Introduction

Figure 1 Willunga and Noarlunga embayments backed by the Willunga and Clarendon – Ochre Cove faults respectively. Modified from Talbot and Nesbitt (1968).

The application of optically stimulated luminescence (OSL) dating techniques to the alluvial and colluvial successions in Sellicks Creek (Bourman et al. 2020) and the recognition of the ‘Pirramimma Sandstone’ of Fairburn (1998) as a discrete and unconfined aquifer (Barnett and Bourman 2022) have provided the impetus for a revision of the Pleistocene stratigraphy of the Willunga and Noarlunga embayments (Fig 1). Contrary to the work of Fairburn (1998) and Fairburn et al. (2010) these investigations demonstrate that the ‘Pirramimma Sandstone’ is not of Pleistocene age, that the Seaford Formation is of Early Pleistocene age, and that the Ngaltinga Formation and the Taringa Formation are separate units of Middle Pleistocene age and penultimate glacial age Marine Isotope Stage (MIS) 6 (160 ± 15 ka) respectively. Furthermore, the Pooraka Formation (the Christies Beach Formation of Ward 1966) is revealed to span the entire last interglacial (c. 132–73 ka ago). In addition, some previously unreported units, an unnamed MIS 3 (42 ± 3.2 ka) subpluvial alluvial fill unit, a late MIS 2 (14.8 ± 1.7 ka) unit and post European settlement aggradation (PESA) sediments have been identified. Finally, establishing a mid-Holocene (MIS 1) age for the Waldeila Formation using radiocarbon and amino acid racemisation dating suggests that the formation of the Ngankipari Sand can be tentatively ascribed to the latter part of the last glacial maximum (MIS 2). No doubt there will be future revisions as further discoveries occur and new dating techniques become available.

Stratigraphic successions

The stratigraphy is discussed below and summarised in Table 1.

Pirramimma Sand

The Pirramimma Sand was originally defined as the ‘Pirramimma Sand Member’ of the Port Willunga Formation of Late Oligocene to Miocene age by Cooper (1977, 1979), whereas Fairburn (1998) renamed it the ‘Pirramimma Sandstone’ and considered it to be of Middle Pleistocene age. He also regarded it as a sand member of the Ngaltinga Formation, reworked from the Port Willunga Formation. In a subsequent revision Fairburn et al. (2010) inserted the ‘Pirramimma Sandstone’ between the Burnham Limestone and the Kurrajong Formation in the Early Pleistocene, while in the most recent version of the map Geology of the McLaren Vale Wine Region (Aldam et al. 2019) the ‘Pirramimma Sandstone’ and the Seaford Formation were grouped together between the Hallett Cove Sandstone and the Burnham Limestone.

Recent investigations by Barnett and Bourman (2022, same MESA Journal issue) concluded that the Pirramimma comprises a discrete stratigraphic unit overlain by Quaternary sediments and underlain by the Port Willunga Formation and older units. They proposed the name Pirramimma Sand to reflect its unconsolidated nature. It is possibly of Late Oligocene (Janjunkian) age (Lindsay 1988; Barnett and Bourman 2022).

Pleistocene terrestrial successions

Burnham Limestone

The Burnham Limestone occurs at the base of the Pleistocene sequence in the Willunga Embayment at Sellicks Beach (May and Bourman 1984), providing an age of c. 2.6 Ma for the base of the alluvial Pleistocene successions on current numeric ages ascribed to the Geological Time Scale (Pillans and Gibbard 2012). The Burnham Limestone (Firman 1976) contains a distinguishing pelagic fossil gastropod, Hartungia dennanti chavani, considered to be an Early Pleistocene marker (Ludbrook 1983), but which may also extend back into the Late Pliocene (James et al. 2006).

Seaford Formation

Figure 2 Pleistocene stratigraphy exposed in coastal cliff at the Sellicks trig site resting on the limestone of the Port Willunga Formation (PWF). Burnham Limestone (BL) interfingers with the Seaford Formation (SF) at its base and is successively overlain by Ochre Cove Formation (OCF), Ngaltinga Formation (NF), possible Taringa Formation (TF) and Pooraka Formation (PF). Thin layers of dolomite (D) occur in the lower part of the section, while BT marks the location of beds back-tilted during land slumping, and B indicates the location of a bench formed during the slumping. Data from Ward (1966) and May and Bourman (1984). (Photo 418476)

The oldest of the Pleistocene alluvial successions, the Seaford Formation as mapped by Ward (1965, 1966), occupies 22 m of a 50 m thick coastal section adjacent to the former Sellicks trigonometrical station at latitude 35°20’21.07” S, longitude 138°26’47.33” E (Fig 2) May and Bourman (1984) described it as consisting of sandy clays and gravel beds ranging in colour from brown, grey-black, and light olive-green to grey, with horizons of cobble-sized clasts. It also contains weak yellow-orange goethitic mottles and layers. Ward (1966) considered that the Seaford Formation is of Pliocene age because he thought that it inter-digitated with the Pliocene Hallett Cove Sandstone at the base of the succession at the Sellicks trig site.

The Seaford Formation does sit directly on the Hallett Cove Sandstone at Ochre Point, the site of the type section (Ward 1966), and at some other localities where the Burnham Limestone is absent. However, May and Bourman (1984) reported the occurrence of Hartungia dennanti chavani at the base of the Sellicks trig section demonstrating that the majority of the Seaford Formation is of an Early Pleistocene age. Earlier, Stuart (1969) had identified fossiliferous sediments at Maslin Bay, which grade into the Seaford Formation, and which were subsequently assigned to the Burnham Limestone. Twidale et al. (1967) had also identified a Pleistocene limestone between the Hallett Cove Sandstone and the Seaford Formation at the same locality. Using magnetostratigraphy, Pillans and Bourman (1996) established that the Seaford Formation occurs within the Matuyama Reversed Chron between 1.07 Ma and the base of the Pleistocene, currently considered to be at c. 2.6 Ma (Pillans and Gibbard 2012).

Given this information, we believe that the unit mapped as the ‘Pirramimma Sandstone’ (Fairburn 1998; Fairburn et al. 2010; Aldam et al. 2019) is actually the Early Pleistocene Seaford Formation, while the Pirramimma Sand is a much older unit of Late Oligocene age (Barnett and Bourman 2022).

Silicified quartzose breccia of the Ochre Cove Formation

Figure 3 Exposure of the silicified breccia of the Kurrajong Formation in the base of Sellicks Creek downstream of Main South Road. It is traceable into the overlying Ochre Cove Formation without a stratigraphic break. Width of field ~4 m. The silicified breccia also extends for some distance upstream of Main South Road where it appears to have been tilted during ongoing tectonic uplift. (Photo 418477)

Near the foot of the Willunga Escarpment, Sellicks Creek has exposed a silicified, resistant breccia up to several metres thick, originally mapped as the Kurrajong Formation (Fig 3) by Ward (1966), who regarded it as being younger than the Ochre Cove Formation. However, it can be traced vertically without break into the overlying more weakly consolidated and strongly mottled Ochre Cove Formation. Pillans and Bourman (1996), noting the transition from the Kurrajong Formation of Ward (1966) into the less consolidated sediments of the Ochre Cove Formation without a stratigraphical break, suggested that the 2 formations might be regarded as facies variants of the same unit, with variations in lithology and induration being influenced by proximity to the foot of the fault escarpment.

Ochre Cove Formation

The Ochre Cove Formation consists of grey coloured alluvial clays, sands, angular gravel clasts and conglomerates with prominent dark red to purple hematitic mottles, occasionally containing maghemite (Wopfner 1972). In the cliffs adjacent to the Sellicks trig site the unit is ~17 m thick, conformably overlying the Seaford Formation. Magnetostratigraphy carried out on the hematitic mottles demonstrates that weathering of the Ochre Cove Formation occurred during the mid-Pleistocene, as both the Brunhes–Matuyama reversal (c. 0.781 Ma) and the Jaramillo subchron (between 1.07 and 0.99 Ma) are preserved within the succession. This suggests sedimentation and mottling developed over approximately 200,000 years (Bourman and Pillans 1997), with sedimentation and weathering proceeding upwards at an average rate of 0.03 m ka^–1. This illustrates well the relationships of the sediment age and the timing of the weathering events based on magnetic remanence; the weathering imprints are younger than the sediments themselves. In a different setting involving the Pleistocene aeolianites of the Naracoorte Range of the Coorong Coastal Plain, Idnurm and Cook (1980) suggested a time lag of 30 ka between calcareous dune formation and the acquisition of chemical remanence.

Figure 4 Cross-section of a coastal landslump in Pleistocene sediments immediately north of the Sellicks trig site revealed by disruption of strata. Modified from Bourman and May (1984) and Cann et al. (2014).

The above paleomagnetic data disprove the contention of Fairburn (1998) that the Seaford and Ochre Cove formations in the Sellicks Beach sections as mapped by Ward (1966) and May (1992) do not exist. Furthermore, the Quaternary stratigraphy, including the Seaford and Ochre Cove formations, was used to demonstrate major coastal rotational land slumping in the Sellicks Beach area (May and Bourman 1984; Bourman and May 1984; Fig 4).

Ngaltinga Formation

The Ngaltinga Formation was originally named the Ngaltinga Clay by Ward (1966), who described the unit as comprising calcareous grey to olive-grey fine plastic clays, with widely disseminated fine quartz sand and sandy clays, including poorly defined lenses of argillaceous sand and white marl. There were also occasional lenses of small gravel and alunite. Marls occur in the upper parts of the unit, with the top metre or so consisting entirely of marl, which grades into the underlying clay without lithological discontinuity. The Ngaltinga Formation also contains small red, ferruginous (hematitic) mottles suitable for paleomagnetic investigation.

Ward (1966, pp 39–41) considered the Ngaltinga Formation to be essentially an aeolian succession as it thins out in an easterly inland direction, with thick deposits at the coast thinning to its depositional edge some 10 km inland, while ranging in elevation from sea level up to 200 m AHD inland. The formation, which contains micro-fauna derived from the Blanche Point Formation (Ludbrook, in Ward 1966), occurs above the non-calcareous Ochre Cove and Seaford formations and older rocks. Furthermore, heavy minerals in similar sediments of the Adelaide area (Keswick Clay) were deduced to have been introduced as wind-blown dust from an external source at the time of Ngaltinga Formation deposition (Aitchison et al. 1954).

Magnetostratigraphic investigations at Hallett Cove, Sellicks Beach and Redbanks on Kangaroo Island (Pillans and Bourman 1996, 2001) identified the Brunhes–Matuyama boundary in all 3 sites at 781 ka in the Ochre Cove Formation. The upper part of the Ochre Cove Formation and all of the Ngaltinga Formation carry the normal polarity of the Brunhes normal chron. The Ochre Cove Formation is overlain by the Ngaltinga Formation, a calcareous, green-grey clay, which in turn is overlain by calcareous sediments of the Taringa and Pooraka formations. This reveals a marked shift from an oxide-dominated to a carbonate-dominated weathering regime at c. 500–600 ka related to a major arid shift in the regional climate and accessions of aeolian dust, particularly during the lower sea levels of glacial periods (Pillans and Bourman 2001). These events correspond with the latter portion of the Early–Middle Pleistocene transition, a period in which the low-amplitude, obliquity influenced (41 ka) glacial cycles ceased to dominate long-term climate change, while the stratigraphical record reveals greater influence of 100 ka cycles, characterised by larger ice volumes at times of peak glaciations (Head et al. 2008).

Trough cross-bedding and gravel lenses in the largely fine-grained sediments have suggested a fluvial origin for the Ngaltinga Formation to some workers (Phillips and Milnes 1988; May 1992). However, Pillans and Bourman (2001) noted that in their 3 study sites the fine-grained clays, with a lack of sedimentary structures and a sharp lower contact with the sandy and gravelly sediments of the Ochre Cove Formation, are all consistent with an aeolian origin. This view is supported by the observation that the formation mantles the landscape in a manner essentially independent of topography and geology. Nevertheless, during accumulation of the aeolian dust the landscape may still have been affected by running water, sporadically introducing coarser particles as well as reworking the aeolian clays.

Taringa Formation

The Taringa Formation (Ward 1966) is a grey and yellowish grey massive alluvial and calcareous clay unit containing lenses of gravelly clay, gravel, sand and marl. It includes dune sands at Aldinga Beach. In places it has a columnar structure and commonly contains angular clasts, suggesting that in places it may be a mudflow deposit.

There are errors in the stratigraphic revision of the Willunga Basin by Fairburn (1998): the failure to distinguish between the Taringa Formation and the Ngaltinga Formation as discrete units in the Sellicks Beach area is one of them. He proposed that the grey clay (Taringa Formation of Ward 1966) underlying the Christies Beach Formation (equivalent of the Pooraka Formation) in Sellicks Creek and other gullies emanating from the Willunga Fault Escarpment does not exist but actually comprises sediments of the Ngaltinga Formation. Consequently, he excised the Taringa Formation from the Pleistocene stratigraphy.

However, the 2 formations are distinctively different. The Ngaltinga Formation consists largely of grey-green stiff plastic self-mulching clays, with small red hematitic mottles grading up into calcareous clays. It is largely an aeolian deposit directly overlying the Middle Pleistocene Ochre Cove Formation, the age of which has been established by magnetostratigraphy (Pillans and Bourman 2001). There are some sandy, fluvial channels in the Ngaltinga Formation (May 1992; Phillips and Milnes 1988), but the Ngaltinga Formation is dominantly of aeolian origin, the deposition of which began about 500 ka ago (Pillans and Bourman 2001).

In contrast, the Taringa Formation contains no strong mottling, possesses a columnar structure and is possibly a mass movement deposit often containing angular rock fragments. It consistently underlies the last interglacial Pooraka Formation in Dry Creek, in the lower Onkaparinga River, in Sellicks Creek and at Tunkalilla Beach. Its age has been established using OSL techniques in 2 separate locations. In Dry Creek, where it immediately underlies the Pooraka Formation, an age of 164 ± 17 ka was determined (Bourman et al. 2010), while near the outlet of Sellicks Creek, again where it also underlies the Pooraka Formation, an age of 160 ± 15 ka was derived (Bourman et al. 2020). Both of these ages fall within the lattermost portion of the penultimate glacial cycle within MIS 6, a period of low sea level. During this glacial stage, periglacial conditions may have prevailed in the study area, with freezing and thawing along with solifluction processes resulting in the deposition of angular fragments of rock in mudflow deposits. Thus, the different physical characteristics of the Ngaltinga and the Taringa formations, together with their numerical ages, demonstrate the existence of both formations as separate entities as originally determined by Ward (1966).

Kurrajong Formation

The type section of the Kurrajong Formation as described by Ward (1966, p. 35) exceeds 14 m in thickness and comprises ‘compact fanglomerates, alluvial gravels, sandstones and clays forming high terraces and fans at the foot of the Willunga and Clarendon scarps’. Ward (1966) also noted that much of the Kurrajong Formation is poorly exposed, being mantled by a thick layer of residual soil. Ward (1966) placed the Kurrajong Formation in the early part of the last interglacial, while Fairburn (1998) mapped the Kurrajong in the Pleistocene, immediately below the Christies Beach Formation and above the Taringa Formation.

It appears that at least 2 alluvial–colluvial successions and associated soils have been incorporated within the Kurrajong Formation. For this reason, the lithified breccia in Sellicks Creek (Fig 3) is considered to be a facies variant of the Ochre Cove Formation, while the fanglomerates and conglomerates are regarded as comprising a younger unit, encompassing non-lithified alluvium and colluvium, mantled by modern soils. We suggest that the name ‘Kurrajong Formation’ should be reserved for this younger unit. No numeric ages are available for this formation, but its character and stratigraphic relationships suggest a time in the latter part of the penultimate glacial (MIS 6), where we tentatively place it.

Pooraka Formation

The Pooraka Formation, equivalent of the Christies Beach Formation of Ward (1966), is a distinctive red-brown alluvial succession of sandy clays, gravel lenses and beds, blanketed with a red-brown earth soil profile containing nodules and cylindroids of calcium carbonate in the B-horizon. It overlies the Taringa Formation and is widespread though the Willunga and Noarlunga embayments, typically forming alluvial fans and high river terraces along major streams such as the Onkaparinga River (Bourman 1969, 2006).

Figure 5 Layer cake arrangement of Pleistocene sediments in the lower reaches of Sellicks Creek showing the last interglacial (MIS 5a) Pooraka Formation at the base, a MIS 3 subpluvial unit sandwiched between 2 paleosols, and a last glacial alluvial deposit (MIS 2) overlain by the grey-black Waldeila Formation and a gravelly PESA unit. The paleosols indicate phases of subaerial exposure and landscape stability. The circles mark OSL sample locations (SK-2, 3, 4). This gully has now been filled and landscaped. (Photo 418478)

Using OSL dating the Pooraka Formation has been demonstrated to span the last interglacial (MIS 5) from Sub-Stage MIS 5e (132–118 ka) to the Interstadial MIS 5a (c. 80 ka). The variable age of the Pooraka Formation appears to relate to different geomorphic settings. In major river valleys such as the River Torrens, Dry Creek, Inman, Hindmarsh and Onkaparinga rivers, and the Burra Creek (Bourman et al. 2010) the Pooraka Formation was deposited during the last interglacial maximum (MIS 5e). On the other hand, in subsiding coastal areas (Ludbrook 1976; Belperio and Rice 1989), in smaller drainage basins and on alluvial fans at the foot of actively uplifting zones such as in Sellicks Creek (Bourman et al. 2020), the deposition of the Pooraka Formation continued until the end of the last interglacial at c. 80 ka (MIS 5e; Fig 5). The alluvial clays of the raised terraces of the Onkaparinga River as shown on Fairburn et al. (2010) are actually sediments of the Pooraka Formation (Bourman 2006). The age attributed to the Pooraka Formation by Ward (1966) extended from the peak to the end of the last interglacial, which, given the lack of dating techniques then available, was remarkably prescient.

Unnamed subpluvial alluvial fill unit (MIS 3)

An unnamed (MIS 3) subpluvial alluvial fill unit (following Nanson et al. 1992), overlying the Pooraka Formation and sandwiched between 2 red-brown earth paleosols in the lower reaches of Sellicks Creek was dated at 42 ± 3.2 ka (MIS 3) by OSL (Bourman et al. 2020; Fig 5). This replicated the age of a similar unit in the Burra Creek (43 ± 3 ka; Bourman et al. 2010), which also overlies the Pooraka Formation, and which contained skeletal remains of the extinct Diprotodon at its base. The subpluvial unit (MIS 3) resembles the older Pooraka Formation in both settings.

Late last glacial succession (MIS 2)

Sediments of late last glacial age (14 ± 1.2 ka, MIS 2) occur as colluvium in colluvium-filled bedrock depressions (CBD deposits) in the upper reaches of Sellicks Creek and as alluvium in the lower section of the creek (15.7 ± 1.7 ka, MIS 2) overlying a paleosol developed on the subpluvial MIS 3 unit (Bourman et al. 2020; Fig 5). During this time sea level was low and as with the MIS 6 penultimate glacial cycle, periglacial conditions involving freezing and thawing probably prevailed, especially in the upper reaches of Sellicks Creek, producing fragmented rocks and mass movements into valley bottoms, resulting in the CBD deposits. Alluvial sediments of the lattermost last glacial maximum (MIS 2) also overlie a paleosol on the MIS 3 subpluvial unit in Sellicks Creek.

Ngankipari Sand

The Ngankipari Sand, which includes weakly consolidated calcareous drift sands locally overlain by mid-Holocene Waldeila Formation alluvium, was originally mapped and described by Ward (1965, 1966), who considered that these sands were deposited about 10,000 years ago as they contain Aboriginal relics and rest on Kartan hearths (Tindale 1957). Last glacial dunes are widespread across South Australia and it is possible that the Ngankipari Sand was formed by the reactivation of pre-existing sands, such as the North Maslin Sand, during the colder, drier and windier conditions of the latter part of the last glacial maximum (MIS2). In this paper the Ngankipari Sand is tentatively ascribed to the latter part of the last glacial maximum (MIS 2).

Fairburn (1998), noting common occurrences of thin unconsolidated sand deposits in the Willunga Embayment, particularly on the crests of hills, considered them to be equivalents of the Ngankipari Sand of Ward (1966). However, the name Semaphore Sand, originally used for present-day coastal dunes (Firman 1966), was extended by Fairburn et al. (2010; Semaphore Sand Member) to include these inland sand sheets. In a subsequent revised map (Aldam et al. 2019), these sands were described as ‘Unconsolidated sand dune, drapes and spreads’ of Holocene age, and ‘Semaphore Sand’ was returned to its rightful place at the coast.

Waldeila Formation

The Waldeila Formation of Ward (1966), which is is a grey-black sandy clay with occasional layers and lenses of gravels, is of mid-Holocene age. It forms low terraces and floodplains in valleys cut into the Pooraka Formation and older units, as along Sellicks Creek. Where it spills out of the valleys it forms levees. Although Fairburn (1998) did not discuss or map the Waldeila Formation, the area mapped as ‘alluvial clay of valleys’ by Fairburn et al. (2010) along the lower Onkaparinga River is one site of the Waldeila Formation. Here, estuarine shells interbedded within the Waldeila Formation at 1 m AHD and 2 km directly inland from the coast were dated at 4,580 ± 160 years BP by radiocarbon techniques (Bourman 2006). Nearby, estuarine shells from an Aboriginal midden site are of a similar, mid-Holocene age of 5820 ± 90 years BP (Tindale 1957). OSL techniques provided an age of 3.5 ± 0.3 ka (MIS 1) from the upper portion of the Waldeila Formation in Sellicks Creek near the hill–plain junction (Bourman et al. 2010) at a depth of 2 m below the surface.

At this locality, the Waldeila Formation, which is more than 6 m thick, contains a pronounced charcoal layer accompanied by thin charcoal bands throughout the section, possibly representing Aboriginal firestick farming practices during the Holocene. Similar Holocene age sediments, with distinct charcoal-rich layers are found deposited in elevated valley-fill peat bogs in the Mount Lofty Ranges (Buckman et al. 2009). These were dated by OSL and radiocarbon dating techniques and were used to report a detailed fire history for the Cleland Valley at Mount Lofty in which there were at least 15 major fires in the past 7,000 years. A sharp transition from siliciclastic sedimentation to peat formation at c. 1200 cal. years BP suggests a return to wetter conditions but were accompanied by an increase in fire frequency, possibly resulting from increased anthropogenic burning or ENSO (El Niño – Southern Oscillation) related factors.

Semaphore Sand

In this paper, the label ‘Semaphore Sand’ has been restricted to the modern coastal dunes as first established by Firman (1966). Fairburn (1998) noted common occurrences of thin unconsolidated sand deposits in the Willunga Embayment, particularly on the crests of hills, which he considered to be equivalent to the Ngankipari Sand of Ward (1966). Subsequently, on the Fairburn et al. (2010) map there seemed to be some confusion about the Semaphore Sand, which referred originally to the well-developed dunes of the present coast (Firman 1966), but which Fairburn et al. (2010) extended well inland as a ‘Semaphore Sand Member’. In a subsequent revised map (Aldam 2019) this appears to have been replaced by a new unit titled ‘Unconsolidated sand dune, drapes and spreads’, some of which may relate to the Ngankipari Sand.

Post-European sediment aggradation

A phase of accelerated erosion and sedimentation was generated with the arrival of European settlers, who cleared, overgrazed and tilled the land inappropriately. In some places in the Willunga and Noarlunga embayments, erosion of soils and sediments from high country ensued with associated deposition both in stream channels and mantling of the upper surfaces of some alluvial fans. These PESA sediments contain European artefacts such as bottles, crockery, plastics, metal chains and other metal objects, while burying fence posts and pre-existing soils.

Table 1 Quaternary stratigraphy correlated with marine isotope stages in the Willunga and Noarlunga embayments

Conclusion

The recognition of the Pirramimma Sand as a discrete stratigraphic unit, which together with the numerical dating of alluvial and colluvial sequences along Sellicks Creek using OSL techniques, and the application of magnetostratigraphy in coastal sections prompted this revision of the Pleistocene stratigraphy of the Willunga and Noarlunga embayments. The use of OSL, in particular, enabled correlation between different fluvial systems as well as facilitating correlation with global marine isotope stages. Two new units, an ‘unnamed subpluvial alluvial fill unit (MIS 3)’ and a ‘late last glacial succession (MIS 2)’ were recognised using OSL. Furthermore, the OSL results revealed the Taringa Formation as a separate penultimate glacial age (MIS 6) deposit and not part of the Ngaltinga Formation as claimed by Fairburn (1998). The paleomagnetic data establish the existence of the Seaford and Ochre Cove formations and their Pleistocene ages in the coastal section at the Sellicks trig site, invalidating the contention of Fairburn (1998) that they were absent there.

What is quite remarkable is that much of the basic stratigraphic interpretation of Ward (1965, 1966), established more than 5 decades ago without the benefit of modern dating techniques, is still valid. He did not identify the Burnham Limestone, but that unit was not formally recognised until later (Firman 1976). In addition, his positioning of the Kurrajong Formation may be misplaced, but otherwise his general stratigraphy appears to be sound. Further clarification of the stratigraphic successions will continue in the future as existing techniques such as OSL are applied to determining the ages of units such as the Ngankipari Sand, while new discoveries and the application of techniques as yet unknown will bring further refinements.

Acknowledgements

Wolfgang Preiss (Geological Survey of South Australia, retired) and an anonymous reviewer are thanked for constructive feedback which improved this paper.

References

Aitchison ME, Sprigg RC and Cochrane GW 1954. The soils and geology of Adelaide and suburbs, Bulletin 32. Geological Survey of South Australia, Adelaide.

Aldam RW, Fairburn WA, Olliver JG and Preiss WV 2019. Geology of the McLaren Vale Wine Region, Brochure 126. Prepared for the McLaren Vale Grape Wine & Tourism Association. Department for Energy and Mining, South Australia, Adelaide.

Barnett SR and Bourman RP 2022. Some revisions of the hydrostratigraphy and structural geology of the Willunga Embayment. MESA Journal 96:22–26. Department for Energy and Mining, South Australia, Adelaide.

Belperio AP and Rice R L 1989. Stratigraphic investigations of the Gillman development site, Port Adelaide estuary, Report Book 89/00062. South Australia Department of Mines and Energy, Adelaide.

Bourman RP 1969. Some aspects of the geomorphology of the Onkaparinga River. Taminga 2:117–143. The Journal of the Geography Association of South Australia.

Bourman RP 2006. River terraces of the Fleurieu Peninsula. South Australian Geographical Journal 105:1–24. Royal Geographical Society of South Australia.

Bourman RP, Banerjee D, Murray-Wallace CV, Buckman S, Panda DK, Belperio AP and Jayawardena CL 2020. Luminescence dating of Quaternary alluvial successions, Sellicks Creek, south Mount Lofty Ranges, southern Australia. Australian Journal of Earth Sciences 67(5):627–647. doi:10.1080/08120099.2020.1722967.

Bourman RP and May RM 1984. Coastal rotational landslump. Australian Geographer 16:144–146. doi:10.1080/00049188408702867.

Bourman RP and Pillans B 1997. The use of the geomagnetic polarity time scale in elucidating Pleistocene landscape evolution: an example from South Australia. South African Geographical Journal 79(2):75–82. doi:10.1080/03736245.1997.9713627.

Bourman RP, Prescott JR, Banerjee D, Alley NF and Buckman S 2010. Age and origin of alluvial sediments within and flanking the Mount Lofty Ranges, southern South Australia – a Late Quaternary archive of climate and environmental change. Australian Journal of Earth Sciences 57:175–192. doi:10.1080/08120090903416260.

Buckman S, Brownlie KC, Bourman RP, Murray-Wallace CV, Morris RH, Lachlan TJ and Cann JH 2009. Holocene palaeofire records in a high-level, proximal valley-fill (Wilson Bog), Mount Lofty Ranges, South Australia. The Holocene 19(7):1017–1029. doi:10.1177/0959683609340998.

Cann JH, Lower CS and Jago JB 2014. Provenance and sediment characteristics of contemporary gravel deposits at Sellicks Beach, eastern shore of Gulf St Vincent, South Australia, Australian Journal of Earth Sciences 61(6):819–836. doi:10.1080/08120099.2014.941400.

Cooper BJ 1977. New and revised stratigraphic nomenclature for the Willunga Embayment. Quarterly Geological Notes 64:2–5. Geological Survey of South Australia, Adelaide.

Cooper BJ 1979. Eocene to Miocene stratigraphy of the Willunga Embayment, Report of Investigations 50. Geological Survey of South Australia, Adelaide.

Fairburn W 1998. The Willunga Embayment — a stratigraphic revision. MESA Journal 11:35–41. Primary Industries and Resources South Australia, Adelaide.

Firman JB 1966. Stratigraphic units of Late Cainozoic age in St Vincent Basin, South Australia. Quarterly Geological Notes 17:6–9. Geological Survey of South Australia, Adelaide.

Firman JB 1967. Stratigraphy of Late Cainozoic deposits age in South Australia. Transactions Royal Society of South Australia 91:165–178.

Firman JB 1976. Limestone at the base of the Pleistocene sequence in South Australia. Quarterly Geological Notes 58:2–5. Geological Survey of South Australia, Adelaide.

Head MJ, Pillans B and Farquhar SA 2008. The Early–Middle Pleistocene Transition: characterization and proposed guide for the defining boundary. Episodes 31(2):255–259. doi:10.18814/epiiugs/2008/v31i2/014. (Open access)

Idnurm M and Cook PJ 1980. Palaeomagnetism of beach ridges in South Australia and the Milankovitch theory of ice ages. Nature 286:699–702. doi:10.1038/286699a0.

James NP, Bone Y, Carter RM and Murray-Wallace CV 2006. Origin of the Late Neogene Roe Plains and their calcarenite veneer: implications for sedimentology and tectonics in the Great Australian Bight. Australian Journal of Earth Sciences 53: 407–419. doi:10.1080/08120090500499289.

Lindsay JM 1988. Preliminary report of Late Oligocene foraminiferal age of fossiliferous sandstone at Gawler, Report Book 88/00031. South Australia Department of Mines and Energy, Adelaide.

Ludbrook NH 1976. The Glanville Formation at Port Adelaide. Quarterly Geological Notes 57:4–7. Geological Survey of South Australia, Adelaide.

Ludbrook NH 1983. Molluscan faunas on the Early Pleistocene Point Ellen Formation and Burnham Limestone, South Australia. Transactions Royal Society of South Australia 107:37–50.

McGowran B, Lemon N, Preiss W and Olliver J 2016. Geological Field Excursion Guide — Cenozoic Willunga Embayment: from Australo-Antarctic Gulf to Sprigg Orogeny, Report Book 2016/00008. Department of State Development, South Australia, Adelaide; and Geological Society of Australia, South Australian Division.

May RI and Bourman RP 1984. Coastal landslumping in Pleistocene sediments at Sellicks Beach, South Australia. Transactions Royal Society of South Australia 108:85–94.

May RI 1992. Origin, mineralogy and diagenesis of Quaternary sediments from the Noarlunga and Willunga embayments, South Australia. Unpublished PhD thesis, Department of Soil Science, Waite Agricultural Research Institute, and the University of Adelaide.

Nanson GC, Price DM and Short SA 1992. Wetting and drying of Australia over the past 300 ka. Geology 20:791–794. doi:10.1130/0091-7613(1992)020<0791:WADOAO>2.3.CO;2.

Phillips SE and Milnes AR 1988. The Pleistocene terrestrial carbonate mantle on the southeastern margin of the St Vincent Basin, South Australia. Australian Journal of Earth Sciences 35:463–481. doi:10.1080/08120098808729463.

Pillans B and Bourman RP 1996. The Brunhes/Matuyama Polarity transition (0.78 Ma) as a chronostratigraphic marker in Australian regolith studies. AGSO Journal of Geology & Geophysics 16(3):289–294.

Pillans B and Bourman RP 2001. Mid Pleistocene arid shift in southern Australia, dated by magnetostratigraphy. Australian Journal of Soil Research 39:89–98. doi.org/10.1071/SR99089.

Pillans B and Gibbard P 2012. The Quaternary period. In Gradstein FM, Ogg JG, Schmitz MD and Ogg GM eds, The Geologic Time Scale 2012. Elsevier, ch 30, pp 979–1010. doi.org/10.1016/B978-0-444-59425-9.00030-5.

Sheard MJ and Bowman GM 1994. Soils, stratigraphy and engineering geology of near surface materials of the Adelaide Plains, Report Book 94/00009. Department of Mines and Energy, South Australia, Adelaide.

Stuart WJ 1969. Stratigraphic and structural development of the St Vincent Tertiary Basin, South Australia. Unpublished PhD thesis, the University of Adelaide.

Talbot JL and Nesbitt RW 1968. Geological excursions in the Mount Lofty Ranges and the Fleurieu Peninsula. Angus and Robertson, Adelaide.

Tindale NB 1957. Culture succession in south-eastern Australia from Late Pleistocene to the present. Records of the South Australian Museum 13:1–49.

Twidale CR, Daily B and Firman JB 1967. Eustatic and climatic history of the Adelaide area, South Australia – a discussion. Journal of Geology 75:237–242.

Ward WT 1965. Eustatic and climatic history of the Adelaide area, South Australia. Journal of Geology 73:592–602.

Ward WT 1966. Geology, geomorphology, and soils of the south-western part of County Adelaide, South Australia, Soil Publication 23. CSIRO.

Wopfner H 1972. Maghemite from Cainozoic sediments at Hallett Cove. Quarterly Geological Notes 43:5–8. Geological Survey of South Australia, Adelaide.

Back to top