Carmen Krapf1, Adrian Costar2, Liliana Stoian1, Mark Keppel2, Georgina Gordon1, Kent Inverarity2, Andy Love3 and Tim Munday4
1 Geological Survey of South Australia, Department for Energy and Mining
2 Water Science and Monitoring, Department for Environment and Water
3 National Centre for Groundwater Research and Training, Flinders University
4 Australian Resources Research Centre, CSIRO
Download this article as a PDF (17.1 MB); cite as MESA Journal 90, pages 4–22
Published June 2019
Introduction
The palaeovalleys in the Anangu Pitjantjatjara Yankunytjatjara (APY) Lands, far northwestern South Australia, are the least studied palaeodrainage systems in the state due to their remoteness and limited drilling and geophysical data coverage. An extensive airborne electromagnetic (AEM) survey with a total of 17,395 line kilometres was undertaken in 2016 as part of the Goyder Institute for Water Research’s Finding Long-term Outback Water Solutions (G-FLOWS) Stage 3 project, which was co-funded by the South Australian government’s PACE initiative and supported by the Geological Survey of South Australia (Heath, Wilcox and Davies 2017). The AEM data revealed intricate and well-defined drainage patterns below the modern day land surface of the APY Lands representing palaeovalleys (Soerensen et al. in press). Although the existence of palaeovalleys in this region has been previously discussed (Rogers 1995; Magee 2009) the AEM data enabled detailed mapping of the palaeodrainage and higher spatial accuracy than previous interpretations.
A drilling program was conducted in mid 2018 under the supervision of the Department for Environment and Water based on results of the AEM survey (Costar et al. 2019, same MESA Journal issue; Keppel et al. in press). Drilling focused on two areas within the Lindsay East Palaeovalley near the community of Kaltjiti (Fregon) in the eastern part of the APY Lands (Fig. 1). The drilling program established a number of hydrogeological control test sites to understand and monitor aquifers to gain insights into the groundwater system and potential of the Lindsay East Palaeovalley. Furthermore, drilling determined the thickness of the palaeovalley fill, real depth to basement and provided a stratigraphic record of the palaeovalley fill. This enabled validation of the AEM geophysical model presented in Soerensen et al. (in press), which included the identification of groundwater-bearing zones within the palaeovalleys (Costar et al. 2019; Keppel et al. in press).
The first drillsite (Site DH1) is located ~5 km southeast of Kaltjiti in close vicinity of the Kaltjiti–Mimili road, which crosses the entire Lindsay East Palaeovalley perpendicular to its mainly N–S-oriented drainage path (Fig. 1). It was selected where the channel fill was expected to be at its thickest based on the AEM data (flight line 503401). Drilling revealed the lithological composition of the valley fill, which helped to assess its stratigraphic and hydrogeological compartmentalisation, potential aquifers and its groundwater potential. The second drillsite (Site S22), located ~5 km north of Kaltjiti near the Kaltjiti–Umuwa road (Fig. 1), targeted a tributary to the main palaeovalley with a topographically high palaeo-interfluve area overlying shallow basement. At each drillsite, a diamond drill core through the entire palaeovalley fill as imaged in the AEM model was acquired, revealing for the first time the nature of the sediment fill of a palaeovalley in the APY Lands.
This article focuses on the stratigraphy and evolution of the Lindsay East Palaeovalley, whereas the complementary article by Costar et al. (2019) focuses on the hydrogeological aspects of the drilling program.
Geological setting
The Lindsay East Palaeovalley is located within the southeastern portion of the Musgrave Province and traverses the eastern part of the APY Lands in a N–S direction (Fig. 1). The Paleo- to Mesoproterozoic Musgrave Province consists predominantly of gneissic rocks of the Birksgate Complex, which were deformed and metamorphosed as well as intruded by granitic plutons of the Pitjantjatjara Supersuite during the province-wide c. 1220–1120 Ma Musgravian Orogeny (Edgoose et al. 2004; Howard et al. 2011; Major and Conor 1993; Smithies et al. 2011). These basement rocks were intruded by mafic rocks of the Giles Complex, during the c. 1085–1030 Ma Giles Event, including dykes of the Alcurra Dolerite (Close, Edgoose and Scrimgeour 2003; Edgoose et al. 2004; Glikson et al. 1996; Howard et al. 2011; Quentin de Gromard et al. 2017; Woodhouse and Gum 2003). Following the Giles Event, the c. 825 Ma Amata Dolerite intruded the rocks of the Musgrave Province (Werner et al. 2018). During the 580–520 Ma Petermann Orogeny, several of the prominent E–W-trending faults were reactivated leading to the development of widespread mylonitic shear zones and the exhumation of the Musgrave Province from beneath the Centralian Superbasin (Edgoose et al. 2004; Howard et al. 2011; Pawley et al. 2014). Coeval with the Petermann Orogeny was the development of the Levenger and Moorliyanna grabens, which are interpreted as pull-apart basins (Edgoose et al. 2004; Hand and Sandiford 1999) that were infilled with clastic sediments derived from the locally exposed Musgrave Province basement (Coats 1962). The faults are also interpreted to have been reactivated during the c. 450–300 Ma Alice Springs Orogeny, with fluid flow resulting in local epidote and silica alteration (Conor et al. 2006).
Following the Alice Springs Orogeny, the Musgrave Province underwent at least one phase of intensive deep weathering and erosion prior to the deposition of clastic sediments of the Mesozoic Eromanga Basin along its eastern margin (Conor 2004).
During the Late Cretaceous and throughout the Tertiary, episodes of intensive chemical weathering and leaching of crystalline and sedimentary surface rocks resulted in up to 100 m deep weathering profiles (Conor 2004). These composite deep weathering profiles are represented by mottled to varicoloured, pallid kaolinitic, siliceous and ferruginous zones, which are often associated with various duricrust developments, e.g. silcrete or ferricrete cappings, as well as intermittent sedimentation and erosion (Krapf et al. 2014).
During the Late Paleogene and Early Neogene, a number of drainage systems including the Lindsay and Serpentine palaeovalleys as well as the Hamilton Basin formed and incised up to 70 m into the weathered cover sediments and basement rocks (Rogers 1995). These palaeodrainage systems had their headwaters in the Musgrave Ranges, with flow directions towards the south. They were subsequently filled during a warmer and wetter subtropical to tropical climate in the Mid to Late Neogene with clastic sediments including alluvial, fluvial, and lacustrine sediments composed of clay, sandy clay, mixed sand-clay deposits, and lenses of coarse sand and gravel (Rogers 1995; Magee 2009).
In the Quaternary, the onset of aridity, with episodes of alluvial and aeolian activity, shaped today’s landscape leading to the formation of alluvial fan fringes along the ranges and alluvial plains, creeks, sandplains and aeolian dunes and dunefields (Krapf et al. 2018).
Extent of Musgrave Province palaeovalleys and thalweg mapping
The mountainous region of the Musgrave Ranges dominates the modern day landscape to the north of the study area with the highest mountain in South Australia, Mount Woodroffe, at 1,435 m AHD (Australian Height Datum, i.e. sea level) being located ~57 km NNW of Kaltjiti. The palaeovalleys, similar to the modern day creeks, emerged from the southern foothills of the ranges, from where they flowed in a southerly direction towards the Eucla Basin.
Among a number of creeks draining the Musgrave Ranges today is the Officer Creek, which traverses the study area in a N–S direction. The main channel of the Officer Creek lies ~2 km to the west of the newly mapped thalweg of the Lindsay East Palaeovalley (Fig. 1). Most of the APY communities are located in close proximity to modern day creeks and town water is commonly supplied from wells close to these creeks.
The palaeovalleys of the Musgrave Province are covered by extensive sandplains (Fig. 2) and dunefields and have no obvious surface expression in today’s landscape. However, occurrences of calcrete and chalcedony mounds within semi-confined low areas within sandplains and dunefields (Fig. 3) coincide in many places with the occurrence of the palaeovalleys in the subsurface.
Figure 2 Landscape around drillsite DH1a showing the extensive sandplain that typically covers the palaeovalleys in the APY Lands. (Photo 417794)
The subtle expression of the palaeovalleys in the modern day landscape calls for use of remote sensing and geophysical datasets in order to map these buried systems.
The palaeodrainage network across the Musgrave Province and Officer Basin was originally interpreted from NOAA–AVHRR (National Oceanic and Atmospheric Administration - Advanced Very High Resolution Radiometer) satellite pre-dawn thermal imagery (Statham-Lee 1994, 1995) and published at 1:1 million scale as part of the Geological Atlas of the Officer Basin, South Australia (Lindsay 1995). Minor updates to this interpretation were implemented as part of the Geological Survey’s Palaeodrainage and Cenozoic coastal barriers of South Australia map (Hou et al. 2007, 2012) and Geoscience Australia’s WASANT Palaeovalley Map (Bell et al. 2012).
In this study, the new AEM data was used to map the spatial distribution of the palaeovalleys in the subsurface as well as to define the palaeovalley thalweg. Conductivity model depth sections inverted from AEM data reveal a complex, well-defined and relatively narrow set of palaeovalleys that contrast with those depicted in the contemporary landscape today (Fig. 1; Munday et al. 2013; Soerensen et al. in press). This enabled mapping of the palaeovalley network in more detail and higher spatial accuracy in comparison to the previous interpretations.
Based on E–W-trending conductivity cross-sections oriented perpendicular to the palaeovalley’s axis, the top of the palaeovalleys were picked from the AEM dataset. This enabled choosing the appropriate conductivity depth slice for palaeovalley mapping in plan view under cover. Furthermore, drillhole DH1a (SA Geodata drillhole no. 313475, unit no. 5344-87) revealed the boundary between the Lindsay East Palaeovalley and the overlying sandplain deposits to be at ~30 m depth. Consequently, the 26.2–31.8 m conductivity depth slice formed the base of our new interpretation of the spatial extent of the Lindsay East Palaeovalley as well as for the entire palaeovalley network in the APY Lands as imaged in the AEM dataset (Fig. 1).
The thalweg is the line of lowest elevation within a valley. Due to the variable incision rate and the undulating palaeotopography of the variably weathered basement, several AEM depth slices had to be utilised for thalweg mapping. Drilling showed that the depth of the basement–palaeovalley interface ranges from ~45 m in drillhole S22i (drillhole no. 313477, unit no. 5344-88) to ~108 m in drillhole DH1a. Thus, AEM depth slices 86.1–100.0 m and 100.0–115.8 m were chosen for the thalweg mapping.
This new thalweg data can be used to identify the deepest part of the palaeovalley that has the highest potential for good quality groundwater with high flow rates (Costar et al. 2019; Keppel et al, in press).
Role of neotectonics on palaeovalley evolution through time
The AEM data reveals the extensive and intricate palaeodrainage network throughout the APY Lands and also indicates that underlying basement structures had a significant influence on its development through time. Despite the main palaeovalley axes following a N–S trend, many tributary channels are oriented more E–W (Fig. 4).
The Musgrave Province is crosscut by a series of major east-trending shear zones, developed during the Musgravian Orogeny, and reactivated during the Petermann Orogeny (Woodhouse and Gum 2003). One of the most prominent structural features within the wider study area is the Woodroffe Thrust, which is a zone of sheared gneiss, mylonite, and pseudotachylite. The Woodroffe Thrust subdivides the Musgrave Province into the northern Mulga Park Subdomain and the southern Fregon Subdomain and represents a major crustal discontinuity (Major and Conor 1993; Pawley and Krapf 2016). The area around Kaltjiti is dominated by a series of unexposed, east-striking, laterally extensive faults that can be traced as low-intensity aeromagnetic lineaments across the study area (Pawley et al. 2014). These include the Wintiginna and Echo faults (Fig. 4).
The map in Figure 4 clearly shows the influence of these basement structures on the palaeodrainage pattern in the Kaltjiti area. Lateral extensions of conductivity spreading out perpendicular to the N–S-oriented main palaeovalley indicates topography-induced ponding effects along east-trending tectonic structures. These ‘ponding’ zones could represent potential targets for groundwater exploration in this region.
Furthermore, these observations also indicate that these basement structures may have been active during and after palaeovalley development. Recent seismic activity shows that the Musgrave Province is still tectonically active with a number of significant earthquakes having occurred in the region in the last 40 years (Pawley and Krapf 2016). In 1986, a 6.0 magnitude earthquake produced a 13 km long fault scarp with a maximum throw of 0.6 m along the Marryat Fault zone and, more recently, the community of Pukatja (Ernabella) recorded two magnitude 5.7 earthquakes both in 2012 and 2013 (Geoscience Australia 2018). This suggests that reactivation of these faults is still occurring, with tectonic movements potentially increasing porosity and permeability along these fault zones via brecciation, but also causing local groundwater ponding due to vertical movements along these structures.
Stratigraphic composition of the palaeovalley fill
Limited stratigraphic information exists about the sedimentary infill of palaeovalleys in the Musgrave region (Magee 2009). Zang and Stoian (2006) identified two palaeovalley units, the Eocene fluvial Pidinga Formation and the Miocene to Pliocene lacustrine Garford Formation, both of which are well known from palaeovalleys located along the northern margin of the Eucla Basin (Hou et al. 2003, 2006, 2008). As part of the Geological Atlas of the Officer Basin, South Australia, Lau et al. (1995) mapped the regional groundwater systems and identified Garford Formation sediments and the lacustrine Mangatitja Formation (Major 1973) as the main stratigraphic units in palaeovalleys that trend southwards from the Musgrave Province across the Officer Basin to the Eucla Basin.
The palaeovalleys are overlain by extensive sandplains (Fig. 2) or sand dunes with occasional outcrops of calcrete and chalcedony mounds (Fig. 3). Optically stimulated luminescence (OSL) dating (Sheard et al. 2006) indicated that the neighbouring dunes of the Great Victoria Desert were deposited around 200 ka ago. Recent OSL dating of dunes and sandplains in the vicinity of the Alberga River in the easternmost part of the Musgrave Province by Krapf et al. (2018) yielded much younger ages between 24 and 64 ka for these deposits.
Sedimentary facies of the Lindsay East Palaeovalley fill from new drillholes
One of the aims of the recent drilling program in the APY Lands was to acquire a diamond drill core of the thickest part of the sedimentary infill of the Lindsay East Palaeovalley, ideally also including the top of the underlying basement (Costar et al. 2019; Keppel et al. in press). Drillhole DH1a was sited for this purpose and thereby tested the AEM data and geophysical model of Soerensen et al. (in press). A second drillhole (S22i) aimed at intersecting and coring the palaeovalley fill of a tributary channel to the main Lindsay East Palaeovalley.
Drill core DH1a
Drill core was successfully retrieved from drillhole DH1a to a depth of 93.4 m. This drillhole intersected three main sandy successions with the lower two separated by a thick interval of mud (Figs 5, 6). No core material could be recovered from below 93.4 m depth due to continued core loss. Successive rotary mud drilling produced cuttings up to a depth of 117 m with the palaeovalley fill – basement contact intersected at ~108 m.
The contact between weathered basement and the overlying, also intensively weathered, palaeovalley fill sediments is not very distinctive in the drill cuttings, which is also reflected in the diffuse boundary seen in the AEM model (Soerensen et al. in press). However, thorough inspection of the cuttings identified a noticeable change of quartz grain morphology with depth. Above ~108 m, quartz grains are mainly subangular to subrounded indicating that the grains have experienced mechanical abrasion during sedimentary transport. Below ~108 m, quartz grains display more angular to subangular morphologies and are interpreted to be in situ within the weathered basement, likely granites of the Pitjantjatjara Supersuite (Pawley and Krapf 2016).
The basal part of the sedimentary palaeovalley fill is ~23 m thick (85–108 m depth). It is composed of highly weathered, semi-consolidated, poorly to moderately sorted, coarse-grained, quartz-rich, kaolinitic sandstones that grade upward into consolidated, massive, more mature, moderately to well sorted, fine-grained, quartz-rich sandstones (Figs 6a, 7a).
Figure 6 Drill core DH1a. (White stars indicate the position of thin section samples shown in Fig. 7)
This sandstone succession is overlain by a ~20 m thick succession of distinctive brown-black organic-, sulfidic- and clay-rich basal muds that are separated from olive-green muds higher up in the succession by a 1.25 m thick massive gypsum layer (Figs 6b, 7b). This mud-dominated interval (from 85 to 65 m) corresponds to a prominent conductive zone in the AEM dataset (Soerensen et al. in press; Keppel et al. in press).
The mud interval is overlain by a second, ~38 m thick (65–27 m depth) sandy succession characterised by partly massive, moderately sorted, fine- to coarse-grained, quartz-rich sands and sandstones, which in parts have intercalations of clay and gravel layers (Figs 6c, 7c). The consolidation and cementation of these sandy deposits is highly variable, ranging from compact sandstones to free flowing unconsolidated sands. The depth interval 65 to 60 m is composed of interbedded mud and coarse-grained sand, which shows no or only weak calcification. Higher up in the core, calcification increases within this sandy unit. The calcified part shows dissolution features resulting in cellular-like calcite veining patterns in parts of the core (Fig. 7c). Root casts are also common in the upper part of this unit. The transition into the overlying uppermost sandy succession is very gradational as its composition is similar to the underlying sandy unit. This is also reflected in the AEM data with no distinct conductivity variations visible within the combined upper sandy interval above the high conductivity mud unit (Fig. 5).
From ~27 m depth to the surface, distinctive up to 50 cm thick pedogenic calcrete horizons (Fig. 7d) occur within semi-consolidated, moderately sorted, fine- to coarse-grained sands (Fig. 6d). Hyperspectral mineralogy data acquired via HyLoggerTM reflect this in a change in clay mineralogy from kaolinite in the upper sandy interval to montmorillonite in the overlying sandplain deposits (Fig. 5).
Figure 7 Thin section photomicrographs of samples from drill core DH1a. (Sample locations shown in Fig. 6)
Drill core S22i
Drillhole S22i is located over a tributary channel to the main Lindsay East Palaeovalley. This tributary channel is less incised and thus has a thinner fill than the main palaeovalley below drillhole DH1a. Drillhole S22i was cored to a depth of 54.34 m (Fig. 8) with highly weathered and mottled crystalline basement being intersected from ~46.3 m (Fig. 9a). As in drillhole DH1a, the basement in drillhole S22i consists of Pitjantjatjara Supersuite granites (Pawley and Krapf 2016).
In contrast to drillhole DH1a, drillhole S22i intersected only sands and sandstones above the weathered basement. The ~46 m thick sandy succession consists of two fining-upward sequences of alluvial/fluvial palaeovalley deposits (10–46 m) that are overlain by sandplain deposits containing pedogenic calcrete (0–10 m). The lower sandy sequence of the palaeovalley fill (22–46 m) is composed of mottled, poorly sorted, coarse- to very coarse-grained, gravelly sandstones that grade higher up into better sorted, medium-grained sandstones. The upper sequence (10–22 m) starts with moderately sorted, coarse-grained sandstones grading upwards into moderately to well sorted, fine- to medium-grained sandstones (Fig. 9b).
Figure 9 Drill core S22i. (a) Core interval 44.92 to 48.59 m showing the contact between deeply weathered and intensively mottled palaeovalley fill sediments and granitic basement at 46.3 m. (Photo 417803) (b) Massive to mottled, well-sorted, fine- to medium-grained fluvial sandstones of the palaeovalley fill between 20.52 and 24.16 m. (Photo 417804)
Sedimentological and depositional environment interpretation of the Lindsay East Palaeovalley fill from new drillholes
The quartz-dominated sands and sandstones of the palaeovalley fill are lacking sedimentary structures and have limited compositional and grain-size variation (no clasts greater than granule size). This can be attributed to the source material of the palaeovalley fill, which was dominated by felsic gneisses and granites that have been intensively weathered to kaolinitic saprolite with residual quartz grains. Quartz-rich sands that possess compositional and grain-size properties similar to the sediments within the palaeovalley also dominate channel sediments of modern creeks in the APY Lands, such as Officer Creek. Hence, the basal and upper sandstone units that are composing the major part of the fill of the Lindsay East Palaeovalley in drillhole DH1a are both interpreted to be of fluvial origin.
The clay-rich muds were likely deposited in a quiescent environment within the palaeovalley. The brown-black basal muds between 75 and 85 m depth are organic-rich and sulfidic indicating that the depositional environment was anoxic and permanently covered with water during mud deposition. The development of a gypsum layer between the basal dark coloured and the overlying lighter coloured muds indicates that the palaeovalley became temporarily dry, before deposition returned to mainly subaqueous conditions in a more oxidising and ephemeral environment, comparable to the conditions in a playa lake.
The sand unit overlying the palaeovalley fill sediments is characterised by multiple up to 50 cm thick pedogenic calcrete horizons. These calcrete-bearing sandy sediments record the change from a fluvially dominated environment during the filling of the palaeovalley to the formation of a semi-arid to arid sandplain landscape dominated by sheetwash and aeolian processes with only minor fluvial activity.
The sedimentary fill of the tributary channel intersected in drillhole S22i is overall compositionally less mature than that of the main palaeovalley fill, reflecting the distality of the tributary within the palaeodrainage system and its proximity to exposed basement, which outcrops ~5 km upstream (i.e. NW) from drillhole S22i. The absence of lacustrine muds as intersected in drillhole DH1a also reflects the distality of this tributary channel to the main trunk of the Lindsay East Palaeovalley and indicates that flooding of the main palaeovalley did not extent upstream into this tributary.
Palynological constraints on the depositional environment and age for the Lindsay East Palaeovalley
Eleven samples from fine-grained, clay-rich intervals of the two drill cores were collected for palynological analyses. Nine samples between 65.2 to 84.5 m depth were taken from the mud-rich interval of drill core DH1a (sample locations shown in Fig. 5). Only two samples were taken from drillhole S22i (at 6.9 and 8.75 m depth) as fine-grained sedimentary material suitable for palynological analysis is rare in this core (sample locations shown in Fig. 8). Samples were processed at MGPalaeo Laboratory in Perth using standard laboratory techniques for palynology: treatment with hydrochloric and hydrofluoric acids followed by digestion using LST (lithium heteropolytungstate in water) at a specific gravity of 2.1, sieving through 10 µm sieve and the residue mounted as kerogen slides. Palynological analysis, interpretation of the depositional environment and dating were undertaken by Liliana Stoian (Geological Survey).
Overall, pollens and spores are rare or absent in the majority of the samples from DH1a and S22i. Nevertheless, the variety of taxa and palynomorphs encountered in the samples allowed interpretation of a depositional environment and age for most samples.
Marine algal flora and marine dinoflagellate cysts such as cf Batiacasphaera sp. and Impagidinium sp. (Figs 10a–d) are present in samples from 65.2 and 81.7 m depth in drill core DH1a. Samples from 68.85, 74.6, 77.25 and 78.32 m depth in drill core DH1a show a mix of freshwater and marine taxa including dinoflagellate cysts and phytoplankton typical for a brackish environment. The freshwater dinoflagellate Saeptodinium sp. (Fig. 10e) as well as a variety of pollen and spores species, including Staurastrum sp. and Myrtaceidites eucalyptoides (Figs 10f–n), are present in samples from 76.83, 80.3 and 84.45 m depth. Staurastrum sp. (Fig. 10f) represents a particular group of algae living exclusively in a freshwater lacustrine environment.
Palynomorphs from the two samples of drill core S22i are derived from clay-rich sands in the uppermost stratigraphic unit (Fig. 8). They contained freshwater algae, which record an exclusively terrestrial non-marine depositional environment. Pollen are very rare. Only the sample from 6.9 m depth had pollen of Myrtaceidites eucalyptoides, which is known from the Tertiary.
Palynomorphs that are indicative of a terrestrial depositional environment identified in samples from DH1a and S22i include mainly salt-tolerant grass taxa, with rare eucalyptus (Figs 10j–m). It is interpreted that as low nutrient soils became increasingly widespread, expansion of Chenopodiaceae and Asteraceae was favoured. Chenopodipollis chenopodiaceoides, Tubuliflorides antipodica, Myrtaceidites eucalyptoides and Rhoipites sp. are consistently present in the lower interval (samples from 76.83 to 84.5 m depth) in DH1a. Phytoliths, plant microfossils of solid opal silica deposited between and within the cells of most plants, are also present in some samples. They are associated with either land or very shallow wetland type environments. They derive from grasses, possible wetland plants, and are present in non-marine sediments. The palynomorphs encountered in the samples could be derived from vegetation that inhabited the palaeovalley, e.g. floodplains or lake shorelines, or may have been washed or blown in from adjacent interfluve areas outside of the palaeovalleys.
The low number of pollen and spores in the majority of the samples from drill cores DH1a and S22i may be linked to a low preservation potential in oxidative and acidic palaeoenvironments. Such conditions may have developed in some parts of the palaeovalley area due to the oxidation of biogenic and/or diagenetic pyrite present in lacustrine to marginal marine sediments.
The palynological record in the samples from drill core DH1a shows that the mud-rich succession was mainly deposited in a subaqueous environment characterised by fluctuating freshwater to brackish conditions in a fluvial to lacustrine depositional environment. Two samples from this mud-rich interval, one near the base at 81.7 m depth and one near the top at 65.2 m depth, contained taxa that are indicative of marine influence and record two intermittent marginal marine to estuarine environmental conditions within the palaeovalley.
The palynological analyses and dating of the core samples from DH1a indicates a Late Miocene – Early Pliocene age (10–5 Ma) for the sampled interval. Palynomorphs from the two samples derived from the uppermost stratigraphic unit in S22i record an exclusively terrestrial non-marine depositional environment. Due to absence of key indicator taxa in drill core S22i, no age could be assigned to the sampled uppermost depositional unit.
The palynology results show that marginal marine to estuarine conditions prevailed in the palaeovalleys far inland from the coastal areas of the Eucla Basin during Late Miocene – Early Pliocene in South Australia (Fig. 11). The marine influence reached as far north as the foothills of the Musgrave Ranges, where alternating freshwater, lacustrine and marginal marine to estuarine conditions during the Late Miocene and Early Pliocene are documented in drill core DH1a. Palynological evidence from other palaeovalley sediments encountered in cores in other parts of the Musgraves and adjacent areas (Fig. 11) also indicate marine influences, e.g. in drillhole Officer 1 (Zang and Stoian 2006). Further to the northeast, palynomorph assemblages of samples from the MAN 1 drillhole can also be attributed to deposition in a marginal marine to estuarine environment during the Late Miocene – Pliocene and include Gramineae and Chenopodiaceae with rare Casuarinaceae and Podocarpaceae. Dinoflagellate cysts, such as Tectatodinium pellitum and Apteodinium australiense, are species which typically characterise warm temperate waters. Palynological analyses and dating of core and cutting samples from a significant number of drillholes in the Eucla Basin also suggest some marine influence in Late Miocene to Early Pliocene sediments. These include samples from drillholes KIN 20, KIN 21, KIN 22, KIN 45, CAR 034, CAR 037, NALARA NR 3, NALARA NR 4, KINPC 2 and Konkaby 1 (Fig. 11). The marine influence is present – at different degrees – in all these locations.
Evolution of the Lindsay East Palaeovalley
Late Mesozoic to Early Paleogene: formation of incised valleys
The Lindsay East Palaeovalley was incised up to 40 m into the underling weathered crystalline bedrock (Fig. 12a) and based on the AEM data this scale of incision also applies to other palaeovalleys in the wider Musgrave Province region. Incision was preceded by deep weathering of exposed basement rocks in the Mid to Late Mesozoic (Alley et al. 1999). The palaeolandsurface was mantled by a deep weathering profile reflecting a warm, humid climate (Alley et al. 1999; de Broekert and Sandiford 2005). Incision was accompanied by the generation of a regional unconformity involving widespread stripping of older sediments and basement rocks well beyond the palaeovalley flanks (Alley et al. 1999; de Broekert 2002).
The time of incision and hence the formation of the palaeovalleys in South Australia, including the Musgrave region, is debated. Hou et al. (2008) consider that the formation of the majority of the Eucla Basin palaeovalleys happened in the Mesozoic as their lower reaches contain sediments of the Early Cretaceous Madura Formation.
Although palaeovalley incision cannot be directly dated, dating of sediments shows that infilling of the palaeovalleys began in the Early to Middle Eocene in the eastern Eucla Basin area (Alley and Beecroft 1993; Benbow et al. 1995; Hou et al. 2006).
Eocene: First fluvial deposition within palaeovalley
Infilling of the palaeovalley began with the deposition of a sandy fluvial succession (Figs 12a, b). The exact age of these sediments is not known but they must be older than the overlying Late Miocene – Early Pliocene mud succession. By comparing the sedimentary succession of the Lindsay East Palaeovalley with those of other palaeodrainage systems in South Australia, this basal sandy unit is tentatively correlated with parts of the Pidinga Formation, which forms the lower part of the stratigraphy in the palaeovalleys adjacent to the Eucla Basin. Consequently, a Paleogene to Early–Mid Neogene age is inferred for the basal sandy interval in the Lindsay East Palaeovalley.
Late Miocene to Early Pliocene: freshwater and marine environmental reversals
The identification of two marginal marine to estuarine intervals within the mud unit of drill core DH1a suggests that the Lindsay East Palaeovalley periodically experienced marine influences with the sea transgressing far inland beyond the coastal margin and wetlands areas of the Eucla Basin (Fig. 12b). The combined effects of a warm and humid climate and a rising sea level accompanied by subsidence and orogenic movements during the Late Miocene (Hou et al. 2008) can explain the presence of marginal marine deposits in the Lindsay East Palaeovalley at the foothills of the Musgrave Ranges more than 300 km NNE of the palaeocoastline.
Inland marine conditions during the Late Miocene to Early Pliocene have also been identified in other areas of South Australia, for example, within palaeovalleys at the northeastern margin of the Eucla Basin, within the Willochra and Pirie basins, and within small intermontane basins located in the Adelaide Hills adjacent to the St Vincent Basin (Stoian 2003a, b, 2004a, b, c, 2005).
The base of the mud-rich interval intersected in drill core DH1a records a first marine incursion, where the deeply incised valleys became flooded to form a large inland estuary system. The marine influence gradually faded and a brackish to freshwater lake occupied the valley floor or parts of it as indicated in the palynology from marine to brackish to freshwater taxa. The organic-rich and sulfidic black muds are indicative of a depositional environment that was anoxic and permanently covered with water. Evaporation caused the temporary drying out of this waterbody leading to the deposition of a gypsum layer (Fig. 12c). After this evaporation event, conditions returned to mainly subaqueous deposition in a more oxidising and ephemeral lacustrine environment, comparable to the conditions in a playa lake. However, palynology indicates brackish conditions for this upper part of the mud succession and near the top the palynomorphs assemblage documents a second marine incursion.
Early to Late Pliocene: Second fluvial deposition within palaeovalley (final infilling)
After the mud-dominated subaqueous deposition phase in the Late Miocene to Early Pliocene, the interior of the palaeovalley changed back to a dominantly fluvial environment leading to the successive infill of the palaeovalley with quartz-rich sands containing minor clay and gravel intercalations (Fig. 12d). This final infill phase may be related to wetter conditions in the catchment area and thus increased water and sediment inflow into the palaeovalley.
Late Pliocene to Early Pleistocene: sedimentation beyond the palaeovalley
Deposition continued after infilling of the palaeovalley. However, with increasing aridity the depositional environment gradually switched from fluvial-dominated to sheetwash- and aeolian-dominated leading to the formation of extensive sheet-like sandplain deposits. These sediments were locally indurated by carbonate or silica forming calcrete and chalcedonic silcrete horizons within them (Fig. 12e).
Early Pleistocene to Holocene: sandplain with pedogenic calcrete development and aeolian dunes and dunefields
Today’s landscape is characterised by extensive sandplains and dune fields with minor creeks. Pedogenic calcretes and chalcedonic silcretes, which have widely formed within the sandplain deposits, are frequently cropping out as resistant mounds in low-lying areas (Fig. 12e).
Conclusion
- The AEM data in combination with drilling provided novel insights into the subsurface distribution of palaeovalleys in the Musgrave Province within the APY Lands and allowed the mapping of the palaeovalleys and their thalwegs. AEM data show the E–W orientation of tributaries to the main palaeovalley drainage, which is likely a function of the E–W structural grain in the Proterozoic basement.
- Two diamond drill cores revealed the lithological composition, depositional environments and stratigraphic architecture of the palaeovalley fill and resulted in a landscape evolution model. Two main episodes of fluvial deposition of sand are separated by a phase of dominantly subaqueous mud deposition under freshwater–lacustrine to brackish-estuarine to marginal marine conditions. Intermittent evaporation and drying up of the palaeovalley led to the deposition of a 1.25 m thick gypsum layer.
- Recurrent marine influences, as indicated by palynomorphs, indicate that marine flooding turned the Lindsay East Palaeovalley into an estuarine system during the Late Miocene – Early Pliocene and marine influence reached close to the foothills of the Musgrave Ranges, more than 300 km inland of the palaeocoastline of the Eucla Basin. Palynological analysis also enabled the stratigraphic correlation of the sedimentary succession of the Lindsay East Palaeovalley with those of the palaeovalleys of the Eucla Basin.
Acknowledgements
We would like to acknowledge the traditional owners of the APY Lands, the Pitjantjatjara, Yankunytjatjara and Ngaanyatjarra people. In particular, we would like to thank Mr Witjiti George, Mr Maxi Stevens, Mr Robert Stevens, Mr Bruce, Mr Frank, Mr Lee and many others for undertaking on-country site inspections. We would also like to acknowledge the work undertaken by the APY Consultation, Land and Heritage Unit, including Ms Charmaine Jones, Ms Cecilia Tucker, Mr Noah Pleshet and Mr Andrew Cawthorn, who facilitated the necessary clearance approval to undertake this program of works within the APY Lands. Silver City Drilling are acknowledged for their drilling services to deliver this program as well as the staff of Regional Anangu Services Aboriginal Corporation (RASAC) in Umuwa for logistical support including accommodation. We also thank APY General Manager, Mr Richard King, and the entire APY Executive Board who were supportive of the G-FLOWS Stage 3 project. Mario Werner, Anthony Reid and Justin Gum (Geological Survey) are thanked for their helpful review of the manuscript.
References
Alley NF and Beecroft A 1993. Foraminiferal and palynological evidence from the Pidinga Formation and its bearing on Eocene sea level events and palaeochannel activity, eastern Eucla Basin, South Australia. Memoir of the Association of Australasian Palaeontologists 15:375–393.
Alley NF, Clarke JDA, Macphail M and Truswell EM 1999. Sedimentary infillings and development of major Tertiary palaeodrainage systems of south-central Australia. Special Publication of International Assessment of Sediment 27:337–366.
Bell JG, Kilgour PL, English PM, Woodgate MF, Lewis SJ and Wischusen JDH comps 2012. WASANT Palaeovalley Map - Distribution of palaeovalleys in arid and semi-arid WA-SA-NT, 1:4,500,000 scale, Geoscientific thematic map. Geoscience Australia, Canberra.
Close DF, Edgoose CJ and Scrimgeour IR 2003. Hull and Bloods Range Special, Northern Territory, 1:100 000 Geological Map Series Explanatory Notes, 4748, 4848. Northern Territory Geological Survey, Darwin.
de Broekert PP 2002. Origin of Tertiary inset-valleys and their fills, Kalgoorlie, Western Australia. PhD thesis, Australian National University, Canberra.
de Broekert PP and Sandiford M 2005. Buried inset-valleys in the eastern Yilgarn Craton, Western Australia: geomorphology, age, and allogenic control. The Journal of Geology 113:471–493.
Edgoose CJ, Scrimgeour IR and Close DF 2004. Geology of the Musgrave Block, Northern Territory, Report 15. Northern Territory Geological Survey, Darwin.
Gallant JC and Dowling TI 2003. A multiresolution index of valley bottom flatness for mapping depositional areas. Water Resources Research 39(12):347.
Geoscience Australia 2018. Earthquakes@GA. Geoscience Australia, viewed April 2018.
Glikson AY, Stewart AJ, Ballhaus CG, Clarke GL, Feeken EHJ, Leven JH, Sheraton JW and Sun S-s 1996. Geology of the western Musgrave Block, Central Australia, with particular reference to the mafic-ultramafic Giles Complex, Bulletin 239. Australian Geological Survey Organisation, Canberra.
Hand M and Sandiford M 1999. Intraplate deformation in Central Australia, the link between subsidence and fault reactivation. Tectonophysics 305:121–140.
Hou B, Alley NF, Frakes LA, Gammon P and Clarke J 2003. Facies and sequence stratigraphy of Eocene palaeovalley fills in the eastern Eucla Basin, South Australia. Sedimentary Geology 163:111–130.
Hou B, Alley NF, Frakes LA, Stoian L and Cowley WM 2006. Eocene stratigraphic succession in the Eucla Basin of South Australia and correlation to major regional sea-level events. Sedimentary Geology 183:297–319.
Hou B, Frakes LA, Sandiford M, Worrall, Keeling J and Alley NF 2008. Cenozoic Eucla Basin and associated palaeovalleys, southern Australia — Climatic and tectonic influences on landscape evolution, sedimentation and heavy mineral accumulation. Sedimentary Geology 2003:112–130.
Hou B, Zang W, Fabris A, Keeling J, Stoian L and Fairclough M comps 2007. Palaeodrainage and Tertiary coastal barriers of South Australia, 1:2,000,000 Series. 1st edn. Department of Primary Industries and Resources South Australia, Adelaide.
Howard HM, Smithies RH, Evins PM, Kirkland CL, Werner M, Wingate MTD and Pirajno F 2011. Explanatory notes for the west Musgrave Province, Record 2011/4. Geological Survey of Western Australia, Perth.
Keppel M, Costar A, Krapf C and Love A in press. G FLOWS Stage 3, APY Lands Drilling Program, north-western South Australia, Technical Report. Goyder Institute for Water Research, Adelaide.
Magee JW 2009. Palaeovalley groundwater resources in arid and semi-arid Australia – a literature review, Record 2009/03. Geoscience Australia, Canberra.
Munday T, Adbat T, Ley-Cooper Y and Gilfedder M 2013. Facilitating Long-Term Outback Water Solutions (G FLOWS) Stage-1: Hydrogeological framework, Technical Report Series13/12. Goyder Institute for Water Research, Adelaide.
Quentin de Gromard R, Howard HM, Smithies RH, Wingate MTD and Lu Y 2017. The deep seismic reflection profile 11GA-YO1 in the west Musgrave Province: an updated view, Record 2017/8. Geological Survey of Western Australia, Perth.
Smithies RH, Howard HM, Evins PM, Kirkland CL, Kelsey DE, Hand M, Wingate MTD, Collins AS and Belousova E 2011. High-temperature granite magmatism, crust–mantle interaction and the Mesoproterozoic intracontinental evolution of the Musgrave Province, Central Australia. Journal of Petrology 52:931–958.
Soerensen CC, Munday TJ, Ibrahimi T, Cahill K and Gilfedder M in press. Musgrave Province, South Australia: Processing and inversion of airborne electromagnetic (AEM) data: preliminary results, Technical Report Series. Goyder Institute for Water Research, Adelaide.
Soerensen C, Munday T, Krapf C, Love A , Costar A, Inverarity K, Gogoll M and Gilfedder M 2018. Uncovering the Musgrave Province in South Australia using airborne EM. ASEG Extended Abstracts 2018:1–5.
Zang WL and Stoian L 2006. Tertiary-Quaternary land evolution of north-western South Australia. Regolith 2006, pp. 372–376.