Lidar-made Dinosaurs

Paleontologists are using 3D laser scanning and computer modeling to create replicas of dinosaurs and to capture data from their information-rich tracks. This resulting data propels our knowledge of the prehistoric to new dimensions.

by Phillip L. Manning, PhD, Tony Purslow, and Karl Bates
Paleontology—the study of prehistoric life, mainly through the analysis of fossils—has been transformed and expanded over the past 20 years by the advent of new surveying technologies. Once viewed by some scientists as the equivalent of stamp collecting, paleontology is now at the cutting edge of multidisciplinary research, able to answer many more questions than in the past. A good example is the excavation, preparation, and exhibition of dinosaur skeletons, which is now part of a broader effort to determine dinosaur composition and biomechanics and to develop new preservation methods.

It is difficult to estimate a dinosaur’s weight or map its tracks found on a steep mountainside. However, using 3D laser scanning technology, a research team at the University of Manchester in the United Kingdom was able, over a 12-month period, to digitally capture dinosaur skeletons in museums and dinosaur fossil track sites in the U.K., Germany, Spain, the United States, China, and Argentina. They then used the point clouds generated by the scanner to create computer models and analyzed them quantitatively. 

Requirements and Equipment

Modeling how dinosaurs moved and functioned requires very precise data on the geometry and morphology of skeletons, tracks, and other paleontological finds. Additionally, the Manchester University team knew that for this project it was going to have to travel around the world to scan fossils in a wide range of environments—from deserts to museums, from mountain slopes to floodplains—and that it would have access to many sites for only short periods of time.

Therefore, the team needed a light and portable scanner with high accuracy and resolution and the ability to rapidly collect and store a huge amount of data. It selected the Z+F IMAGER 5006i, a medium-range, phase-based laser scanner with a resolution range of .1 millimeters, a data capture rate of 500,000 points per second, and a 60-gigabyte internal hard drive, manufactured by Z+F, a German company. Vast amounts of data were quickly generated and stored on the scanner’s internal hard drive and an attached notebook. Even the notebook was not required when creating scans beneath skeletons make it awkward to be connected.

For the year of travel, the University of Manchester research team adapted the scanner by mounting it on a carbon-fiber television camera tripod so that they could haul only that single tripod up the mountainside each day. 

3D Dinosaur Models

Paleontologists are interested in techniques to acquire data on the body mass of dinosaurs because body mass is an important factor in understanding the locomotion and behavior of both living and extinct species. Body mass can be estimated with great accuracy by scanning complete or near-complete mounted skeletons and using the data to build 3D digital models. These models can then be imported into image-processing software that enables users to estimate mass on the basis of the relationship between body volume and density.

Thankfully, vertebrates are rather consistent when it comes to the density of their flesh and bone. The scanned skeleton provides a body volume from which you can derive its mass, and then you just need to add weight to account for the flesh. The virtual body volume can also be manipulated to account for subtle changes in body mass that might affect locomotive ability and a species’ center of mass.

In several museums, including Berlin’s Humboldt, Patagonia’s Carmen Funes, and the Manchester Museum, the research team scanned mounted skeletons from a variety of perspectives to provide full 3D coverage and eliminate shadows in the data set. They aligned the segmented right side of each skeleton with the x-axis in Autodesk’s Maya (software for modeling and animation) and mirrored it to produce complete symmetrical models.

Many scans were produced and registered via cloud-to-cloud processes and real-time views, which provided in situ verification of data capture. The team then generated 3D models to produce a high-resolution skeletal framework for use in the locomotion software.

The study showed that the estimated total body mass of four predatory dinosaurs ranged from about 932 pounds (423 kilograms) for the struthiomimus dinosaur to about 8.5 tons (7,655 kilograms) for the tyrannosaurus rex. The largest laser-scanned models represented an increase of 21% to 29.8% in total body mass over best-estimate predictions, while the smallest were 8.2% to 9.9% lighter than predicted. 

Mapping Dinosaur Tracks

Rapid 3D data capture via laser scanning has also benefited the emerging field of geological heritage conservation (geoconservation) by improving site documentation. The University of Manchester team has been conducting research since 2007 at the Fumanya excavation site in northern Spain using a combination of both long- and mid-range lidar scanning units to generate 3D digital outcrop models. These models provide valuable data on the yearly weathering of the site and enable researchers to interpret it better.

The largest track sites in Fumanya are nearly vertical and consist of weak and crumbly mudstones. Earlier attempts to map them used traditional surveying techniques that entailed roping down the face of the track sites, causing too much damage and yielding data of comparatively low accuracy. By contrast, using long-range lidar, researchers were able to map the tracks with much higher accuracy and without affecting their surface.

At one track site at Fumanya, however, the tracks are very shallow. Here, the approximately 5mm accuracy of the long-range lidar unit used was not sufficient to map them reliably, but the higher resolution scanning capability of the Z+F IMAGER 5006i was. The low-relief (approximately 1mm to 3mm) trackway of a baby sauropod is now conserved virtually for future generations to review and analyze.

Researchers produced data from the scanner in ASCII format, co-registered it using LFM Register software, and imported it into the University of Manchester’s CAD system. This enabled the team’s paleontologists, working jointly with computational biologists, to generate a 3D model of each dinosaur that had virtual muscle groups. These were rigged so that the muscle activation patterns could be predicted by a physics simulator. These models created, for all the species studied, detailed 3D virtual dinosaurs able to walk and run much like the real-life ones might have done more than 60 million years ago.

3D Interest

This project generated significant interest in research on how dinosaurs moved and their maximum running speeds. It is included in a six-part series, Jurassic CSI, commissioned by the National Geographic Channel, that explores the application of 21st-century science to paleontology and includes the work of leading scientists from around the world.The University of Manchester research team has just begun to explore the potential of laser scanning in the field of paleontology. If centrally archived, the data it is collecting can transform 2D publications into full 3D ones that leap out at readers from their computer screens. This would sharpen the interpretation of new dinosaur fossils, especially those based on fragmented material, at times correctly indentifying a few previously misidentified species. It opens the possibility of an online museum of 3D virtual dinosaurs for everybody.

Surveyors interested in new markets for their 3D scanning technology might want to consider this possibility. Start by investigating archeology and paleontology research groups associated with universities and museums near you.

Phillip L. Manning, PhD, is with the School of Earth, Atmospheric and Environmental Science at the University of Manchester, Manchester, UK.

Tony Purslow is the UK sales account manager at Z&F UK LTD, Manchester, UK.

Karl Bates
is with the School of Biomedical Sciences, University of Liverpool, Liverpool, UK.

» Back to our April 2012 Issue