Putting forensic science on a sound statistical footing is the aim behind three grants totaling over $2.2 million awarded to the UC Davis program in forensic science by the U.S. Department of Justice.

The grants: $1.4 million over three years to develop a bullet-matching database; $700,000 over two years for studying the impressions left on cartridge cases by firearms; and $150,000 over two years to establish whether torn pieces of duct-tape can be reliably matched.

The broad aim of all three grants is to develop methods to associate probabilities with forensic evidence, said David Howitt, professor of chemical engineering and materials science at UC Davis. He is co-principal investigator on the two firearms-related grants with staff researcher Karen Cebra and Fred Tulleners, director of the campus's graduate program in forensic science.

"Based on the extent of similarities, is the likelihood of a match one in a hundred, one in a thousand, one in a million?" Howitt said.

In a report published earlier this year, the National Academies of Science called for more research to establish the scientific basis of forensic methods and develop ways to measure the uncertainty in forensic conclusions. The UC Davis program aims to meet that need, Tulleners said.

In the largest grant, the researchers will develop a database of 10,000 bullets from Northern California crime labs. They will use a technique called confocal microscopy to study the surface features of bullets, and identify key characteristics that can be used to compare bullets in a statistically accurate way.

Comparing photographs of bullets can be misleading because of differences in how the pictures are taken, Howitt said. And digital images contain so much data that the files become very large, requiring more powerful computers and causing statistical problems.

Instead, the UC Davis team plans to identify key elements that describe a bullet and that can then be entered in a database and compared with the "signature" of another bullet.

"It's much easier to make comparisons if you know what you're looking for," Howitt said.

The $700,000 grant will look at the impressions left on cartridge cases by the breech face of a gun. Although guns are mass-produced items, Tulleners said that differences in the finishing process mean that they expect to find variations in the marks left by breech faces of different guns, even when the same parts were manufactured consecutively under the same controlled conditions.

In the third grant, Tulleners and Jerome Braun, senior statistician in the Department of Statistics, will look at whether torn pieces of duct tape can be matched reliably.

Duct tape is often used to bind kidnap or murder victims, and remnants of tape can be recovered from crime scenes. At present, duct tape comparison is purely subjective.

"We want to provide some statistics that prove a duct tape tear is unique," Tulleners said.

The graduate program in forensic science offers a research-oriented Master of Science degree taught by an interdisciplinary group of UC Davis faculty and outside experts. The program provides a thorough training in the latest scientific techniques and forensic methods, equipping graduates to work in forensic science laboratories. It is the only graduate degree program of its kind in the state and currently has an enrollment of about 85 students.