Many of Ausralia's finest red wines are made in Coonawarra and it is one of the country's few world class wine producing areas. Coonawarra, like most of the world's great vineyard areas, is a clearly defined, localised, geographic setting. This setting and the other influences that make up the terroir may explain the particular wine flavours that make Coonawarra wines so popular with consumers. All vineyard areas can make delicious and affordable wine, but very few have the terroir to make wines for which consumers will pay greatly increased prices.

The Coonawarra vineyards lie on a flat plain in the South-eastern corner of South Australia on the Limestone Coast. The Limestone Coast formed during the last one million years in a geologic period called the Quaternary. This large area of the coast, about 350 km by 100 km, is underlain by limey soils and rocks. Many vineyards grow on the Limestone Coast, usually associated with red soils known locally as terra rossa.

Coonawarra is characterised by these terra rossa soils, forming a thin, linear shape more than 20 km long. The Coonawarra soils differ from the other vineyard areas on the Limestone Coast in this thin, continuous shape and in the colour intensity of the red soil. Coonawarra barely rises out of the surrounding plains. Most of the other vineyard sites on the Limestone Coast stand well above the plain. Clearly, Coonawarra has a different origin and it seems that the conditions that created this area occurred only once.

Compared to the classical Old World vineyards, with vines clinging to steep hillsides, Coonawarra appears a most unlikely site for great red wine. It is the high quality of grapes that grow on the terra rossa soils that makes Coonawarra so interesting for winemakers. Consumers also are curious as to why such small areas of red soil are associated with such special wines.

Why do these soils occur here; why are they confined to a narrow band; and when did these strange soils form? The scope of this paper is to explain the origin of the Coonawarra land surface and it does not deal with vineyards, wineries and wine quality. It was thought worthwhile, though, to include a short summary of the history of settlement and farming at Coonawarra, taken from current accounts, which includes a table of the vineyard plantings at 2000, to give focus to these important issues.

A BRIEF HISTORY OF WINE GROWING AT COONAWARRA

By 1850-1860 settlers at Penola, on the southern end of Coonawarra soil, found the climate and soils favourable for fruit trees.No doubt a few vines were planted.

In 1861 John Riddoch purchased the 14,170 ha pastoral property Yallum Park which covered a large area north of Penola. By 1880 Riddoch had built an imposing homestead a few kilometres to the west of Penola. He planted fruit trees and vines around the homestead. The area of Yallum Park continued to expand and by 1880 extended from Coonawarra to Mount Gambier, an area of 695 square kilometres.

In 1890 Riddoch founded the Penola Fruit Colony and in 1891 commenced subdivision of 465 ha of Yallum Park, north of Penola, centred on what is now the town of Coonawarra. Twenty-six settlers took up blocks, from 4 ha to 32 ha, and the planting of fruit trees, mostly apples, and vines commenced.By 1897 the settlers had 89 ha of vines planted and Riddoch had planted 52 ha. The main grape varieties were Shiraz, Cabernet Sauvignon, Malbec and Pinot Noir.

The first vintage was made at the Riddoch cellars in 1897, in the building which is now the Wynns winery. At about this time Riddoch changed the name of the Penola Fruit Colony to the Coonawarra Fruit Colony. The initial orchardists did not prosper but the history of viticulture in the region had begun.

In 1901, 14-year-old Bill Redman arrived in the district. In 1908 he purchased 16 ha from the Riddoch estate and went on to complete 45 vintages. These were difficult times. By the mid 1930s the Coonawarra vineyards had been reduced to 121 ha, from 365 ha, at the turn of the century. Most of the grapes were distilled.

In 1951 Samuel Wynn purchased the old Riddoch winery and vineyards. It was about this time that a small number of winemakers and consumers recognised the region's potential, but had to wait for a larger number of Australian consumers to discover wine before the area could expand.

Recognition by wine consumers of the area's huge potential came as late as the mid 1970s to early 1980s.

The current vineyard area within the proposed Coonawarra Geographic area at the end 2000 is:

COONAWARRA: A PART OF THE LIMESTONE COAST

Coonawarra is part of the Limestone Coast, a South Australian coastal plain developed over the last one million years. Coonawarra lies towards the eastern edge of this plain. Across the limestone coastal plain lie a series of parallel fossil sand dunes, about 15 metres high, trending northwest/ south-east. These dunes make the Limestone Coast a complex record of the rise and fall of the world's oceans. The onset of the current cooler period in the Earth's history, which commenced about 2.5 million years ago, triggered a series of ice ages causing sea levels to repeatedly rise and fall.

Coonawarra sits between two dune ranges, the West Naracoorte Range to the west and the Harpers-Stewarts Range to the east. The Harpers-Stewarts Range dune formed some 680,000 years ago. The West Naracoorte Range may well be made of a number of dunes that lie on top of one another. The youngest of these has a date of less than 780,000 years.

A quarry exposure of the West Naracoorte Dune before it was excavated to form the Gartner Winery. Note the blocky nature of the sediments and the relatively steep dip of the dune sediments. (Reference point 1 on map).

The unconsolidated sediments that underlie the black soil plains to the West of Coonawarra. Excavated rubble at the surface which overlies black soil rubble and seams of black soil lower down. (Reference point 2 on map).

An important feature of the West Naracoorte Range is its partial abutment against a cliff, termed the Kanawinka Escarpment. This escarpment separates two dunes, the East and West Naracoorte Ranges. It marks a fault line that first moved millions of years ago. This fault shifted again about 780,000 years ago. The country to the west of the fault fell about 40 metres, perhaps under the sea. It was against this cliff face that the Southern Ocean deposited the dunes comprising the West Naracoorte Range.

The fault line can be dated using the magnetic polarity of the East and West Naracoorte Ranges. Tiny particles of magnetite within the dunes aligned themselves with the Earth's magnetic field. The Earth's magnetic field frequently reverses; 780,000 years ago (+/- 20,000 years) a reversal, the Brunhes-Matuyama, named after early pioneers of this dating method, occurred. As the East and West Naracoorte dunes have a different magnetic polarity, the Brunhes- Matuyama reversal can be used to date the fault. This date helps anchor all the dates for dune ranges to the west of the fault. The East Naracoorte Range is older than 780,000 years, and may be as old as 860,000 years. This fault line marks a natural divide between the Limestone Coast and its vineyards and the Naracoorte tableland, on which the Wrattonbully vineyard district has grown.

From the period of warmth, during which the West Naracoorte Range dune was deposited at the base of the Kanawinka Escarpment, the Earth began to cool. As the ice caps grew, the ocean retreated westward over the plain on which Coonawarra now sits. The ocean shore re-formed far to the south-west, well out on the continental shelf.

As the Earth's weather again began to warm, the ice sheets at the poles began to melt. The sea level rose and water once more covered the limestone plain. While the sea was in retreat the Australian crust in this area had been slowly rising. Since the formation of the West Naracoorte Range, this dune and the surrounding country had risen about 7 metres.

Hence the new seashore was not as before the West Naracoorte Range but 10 km to the south-west. Here, a new coastal dune, the Harpers- Stewarts Range, began to form.

While the age of the lagoonal sediments east of the Harpers-Stewart Range is similar to the age of that range, 680,000 years, marine sedimentation would have ceased with the retreat of the ocean and the formation of the next westerly dune, the Peacock-Woolumbool Range, at 570,000 years.

UNDERSTANDING COONAWARRA

To begin to understand Coonawarra the best place to start is on top of the Harpers- Stewart Range dune. The dune rises about 15 metres above plains to the east and west. The view eastward towards Coonawarra is of a flat, dullgrey, treeless surface, passing over Coonawarra to the fading, grey-green hills of the West Naracoorte Range. With the winter rains these plains become waterlogged and boggy. Closer examination of the plains will show small scoops and depressions, some of which are rimmed on the eastern edge with a slight rise. These have been hollowed out by wind. The larger lagoons may be rimmed by sediments 10-20 metres high.

Driving across the plain towards Coonawarra, the soils are a dirty grey-black colour, about a half to two metres thick. Below are 10-15 metres of friable, weakly bedded, chalky sediments, so soft that the term bedrock does not seem appropriate. These chalky sediments lie on older, well-washed, creamy-white, sandy, unconsolidated marine sediments.

Beneath these sediments lie the basement rocks.

The dull, grey-black soil of the plain is littered in some areas with fragments, small discs and plate-sized stones of white calcrete. This tough calcareous-rich rock is a feature of Coonawarra. These areas of calcrete rubble are usually, but not exclusively, associated with a slight rise of 1-1.5 metres. In most places, the unconsolidated chalky sediment grades directly upward into the black soil. However, in places it is thinly veneered with a calcrete cap, millimetres to centimetres thick.

It is rare to find a pronounced cap up to 30 cm thick. Fragments of this cap work their way to the surface, probably through farming activities, forming areas of calcrete rubble.

The soil associated with these rises covered in calcrete rubble is usually grey-black, but in places a noticeable browning or reddening may occur. The elevated rises begin to take on the ferruginous tint of terra rossa. A few of these rises look promising enough to plant with vines.

Beach sediments exposed in a quarry east of Penola. These are overlain by C colour red-brown soils, some of which have been rolled into mud balls by water and are surrounded by paler wind blown siliceous sandy soils swept into the water at the same time. The soil and rubble at the top is from recent excavation. (Reference point 3 on map).

The Western end of Drain C looking south east. At this point C colour soils grade into B colour soils which coincides with a thickening of the calcrete cap and a 1 to 1.5 metre lift in elevation. (Reference point 4 on map).

Indeed, in the middle of this plain a vineyard has been positioned on a local rise where the soils show a noticeable reddening. In many quarry exposures a slight browning-reddening of the soil can be seen at the interface of the calcrete and the soil. Small patches of red-brown soil, several centimetres thick, are not infrequent.

Approaching Coonawarra, the sediments slowly change. The unconsolidated, crumbly, creamy to dirty grey limestone begins to show faint, but distinct, bedding lines. And it is always thinly veneered with a calcrete cap, which thickens quickly as the soil reddens.

Continuing the journey west, the land begins to rise after the Mount Gambier rail line is crossed. As the ground lifts the colour of the soil, over several hundred metres, changes to a grey, brown-red, then a brighter red-brown and then, opposite Wynns, a fierce rust colour. Coonawarra is identified with the famous Wynns Winery. Wynns and the town of Coonawarra sit in the centre of the famous red soils about 1.5 to 2 metres above the plain. Here plates of calcrete are common at the surface. The land now levels out and travelling past Coonawarra, a darkening and toning down of the soil colour occurs. Crossing the main north-south road, not far from the Brands winery, the soil again brightens to a rust colour and then fades to a brown-red chocolate colour, before quickly fading to brown again and finally returning to the grey-black colour of the plains. The width of this strip of red soils is about 1.6 km. The ground does not dip again, but rolls into the flat plains of eastern Coonawarra. The land surface seems different to the western plains.

The western edge of Coonawarra has a distinct boundary. The 1 to 2 metre lift from the grey-black soil plains is often sudden. In contrast the eastern boundary dips and rolls away into small gullies. In other places the brown soils level off into a flat, black coloured soil plain.

MAPPING THE RED SOILS OF COONAWARRA

The soils of Coonawarra have been mapped by their colour (see map at right). The divisions on this map are not those that experts in soils would use. They were selected to highlight, if it existed, the underlying geological grain of Coonawarra. In this way the width, length and trend of Coonawarra is revealed and magnified. The colour boundaries are quite precise and the change, in general, occurs over a few metres. Five soil colours have been identified. They are:

A colour soils - Soils with distinctive terra rossa colour associated with Coonawarra. Brown-red to a bright iron-rust red, distinctive, at times almost glowing with brightness. Frequently associated with rubbly calcrete. Much of this calcrete may have been brought to the surface by deep ripping prior to vineyard planting. This ripped calcrete is angular. Calcrete brought to the surface naturally is rounded and smooth and occurs in areas where the soil cover is thin, often less than 10 centimetres. The amount of calcrete brought to the surface by vineyard preparation depends on the soil depth and the length of the ripping device. The A coloured soils occur at the high point of the Coonawarra rise.

B colour soils - Browner than the A colour soils, although patches of A colour occur. Likewise, patches of B coloured soils are seen in the A zone, usually associated with slight dips and lower scoops. The soils range from rusty brown to deep red- brown and are deeper in colour than the Acolour soils. In places, near the boundary between the A and B zones, A coloured soils have been washed onto the B coloured soils. There is less calcrete rubble on the surface and large areas have no surface rubble.

C colour soils - C coloured soils are dominated by brown and grey colours. Soil redness is never present for long stretches, although slight ground rises often show a weak reddening. The soils are darker than the A and B colours. Calcrete is common as tiny fragments and rubbly areas are small. The C coloured soils represent a transitional soil to the plains on either side of Coonawarra.

D colour soils - These are the soils of the black limestone plains. Black to grey in colour and mostly without surface calcrete, except on small rises. A weak reddening in colour may be observed as the black soil passes into the thin calcrete cap or the underlying soft bedrock. Many of the new vineyard plantings on this soil show extensive calcrete rubble from the deep ripping preparation. Where the soils change quickly from C coloured soils, or rapidly through the sequence from A to D, there is a drop in elevation. This is most readily observed along the western edge of the Coonawarra area.

E colour soils - A grey, light- grey to dirty-white coloured soil, generally seen east of a central line through Coonawarra. These soils are wind-blown sands that overlie other soil colours. Because of the patchy nature of this soil it has not been mapped separately. The E colour soils vary in thickness from 10 centimetres to several metres and thicken rapidly to the east of Coonawarra. This soil is not contiguous, but lines the dips and gullies along the eastern edge of Coonawarra. In a quarry east of Penola mud balls of C colour soil are found embedded in E colour sandy soils. These mud balls probably formed in a shallow water-filled depression. When the water dried, the mud cracked and flaked and was rolled into balls by wind and water action. The wind-blown E soils swept in at a similar time.

DISCUSSION OF THE COONAWARRA SOIL COLOUR MAP

The red soils of Coonawarra can be traced for 23 km along a north-west line. The maximum width of the red soils measuring to the C coloured soils is about 1.75 km and stretches for 14 km.

The brightly coloured A soils stretch for 17 km. At the southern end of the area the red soils pinch out at Penola. The northern end can be traced to the Sharefarmers block, where brown colours begin to dominate the red. The soils may continue north although calcrete is poorly developed, indicating the trend is fading quickly. The northern end also shows an arm trending to the north-east. The reason for this arm and whether it has a geological significance is not known. The S. Kidman vineyard and the trend of its red soils is also uncertain. It is not known whether this vineyard lies on the north-east split although this is unlikely as such a connection would cut across the geological trend revealed by the alignment of salt pans and marshes. Along the eastern edge, three vineyard areas trend to the east. The most northerly, the Penley block is on soils that are part of a weak east-west ridge. Much of this ridge is covered with windblown E coloured soils.

Looking East along Drain C. Lagoonal sediments merge upwards into thick calcrete and A colour soils. (Reference point 5 on map).

Beach sediments exposed in the Penley quarry east of the Penola- Coonawarra road, near the Penley Estate winery. (Reference point 6 on map).

The reason for the other two easterly protrusions of red soil to the south is more difficult to understand. The ridge extending east of the Majella block is sculptured into smooth gullies and numerous low-lying pans, evidence of wind and water erosion. The ridge continues to the east and is covered in wind-blown sands. The other prominent high area extends east of Bowen Estate. It also has been sculptured by wind and water erosion, though wind-blown sands are less of a feature. Looking at the map (Page 33), there is a weak indication that the vineyards along the eastern side, associated with isolated patches of C coloured soil, have a common geological control. The new developments south of Penola occur on patches of soil that are similar to the C coloured soils developed to the east of Coonawarra. Possibly these are all linked in some manner.

On the western side of Coonawarra, over the rail line, higher areas of ground are common. Many of these have reddened to a brown-red C coloured soil. There are also rises on the plain, several kilometres further west. Many show strong calcrete development which has helped preserve them from erosion.

The map highlights the extent of all vineyard development either on the Coonawarra ridge or to either side. The data is several years old and plantings have expanded over the last few years. Many plantings extend onto D coloured soils, though usually where associated with a slight elevation increase. Plantings now occur well away from the terra rossa soil, characterising the central rise of Coonawarra. Most of these plantings will, though, lie within the new boundary of the Coonawarra wine region.

A CLOSER LOOK AT THE SOIL RELATIONSHIPS OF COONAWARRA

On the surface, and on the map, the soil colours to the east and west of Coonawarra appear identical. However, examining pits and quarries on the eastern side of Coonawarra shows the underlying bedrock to be a mottled, creamybrown coloured, clayey, unstratified, unconsolidated sediment, different to that of the western side. Also surface calcrete is uncommon east of Coonawarra. A recent publication on the land systems of Coonawarra also notes the change in landforms to the east and west of Coonawarra.

A feature of the C and D coloured soils to the east is the presence of pisolitic iron stones. These are tiny, pea-sized, iron-rich concretions.

They may join together to form lumps. The new vineyards south of Penola have uncovered large football-sized clumps of these pisolites. Pisolites are very rare west of the A coloured soils. This change in the soil type to the east of Coonawarra suggests that the underlying bedrock has changed under the rise of Coonawarra. The map shows a continuous stretch of red soils some 23 km long, with 15 km having a width of about 1.7 km; a continuous soil feature which straddles the change in the underlying bedrock. This cannot be a coincidence.

The Clues from a Cutting Across Coonawarra know as Drain C

There is an east-west cutting across Coonawarra that allows us to examine the soils and sub-surface in great detail. Drain C, as this cutting is called, crosses Coonawarra north of the town. The 2 km walk along this drain is the most informative walk any student of the wines of Coonawarra can take.

Starting on the grazing land to the west of the Mildara Vineyards, C coloured soils are underlain by soft, friable, chalky muds. The calcrete horizon between the two is poorly formed. The change from C coloured soils to B coloured soils occurs over a few metres and coincides with a lift in the surface of about 1.5 metres. It also coincides with the start of vineyard plantings.

Strolling eastwards, the faintest bedding planes appear in the chalky muds. Small pellets of limey mud often occur on the same horizon. The calcrete cap separating the red soil and the chalky mud thickens.

The bedding planes become more distinct as one moves further east. Pelletal, limey muds form continuous lines and link together to form nodular, platey, firm layers that separate softer muds. Under the A coloured soils the calcrete cap is half a metre thick and is spectacularly eroded in places, allowing long pipes of brilliant-coloured red earth to penetrate down. East of the Acoloured soils and close to the main road the bedrock is now slabby and stratified and passes into strongly stratified, nodular, hard limestone east of the main road.

The characteristics of this limestone are best seen in quarries, like that on Penley Estate, and along the eastern side of Coonawarra. These quarries expose a flat-lying, slabby, nodular, limestone of creamy to yellow-brown colour with rust staining. Gently dipping beds are common.

The rock is densely fossiliferous with gastropods, scallops and other bivalves of the sort found on beaches today. This limestone is compacted enough to form metre-wide slabs.

The Bedrock Controls The Soils

As can be seen from the cross-section in drain C, Coonawarra coincides with a change in the underlying bedrock. The transitional sediment type under Coonawarra assisted the formation of a much thicker calcrete cap, than that formed on sediments to the west or east. The thick calcrete cap would have acted to preserve the underlying sediments from erosion. This calcrete crust is thickest under the A coloured soils.

Northern vineyards, like Sharefarmers, while directly on the Coonawarra trend, appear to be at the fading influence of this bedrock change. Towards the south the rise from the C to B coloured soils in the west is gentle, suggesting that the effect of bedrock changes is fading. It is unlikely the bedrock change continues south of Penola.

The lens-like shape of Coonawarra does not seem related to topography, as there is a steady rise of about 6 metres from Sharefarmers to Penola. Such a rise is significant in this flat landscape. This height difference supports the view that the soil colours and the location of Coonawarra are not due to superficial weathering influences, such as topographical influences and the introduction of wind-blown soil, but have an underlying geological control.

ASSEMBLING THE CLUES ABOUT HOW COONAWARRA WAS FORMED

In 1983 Schwebel studied the Limestone Coast inland from Robe. His descriptions are helpful for understanding Coonawarra. He says in reference to sediments formed behind a dune, "They were deposited around the margins of the lagoon and represent low energy beach and near-shore conditions. The framework most commonly consists of either whole or fragmentary bivalves and gastropods." Later he continues, "Asystematic distribution of lithologies is controlled by the energy conditions prevalent in different parts of the lake. Coarser grained, better sorted sediments occur around the leeward margins of the lakes where wind-generated waves winnow the sediment in beach environments. Pelletal mudstones tend to occur on those shallow parts of the lakes where seiching periodically removes the fine disaggregated mud from the lake surface and promotes the formation of laminated pelletal mudstones. Amorphous mudstones appear to be representative of the more central lacustrine conditions where surface water persists through severe seasonal evaporative periods." While technical, it is hard to imagine a better description for the setting of Coonawarra.

The richly fossilised, stratified limestones that occur along the eastern edge of Coonawarra are beach deposits. This and the sediment gradation westward, under Coonawarra and beyond, imply deposition in a lagoon formed behind the Harpers-Stewarts Range. Fifteen kilometres south of Penola the Harpers-Stewart Range swings to the west and fades as it approaches the Mount Burr group of volcanoes. This group of volcanoes is about one million years old. They may have been islands at the time of formation of the Harpers-Stewart Range. It seems possible that the ocean entrance to the lagoon was between the Mount Burr volcanic group and the southern end of the Harpers-Stewart Range. How long the lagoon had an ocean entrance is not known.

Further evidence for this lagoon, and perhaps its size and shape, comes from recent radiometric mapping of the Coonawarra region. These surveys measure radioactive emissions from potassium, thorium and uranium. The map shows a large eliptical shape that takes as its eastern side a curve just to the east of the Coonawarra rise, circles around Bool Lagoon and closes after a wide loop west on Penola. Could this be the outline of the lagoon that existed west of the Coonawarra rise and is now reduced to Bool Lagoon? The implication might be that a lagoon of greater area, that had a seaward entrance to the south, was reduced to the size suggested by the radiometrics after the seaward entrance closed.

The Coonawarra can be compared to the Coorong, which today has formed behind the coastal dune of the Younghusband Peninsula and has an entrance to the sea at the mouth of the Murray River.

EXPLAINING THE FORMATION OF COONAWARRA

The Harpers-Stewart Range was formed 680,000 years ago. This dune marks the advance of the Southern Ocean, after melting of continental ice sheets during global warming. It forms a continuous, north-west trending range, stretching more than 100 km along the Limestone Coast. Behind the southern end of this dune a large saltwater lagoon was impounded. The lagoon was initially open to the sea at its southern end. Aradiometric map of the Coonawarra region suggests that later the lagoon was oval-shaped and about 40 km long and 10-12 km wide. The freshwater Bool Lagoon, north of Coonawarra, may be a remnant of this lagoon. Along the central and south-eastern margin of this large lagoon a beach deposit many hundreds of metres wide was formed. Waves stirred the near-shore sediments and gently rolled the limey muds into laminated pelletal mudstones. This steady stirring washed the finer clay particles deeper into the lake, beyond wave action.

The red soils of Coonawarra overlie many rock types: the beach deposits (overlain by the Eastern C coloured soils); the shallow water, bedded peletal deposits (overlain by the Eastern B coloured soils); the deeper water, poorly bedded, chalky muds (overlain by the A coloured soils); and the unconsolidated, chalky muds further west in the deeper parts of the lagoon (overlain by the Western B and C coloured soils). The relationship of individual soil colours to underlying rocks is only a guide, as geological processes are far too complex for simple solutions.

However, the near-shore depositional environment of the bedrocks is the key to understanding the soils and how they formed. Sorting of the sediments by waves has created varying physical and chemical properties in the bedrock in different places. This change has allowed thicker calcrete to form under the Acoloured soils. Similar marine, lagoonal beach sands and deeper water deposits occur inland from Robe but none have the length of Coonawarra and more importantly they have not had the time to express themselves, like Coonawarra.

The soils to the east of Coonawarra have formed on sediments deposited in swamps, marshes and intermittent lakes.As the continent dried over the last 500,000 years, significant amounts of siliceous wind-blown sand has been added to the sediments. As the country slowly rose the lagoon became brackish and finally fresh, before drying out.

Intermittent lakes would have appeared on the plains during wetter periods. Numerous small depressions, scalloped out by wind, occur east and west of Coonawarra. These occur in rows that follow the trend of the Coonawarra soils and thus reflect their underlying bedrock.

As the lagoonal muds became exposed to the air the soil forming process began. The soft sediments of this area did not require the intensive chemical and physical attack that is required to make soil on a hard granitic surface. Along with soil development, calcrete formation occurred.

A feature of soft, calcium rich sediments is a reaction with water to form weak, carbonic acid. This dissolves the soft limestone which re-deposits as the smooth, firm, crusty skin, termed "calcrete". This weather-resistant capping has preserved the Harpers-Stewart Range and is of course a feature of the inter-dunal areas where solution and re-deposition builds tough, flat slabs of calcrete at the junction of the soil and the bedrock. Calcrete formation is assisted by fluctuating water tables, frequent flooding of the plains and periods of aridity.

Wind has been a powerful factor in modifying the landscape of the Coonawarra region for a long time. A visit to Bool Lagoon shows the power of wind erosion. On the leeward shore a crescent-shaped dune 10 to 20 metres high has been formed of sediment scooped from the lake. Lunettes fringing the clay-salt pans of this area are common. The wind blows predominately from the west and northwest. The wind exposed the Coonawarra ridge by removing the plains sediments to the west, in a process called deflation. The calcrete cap has protected the bedrock of Coonawarra from the corrosive effects of water erosion and stabilised it against wind erosion. In dry periods when wind erosion was most active, wind was unable to cut into the Coonawarra ridge.

Wind carrying siliceous sand particles, the E coloured soils, flowed over the Coonawarra ridge and dumped some of the load along the gullies and ridges that had been sculptured along the eastern side of Coonawarra. The rises on the western plain and to the east of Coonawarra are erosional remnants of the former surface. The active role of water in this region is more difficult to understand as gradients are small and there is no outlet for streams to remove sediment. Fluctuating watertables would have assisted wind erosion by creating weaknesses in the bedrock. Wind and water erosion working together seem to be the cause of the unusual rounded gullies that exist along the eastern side of Coonawarra.

The reddening of the soils at Coonawarra and on the surrounding plains is associated with elevation. The low-lying grey-black soils become waterlogged during wet winters leaving them in an anaerobic state in which iron cannot be oxidised. Rises in the landscape allowing ferruginisation to begin escape this waterlogging. Not all rises have become ferruginised and the reason for this is not known. The exposure of Coonawarra, as a rise of one to two metres above the western plains, was enough for the soils to avoid waterlogging and become ferruginised.

The age at which the Coonawarra ridge became exposed is not known, nor is the age of the colour change. It is likely that both processes began around the same time. The exposure of Coonawarra may have commenced hundreds of thousands of years after the sediments were deposited. Red soils can develop quickly as can been seen on the young dunes at Robe (80,000 years old). Evidence from coastal dunes in Victoria suggests about 30,000-40,000 years is needed in forming red soils. However, these soils are not as deep and rich as those at Coonawarra which suggests a much longer period was needed to create the Coonawarra soils.

As noted, the limited range over which the red soils occur and the unusual red colour suggests that chemical differences in the underlying sediments have played an important part in soil formation*.

CONCLUDING THOUGHTS

This explanation of the origin of the soils of Coonawarra is unlikely to be the final word. The discussion does, though, tighten the boundaries for future research and suggests areas for detailed investigation. Hopefully it will also assist those with vineyard properties in the area to market Coonawarra in a smarter way. No doubt all the world's great wine areas have an exciting story to tell. How many, though, will have undergone such a wonderful, complex series of events just to make the soils? Think about the sequence of soil formation. Ice ages were needed to make the bedrock, unusual marine, estuarine conditions were required to make bedrock, gentle crustal lifting followed to preserve the bedrock, fluctuating water tables and periods of aridity were needed to form the protective calcrete caps, then wind and water erosion left a long narrow ridge exposed. And all of this had to occur in the right sequence for the distinctive terra rossa soils to develop.

Knowing how these valuable soils developed does not make the wine taste any better, but a profound understanding of the site where wine was made transfers into a greater pleasure and understanding of the wine. Also, this study may help the vignerons of Coonawarra focus on those parts of the vineyard that make the most complex and interesting wines.

In a future paper it is intended to relate features of the terroir of Coonawarra, including the different soils, to wine quality. It will be interesting to see how much of the unusual terroir of Coonawarra is expressed in the taste.

SELECTED REFERENCES

Bestland, E.A. (2002) The Dirt on the Coonawarra (for the Flinders Journal), Press release from Flinders University, January, 2002. Billing, N.B. and Cann, M.A. (2000) Land Systems of Coonawarra, Preliminary map. Primary Industries and Reources, South Australia. Drexel, J.F. and Preiss, W.V. (Eds) (1995) The geology of South Australia. Vol. 2, The Phanerozoic. South Australia. Geological Survey. Bulletin 54. Halliday, J. (1983) Coonawarra the history, the vignerons & the wines. Yenisey. Mackenzie, D.E. (1999) Coonawarra wine growing region, Interpretation of soils from airborne radiometric data. Unpublished map, Australian Geological Survey Organisation, Canberra. Schwebel, D.A. (1983) Quaternary dune systems. In: Tyler, M.J., Twidale, C.R.,Ling, J.K. and Holmes, J.W. (Eds). Natural history of the South East. Royal Society of South Australia. Occasional Publications, 3:15-24. Tyler, M.J., Twidale, C.R., Ling, J.K. and Holmes, J.W. (Eds). (1983) Natural history of the South East. Royal Society of South Australia. Occasional Publications.