Detrital or reset? ⁴⁰Ar/³⁹Ar dating of mica from the Lower Jurassic Precipice Sandstone and Evergreen Formation in the Surat Basin

Abstract

The U–Pb detrital zircon record of the Surat Basin, an important part of the Great Australian Superbasin, has already revealed important insights about sediment source terranes. However, owing to the high closure temperature of zircons, low-temperature thermal events that might have impacted the sediment are not recorded. Here, new ⁴⁰Ar/³⁹Ar detrital mica ages, which record low-temperature events as a result of isotopic resetting, are paired with published U–Pb detrital zircon ages from the same samples to provide a more complete interpretation of the tectonic and thermal history of the Jurassic-age Precipice Sandstone and Evergreen Formation. The ⁴⁰Ar/³⁹Ar geochronology of mica grains from five wells reveals two broad groups with distinct age populations: 1500–180 Ma and 150–45 Ma. Micas older than about 180 Ma are sourced from multiple terranes. The slight discrepancy in ages between the ⁴⁰Ar/³⁹Ar and the U–Pb systems of the same samples may represent differences in closure temperature. However, some micas, such as those dating to ca 180 Ma, may also reflect a thermal reset event. Similarly, the younger group of micas, split into Cretaceous and Paleogene populations, reflect the impact of post-depositional thermal events on the basin. Isotopic resetting of the micas was likely the result of hydrothermal fluids migrating through reactivated faults, fractures and/or porous and permeable sediments. The origin of the fluids during the Cretaceous can be linked to an eastern subduction zone and subsequent igneous underplating resulting in uplift and denudation. The exact source of the hydrothermal fluids for the micas of Paleogene age, recorded in samples collected from the base of the Evergreen Formation, however, remains uncertain. Importantly, ⁴⁰Ar/³⁹Ar dating of mica from sandstones permits the detection of post-depositional thermal events that may have implications for tracing fluid migration throughout the basin and reconstructing the Cretaceous–Paleogene tectonic history of the basin.

KEY POINTS

1. Combining ⁴⁰Ar/³⁹Ar dating of micas with U–Pb dating of zircons reveals a more complete tectonic and thermal history of the Jurassic-age Precipice Sandstone and Evergreen Formation of the Surat Basin.

2. ⁴⁰Ar/³⁹Ar dating of detrital micas reveals two broad groups with distinct age populations: 1500–180 Ma and 140–45 Ma.

3. Ages older than ca 180 Ma are linked to multiple source terranes, including the Thomson and New England orogens, and the contemptuous magmatic arc.

4. Ages younger than 150 Ma are most likely the result of post-depositional thermal resetting events, which may have implications for fluid movement throughout the basin.

Keywords:

• ⁴⁰Ar/³⁹Ar dating
• detrital geochronology
• Evergreen Formation
• mica
• provenance
• Precipice Sandstone
• thermal resetting
• Surat Basin

Introduction

Sedimentary basins play a unique role in understanding the tectonic and thermal history of an area (e.g. Dickinson, 1988; Dickinson & Suczek, 1979; Haines et al., 2001; Haughton et al., 1991). They record the origins of sediment and other geological events that occurred near the time of stratal accumulation (e.g. Bryan et al., 1997; Fedo et al., 2003; Zhang et al., 2019). Sedimentary basins can also encode the post-depositional thermal history of sedimentary strata (e.g. Haines et al., 2004; White et al., 2002; Zhang et al., 2019). Understanding and constraining the tectonic and thermal history of sedimentary basins is particularly critical for resource assessment and aquifer modelling because it affects porosity, permeability, and fluid flow pathways (e.g. He et al., 2021; Sobczak et al., 2022).

Dating of detrital minerals has become an invaluable tool for unravelling the tectonic and thermal history of sedimentary basins (e.g. Raza et al., 2009; Sobczak et al., 2022; Zhang et al., 2019). Uranium–lead (U–Pb) dating of heavy minerals (e.g. zircon) is the most commonly used method for determining sediment provenance. Zircons can record ‘long-term’ tectonic events as a result of their resistance to weathering and high closure temperatures (>900°C; e.g. Carrapa et al., 2009; Haines et al., 2004). However, these characteristics also prevent zircons from recording lower-temperature events that occur after deposition, such as uplift and re-heating.

While exact closure temperatures for argon in mica grains vary (∼200–500 °C) depending on their composition, they are significantly lower than for the U–Pb system in zircon, averaging 350 °C for muscovite and 300–350 °C for biotite (Harrison et al., 2009; McDougall & Harrison, 1999; Reid et al., 2022; Reid & Forster, 2021). Therefore, detrital mica grains are unlikely to record repeated orogenic events (McDougall & Harrison, 1999). Detrital muscovite is generally derived from granite or low- to medium-grade metamorphic rocks (Carrapa et al., 2009; Hicks et al., 1999; Miller et al., 1981; Van Hoang et al., 2010; Warren et al., 2012). The lower closure temperature of muscovite can record cooling ages and/or detects events that have caused isotopic resetting of ages (McDougall & Harrison, 1999). Therefore,⁴⁰Ar/³⁹Ar dating of detrital muscovite and biotite is becoming increasingly important in provenance studies and basin thermal history reconstruction (e.g. Haines et al., 2004).

Eastern Australia hosts a number of sedimentary basins that have provided insight into the tectonic and thermal history of eastern Gondwana (e.g. Adams et al., 2021; Bianchi et al., 2018; Foley et al., 2020, 2021; He et al., 2021; Henderson et al., 2022; Korsch, 1984; Korsch et al., 1989; Korsch, Adams, et al., 2009; Korsch, Totterdell, et al., 2009; La Croix et al., 2020; Martin et al., 2018; Raza et al., 2009; Sobczak et al., 2022; Tucker et al., 2016; Wainman et al., 2018, 2019). The Surat Basin, which is located in northern New South Wales and Queensland, Australia, has recently been the subject of a number of studies aimed at absolute and relative age-dating to improve stratigraphic correlation and provenance interpretation (Andrade et al., 2023; Bianchi et al., 2018; Foley et al., 2022; He et al., 2021; La Croix, Hannaford, et al., 2019; La Croix et al., 2020, 2022; Sobczak et al., 2022; Wainman et al., 2015, 2018). Raza et al. (2009) used apatite fission track and vitrinite reflectance to investigate the burial history of the Surat Basin and showed that uplift and denudation coincided with continental extension at ca 95 Ma. However, many of the thermal history events that impacted the Surat Basin since ca 95 Ma remain unknown or unconstrained.

Here we present the first ⁴⁰Ar/³⁹Ar ages of detrital muscovite and biotite from sandstones in the Mesozoic sections of the Surat Basin: the Precipice Sandstone and Evergreen Formation. Detrital zircons from these samples have previously been dated using U–Pb, which revealed that the lower Precipice Sandstone and Evergreen Formation had a complex depositional history, with the sediment derived from multiple source terranes located within the Tasmanides (Sobczak et al., 2022). Pairing ⁴⁰Ar/³⁹Ar ages with the U–Pb detrital zircon ages from the same samples offers a powerful tool in unravelling the tectonic and thermal history of the Surat Basin.

Geological background

The eastern Australian Tasmanides consist of orogenic belts and associated sedimentary basins that formed during the early Phanerozoic (e.g. Glen, 2005; Rosenbaum, 2018). The Tasmanides formed as a result of tectonic plate convergence (Collins, 2002). The three orogenic belts of relevance to the study area are the Thompson, Lachlan, and New England orogens (Figure 1).

Figure 1. Left: outline of eastern Australia showing the different orogens comprising the Tasmanides, three of which are partially overlain by the Surat Basin (modified after Sobczak et al., 2022). The map also shows the location of the Hoy lava-field, the Whitsunday Volcanics, and the Moreton Igneous Association (Henderson et al., 2022; Johnson, 1989). Right: map showing the main structural features within the Surat Basin and the location of the wells sampled in this study (modified after La Croix et al., 2020).

The 327, 000 km² Surat Basin unconformably overlies the Bowen Basin, with basement floor structures reflecting those of the Bowen Basin (Exon, 1976; Green, 1997; La Croix et al., 2020). It is bound by the time-equivalent Clarence-Moreton Basin in the east across the Kumbarilla Ridge (Figure 1; Exon, 1976; Green, 1997). In the west, the Surat Basin is bound by the Eromanga Basin across the Nebine Ridge (Figure 1; Exon, 1976; Green, 1997).

Other key structural features of the Surat Basin include the north–south-trending Mimosa Syncline, which broadly defines the basin axis (Figure 1; Exon, 1976; Fielding et al., 1990). Major faults include the Moonie-Goondiwindi and the Burunga-Leichhardt fault systems, which are located east of the Mimosa Syncline, and the Hutton-Wallumbilla fault system, located west of the Mimosa Syncline (Figure 1). The Moonie-Goondiwindi and the Burunga-Leichhardt fault systems are thought to have been reactivated in the geological past (Fielding et al., 1990; He et al., 2021; Raza et al., 2009). However, it is unclear whether this reactivation was syn-depositional or post-depositional (Babaahmadi & Rosenbaum, 2014; Sliwa et al., 2018).

The Lower Jurassic to Lower Cretaceous sediments of the Surat Basin, the focus of this study, are generally flat-lying, with local gentle folding and faulting (Martin et al., 2013; Raza et al., 2009). Sedimentation ceased in the mid-Cretaceous owing to uplift and erosion (Martin et al., 2013; Raza et al., 2009). The Precipice Sandstone has a maximum thickness of approximately 175 m (La Croix et al., 2020). It is separated into two sections: the ‘lower’ Precipice Sandstone, which is dominated by thick-bedded, coarse-grained sandstone, and the fine to medium-grained ‘upper’ Precipice Sandstone (La Croix, Wang, et al., 2019; Wang et al., 2019).

Overlying the Precipice Sandstone are the fine-grained sandstones and mudstones of the Evergreen Formation (Figure 2), which have a maximum thickness of approximately 300 m (La Croix et al., 2020). The Evergreen Formation is late Sinemurian or Pliensbachian to Toarcian (Figure 2; La Croix et al., 2022; Price, 1997). Recent U–Pb dating of depositional zircon from rare tuff layers supported the previously assigned age for the Evergreen Formation (Sobczak et al., 2022).

Figure 2. Stratigraphy of the Lower and Middle Jurassic series in the Surat Basin, incorporating age assignments from La Croix et al. (2022).

The Precipice Sandstone has been interpreted to have been deposited initially by a braided river system shifting towards meandering fluvial and fluvial–lacustrine systems through time, reflected by the overlying Evergreen Formation (e.g. Exon, 1976; Green, 1997; Ziolkowski et al., 2014). However, marine incursions into these systems are known to have occurred during deposition of the Evergreen Formation (Bianchi et al., 2018; La Croix, Hannaford et al., 2019; La Croix et al., 2020; Martin et al., 2018).

Samples analysed in this study are derived from the Precipice Sandstone and the Evergreen Formation and were subjected to earlier U–Pb dating of detrital zircon (Sobczak et al., 2022). Sobczak et al. (2022) identified three broad sediment sources. The oldest group, which has ages greater than 500 Ma, was derived from recycled sedimentary and metasedimentary basement rocks of the Thomson Orogen. The middle group, ca 450–200 Ma, was linked to volcanic and plutonic basement rocks of the Thomson and New England orogens. The youngest group, which produced ages younger than 200 Ma, has been linked to contemporaneous subduction-related volcanism. Here, we used detrital micas from the same samples to gain additional provenance information, as well as to determine whether these samples recorded subsequent thermal events, in particular, those associated with Cretaceous spreading or hot-spot activity.

In this research, and similar to Sobczak et al. (2022), the sequence-stratigraphic framework of Wang et al. (2019) was adopted (Figure 2). The Precipice Sandstone interval occurs between the J10 and MFS1 surfaces (lowstand and transgressive systems tract of sequence SQ1). The Evergreen Formation interval occurs between the MFS1 and J30 stratigraphic surfaces (highstand systems tract of sequence SQ1, plus the two overlying sequences).

Sampling

Muscovite grains were collected from the same sandstone samples for which U–Pb ages have been determined via LA-ICP-MS analyses of detrital zircon grains (Sobczak et al., 2022). Specifically, 13 sandstone samples from the Evergreen Formation and Precipice Sandstone were examined from five wells: Taroom 17, Chinchilla 4, Kenya East GW7, Moonie 34, West Moonie 1, to ensure wide coverage of the basin (Table 1; Sobczak et al., 2022). These samples were selected for ⁴⁰Ar/³⁹Ar geochronology based on their muscovite and/or biotite content (Sobczak et al., 2022).

Table 1. Details of the sandstone samples selected for ⁴⁰Ar/³⁹Ar detrital mica dating.

Sample	Well	Depth from (m)	Depth to (m)	Stratigraphic position	Formation^a	Age range (Ma) error (2sig)
C4-D3	Chinchilla 4	1075.65	1075.05	MFS1-J30	Evergreen	364.7 ± 0.5 to 1502.5 ± 1.3 (n = 10)
C4-D4	Chinchilla 4	1140.65	1140.05	MFS1-J30	Evergreen	277.8 ± 0.7 to 1417.1 ± 1.4 (n= 9)
C4-D5	Chinchilla 4	1189.65	1189.1	J10-TS1	Lower Precipice	280.8 ± 0.5 to 406.4 ± 0.6 (n= 9)
C4-D6	Chinchilla 4	1199.45	1199.85	J10-TS1	Lower Precipice	360.4 ± 0.6 to 1486.0 ± 2.1 (n= 10)
C4-D7	Chinchilla 4	1224.4	1223.9	J10-TS1	Lower Precipice	303.1 ± 0.3 to 447.8 ± 0.4 (n= 12)
KEGW7-D1	Kenya East GW7	994.3	993.9	MFS1-J30	Evergreen	93.789 ± 0.4 to 160.6 ± 0.2 (n= 9)
KEGW7-D3	Kenya East GW7	1139	1138.7	MFS1-J30	Evergreen	46.8 ± 0.5 to 179.8 ± 0.3 (n= 6)
KEGW7-D3bt	Kenya East GW7	1139	1138.7	MFS1-J30	Evergreen	53.0 ± 0.1 to 95.5 ± 0.3 (n= 7)
KEGW7-D7	Kenya East GW7	1226	1225.7	J10-TS1	Lower Precipice	80.3 ± 0.1 to 175.7 ± 0.2 (n= 10)
M34-D2	Moonie 34	1777.8	1777.45	J10-TS1	Lower Precipice	94.6 ± 0.1 to 118.2 ± 0.1 (n= 8)
T17-D1	Taroom 17	295.5	295	MFS1-J30	Evergreen	96.1 ± 0.1 to 134.5 ± 0.2 (n= 9)
T17-D5	Taroom 17	432.75	432.2	J10-TS1	Lower Precipice	82.8 ± 0.1 to 299.2 ± 0.6 (n= 8)
WM1-D1	West Moonie 1	2258.77	2258.67	MFS1-J30	Upper Precipice	464.7 ± 0.6 to 506.0 ± 0.4 (n= 8)
WM1-D2	West Moonie 1	2269.5	2269.4	J10-TS1	Lower Precipice	317.2 ± 0.8 to 401.0 ± 0.4 (n= 11)

^a Formation allocation is based upon the work of Sobczak et al. (2022), following Wang et al. (2019).

Methods

The light fraction of samples was previously separated for zircon analysis (Sobczak et al., 2022) and preserved. The muscovite and biotite grains were then selectively picked using a binocular microscope to ensure minimal degree of weathering. The grains were then rinsed in acetone and distilled water, to remove any adhered organic particles/dust, weighed and wrapped in aluminium foil envelopes, and placed into quartz glass vials together with interspersed aliquots of the flux monitor Fish Canyon Tuff sanidine (age = 28.176 ± 0.005 Ma; Phillips et al., 2022). The package (UM#96) was then encapsulated in an outer sealed glass vial and irradiated in the CLICIT facility of the Oregon State University TRIGA Reactor for 50 MWh. Following irradiation and cooling, ⁴⁰Ar/³⁹Ar analyses were undertaken in the Noble Gas Laboratory at The University of Melbourne.

Single-grain fusion analyses were carried out on single mica grains. Grains were placed into the sample chamber of a gas-handling system equipped with a Photon Machines Fusions 10.6 CO₂ laser and connected to a Thermo Fisher Scientific ARGUSVI mass spectrometer at The University of Melbourne. Apparatus details are given in Phillips and Matchan (2013) with updated Faraday detectors now equipped with 1 × 10¹³ Ω resistors as described in Heath et al. (2018). Analytical methods follow those described by Matchan and Phillips (2014) and Heath et al. (2018). The analyses were carried out using a focused beam (0.15 mm) at 3 or 10% laser power.

All results are corrected for system blanks, mass discrimination, radioactive decay and reactor-induced interference reactions. Correction factors vary for each irradiation canister and are supplied in the supplemental data (TablesTables A1 and A2). Mass discrimination was monitored by analysis of standard air volumes, assuming the air argon isotopic composition of Lee et al. (2006). Analytical results are summarised below, with the full dataset provided in the supplemental data (TablesTables A1 and A2). Inclusion of uncertainties in the J-value and age of Fish Canyon Tuff sanidine have a negligible impact on uncertainties. Decay constants are those of Steiger and Jäger (1977). The ⁴⁰Ar/³⁹Ar dating technique is described in detail by McDougall and Harrison (1999).

Results

The range of apparent ages for each sample are provided in Table 1. The full results for the single crystal fusion ages for each of the 13 samples are provided in the supplemental data (Table A1). Bivariate plots that show the variability in apparent ages for each individual sample are plotted in the supplemental data.

Taroom 17

Two samples were analysed from the Taroom 17 well. One sample was dated from the Evergreen Formation (T17-D1) and one from the Precipice Sandstone (T17-D5). Single grain fusion ages (n = 9) for muscovite from T17-D1 returned apparent ages ranging from 134 to 96 Ma (Figure 3; Table 1). There is a prominent population (n = 5) of grains with ages in the range 119–115 Ma, including two indistinguishable ages at ca 118 Ma. Apparent ⁴⁰Ar/³⁹Ar ages (n = 8) for muscovite fusions from sample T17-D5 span an age range of ca 220 Myr from 299 to 83 Ma (Table 1). The younger population of grains (n = 6) records a more limited age range of 124–83 Ma.

Figure 3. Box-and-whisker plot of ⁴⁰Ar/³⁹Ar ages from samples from (a) the Evergreen Formation, (b) upper Precipice Sandstone and (c) lower Precipice Sandstone ages. The colour dots represent individual data points that are outliers.

Chinchilla 4

Four samples were analysed from the Chinchilla 4 well. Apparent ages (n = 10) for muscovite grains from sample C4-D3 suggest that there are two main age populations. The younger population has apparent ages ranging from 396 to 365 Ma. The older population has apparent ages spanning 494–478 Ma (Table 1). A single apparent age of 1502.5 ± 1.3 Ma was recorded.

Similar to C4-D3, apparent ages (n = 9) for muscovite aliquots from sample C4-D4 also indicate that there are two age populations, although the age ranges are greater (Figure 3). The younger population has apparent ages spanning 510–278 Ma, which contrast strongly with the older population, which spans some ca 500 Ma from 1417 to 1072 Ma.

The majority of apparent ages (n = 9) of muscovite for sample C4-D5 fall between 406 and 378 Ma, with one grain yielding a notably younger age at 281 Ma (Figure 3; Table 1). The former population has apparent ages of four grains centred around ca 395 Ma.

All but one of the apparent ages (n = 10) of muscovite for sample C4-D6 lie in the range between 439 and 360 Ma, with one grain yielding a considerably older age at 1486 Ma (Figure 3; Table 1). All ages in the dominant, younger population are distinguishable at the 2σ uncertainty level (n = 9). Similarly, all but one of the apparent ages (n = 12) for sample C4-D7 range in age between 448 and 385 Ma (Figure 3; Table 1). One grain yielded an apparent age of ca 300 Ma (Table 1).

Kenya East GW7

Three samples were analysed from the Kenya East GW7 well. Apparent ages (n = 9) from muscovite analyses for sample KEGW7-D1 range from 161 to 94 Ma (Figure 3; Table 1). All single crystal ages are distinguishable at the 2σ level, although there are four grains with apparent ages centred at ca 120 Ma.

Both muscovite and biotite grains were analysed from Kenya East GW7-D3. Four analyses were excluded from the KEGW7-D3 dataset owing to a low ⁴⁰Ar signal and/or low %⁴⁰Ar*, and the remaining six single crystal ages range from 180 to 47 Ma (Figure 3; Table 1). A younger population is present between 57 and 47 Ma, with the three older grains spanning 180–126 Ma. The KEGW7-D3bt sample represents the biotite equivalent of KEGW7-D3. In this instance, three analyses were excluded from the KEGW7-D3bt dataset owing to a low ⁴⁰Ar signal or low %⁴⁰Ar*. The remaining seven single grain fusion analyses yield two distinct populations. Four grains have apparent ages ranging from ca 55 to 53 Ma, and the other two grains are markedly older at 96 and 86 Ma (Table 1). These younger ages are similar to those from muscovite analyses in KEGW7-D3.

Muscovite apparent ages (n = 10) from sample KEGW7-D7 range from 176 to 80 Ma with four distinct populations, defined by two to three grains each (Figure 3; Table 1). Populations cluster at ca 80 Ma, ca 116 Ma, 131 Ma (two indistinguishable ages) and 139 Ma (two indistinguishable ages). The age of one grain is distinctly older than the others at 176 Ma.

Moonie 34

One sample was analysed from the Moonie 34 well. The majority of analysed muscovite grains (n = 6) in sample M34-D2 returned apparent ⁴⁰Ar/³⁹Ar ages in the range 100–95 Ma, with all aliquots in this range distinguishable at the 2σ level (Figure 3; Table 1). Two grains returned notably older ages at ca 118 Ma (Table 1).

West Moonie 1

Two samples were analysed from the West Moonie 1 well. Three aliquots were excluded from the final dataset for WM1-D1 owing to having both a low yield of ⁴⁰Ar and radiogenic argon (<75% ⁴⁰Ar*). With these data excluded, the remaining apparent ages (n = 8) extend from 506 to 465 Ma (Figure 3; Table 1). A subset of grains ranges from 482 to 475 Ma, while the two oldest grains are indistinguishable at ca 506 Ma.

Muscovite grains (n = 11) from sample WM1-D2 have apparent ages ranging from 401 to 317 Ma (Figure 3; Table 1). Notably, four grains define an indistinguishable population at around 322 Ma, and the two oldest grains are also indistinguishable at ca 401 Ma.

Summary

In summary, the new ⁴⁰Ar/³⁹Ar detrital micas show a wide range of ages. However, there appear to be two broad age groups (Figure 3): 1500–180 Ma and 150–45 Ma.

Discussion

There is a significant range of apparent ages for each sample, which is unsurprising given the detrital nature of the mica grains. The variability in ages could represent (1) differences in sedimentary provenance, including various thermal events in source areas prior to deposition in the Surat Basin, (2) post-depositional thermal events that led to an isotopic reset of the ⁴⁰Ar/³⁹Ar system in detrital mica grains, and/or (3) radiogenic argon loss owing to deformation and/or alteration (younger ages), whereas some of the older ages might reflect incompletely reset detrital and/or composite muscovite grains (e.g. Blewett et al., 2019; Blewett & Phillips, 2016; and references therein). These points are discussed below.

Detrital zircon dating revealed that the provenance of the Precipice Sandstone and Evergreen Formation is complex with three broad populations (Sobczak et al., 2022). The oldest population, older than 500 Ma, was derived dominantly from recycled sedimentary and metasedimentary basement rocks from the Thomson Orogen. The middle population ranges in age between 500 and 200 Ma and was derived from plutonic and volcanic basement rocks of the Thomson and New England orogens. The youngest population is less than 200 Ma and is related to an active magmatic arc on the eastern margin of Gondwana (Foley et al., 2021; Sobczak et al., 2022). However, many of the younger tectonic and thermal events that impacted the Surat Basin strata remain largely undocumented.

Detrital provenance ages

The ⁴⁰Ar/³⁹Ar detrital mica ages that fall within the three broad populations, defined by U–Pb zircon ages (Sobczak et al., 2022), are derived dominantly from the Chinchilla 4 and the West Moonie 1 wells (Figures 3 and 4). Rare ⁴⁰Ar/³⁹Ar detrital mica ages older than 200 Ma are also observed in samples from the lower Precipice Sandstone intersected by the Taroom 17 well (Figure 5). Here the new ⁴⁰Ar/³⁹Ar detrital mica ages are compared with the U–Pb detrital zircon ages in each broad population.

Figure 4. Kernel Density Estimation plots of the samples from the Chinchilla 4 and West Moonie 1 wells, separated into the Evergreen Formation or upper Precipice Sandstone, and the lower Precipice Sandstone. The Kernel Density Estimation of the ⁴⁰Ar/³⁹Ar ages has not been scaled for sample size in order to show where the ages overlap.

Figure 5. Kernel Density Estimation plots of the samples from the Taroom 17 well, separated into the Evergreen Formation and the lower Precipice Sandstone. The Kernel Density Estimation of the ⁴⁰Ar/³⁹Ar ages has not been scaled for sample size to show where the ages overlap.

Pacific-Gondwana, Grenvillian and older Proterozoic (>500 Ma) age population

Detrital micas from the Chinchilla 4 and West Moonie 1 wells produced ⁴⁰Ar/³⁹Ar ages that ranged from ca 1500 to 480 Ma, which overlap with the Pacific-Gondwana (650–500 Ma), Grenvillian (1200–900 Ma) and Mezoproterozoic (1600 to >1450 Ma) sub-populations recorded for the detrital zircons from the same samples (Sobczak et al., 2022). Grenvillian and Mesoproterozoic aged detrital micas are, however, only present in the Chinchilla 4 well. Although mica grains tend not to be as durable as zircons in surviving multiple orogenic events over long time periods (McDougall & Harrison, 1999), the occurrence of Grenvillian and Mesoproterozoic aged zircons in the same samples (C4-D3, C4-D4, C4-D6) supports the view that the two detrital minerals share the same provenance. The large distribution of ages from a single sample is not uncommon. For example, Blewett et al. (2019) reported ⁴⁰Ar/³⁹Ar mica populations greater than 700 Ma, in samples where the main population was ca 250 Ma, and Haines et al. (2004) found several distinct Proterozoic populations in samples of Cambrian sedimentary rocks of the Adelaide Rift Complex.

The older (>500 Ma) population of mica is more common in samples from the Evergreen Formation than from the Precipice Sandstone (Figure 4). For example, the Pacific-Gondwana aged detrital micas are present in the samples from Evergreen Formation from the Chinchilla 4 well and the samples from the upper Precipice Sandstone at the West Moonie 1 well but absent in the samples from lower Precipice Sandstone in both wells (Figure 4). Dating of the U–Pb detrital zircons also supports these findings, with the Pacific-Gondwana sub-population increasing up section (Sobczak et al., 2022). This pattern might reflect the different depositional environments of these two formations. For example, the lower-energy meandering fluvial and fluvial–lacustrine environment of the Evergreen Formation (e.g. Exon, 1976; Green, 1997; Ziolkowski et al., 2014) may have allowed more mica to remain in the system than the high-energy braided river system of the Precipice Sandstone.

Sobczak et al. (2022) noted that the Pacific-Gondwana sub-population increase and the Grenvillian and Proterozoic sub-populations decreased up section, which they attributed to a change in sediment provenance from the northern Thomson Orogen basement blocks (Charters Towers and Anakie Block) to the central Thomson Orogen (notably Roma Shelf) with time. Rare Grenvillian ages are present in the mica samples from the Evergreen Formation but are absent in the samples from the lower Precipice Sandstone (Figure 4). Similarly, rare Mesoproterozoic aged micas are present in the samples from the Evergreen Formation with only a single Mesoproterozoic grain present in samples from the lower Precipice Sandstone (Figure 4).

Samples from the Kenya East GW7, Moonie 34 and Taroom 17 wells (Figure 1) did not produce micas ages older than 480 Ma (Figures 5 and 6). There are two likely explanations for the lack of older ages in these wells: (1) compared with the detrital zircon study, the mica dataset is relatively small (n = 10) and may not have captured the full range of ages present in these samples; and (2) the detrital micas in these wells might have been reset by other younger thermal events. If the resetting event(s) were centred on crust largely to the east of the Surat Basin, they may have affected strata intersected by the eastern wells (KE-GW7 and M34), whereas those to the west (intersected by wells C4 and WM1) escaped overprint.

Figure 6. Kernel Density Estimation plots of the samples from the Kenya East GW7 and Moonie 34 wells, separated into the Evergreen Formation and the lower Precipice Sandstone. The KDE of the ⁴⁰Ar/³⁹Ar ages has not been scaled for sample size to show where the ages overlap.

Population of volcanic and plutonic basement rocks (450–210 Ma)

Based on U–Pb detrital zircon ages, this population has at least four sub-populations. Sobczak et al. (2022) identified a small sub-population of detrital zircons aged 455–450 Ma from samples from the lower Precipice Sandstone (J10-TS1 interval). The source of these detrital zircons was suggested to be igneous rocks, including the Granite Springs Granite (Eulo Ridge: 456–406 Ma; Cross et al., 2018; Purdy et al., 2016, 2018; Sobczak et al., 2022), Fat Hen Complex (Charters Towers Block: 495–450 Ma; Cross et al., 2018; Sobczak et al., 2022) and granites of the Nebine Ridge (458–450 Ma; Siégel et al., 2018), that have intruded the metasedimentary basement rocks of the Thomson Orogen (Sobczak et al., 2022). ⁴⁰Ar/³⁹Ar dating of detrital muscovite from the samples from the lower Precipice Sandstone from the Chinchilla 4 well produced ages that overlapped with the detrital zircon ages, but also extended to younger ages: 455–430 Ma (Figure 4). It is likely that these younger ages represent the same sediment source as the zircon population, but with the Ar–Ar system in mica cooling through its closure temperature (∼350 °C) later than the U–Pb system in zircon (900 °C).

Detrital mica ages in the 424–360 Ma range from samples from the lower Precipice Sandstone in the West Moonie 1 and the Chinchilla 4 wells overlap closely with the corresponding age population of the detrital zircons. However, the muscovite ages from the lower Precipice Sandstone sample from the West Moonie 1 well (WM1-D2) show a slightly younger age range (Figure 4). This may again represent the differences in the closure temperature of the two isotopic systems.

The Early to Middle Devonian sub-population produced by dating of the detrital zircons was interpreted to be recycled from the locally exposed Moolayember Formation or Eddystone Beds, which are stratigraphically below the Precipice Sandstone and straddle the boundary between the Bowen and Surat basins successions (Sobczak et al., 2022). The source was suggested to have been isolated from the Thomson Orogen sources further to the north and west owing to reactivation of the Moonie-Goondiwindi Fault Zone, which includes the Leichhardt-Burunga Fault System (Figure 1; Sobczak et al., 2022). The younger (Middle to Late Devonian) part of this sub-population has been suggested to be sourced from the Texas and Silverwood blocks of the New England Orogen (Sobczak et al., 2022). Alternatively, the variation in age in this sub-population as represented by the detrital mica ages could be due to the difference in isotopic closure temperatures. It could also represent sourcing from the Texas and Silverwood blocks, as U–Pb dating of the Silverwood block indicates that deposition and accretion occurred in the Late Devonian (383–359 Ma; Rosenbaum et al., 2021; Sobczak et al., 2022).

A middle to late Carboniferous (335–317 Ma) sub-population was recorded in the detrital muscovite grains sampled from the lower Precipice Sandstone (J10–J30). This sub-population was also evident in the detrital zircons from the same samples from the lower Precipice Sandstone (J10–J30: Sobczak et al., 2022). Igneous rocks located in the northern New England Orogen were suggested as a possible source as the detrital zircons of the samples (Sobczak et al., 2022).

The Permian–Triassic (266–240 Ma) sub-population of detrital zircons increases up section through the Precipice Sandstone to the Evergreen Formation. This sub-population was linked to a source region from the southern New England Orogen (Sobczak et al., 2022). Permian–Triassic-aged detrital micas, however, are rare in both the lower Precipice Sandstone and the Evergreen Formation. This likely reflects the small Ar–Ar mica sample size. However, the Permian–Triassic population in the much larger zircon dataset is also only a minor population.

Early Jurassic population (<200 Ma)

The youngest age population observed in both the mica and zircon datasets is Early to Mid-Jurassic. Both the detrital micas and zircons of Early to Mid-Jurassic were found to be more common in the sampled stratigraphy above the upper Precipice Sandstone (TS1; Sobczak et al., 2022). However, a single detrital muscovite grain from the lower Precipice Sandstone was dated at 175.7 ± 0.2 Ma. This age is younger than the detrital zircon crystals of the same sample (>200 Ma). Similarly, in the more common age populations from the Evergreen Formation, the detrital muscovites (179.8 ± 0.1 to 152.7 ± 0.2 Ma) are younger than the detrital zircons (197.6 ± 12.9 to 183.8 ± 12.2 Ma). This suggests that the micas progressed through the isotopic closure temperature after the zircons and are, therefore, likely to be derived from the same source. A similar conclusion was reached by Haines et al. (2004) for the Proterozoic Niggly Gap beds in South Australia, where they suggested the muscovite population, dated at ca 1536 Ma, originated from the same source, but reached isotopic stability after the zircons, dated at ca 1590 Ma.

Surat Basin detrital zircons that were younger than 200 Ma were interpreted as sourced from contemporaneous subduction-related volcanism (Sobczak et al., 2022). A subduction zone has been argued to have been active along the eastern margin of Australia during this time (Adams et al., 2021; Foley et al., 2021; Reid et al., 2009; Tucker et al., 2016; Wainman et al., 2019). This proposal is further supported by the geochemical signature of zircons sampled from the Surat Basin, which displays subduction-like signatures rather than intraplate signatures (Wainman et al., 2019). Like the zircons, the detrital micas are more common up section but are only evident in one location: the Kenya East GW7 well, in the eastern part of the basin.

However, it is also possible that the dated micas were not derived from this subduction-related source region but reflect a thermal reset. Early Jurassic ages have been attributed to hydrothermal activity in southeast Queensland associated with the North Arm Volcanics (Ashley & Shaw, 1993). Hydrothermal activity can also be expected to be overprinted by the volcanic provinces of early Jurassic age in the Eromanga Basin to the west (Hardman et al., 2019). Within the samples from the Kenya East GW7 well, there are also mica ages in the 140–47 Ma range, which is significantly younger than the depositional age. These younger ages offer an insight into post-depositional events that have resulted in the resetting of the detrital micas.

Reset ages

The Evergreen Formation has a depositional age that extends to ca 180 Ma (Sobczak et al., 2022). ⁴⁰Ar/³⁹Ar dating of the detrital micas revealed two broad age populations that are younger than the depositional age. A Cretaceous population was recorded for samples from both the Precipice Sandstone and Evergreen Formation, and a Paleocene/Eocene population was recorded in the samples from the Evergreen Formation (Figures 4, 6 and 7).

Figure 7. Schematic linking young ⁴⁰Ar/³⁹Ar ages from detrital micas with key tectonic events occurring in eastern Australia between 200 and 25 Ma (modified from Henderson et al., 2022; Raza et al., 2009; Sliwa et al., 2018).

These young populations dominantly occur in samples from the eastern part of the basin (Kenya East GW7 and Moonie 34 wells), as well as in samples from the north (Taroom 17; Figure 1b). One possibility that could account for these young ages is clay alteration (e.g. Arostegui et al., 2001). However, the grains appeared fresh, and both the detrital muscovite and biotite from the sample produced a similar young age range (Figure 3). Therefore, it is more likely that these younger ages recorded different thermal events within the basin, rather than being the result of low-temperature alteration.

A number of mechanisms can raise the temperature in sedimentary basins, including burial, basal heat flow from magmatic activity and hydrothermal fluids (Raza et al., 2009). Burial was shown to have increased the temperature of the Surat Basin slightly (Raza et al., 2009). However, burial generally only raises the temperature between 200 and 300 °C, at about 10 000 m depth, and as the Surat Basin has a maximum thickness of 2500 m (Exon, 1976), it is very unlikely that the burial temperature reached the required temperature to reset micas (∼350 °C: Spear, 1995).

The population of Cretaceous ages can be broken down into three sub-populations based on age gaps: approximately 140–114 Ma, 103–95 Ma and 83–80 Ma. The Early Cretaceous age (ca 140–114 Ma) sub-population occurs in samples from the eastern Kenya East GW7 well from both the lower Precipice Sandstone and the Evergreen Formation, and from the Evergreen Formation from the northern Taroom 17 well (Figures 6 and 7). The middle age sub-population is observed in samples from the lower Precipice Sandstone from the Taroom 17 well in the north, and samples from the Moonie 34 well in the south of the basin (Figures 6 and 7). This sub-population (103–96 Ma) is also observed in samples from the Evergreen Formation from the Taroom 17 well and samples from the Kenya East GW7 well in the east (Figures 6 and 7).

These two ⁴⁰Ar/³⁹Ar age sub-populations are coeval with subduction along the eastern Australian margin (Foley et al., 2021; Henderson et al., 2022). The Early Cretaceous (140–114 Ma) sub-population overlaps with the youngest U–Pb ages of the Moreton Igneous Association (145–140 Ma), and a range of ages from the Whitsunday Igneous Provenance (132–95 Ma; Figure 7; Henderson et al., 2022). The middle Cretaceous sub-population of 103–96 Ma overlaps with the Whitsundays Igneous Province only. If there were an association between these ⁴⁰Ar/³⁹Ar ages and the coeval igneous activity to the east of the Surat Basin, the micas may have been reset by crustal heat flow from more extensive magmatic activity and/or related hydrothermal fluids.

Raza et al. (2009) noted that large-scale basal heat flow owing to magmatic activity is unlikely to apply to the Surat Basin because it is not recorded in apatite fission track and vitrinite-reflectance data from the samples examined in that study. If hot hydrothermal fluids and metasomatic fluids, which can reach temperatures ranging from about 50 °C to granite melting temperatures (Humphris, 1984; Saunders, 1984), isotopically reset the micas, they would have also reset the apatite fission track data. However, it remains a possibility that hydrothermal activity was limited to certain parts of the basin, exclusive of those documented by Raza et al. (2009).

Several lines of evidence support the suggestion that hydrothermal fluids have percolated through the basin. The mineralogy of fracture fills in cores of the Evergreen Formation suggested that they were potentially formed from high-temperature hydrothermal fluids (Pearce et al., 2019). Golding et al. (2016), through the study of carbonate cements, also supports that view. The presence of dickite, a mineral commonly associated with hydrothermal alteration (e.g. Inoue, 1995; Murray, 1988; Murray & Keller, 1993), is more common in the basal part of the Precipice Sandstone, but less common in the Evergreen Formation (Farquhar et al., 2013).

Fault reactivation may have facilitated the transport of hydrothermal fluids through the basin (Middleton et al., 2014; Uysal et al., 2000). The Kenya East GW7 well is located approximately 30 km southeast of Chinchilla 4 (Figure 1) and where the detrital mica dated from the Evergreen Formation ranges in age from 1500 to 277 Ma (Figure 3). The area around the Kenya East GW7 well, however, is characterised by a normal faulting (OGIA, 2020), but involvement of these structures in hydrothermal fluid transport awaits investigation.

The youngest Cretaceous sub-population is present in samples from both the Precipice Sandstone and the Evergreen Formation. Detrital micas from the Precipice Sandstone date from 83 to 80 Ma (samples from the Kenya East GW7 and the Taroom 17 wells). For the Evergreen Formation, the samples date from 86 to 80 Ma but only in the samples from the Kenya East GW7 well. However, the limited occurrence of the youngest Cretaceous sub-population may be the result of the small sample size.

The youngest Cretaceous detrital micas produced ages that post-date the mid-Cretaceous cessation of subduction in eastern Australia (Figure 7) but do correlate with uplift of the eastern Australian highlands, following the uplift and denudation in the Surat Basin (Jones & Veevers, 1983; Müller et al., 2016). Lister and Etheridge (1989) proposed that uplift of the eastern Australian highlands was caused by igneous underplating, and the resetting of Surat Basin micas may have been induced by related crustal heating. However, Raza et al. (2009) found that a normal geothermal gradient applied for the wells they studied, with no evidence of heat generated by underplating. On the other hand, it is possible that hydrothermal fluids generated by deep crustal heating owing to underplating were of limited distribution migrating through reactivated fault systems (e.g. Middleton et al., 2014) or locally through the contact zone between the porous basal Precipice Sandstone and the underlying, less permeable Moolayember Formation (Farquhar et al., 2013).

The youngest thermal event produced ⁴⁰Ar/³⁹Ar ages that ranged from 57 to 47 Ma (Paleocene/Eocene population). These ages were from one sample (KEGW7-D3) from the base of the Evergreen Formation intersected by the Kenya East GW7 well (Figure 7). Like the Cretaceous population of detrital micas, it is likely that these micas were isotopically reset by the movement of hot hydrothermal fluids. The highly faulted and fractured nature of the Kenya East area would have facilitated fluid transport through the sediments (e.g. Middleton et al., 2014).

The origin of the potential hydrothermal fluids that caused the reset of these youngest mica ages remains a question for further investigation. A possible source is associated with magmatic transport through the crusts by the eruption of the eastern Australian lava-fields, dated from ca 70 Ma (Cohen, 2007, Crossingham et al., 2023; Sutherland, 2003). The Hoy lava-field, located close to the Surat Basin (Figure 1), has eruptions occurring at approximately 67 Ma, 32–31 Ma and 19–21 Ma (Cohen, 2007; Crossingham et al., 2023). However, as only about five of the 70 volcanic plugs related to the lava field have been dated (e.g. Cohen, 2007; Crossingham et al., 2023), it is possible that eruptions were also occurring in the 57–47 Ma interval when the isotopic resetting of the mica from older units of the Surat Basin occurred. Further ⁴⁰Ar/³⁹Ar dating is required to understand the likely cause of this Paleogene event recorded in the basin.

Conclusion

New ⁴⁰Ar/³⁹Ar dating of detrital mica reveals a large range of ages (ca 1500–45 Ma) that are separated into two broad age groups: 1500–180 Ma and 150–45 Ma. Ages older than approximately 180 Ma, when coupled with the U–Pb detrital zircon ages, are linked to multiple source terranes for the sediment (Thomson and New England orogens and the contemptuous magmatic arc on the eastern margin of Gondwana). These sources include recycled sedimentary and metasedimentary basement rocks older than 500 Ma, volcanic and plutonic basement rocks aged 450–200 Ma and volcanic rocks younger than 200 Ma linked to contemporaneous subduction (Sobczak et al., 2022). The ⁴⁰Ar/³⁹Ar detrital mica ages, however, commonly trend toward slightly younger ages than the U–Pb ages of detrital zircons from the same samples. The younger mica ages are likely to reflect the same source terranes as their corresponding zircons, and the discrepancy in ages represents differences in the closure temperature of the two isotopic systems. Post-depositional thermal resetting of the micas by igneous and related hydrothermal activity influencing the Surat Basin may have also contributed to the detrital mica age distribution of this population.

The younger group of ⁴⁰Ar/³⁹Ar dated detrital micas includes Cretaceous and Paleogene populations. These ages post-date deposition and, thus, are interpreted to represent thermal resetting of micas in the Precipice Sandstone and Evergreen Formation. The likely cause of the thermal resetting of the micas was the percolation of hydrothermal fluids through the basin, facilitated by the reactivation of faults and fractures; and/or through more porous and permeable sedimentary layers.

The Cretaceous detrital micas ages may reflect several possible events occurring in the region at the time, including subduction along the eastern Australian margin, and igneous underplating and related generation of hydrothermal fluids. Tectonic event(s) that triggered resetting of the youngest Paleogene population of detrital micas ages remains unconstrained and requires further investigation. This apparent resetting event is only recorded in samples from the base of the Evergreen Formation obtained from the Kenya East GW7 well (eastern Surat Basin). Further ⁴⁰Ar/³⁹Ar dating of micas from the Kenya East GW7 and neighbouring wells is required to determine the age range and impact of potential Paleogene thermal events. Similarly, geochemical fingerprinting of the minerals within fault planes would also help to determine the influence of any hydrothermal fluids. Nevertheless, this study has shown that ⁴⁰Ar/³⁹Ar dating of micas from the Precipice Sandstone and Evergreen Formation reveals different thermal events that have implications for tracing the migration of fluids throughout the Surat Basin.

Acknowledgements

We thank Stan Szczepanski for his expertise in the AuScope-supported ⁴⁰Ar/³⁹Ar dating laboratory at The University of Melbourne. We thank: Nick Hall and the late Rob Heath from CTSCo for providing access to the West Moonie 1 core; Grant Dawson and Sue Golding for their assistance with sampling; Paulo Vasconcelos for his advice on mica grain selection; and Stratum Reservoir and Exploration Data Centre staff for their technical support. We thank Bob Henderson and an anonymous reviewer for their comments and suggestions.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

We would like to thank AuScope for funding the National Argon Map initiative, led by the Australian National University, which supported the ⁴⁰Ar/³⁹Ar analysis. We acknowledge funding to the Centre for Natural Gas at the University of Queensland for funding integration of the ⁴⁰Ar/³⁹Ar results with previous data. The initial sampling was funded through Australian National Low Emissions Coal Research and Development (ANLEC R&D) supported by Low Emission Technology Australia and the Australian Government through the Department of Industry, Science, Energy and Resources.