Dino soars in the farmyard

Palaeontologist David Norman explains how working on a TV series recreating dinosaurs on computers threw up some unexpected insights

There can be little doubt that nowadays dinosaurs are big news. Since the release of the blockbuster film Jurassic Park, with its amazing computer-animated creatures, the media has paid a lot of attention to dinosaur-related news: whether it be about feathered fossils in China, nests of eggs in Mongolia, gigantic remains in Argentina or bizarre-looking dinosaurs from Africa and England.

But while this latest wave of media interest has given palaeontologists the chance to air their research to a wider audience, this is by no means the first time dinosaurs have grabbed the headlines. In 1854 dinosaurs burst on to the public scene when huge life-like models were displayed for public view in the landscaped grounds surrounding the re-opened Great Exhibition pavilion, "The Crystal Palace" in South London.

This publicity coup was masterminded by the sculptor Benjamin Waterhouse Hawkins and Professor Richard Owen - the scientist who "invented" dinosaurs in 1842. It is now hard to imagine, but tens of thousands of people from all over the world paid good money to come and gaze in awe and wonder at these marvellous beasts. Dinosaurs entered the popular language and dinosaur names even entered the novels of Charles Dickens.

We can still see those Victorian creations (Iguanodon, Megalosaurus and Hylaeosaurus) in Crystal Palace park and be amused by their appearance - they look benign rather than fearsome as their names imply and a little quaint. They are inaccurate in the light of modern knowledge. As time moves on I wonder if that is how future generations will regard the Jurassic Park icons: Velociraptor, Brachiosaurus and Tyrannosaurus.

The foundation for any image of a long-extinct creature has to be the fossil remains and, built upon those, the framework of interpretation provided by a combination of palaeontological science and technology. Our Victorian predecessors had rather few scraps of bone to work on, little technology to help them and no "blueprints" for comparison. Today we have complete skeletons, equipment to help prepare them and comparatively sophisticated technology to assist our interpretation and understanding.

I feel confident that future generations will not be able to make patronising comments about the appearance of our dinosaurs - although I am sure that they will be sympathetically aghast at our rather primitive mode of display compared with their free-moving "cyborg" dinosaurs programmed to live in Mesozoic theme parks, or their interactive holographic documentaries.

Palaeontologists have recently been playing a key role in ensuring that the images of dinosaurs presented to the public are as accurate as possible - rather than being the mental concoctions of uninformed television producers. This autumn, a six-part natural history series Walking with Dinosaurs will be broadcast by the BBC. It was produced in collaboration with partners in the United States, Germany and Japan. A team including computer animators and palaeontologists from England and the US has recreated dinosaurs behaving naturally in their world.

As a dinosaur palaeontologist (one of very few in England), working on the programmes has been a fascinating learning experience for me. In my research I seek to answer questions such as: how are bones shaped and do the shapes tell us how the dinosaurs moved and lived? Transforming my two-dimensional reconstructions of dinosaurs into 3D objects that move and behave in the virtual environment provided by a workstation proved surprisingly challenging.

Taking detailed, jointed, wire-frame abstractions of my 2D drawings as a starting point, the TV series' "skeletons" were animated by being made to move through a simple walking or running cycle. Factors such as the positioning of the shoulder bones, the curvature of the neck, the flexibility of the spine, the angulation of the hip joint, knee and ankle, and the positioning of the feet and toes all interacted in subtle and unexpected ways and required modifications that I had previously not considered sufficiently.

These interactive sessions between palaeontologist and designer resulted in a significantly more life-like wire-frame skeleton. This was later given a remarkably life-like skin and then introduced into appropriately chosen landscapes, again with considered advice from the palaeontologists concerning the behaviours that were being portrayed. It filled me with envy to see the powerful software and computer hardware utilised on the project, compared with my facilities in Cambridge.

Creating accurate animations of some dinosaurs is an interesting aspect of research work, but the material basis has to be the bare bones upon which we work as palaeontologists, and other (slightly less obvious) traces of dinosaur activity.

Preserved footprints indicate not only how, but where dinosaurs moved: eggs, nests and occasional babies provide surprisingly domestic information about the way of life and growth of some dinosaurs. These are combined in what I like to call (slightly tongue in cheek) a "forensic" research programme. Parallels between palaeontology and forensic pathology are quite striking: both disciplines attempt to reconstruct accurately the appearance, behaviour and circumstances leading up to the death of a creature, and do so using a surprisingly broad range of scientific methodology.

For instance, CT scanners have proved to be of immense value to medical practice as a non-invasive method of probing the human body. In some circumstances CT scanners can also aid palaeontologists. An imperfect but well-preserved skull of the large carnivorous dinosaur, Allosaurus, was discovered recently in North America. This proved amenable to scanning and can now be investigated as a 3D object from serial CT slices.

Emily Rayfield in my research group has been using this information to build a mechanical abstraction of the skull in two and three dimensions as a mesh of finite elements that can then allow the skull to be subjected to stress and consequent strain in ways that would be impossible in the petrified and therefore very brittle and fragile original skull.

This type of investigation adds an important dynamic component to the investigation of the biology and way of life of large predatory dinosaurs and can allow the mechanical behaviour of parts of the skull to be tested in the same way that an engineer can test his design for a new bridge.

But new and revealing insights also come from the bare bones. In recent years, new discoveries have begun to fill the evolutionary gap between dinosaurs and living birds to a remarkable degree. First, dinosaurs of various types have been shown to possess many supposedly distinctive bird-like pieces of anatomy: the supposedly unique "wishbone" or furcula is now known to be present in many dinosaurs; the uniquely bird-like arrangement of hip bones is common among dinosaurs; the way that the legs and tails of birds work is also seen in some dinosaurs. But the real clincher has been the discovery in the past two years of dinosaurs with that archetypically bird feature: feathers. Some rare examples preserved at sites in Liaoning, in the northeast of China, have shown either clear feathers or in other instances very downy coverings in dinosaurs named Sinosauropteryx, Caudipteryx and Protoarchaeopteryx.

It is clear that dinosaurs are still with us today in the form of their feathery descendants that we call birds. So birdwatchers know a thing or two of dinosaur behaviour today - 65 million years on admittedly, and probably highly modified because of their aerial habits. But look a big rooster in its glittering eye, then glance at its formidable scaly legs and taloned feet and deny, if you can, some distant kinship with the dinosaurs.

David Norman is director of the Sedgwick Museum, University of Cambridge. He will deliver the talk, "Did British dinosaurs behave themselves?", at the British Association meeting in Sheffield next week. Walking with Dinosaurs begins on BBC1 in October.