Preliminary Report Suggests Baltimore Aerial Investigation Research Pilot Was Associated with Small Improvements in Solving Crimes

A preliminary report about an effort to use aerial surveillance to aid police investigations in Baltimore finds that the effort was associated with small increases in the rate at which police solved serious crimes, but an overall evaluation of the program will require a wider review of citywide police efforts, according to a new RAND Corporation report.

The report does not draw conclusions about the effectiveness of the Aerial Investigation Research pilot that ran from May to October 2020.

RAND researchers concluded that preliminary evidence on case outcomes suggest the program may have helped police solve an additional 11 serious crimes during the six-month trial.

Making conclusions about the overall effectiveness of the program will require an analysis of citywide crime and policing outcomes—information that was not yet available for the preliminary evaluation. RAND researchers expect to release an assessment of the overall effectiveness of the program in April 2022.

“Our report outlines a few preliminary findings based on case outcomes to date, but some of these outcomes may change as the Baltimore police continue to investigate crimes that occurred during the pilot period,” said Andrew Morral, lead author of the report and a senior behavioral scientist at RAND, a nonprofit research organization. “This is an innovative intervention and our evaluation of how crime and police investigations changed in Baltimore from before to during and then after the pilot will require another year to complete.”

The Aerial Investigation Research program involved having airplanes equipped with powerful cameras fly above the city during daylight hours, capturing a second-by-second record of outdoor events.

Images taken from the aircraft could be used to support crime investigations by allowing analysts to track suspects as they approached and departed the crime scene. Although image resolution was too poor to identify individual's characteristics, by tracking them through the city analysts were often able to identify when suspects or their vehicles passed CitiWatch closed-caption television cameras or police license plate scanners, which could provide more detailed information about suspects' identities.

The Aerial Investigation Research pilot program was designed to help police investigations of four types of crime—homicides, nonfatal shootings, armed robberies, and carjackings.

Although the aerial surveillance program was planned to operate with three aircraft monitoring daytime events across most of Baltimore city, the pilot launched with just a single plane, a second was added nearly two months into the pilot, and a third was never employed in the surveillance activities.

In part for these reasons, and in part because the planes only operated during the daytime, the program was able to provide evidence on just 10% of the crimes it was designed to help investigate.

A large fraction of crimes that are solved in Baltimore end in an arrest either at the crime scene or shortly after the crime occurs. These less-difficult cases are not crimes that typically require expanded investigatory tools like aerial surveillance.

When researchers focused on just the subset of crimes that were not solved within one day, they found that those with aerial evidence were cleared by police at a rate 7 percentage points higher than similar cases that received no aerial evidence. This is a statistically significant difference in clearance rates.

To evaluate the citywide effect of the Aerial Investigation Research pilot program on crime and clearance rates, RAND researchers will need to examine the program's effect on all crimes and their investigations. This is because the aerial program might affect when crimes are committed (for example, shifting crime to nighttime hours when the program did not operate) or it could improve or reduce the efficiency of police investigations, increasing or decreasing the policing resources available for cases that did not receive aerial evidence.

Support for the RAND research was provided by Arnold Ventures, which also provided funding for the Aerial Investigation Research pilot program.

The report. “Preliminary Findings from the Aerial Investigation Research Pilot,” is available at www.rand.org. Other authors of the report are Terry L. Schell, Brandon Crosby, Rosanna Smart, Rose Kerber, and Justin Lee.

The RAND Justice Policy Program conducts research across the criminal and civil justice system on issues such as public safety, effective policing, drug policy and enforcement, corrections policy, court reform, and insurance regulation.