Category Archives: Landscape evolution

Surficial albitization – palaeoweathering preserved beneath the extensive Triassic unconformity in Western Europe

One form of albitization consists of the replacement (albitization) of Ca-bearing plagioclase by secondary Na-plagioclase (i.e. albite) and is typically accompanied by alteration of biotite to chlorite and the staining of the rock by iron oxide pigments.  In crystalline rocks, albitization is rather widespread but is traditionally considered result from deep and high-temperature metasomatic alteration linked to magma cooling during exhumation.  Studies of albitization have principally been undertaken by petrologists who are mostly interested in the origin and emplacement of granites and naturally focus on aspects of their petrography and the details of their parageneses in this context.  They interpret albitization to be a ‘window’ to post magmatic processes. 

But albitization can also be a consequence of surficial weathering.  The example we studied in Spain (Fig. 1) is widespread and spatially and temporally associated with a Triassic palaeosurface.  Its occurrence forces a completely different point of view about this form of alteration.  A major gap separates the two points of view about albitization and our findings have exposed a ‘cultural barrier’ that has to be breached in order to advance knowledge about this process, particularly where it is found in crystalline rocks.  Key foci in the scientific exchange ought to include the geometry of the alteration in the field, petrographic observations detailing stages in the alteration, and the dating of these stages.

Figure 1. Field locations of albitized granites below the Triassic unconformity in Spain. (a) Guilleries Massif. (b) Roc de Frausa Massif. CCR: Catalan Coastal Ranges. From Fabrega et al. (2019).
Figure 1. Field locations of albitized granites below the Triassic unconformity in Spain. (a) Guilleries Massif. (b) Roc de Frausa Massif. CCR: Catalan Coastal Ranges. From Fabrega et al. (2019).

The outcrop scale

The albitization we have studied in granites is characterised by reddening due to impregnation of the feldspars by iron oxides (Fig. 2).  It results in: (1) a pervasive coloration distributed rather homogeneously within the rock volume, (2) fracture-bound red to pinkish granite facies or (3) spotting when the alteration is weak, resulting in pinkish spots distributed within non-reddened rock.

Figure 2. Strongly albitized Variscan red granites below the Triassic unconformity in the Guilleries Massif, NE Spain.
Figure 2. Strongly albitized Variscan red granites below the Triassic unconformity in the Guilleries Massif, NE Spain.

In terms of structure, a vertical profile of albitized granite, with a typical thickness around 150-200 m, occurs below the Triassic unconformity in the Guilleries and Roc de Frausa Massifs in NE Spain (Fig. 3).  The most intense alteration occurs at the top, immediately below the unconformity, and progressively decreases in intensity with depth.  From the top to the bottom, the alteration facies change from red to pink and ultimately spotted.  Unaltered granites are found below (Fig. 3).

Figure 3. Vertical structure and facies evolution of the albitization profile. Intensity of alteration decreases with depth. (a) Mesozoic sedimentary cover. (b) Red facies. (c) Pink facies. (d) Spotted facies. After Parcerisa et al. (2010).

The microscope scale

Microscope observations of the albitized granites show a suite of reactions in the major minerals including plagioclase albitization, K-feldspar microclinization, biotite chloritization and quartz recrystallization.  Albitization affects the primary plagioclase, where the Ca-bearing cores of the crystals are pseudomorphosed by secondary albite along twin planes and micro-fractures (Fig. 4a, b).  Within this secondary albite, part of the calcium released during plagioclase replacement forms small patches of secondary calcite.  Primary K-feldspar crystals are partially pseudomorphosed by secondary microcline (Fig. 4a).  Biotite and is replaced by secondary chlorite (which contains large numbers of apatite inclusions compared with primary biotite; Fig. 4c), secondary titanite and anatase.  Clusters of rare-earth carbonate minerals, probably synchisite-Ce, can be seen inside the secondary porosity of albitized plagioclase and microclinized orthoclase in backscattered electron-microscope images (Fig. 4d).  Secondary quartz partially replaces primary quartz in the upper part of the profile (Fig. 4e).  Primary and secondary monazite occur at the reaction front between albitized and unaltered granite (Fig. 4f).

Figure 4. Cathodoluminescence and SEM images of albitized granite. Ab1: Primary plagioclase. Ab2: Secondary albite. Ap: Apatite. Cal2: Secondary calcite. Chl2: Secondary chlorite. Kfs1: Primary K-feldspar. Kfs2: Microclinized K-feldspar. Mnz1: Primary monazite. Mnz2: Secondary monazite. Qtz1: Primary quartz. Qtz2: Secondary quartz. Syn2: Secondary Synchysite-(Ce). Tit2: Secondary titanite. From Fabrega et al. (2019).

Age dating

Dating of albitized profiles has been conducted at different sites through Europe using a variety of dating methods, including: (1) K-Ar dating of albite and K-feldspar (Bonhomme et al., 1980; Schmitt et al., 1984); (2) palaeomagnetism of Fe-oxides (Ricordel et al., 2007; Parcerisa et al., 2010; Thiry et al., 2011; Yao et al. 2011, Franke et al., 2011, Yao, 2013) and, most recently, (3) K-Ar microprobe dating of albite and K-feldspar and electron microprobe U-Th-Pb dating of monazite (Fabrega et al., 2019).  All have established that the albitization and associated alteration occurred during the Late Permian to Early Triassic, around 250 Ma, while the Variscan granites were exposed at the landsurface. 

Thus, the red-stained albitization relates specifically to the Triassic palaeosurface and was preserved beneath it by a Mesozoic and Cenozoic sediment cover.  The alteration profiles are palaeoweathering features relating to surficial weathering and were nor much eroded.  The  geopetal organization (decreasing intensity of albitization with depth ending in alteration confined to fractures) is characteristic of weathering and alteration under the influence of geomorphology, climate, and groundwater.

Palaeoenvironmental setting of the alteration

The morphology of the alteration profile and its proximity to the Permo-Triassic palaeosurface suggests that the responsible fluids were Na-rich groundwaters (Fig. 5a).  The depth of the alteration profile, the long-lasting tectonic stability in the region, and the continental scale abundance of saline environments in the Permo-Triassic period were probably key factors in the development of this near-surface weathering event across Europe.  The albitization profiles remained preserved below the Triassic sedimentary cover or, in some cases (Fig. 5b), directly at the surface.  In contrast to the non-albitized granites usually decompose to sandy material when exposed to surface conditions, the albitized granite has an increased resistance to weathering and erosion.

Figure 5. Permo-Triassic setting that facilitated the development (a) and preservation (b) of albitization. From Fabrega et al. (2019.

Extent of the Late Permian to Early Triassic palaeoweathering

Thus far, this form of palaeoweathering has been observed in Morocco, the Catalan Coastal Ranges, Eastern Pyrenees, Galicia and La Mancha in Spain; Corsica, the Maures, French Central Massif (Fig. 6), Morvan, the Alps, the Vosges and Brittany in France; in Germany; Great-Britain and Ireland; Sudetes in Poland; and in the Oslo Fiord in Norway.


We have demonstrated that the albitized facies are of surficial origin and bound to the weathering (geochemical) environment linked to the Triassic palaeosurface.  This dramatically changes ideas about the geomorphic evolution of the basement areas over a large part of western Europe and is a major contribution to the geodynamic modeling of the region.Indeed, reconstruction of palaeosurfaces is a unique tool in unravelling the evolution of ancient continents.  Extending age data from the better-known sedimentary basin deposits to adjacent crystalline basement landscapes (particularly of palaeoweathering features) provides the possibility to set the correct framework for geodynamical models (Fig. 6).

Figure 6. Occurrence of dated palaeoweathering profiles on the French Massif Central crystalline basement. After Thiry et al. (2012).


  • Bonhomme, M., Yerle, J.-J., Thiry, M., 1980. Datation K-Ar de fractions fines associées aux minéralisations. Le cas du bassin uranifère permo-houiller de Brousse-Broquiès (Aveyron, France).- C.R. Acad. Sc., Paris, 291, sér. D, p. 121-123.
  • Fàbrega, C., Parcerisa, D., Thiry, M., Franke, M., Gurenko, A., Gòmez-Gras, D., Solé, J., Travé, A., 2019. Permian–Triassic red‑stained albitized profles in the granitic basement of NE Spain: evidence for deep alteration related to the Triassic palaeosurface. International Journal of Earth Sciences,
  • Franke, C., Yao, K.F.E., Thiry, M., Gomez-Gras, D., Ihlen, P., Kadzialko-Hofmokl, M., Jelenska, M., Parcerisa, D., Fabrega, C., Lagroix, F., Turniak, C., Szuskiewicez, A., 2011. Réstitution de la paléosurface triasique par datation des ré-aimantations inscrites dans les massifs paléozoïques européens. 13ème Congrès français de Sédimentologie, Dijon, Publ. ASF, Paris, 68, p. 137.
  • Parcerisa, D., Casas, L., Franke, F., Gómez-Gras, D, Lacasa, G., Nunez, J.A., Thiry, M., 2010. Geomorphological stability of Permo-Triasic albitized profiles – Case study of the Montseny-Guilleries High (NE Iberia). Geophysical Research Abstracts, 12, EGU2010, EGU General Assembly 2010, 2 p.
  • Parcerisa D, Thiry M, Schmitt JM (2010b) Albitization related to the Triassic unconformity in igneous rocks of the Morvan Massif (France). Int J Earth Sci 99:527–544. doi: 10.1007/s00531-008-0405-1.
  • Ricordel, C., Parcerisa, D., Thiry, M., Moreau, M.‑G., Gómez-Gras, D., 2007. Triassic magnetic overprints related to albitization in granites from the Morvan massif (France). Palaeogeography, Palaeoclimatology, Palaeoecology, 251, p. 268-282, doi:10.1016/j.palaeo.2007.04.001.
  • Schmitt, J.-M., Baubron, J.-C., Bonhomme, M.-G., 1984. Pétrographie et Datations K-Ar des Transformations Minérales Affectant le Gîte Uranifère de Bertholène (Aveyron—France). Mineralium Deposita, 19(2), 123-131.
  • Thiry, M., Franke, C., Vercruysse, C., Kissel, C., Edel, J.-B., Bruhlet, J., 2011. Datation des paléoaltérations du massif cristallin des Vosges : implications pour l’évolution géodynamique du massif. 13ème Congrès français de Sédimentologie, Dijon, Publ. ASF, Paris, 68, p. 318
  • Thiry, M., Ricordel-Prognon, C., Franke, C., Brulhet, J., 2012. Modernité des paléosurfaces : leur rapport à la géodynamique. in : Cojan I., Grosheny D., Parize O. (eds) Expression de l’innovation en géosciences, Une journée avec Bernard Beaudoin, Paris, Presses des Mines, Collection Sciences de la Terre et de l’Environnement, p. 113-125.
  • Yao, K.F.E., 2013. Albitization and oxidation of the granitoïd rocks related to the Triassic paleosurface in the Sudetes (SW Poland). Thesis École Nationale Supérieure des Mines de Paris in Paris (France) and Państwowy Instytut Geologiczny – Państwowy Instytut Badawczy in Warsaw (Poland). 164 p.
  • Yao, K.F.E ., Franke, C., Thiry, M., Aleksandrowski, P., Szuszkiewicz, A., Turniak, K., 2011. Albitization as record of the Triassic Paleosurface in the Sudetic Crystalline basement (Poland). Geophysical Research Abstracts, 13, EGU 2011, EGU General Assembly 2011, Vienna, Austria, Geophysical Research Abstracts, Vol. 13, EGU2011-5930.

Dr Carles Fàbrega (Departament d’Enginyeria Minera, Industrial i TIC, Escola
Politècnica Superior d’Enginyeria de Manresa, Spain)

Boulder lags in Rosetta Bay at Victor Harbor, South Australia


The significance of strewnfields of large granite erratics throughout the Inman Valley was discussed in an earlier note (Milnes, 2019).  They essentially pinpoint outcrop or subcrop of in-situ glacigene diamictite from which they have been exhumed, or within which they still remain partly encased.  The diamictite is generally plastered over smoothed and striated Cambrian Kanmantoo Group bedrock.

Good examples of these boulder lags occur on the beach at Rosetta Bay and were recognized early on as glacial erratics eroded from glacial till (for example, Howchin (1910; Fig. 1).  Milnes (2019) located and photographed remnants of the source diamictite underlying these boulder lags (Figs 2, 3).  However, recent drone photographs taken in the area by Isaac Forman ( have shown that significant boulder lags occur on the landward side of Wright Island (Fig. 4) and, based on the forms visible beneath the sea surface, may also exist on the seabed in the Bay between the shore and the Island.

One implication of this observation is that the bedrock (Petrel Cove Formation, Cambrian Kanmantoo Group) is at a relatively shallow depth and that the glacial till from which the boulders were eroded (or within which they may be partly embedded) remains close to the sea floor.  Another is that the early held view that the granite landforms (particularly Wright Island and Rosetta Head) were ice-moulded during the Permian glaciation about 300 million years ago is probably correct.  And, finally, it may still be possible to find direct evidence of glacial pavements on parts of these granite landforms, like that discovered fortuitously many years ago at Port Elliot (Milnes & Bourman, 1972).

Fig. 1 (L) – Map from Howchin (1910) showing location of erratics in Rosetta Bay. Fig. 2 (C) -Boulder lag of mainly granite erratics on the beach near the boat-ramp, Rosetta Bay.  Fig. 3 (R) – Boulder lag of granite and other rock-type erratics on the beach at Petrel Cove.


Fig. 4 – Isaac Forman ( drone photo of Wright Island and the seabed on its landward side showing the boulder lag of granite erratics on the landward side of the Island and similar forms on the seabed (centre left).


Howchin, W., 1910, The glacial (Permo-Carboniferous) moraines of Rosetta Head and King’s PointTransactions and Proceedings and Report of the Royal Society of South Australia 34, pages1-12; plates 1-17.

Milnes, A.R., 2019, What’s the significance of the large granite erratics scattered through the Inman Valley in South Australia?

Milnes, A.R.,  BOURMAN, R.P., 1972), A Late Palaeozoic glaciated granite surface at Port Elliot, South AustraliaTrans. R. Soc. S. Aust. 96, 149-155.


Dr Tony Milnes

What’s the significance of the large granite erratics scattered through the Inman Valley in South Australia?


The earliest studies of glacial sediments and landforms of the Inman Valley, starting with Selwyn (1859), made much of the smoothed and striated bedrock pavements (which we now know to have been generated by the westward movement of ice sheets from continental regions that abutted southern Australia around 300 million years ago), and the large granitic and other erratics scattered throughout the valley.   Much of the available information at the time, and hypotheses attached to it, were summarized by Professor Walter Howchin in 1926.  Howchin’s map of the Inman Valley showed the locations of large boulder erratics (principally coarse grained, porphyritic granite similar to that exposed along the coast at Encounter Bay) and striated bedrock pavements, on which the directions of movement of the glacial ice could be measured.  As might be expected, in-situexposures of glacigene sediments were observed directly overlying the pavements in some of these localities.  Howchin remarked on the fact that changes in the courses of the Inman River and subsidiary streams over time continued to variously expose and also obscure rock pavements and overlying glacial deposits, and this situation has continued to the present day.

As well as at sites at which conspicuous large erratics occurred and striated bedrock pavements were located, Howchin assigned the sedimentary fill throughout the Inman Valley and adjacent areas to ‘Permo-Carboniferous’ glacigene deposits. This was partly due to borehole data that indicated diamictitic fill above bedrock in parts of the valley, but also because there were exposures of diamictite associated with the scattered striated pavements and erratics shown on his map (Fig. 1).  This view was promulgated by successive geologists and dominates even the most recent geological maps of the area.  However, as pointed out by Bourman & Milnes (2016), the fill in the valley is complex.  It includes the remnants of glacigene deposits extensively eroded and reworked during Mesozoic and Cainozoic times as well as younger fluvial and alluvial deposits such as the peaty sediments of Pleistocene marshlands and sandy sediments resulting from post-European settlement erosion and aggradation.  The soils map of the area (Fig. 2) is a good indication of this complexity as it presents at the modern landsurface.

171009_Howchin erratics map_enhanced_cropped
Enhanced map of ‘greater’ Inman River valley by Howchin (1926) showing his locations of erratics (red dots) and striated pavements (with directions of ice movement – purple arrows). Region coloured in yellow was assigned to ‘Permo-Carboniferous glacial’ deposits; other coloured areas are bedrock of various types and ages.
171005_erratics vs soils_cropped
Map of soils in the Inman Valley & surrounds showing locations of erratics (red dots – Howchin 1926; purple dots – recent field observations). Yellow indicates the dominant soil type – ‘G3: Thick sand over clay’ which corresponds closely to Howchin’s ‘Permo-Carboniferous glacial’ deposits and is promulgated on recent geological maps on which the soil mapping was based. Red = areas of ‘L1: Shallow soil on rock’ where bedrock is exposed or close to the surface on the steep slopes. Green = areas of ‘K: shallow to moderately deep acidic soils on rock’. Brown colours = areas of ‘F2: Sandy loam over poorly structured brown or dark clay’ soils, ‘E3: Brown or grey cracking clay’ soils, and F1: Loam over brown or dark clay in the modern stream valleys.

Erratic strewnfields

Recent field observations demonstrate that the strewnfields of large, mostly granite erratics (Howchin’s 1926 map, to which there are more recently added occurrences shown in Fig. 2), essentially pinpoint outcrop or subcrop of in-situglacigene diamictite from which they have been recently exhumed, or within which they still remain partly encased.  The diamictite is generally plastered over smoothed and striated Cambrian Kanmantoo Group bedrock and may be ‘lodgement till’.  Good examples of this are on the beach at Rosetta Bay, in dam excavations east of Mt Alma road, in the Inman River channel at Inman Valley township and for some kilometres eastwards, and at the site known as ‘Glacier Rock’.  Elsewhere, Permian glacigene sediments have long been eroded and substantially reworked, and the granite erratics that have been exhumed from them have largely disintegrated and the weathering products dispersed.  The stages in this process can be observed in some granite erratics now being exhumed from glacigene diamictite. Weathering and disintegration of granite erratics in association with Permian glacial diamictites does not occur to anywhere near the same extent in coastal environments, such as on beaches at Rosetta Bay, just north of Port Vincent and at Port Moorowie on Yorke Peninsula, and at Hallett Cove south of Adelaide.


Howchin (1926) referred to the diamictite as ‘.. glacial sandstone and boulder clay..’ and it is quite distinctive, as shown in the accompanying photographs.  In locations near the coastline, for example in Rosetta Bay, the diamictite tends to be bluish in colour, with some bleaching and iron-staining, suggesting that weathering is not pervasive.  In the main part of the Inman Valley, however, and particularly in the river channel, exposures are generally yellowish-white and bleached of colour.  In the earliest reports (Tate et al., 1898), the glacigene sediments were described as being to be so dark in colour that they were thought to be potentially coal-bearing, and this led to exploration drilling of three bores in Back Valley by the Victor Harbour Coal Company.  Carbonaceous glacigene sediments are known elsewhere, for example at Port Moorowie on southern Yorke Peninsula, but have not been observed recently in the Inman Valley. The beds of glacial origin that Howchin (1926)2referred to as being typically ‘.. tenacious blue clays..’  have also not been observed recently although the diamictite that is periodically exposed at low tide in Rosetta Bay and which underlies the conspicuous lags of large granite erratics, is bluish in colour.

The sandy-clay matrix of bluish-coloured diamictite is dominantly quartz, with feldspars and muscovite or biotite.  Unexpectedly, in the samples examined so far, the clay is dominated by poorly crystalline 14Åmontmorillonitic material: no kaolinite was observed.



1.  Strewnfield of large granite erratics west of Mt Alma.  2.  Lag of granite & metamorphic rock erratics on diamictite, Rosetta Harbor.  3.  Lag of granite & other erratics over diamictite, bed of Inman River, east of village.  4.  Disintegrating granite erratic eroding from diamictite, Strangways Hill.  5.  Large granite erratic in diamictite on glaciated pavement, Glacier Rock.  6.  Large granite & other erratics embedded in diamictite on glaciated pavement, bed of Inman River, east of village.  7.  Bluish sandy-clay diamictite (‘lodgement till’) beneath lag of granite erratics, Rosetta Harbor.  8.  Bluish sandy-clay diamictite with embedded granite & other clasts, Rosetta Harbor.


The strewnfields of large granite and other erratics in the Inman Valley are considered to represent vestiges of extensive Permian glacial diamictite.  Remnants of these sediments in localities along the north-central and eastern parts of the valley have been exposed close to the ice-moulded bedrock walls and floor that have been progressively exposed by erosion.  As downwasting continues, it is expected that the large granitic erratics now exposed in the boulder lags will gradually weather and disintegrate, as is common in terrestrial environments.  New occurrences could emerge if riverine erosion exposes more of the original bedrock valley. On the other hand, rising sea-levels may trigger aggradation and the burial of the now exposed strewnfields of erratics, the associated diamictites, and the underlying glaciated bedrock pavements.

Although there have been many investigations of facets of the Permian glaciation, including landforms and sedimentary deposits, starting as early as Selwyn (1859)1, evidence of post-Permian geological processes in the Inman Valley up until the Quaternary has not been recognised.  Some soil mapping linked with Howchin’s (1926) observations and more recent data reported by Bourman & Milnes (2016)2is the most recent information.  Opportunities to discover more of the history of this complex landscape clearly exist.


Selwyn R.C., 1859, Geological notes of a journey in South Australia from Cape Jervis to Mount Serle, No. 20, p. 4.

Howchin W., 1926, Geology of the Victor Harbour, Inman Valley and Yankalilla districts, with reference to the great Inman Valley glacier of Permo-Carboniferous age. Transactions of the Royal Society of South Australia, 50, p. 89-116.

Bourman, R.P. & Milnes, A.R., 2016, The geology and landforms of the Inman River Catchment.  Report to Inman River Catchment Landcare Group, Government of South Australia Department of Environment, Water and Natural Resources, December 2016.  237pp.

Tate R., Howchin W., David T.W.E., 1898, On the evidence of glacial action in the Port Victor and Inman Valley districts, South Australia.  Report of Research Committee No. 5, Australasian Association Advancement Science, 7thmeeting, Sydney 1898, p.  114-127.

Dr Tony Milnes

The beach cliffs north of Stansbury

North of Stansbury on Yorke Peninsula (South Australia), towards Port Vincent, there is an interesting hike along the beach at low tide.  Prominent cliffs of yellowish fossiliferous limestone overlain by reddish sand and mottled clay are in places capped by white carbonate-rich silts and hard limestone called ‘calcrete’ (1).  The cliffs are up to 20m high except where broken in a few places by gullies that mark once active, short streams.  At high tide the sea laps at and erodes the base of the cliffs across a shore platform cut into the limestone. Version 2

The limestone beds (Port Vincent Limestone) were deposited in Oligocene to Miocene times, roughly 20-30 million years ago.  The overlying reddish sands and clays (Hindmarsh Clay) are probably around 700,000 years old (Pleistocene age) and the white carbonate (lime) capping is even younger.

At the Stansbury jetty, the cliffs are mostly of Port Vincent Limestone.  The original bedding in the limestone is outlined in places by thin rubbly layers and elsewhere by shell-rich beds.  The limestone was deposited in shallow seas that once occupied the St Vincent Basin. Conspicuous amongst the fossils are whole-shell sea-urchins (echinoderms), bivalves (clams and oysters) and bryozoa in a sandy matrix that is largely made of shell fragments (4).  The cliffs show spectacular evidence of former caves, sinkholes, pipes and other solution features: these are obvious because they are filled with mottled green, yellow and red sandy clays (2, 3).   Undercutting by the sea has progressively collapsed the limestone to expose these structures and in places wash out the clays.  The caves and pipes characteristically have smooth surfaces dating from a time when lime-rich solutions seeping over their walls precipitated calcite on evaporation. Version 2

Version 2

Version 2

In the section from Beach Point (Long Beach) northwards, the Port Vincent Limestone is overlain by Hindmarsh Clay, a thick band of mottled green-yellow clays and deep red sandy clays.  In contrast to the marine origin of the limestones, these were deposited during the Pleistocene in fluvial environments.  The boundary between the two formations is important as it represents a large time break of several million years, during which the seas retreated from St Vincent Gulf to well south of Kangaroo Island.  As this was happening, in the Late Tertiary period, the changing conditions exposed the limestones at the landsurface and subjected them to erosion, weathering and dissolution to generate a significant karst (caves, sinkholes, pipes, and the like) landscape.  Major riverine and lake environments later became established in the former gulf.  The river systems drained highlands to the north and generated widespread alluvial and lacustrine deposits of sand and clays over the karst limestone, filling the caves and sinkholes to produce the remnants we see in the cliffs today.

It’s interesting to speculate as to whether the remains of any fauna (including megafauna) occur within this cave and sinkhole landscape.  It did, after all, form at about the same time as the extensive cave systems in the limestones of the South East and the Nullarbor which contain abundant mammalian and other fossils.

A feature of the Hindmarsh Clay is the occurrence in places of white bands of the mineral alunite (potassium aluminium sulfate).  Towards the northern end of Long Beach it was sufficiently abundant for local entrepreneurs to begin to mine it to produce potassium sulfate fertilizer (according to a report by RL Jack, Assistant Government Geologist, in 1918),  which was in short supply during the First World War, but the venture failed.   Remnants of mining operations can still be seen at the base of the cliffs behind sand dunes (5).  The alunite was formed from a reaction between acid groundwater and potassic clays in the sediments.  The process in detail is unclear, but the occurrence of alunite in superposed near-horizonal seams suggests that appropriate conditions may have related to flow of local acid groundwater.Version 2

Above the Hindmarsh Clay in most of the cliff-line is a ‘blanket’ of younger Pleistocene lime deposits.  These consist of unconsolidated silts with interlayers of hardened calcrete.  The youngest calcrete is exposed as sheets over large areas of the landsurface on this part of Yorke Peninsula, and was exploited in earlier days for the local production of quicklime.  It remains an impediment to cropping but is progressively being crushed and disaggregated by farmers using heavyweight rollers hauled behind large tractors.  The unconsolidated carbonate silts are considered to be a loess-like aeolian deposits whereas the calcrete bands are likely to represent the remnants of ancient soils that formed at successively younger times.

What else can we note about the geology of the Stansbury cliffs?

Projecting the top of the cliffs seawards provides an indication of the former extent of the Pleistocene landscape that has now been disrupted and incised as the St Vincent ‘valley’ was drowned by rising seas.

Fallen blocks accumulated at the base of the cliffs is the result of mass collapse of limestone and calcrete in the cliffs caused by back-wasting under the attack of waves.  Some rock fragments are very large; others are small and part of a lag concentrated by tides at the back of beach.  Many are angular, which indicates that they have relatively recently fallen from the cliff; some are rounded and smoothed which indicates that they have been washed around in the surf zone for a long period of time.   It’s interesting to speculate about the time that it takes the cliff to retreat under the attack of the rising sea.  Only observations over the long-term (mapping of fallen blocks or reference to old photographs to compare with the present) might answer this question.  However, most cliff erosion is likely to take place at comparatively rare times of severe storms and powerful wave and wind attack, whereas little change occurs under usual weather conditions.

The cliffs in some places are coated with a ‘wash’ of sediment from higher in the sequence, and this sometimes obscures the geology and makes it difficult to pinpoint breaks in the rock sequence. This coating varies from a thin surface veneer to substantial talus deposits, and usually occurs in places where the cliff is less exposed to wind and waves.  As well, in places, the red sand and clay fill in the old caves and pipes has spilled out from the cliffs where erosion has exposed them.

Collapsing masses of rock, spilling of sands and clays from less coherent deposits, breaking apart and exposure of karst features, and wash over cliff surfaces, are all components of cliff retreat in this region.

It’s interesting that the cliffline can be traced from actively eroding beach cliffs north of the jetty at Stansbury to a subdued hillslope at the back of the town to beach cliffs again south of the cemetery. The townsite is actually an embayment that was filled by shallow water shelly limestones (deposited around 10,000 years ago in seas that were at a slightly higher level than today) and covered by more recent sand dunes.  It seems to have been a quirk of nature that preserved the embayment, in much the same way that winds, tides and seawater circulation on this eastern side of the gulf produced distinctive, recurved sand and mud spits that protect other bays like Stansbury.

Further reading.

RP Bourman, CV Murray-Wallace, N Harvey (2016)  Coastal Landscapes of South Australia.  (University of Adelaide Press: Adelaide).  423pp.

AR Crawford (1965)  The Geology of Yorke Peninsula.  Bulletin No. 39.  Department of Mines, Geological Survey of South Australia. 139pp.

RL Jack (1918)  Alunite deposits, Section A. Hundred of RamsayInSouth Australia Department of Mines, Mining Review No. 28 for the half-year ended June 30th, 1918.  pp51-53.

AR Milnes, JT Hutton (1983)  Calcretes in Australia – a Review.  In ‘Soils: an Australian Viewpoint’, Chapter 10, 119-162. (CSIRO, Melbourne/Academic Press, London).

WJ Stuart (1970)  The Cainozoic stratigraphy of the eastern coastal area of Yorke Peninsula, South Australia.  Transactions of the Royal Society of South Australia 94, 151-178.

Dr Tony Milnes – Earth Sciences, University of Adelaide

Gulf St Vincent & Adelaide beaches

Over the last million years in southern Australia, sea levels have fluctuated between about 120m below present sea level, during the Ice Ages, to about 2m above during the interglacials.  As humans first arrived in Australia more than 50,000 years ago, the sea level fluctuated between 60 and 40m below present, and vast coastal areas were exposed for settlement.  Gulf St Vincent has a maximum depth of 40m so, during those times, it had the form of a flat landscape of alluvial plains veneered with calcrete and covered by thin sandy soils and sporadic desert dunes gently sloping towards a central oblong saline swamp (Fig. 1).

Around 30,000 years ago the Earth plunged into a severe Ice Age that lasted until about 15,000 years ago, with sea levels dropping to 120m below present.  A warming climate then caused rising sea levels, so that by around 9,700 years ago the sea encroached into the land to form the Gulf (Fig. 2).  Continued climate warming and further sea level rise progressively filled the Gulf.  The warm shallow seas enabled prolific growth of algae and marine grasses that were nurseries to a rich ecology of fish, shellfish, starfish, sponges, calcareous algae and some corals.  Sand-sized foraminifera were abundant, especially in the nearshore and intertidal zones (corresponding to the mangrove-rich areas north of Adelaide).

By around 6,500 years ago the sea had reached its present level and the Gulf achieved its current shape.  Abundant shells and shell grit were deposited to form a meter-thick veneer of calcareous sediment covering the now drowned landscape.  Predominant south-southwesterly winds created waves that eroded these sediments, for example around Port Norlunga and Maslins Beach, and quartz sand winnowed from them mixed together with more recent shell grit to create the modern beaches (Figs 3, 4).  Some quartz sands came from more ancient deposits that are now exposed at Hallett Cove.

Frequent southwesterly storms drove coastal sands northward from Marino Rocks and Glenelg to form a coastal dune series that enclosed the Patawalunga and Westlake depression.  This wave-dominant dune system continued to grow northward creating the present Le Fevre Peninsula where around 10m of sand has accumulated (Fig. 5).

European settlement since 1836 has altered the coastal zone by building on or removing most sand dunes, thus allowing storms to attack roads and residences. Placement of rocky rip-rap with the aim of protecting the shoreline has in fact increased wave turbulence and lowered the beach profile by removing additional sand.  Replenishment of beach sands to maintain amenity is a continuing Government-funded activity.  Tennyson Dunes is one of a few tiny remnants that provide a glimpse of the original Adelaide metropolitan coastline.


Bourman RP, Murray-Wallace CV, Harvey N. 2016. Coastal Landscapes of South Australia.Available as a free ebook from

Fuller MK, Gostin VA. 2008. Recent coarse biogenic sediments of Gulf St Vincent.InSA Shepherd et al. (eds) Natural History of Gulf St Vincent.  Ch.3 , 29-37. (Royal Society of South Australia).

Harvey N, Belperio AP, Bourman RP. 2001.  Late Quaternary sea-levels, climate change and South Australian coastal geology.  InV Gostin (ed.) Gondwana to Greenhouse: Australian Environmental Geoscience. Geol. Soc. Aust. Spec. Publ. 21, 201-213.  (Geological Society of Australia: Sydney).

James NP, Bone Y. 2011. Neritic Carbonate Sediments in a Temperate Realm, (Springer Science+Business Media).

Author: Dr VA Gostin

New report

Professor Bob Bourman and I have just submitted a report to the Inman River Catchment Landcare Group (southern Fleurieu Peninsula, South Australia) entitled ‘The geology and landforms of the Inman River Catchment‘.  Some funds in support of the project came from the Regional Landcare Facilitator Programme, an initiative of the Australian Government’s National Landcare Programme.  Our time in researching the subject and writing the report was a voluntary effort.  front-page

The general aim of the project was to prepare an overview of the geology and geomorphology of the Inman Catchment.  This was to provide a basis for improving local knowledge and awareness of how landscape and landforms have changed (and continue to change) according to landuse and land management practices.  We enlisted the help of landowners and gained new insights via their responses to a wide-ranging questionnaire.

The report can be downloaded via the following link:

Dr Tony Milnes, Earth Sciences, University of Adelaide