Did the River Murray Ever Cross the Mount Lofty Ranges?

A theory as old as the hills – but how old is that, exactly?


The Problem of Morgan

The River Murray descends from the Great Dividing Range at Albury, and for the next 800km follows a fairly steady northwest course.

That changes at Morgan, where the river makes a sharp left turn towards Goolwa and the sea.

The bend at Morgan has been attributed to the Morgan Fault, which lies just west of the river. But if the fault created the bend, where did the river go before the fault was formed?

Did it keep going, and where to?

A western route for the river seems crazy, not least because just west of Morgan lies the impassable barrier of the Mount Lofty Ranges. But a western route was proposed by Williams and Goode in 1978 who searched the ranges for the abandoned river all the way to Spencer Gulf.

The Route Across the Ranges

The western route theory proposes that before the Mount Lofty Ranges were formed, the River Murray continued northwest and discharged to Spencer Gulf. The river was defeated when the ranges were raised and flow was deflected south by the Morgan Fault. Clues to the original route can still be found across the landscape.

The first clue lies in Burra Creek, a small ephemeral stream that drains the western Mount Lofty Ranges and joins the Murray at Morgan. The alignment of the creek is key, because it continues northwest beyond Morgan. Could this be the former channel of the Murray, now flowing in reverse?

The second clue is in the ranges. Burra Creek lines up with the Broughton River, which continues the northwest route on the other side of the range.

The shape of the Broughton River channel is peculiar. Today it is a small, seasonal watercourse – not much more than a creek. But it flows through a channel that is very broad and deep, indicating it was once the course of a major drainage system.

The Broughton River near Spalding

The final clue is in the gulf, where the Broughton River snakes its way across an oversized delta. Although the river small and seasonal, its delta stretches 40 km from Port Pirie to Port Broughton and bulges 18 km into the sea. According to the theory, this is the ancient mouth of the River Murray.

Debunking the Theory

The western route of the River Murray was always hard to believe, but that doesn’t make it easy to disprove. The theory hinged on the River Murray being older than the ranges which, even in 1978, was not clear. It took another 10 years to sort things out.

The Murray Basin formed when Australia separated from Antarctica, 50 million years ago. The sediments that first filled the basin, and have filled it ever since, stop at the foot of the ranges. Therefore ranges have always presented a barrier at the western extent of the Murray Basin.

The sediments washing off the ranges also indicate their age. These date back at least to the Eocene.

As far as the Broughton River delta goes, it’s big, but it isn’t big enough. If you compare it with the vast volumes of sediment carried by the Murray drainage system, the delta is far too small. In fact, the Broughton River delta isn’t much bigger than similar alluvial deposits from the ranges at Noarlunga and Willunga. In the past the Broughton River drained a wetter catchment and would have carried more sediment. And the shallow floor of the gulf contributes to the delta’s broad extent.

In the same way, the oversized channel of the Broughton River is more easily explained by a wetter climate in the geological past than by a westerly flowing River Murray.

So What does the Bend at Morgan Mean?

More recent work has shown that the course of the River Murray through South Australia is actually quite new.

Previously, the Murray Basin drained through the Douglas Depression in western Victoria and reached the sea near Portland. This route was blocked by faulting 2.4 million years ago, which dammed the Basin and created the vast freshwater Lake Bungunnia.

Adapted from McLaren et al. (2011)

The present day route of the Murray formed only 700,000 years ago when Lake Bungunnia was breached. Water spilt southwards from the lake at Swan Reach and carved the Murray Gorge.

Having cut a route downstream, the gorge continued to erode upstream towards Morgan. This route was probably guided by fractures in the underlying limestone that run parallel to the Morgan Fault.

So the deflection of the Murray River at Morgan is an illusion. It doesn’t represent the redirection of a flow moving downstream, but rather the path of the river channel eroding upstream.

References and Further Reading

McLaren, S., Wallace, M.W. and Reynolds, T. (2012). The Late Pleistocene evolution of palaeo megalake Bungunnia, southeastern Australia: A sedimentary record of fluctuating lake dynamics, climate change and the formation of the modern Murray River. Palaeogeography, Palaeoclimatology, Palaeoecology 317-318: 114-127.

Stephenson, A.E. and Brown, C.M. (1989). The ancient Murray River system. BMR Journal of Australian Geology and Geophysics 11: 387-395.

Twidale, C.R, and Bourne, J.A. (2009). Course of the lower River Murray in South Australia: effects of underprinting and neotectonics? Proceedings of the Royal Society of Victoria 121: 207-227.

Williams, G.E. and Goode, A.D.T. (1978). Possible western outlet for an ancient Murray River in South Australia. Search 9: 442-447.


The Gorge, the Lake and the Abandoned Channel of the River Murray


From Overland Corner to Wellington, over three hundred and forty kilometres, the River Murray is carved deep into the landscape exposing thirty metre high pale yellow limestone cliffs.

The gorge is much deeper and wider than today’s relatively small river could create. Its origins lie in a wetter climate, a massive fault block and the palaeo megalake Bungunnia.

The Douglas Depression

In western Victoria an abandoned river connects the Sunset Country to the coast.

Now mostly filled in by sediment, the channel is hard to recognise on the ground. But this river used to drain the Murray-Darling Basin to the sea.

The Douglas Depression is over two hundred kilometres long. Commencing north of Wyperfeld National Park, it runs south to Douglas. It curves west around the resistant sandstone of Mount Arapiles, then continues south-west towards the coast. The depression is up to twenty kilometres wide but shallow – only thirty to fifty metres lower than the surrounding landscape. But it strongly influences present-day drainage systems. The Wimmera River, when it reaches the depression near Natimuk, turns north. The Glenelg River reaches the depression at Harrow and follows it south-west. Along its length are scattered shallow saline wetlands and lakes including Mitre Lake and White Lake in the south and the Wimmera Terminal Lakes in the north: Lake Hindmarsh, Lake Albacutya, Lake Agnes and the Wirrengren Plain.

Oblique Google Earth image of the Douglas Depression looking north from Balmoral.

Below the surface, the depression is carved up to forty metres into the surrounding geology. It cuts through the dune ridges of the Loxton Parilla Sands and deeper into the underlying limestone.

The western shore of Lake Hindmarsh where the Douglas Depression eroded a cliff into the Loxton Parilla Sands. Photo Lance Lloyd.

The depression was created over five million years ago and discharged to the sea near Portland.

Lake Bungunnia

This ancient River Murray was defeated by a tectonic shift.

The Padthway High is a fault block that stretches from the Grampians to the Mount Lofty Ranges, parallel to the coast.

Adapted from McLaren et al. (2011)

The block was raised 2.4 million years ago, raising with it the bed of the Douglas Depression. Flow from the Murray-Darling basin was gradually choked. With no outlet to the sea a vast freshwater lake was created inland.

Lake Bungunnia is the name given to the lake that extended over the lower Murray-Darling Basin between 2.4 million and 700,000 years ago. The lake initially extended from Swan Hill to Swan Reach, north to Menindie and south to Pinnaroo. At its greatest extent the lake covered 90,000 square km. Today’s largest freshwater lake, Lake Superior, is 82,000 square km.

We know the extent of Lake Bungunnia by the erosion of ancient shorelines and the characteristic grey-greenish clay of the lake bed. The clay has since mostly been buried by windblown sands.

But the lake continues to influence ecosystems today: Lake Tyrrell, the Raak Plain and Noora Basin are all formed on the exposed bed of Lake Bungunnia.

The River Murray Gorge

The maximum extent of Lake Bungunnia was determined by the crest of the Padthaway High. The lake reached a maximum extent in the wet climatic period near the start of its life, and peaked at seventy metres above current sea level. At this level the lake just overtopped the fault block at Swan Reach. This was enough to initiate flow to the sea and to start the erosion of a channel for the future River Murray.

Drier climatic conditions followed and the lake level dropped below the sill. But over the next 1.7 million years, during wet periods, the lake periodically overtopped the sill and reactivated and eroded the channel.

Adapted from McLaren et al. (2012)

At this time the sea level was at least one hundred metres lower than today and the coast was far off the shore of today’s Encounter Bay. This provided a steep gradient from the high point at Swan Reach to the river mouth. Lake outflows cut into the limestone of the fault block to create the Murray Gorge.

Oblique Google Earth image of the River Murray gorge looking upstream near Mannum. Vertical exaggeration x 3.

The gorge finally cut down to the level of Lake Bungunnia 700,000 years ago and drained it completely. But the river continued cutting upstream into the bed of the former lake, completing the gorge at Overland Corner.


The ecosystems we see today are the slowly evolving products of geological and geomorphic processes. We have a much better understanding of how ecosystems work and how to manage them when we understand their environmental setting.

See Also

Mallee Dunefields

The Wirrengren Plain

Further Reading and Sources

Bowler, J. M., Kotsonis, A., and Lawrence, C. R. (2006). Environmental evolution of the mallee region, western Murray Basin. Proceedings of the Royal Society of Victoria 118: 161–210.

McLaren, S., Wallace, M.W., Gallagher, S.J., Miranda, J.A., Holdgate, G.R., Gow, L.J., Snowball, I. and Sandgren, P. (2011). Palaeogeographic, climatic and tectonic change in southeastern Australia: the Late Neogene evolution of the Murray Basin. Quaternary Science Reviews 30: 1086-1111.

McLaren, S., Wallace, M.W. and Reynolds, T. (2012). The Late Pleistocene evolution of palaeo megalake Bungunnia, southeastern Australia: A sedimentary record of fluctuating lake dynamics, climate change and the formation of the modern Murray River. Palaeogeography, Palaeoclimatology, Palaeoecology 317-318: 114-127.

Hill, P.J., De Deckker, P., von der Borch, C. and Murray-Wallace, C.V. (2009). Ancestral Murray River on the Lacepede Shelf, southern Australia: Late Quaternary migrations of a major river outlet and strand line development. Australian Journal of Earth Sciences 56: 135-157.