, artificial groundwater discharge by pumping from wells is negligible for the Galilee Basin units. However this is set to change when dewatering of the upper Permian coal measures begins due to the proposed coal resource development. Further information regarding artificial discharge is reported in companion product 2.6.2 (Peeters et al., 2018) and product 2.5 (Karim et al., 2018a) for the Galilee subregion.
As mentioned previously, a component of flow in the upper Permian coal measures partial aquifer and the Joe Joe Group aquitard discharges eastwards towards the Belyando River valley, but the majority of the groundwater flux in these formations is towards the west. It appears that the groundwater discharge from these formations is dominantly vertical upwards leakage into the overlying formations at the western margins of the Galilee Basin where the strata pinch out. This also appears to be the case for the Triassic formations of the Galilee Basin sequence.
Further discussion on types of discharge from groundwater systems and springs in the Galilee Basin is provided in Section 2.3.2 of companion product 2.3 (Evans et al., 2018). Companion Product 2.7 for the Galilee subregion (Ickowicz et al., 2018) and companion product 3-4 (Lewis et al., 2018) provide further detail on the hydrogeology and ecology of the Doongmabulla Springs complex, as well as other springs with source aquifers in the Galilee Basin.
Figure 86 Springs near the eastern margin of the Galilee subregion
Data: Queensland Herbarium, Environmental Protection Agency (Dataset 27), Bioregional Assessment Programme (Dataset 28, Dataset 29), Geoscience Australia (Dataset 30)
