Australasian Science: Australia's authority on science since 1938

Dinosaur Stampede Stopped in its Tracks

The large Lark Quarry footprints were made by a herbivorous dinosaur.

The large Lark Quarry footprints were made by a herbivorous dinosaur, perhaps one similar to Muttaburrasaurus. Credit: Anthony Romilio

By Anthony Romilio

A forensic analysis of ancient footprints has cast doubt on claims that a dinosaur stampede took place at Lark Quarry.

Only living animals can make footprints, so the presence of more than 3000 individual fossil footprints made by more than 150 dinosaurs is cause for excitement for dinosaur fans and researchers alike. Known as Lark Quarry, the tracksite in central-western Queensland has the potential to reveal the actual behaviour of these ancient creatures when they lived in Australia around 93 million years ago.

Most of the footprints are aligned in a single orientation and were made by small dinosaurs no bigger than chickens and turkeys. These small-bodied dinosaurs were all moving in one direction, but why? This has been a question pondered by scientists since the late 1970s as well as whether the vital clue comes from a trackway made by a much larger Lark Quarry track-maker – a predatory dinosaur measuring more than 2.6 metres tall at the hips.

Now scientists examining these fossil footprints using 3D digital technology have toppled the answer proposed decades ago, with evidence suggesting that some dinosaurs went to extraordinary lengths to leave their mark at Lark Quarry.

Conventional wisdom suggests that most of Lark Quarry’s footprints were made during a single event. If this is correct then more than 150 small-bodied carnivorous and herbivorous dinosaurs stampeded together as one big herd after being frightened by the approach of a large predatory dinosaur.

This extraordinary dinosaur behaviour was described in detail by Drs Thulborn and Wade in 1984, a time when dinosaurs were popularised as sluggish and solitary reptiles. But footprints don’t lie, and these researchers proposed that the Lark Quarry footprints were made by social animals capable of high activity behaviours more typically seen in modern birds and mammals. The research by Thulborn and Wade was ground-breaking when it was published, and a step towards a more modern perception of dinosaur behaviour.

Thirty years on, the story of “dinosaurs stampeding from a large predator” is still being retold time and time again. But it’s not only the professional guides that like retelling it to tourists visiting Lark Quarry. Movie director Steven Spielberg liked it so much that he included a dinosaur stampede in Jurassic Park.

The Australian government liked the story too. In 2004 it protected Lark Quarry with a National Heritage Listing, covered the site with a purpose-built building and renamed it The Dinosaur Stampede National Monument.

But does the original Lark Quarry interpretation hold true? Were the tracks all made at the one time, or were they recorded over a greater period of time? After all, modern sheep and cattle farms have hundreds of footprints around dams and rivers from repeated visits by livestock over periods of days, weeks or even months. Could the same be true for Lark Quarry?

And there are other questions. The stampede apparently comprised a mixed herd of small-bodied herbivorous and carnivorous dinosaurs. Why would plant-eaters and meat-eaters amicably herd together?

And what about the big predator? How do we know that the footprints were made by a predator? And why are its tracks heading in the opposite direction to the ones made by the smaller dinosaurs, so that the small dinosaurs would have had to stampede towards the predator?

To unravel the Lark Quarry mystery Dr Steve Salisbury of The University of Queensland helped me to examine the large footprints as part of my PhD studies. Thulborn and Wade had concluded that these large footprints closely resembled Tyrannosauropus tracks from the USA. This impressive name suggests that these were the footsteps of T. rex, but for decades science has accepted that footprints called Tyrannosauropus were not made by T. rex – they weren’t even made by a predatory dinosaur. In fact, the American Tyrannosauropus footprints were made by large bipedal herbivorous dinosaurs called ornithopods.

Does this mean that the large Lark Quarry footprints were also made by a plant-eater?

The conventional approach has been to document dinosaur footprints as 2D outlines, but this technique immediately loses all the track depth information, and the drawn track outlines are highly reliant on what the investigator perceives to be the footprint’s shape. A more objective method is to take many photographs of each footprint and then stitch them together to form a digital 3D model.

We made digital 3D models of the footprints when we visited the tracksite in 2013, as well as 3D models of the 1970s replica of the tracks that were made soon after the site was excavated.

We found that the large footprints were highly variable in shape, more so than what Thulborn and Wade had portrayed them to be in their 1984 publication. Perhaps a little disturbing was our finding that Thulborn and Wade appear to have mistaken cracks in the track surface as footprint features (such as toe impressions) and that the actual footprint size was only three-quarters of the size documented in 1984. This meant that instead of a track-maker 2.6 metres tall at the hips, it was a good deal shorter: only 1.9 metres tall.

However, these weren’t the only differences. The best-preserved footprints had fat, round toe impressions typical of ornithopods, but these didn’t resemble the herbivorous Tyrannosauropus tracks from the USA. Instead they appeared more like Iguanodontipus tracks, a different footprint type that was made by a more primitive type of ornithopod like Iguanodon.

At this stage our research indicated that an ornithopod dinosaur made the large (or rather not-so-large) Lark Quarry tracks. Could this herbivore have scared the little dinosaurs into a stampede?

To address our query into the timing between different track-making events, we returned to our 3D models. We found information inside and outside of the tracks that related to the sequence by which the dinosaurs crossed the site.

Inside the large footprints are ridges formed when sediment stuck to the underside of the dinosaur’s foot as it left the track. Sediment ridges like these can only form when the ground is saturated but not covered by water. This told us that the large ornithopod was moving on soggy ground, but the surface was exposed to air.

Surrounding the outside of the footprints are pressure bulges that formed around the dinosaur’s foot when it plunged into the damp ground. However, these bulges are cut by 10 cm wide vegetation drag marks that stretch the length of the tracksite. These marks were made when a branch or root of partially floating plant matter dragged along the river bottom as it was pulled along by the current.

This information tells us that some time after the large ornithopod traversed the tracksite, the surface was covered in water. The time needed for water to return to the riverbed, and become deep enough to drag large pieces of plant material, may have been considerable – perhaps hours, days or even weeks!

And where were the small dinosaur footprints in all this? Their footprints overprinted both the large ornithopod tracks and the vegetation drag marks. This means that those smaller dinosaur track-makers were only at the site when the larger ornithopod was long gone.

However, the question still remains: did the small-bodied plant-eating ornithopod and meat-eating theropod stampede at Lark Quarry?’

The Lark Quarry ornithopod tracks called Wintonopus – named after the town of Winton located ~100 km away from the tracksite – have three short toe impressions, and lack a “heel” impression. The small theropod tracks called Skartopus (meaning “nimble foot”) look similar to a chicken footprint, having three long toe impressions, but these tracks also lack the “heel” impression.

Wintonopus and Skartopus track types are quite different , and when examining their trackways we noticed that some began as Wintonopus and then changed to Skartopus, and vice versa.

What was going on? Had we mixed up trackways?

The answer was in the 3D models of the tracks we had constructed. When viewed from above, the classic Skartopus tracks look like a chicken’s footprint, but the side-on view shows it to be a short footprint towards the rear of the track and sloped (nearly at 45°) towards the front. These different front and rear parts meant that Skartopus was a movement trace, the combination of the short-toed Wintonopus footprint (the rear section of the track) that had dragged forwards and out of the track (the front section of the track). My research thus showed that there were not two types of small-bodied dinosaurs, just one: that of the ornithopod Wintonopus track-maker.

Whether all the Wintonopus footprints were made during one or many events can’t be determined with the overprinting method used on the vegetation drag marks above, partly due to the small size of the tracks. These footprints all look very similar to each other, so how could one know if these small dinosaurs moved seconds apart or days?

We realised that the solution came from acknowledging that Lark Quarry had been a riverbed that experienced fluctuations in water levels at the time the tracks were made.

When examining the small tracks, we found classic examples of dinosaur swim traces: three parallel scratch marks. These represent the claw marks from the three toes on the dinosaur foot as it clawed the river bottom while afloat. Other swim traces had the toe impressions aligned vertically, perhaps caused by a nearly fully-buoyed dinosaur punting the river bottom with its feet.

Since footprint size can be used to calculate a dinosaur track-maker’s hip height, if the swimmer’s leg can reach the river bottom, then these dinosaur swim traces can be used to estimate the water depth at the time when they were made. My results showed that some of the small dinosaurs were swimming during times of various water depths, some being around 14 cm, others when it was greater than 40 cm, and still others swam at depths in between. Consequently, the Wintonopus tracks were not made during a single event, but by ornithopods crossing the area at different times under conditions of fluctuating water levels.

Mind you, not all the small-bodied dinosaurs were swimming. Some may have been wading, others may have run when the water was shallow or even when the ground was fully exposed to air. This means that hours or even days may have passed between track-making events. While it wasn’t a stampede, Lark Quarry may have been a suitable river crossing or a dinosaur highway available for use in both dry and wet conditions,

Lark Quarry is a national treasure, a site where the amazing behaviour of some of Australia’s dinosaur track-makers is written in stone, and one that everyone can visit it. This April marks the opening of upgraded Lark Quarry visitor facilities. These include a second viewing platform that allows more tracks to be seen. While you may still hear the tale of small-bodied plant- and meat-eaters joining forces as a herd to stampede away from a predatory dinosaur, keep in mind that “it is a great story”.

Anthony Romilio works at The University of Queensland researching footprints along Western Australia’s “dinosaur coast” near Broome.