Forensic Fire Investigations Need More Scientific Input, AAAS Report Finds

Visit the AAAS Force for Science website to follow the latest updates related to AAAS advocacy activities.


A comprehensive AAAS forensic science report pinpoints gaps in the science behind fire investigations, outlines topic areas in need of research. The report covers both on-scene investigation practices and the laboratory analysis of debris. | State Farm/Flickr (CC BY 2.0)

Determining if a fire is accidental or due to arson is a process often plagued by poorly understood science, subjective judgments on the part of investigators and inadequately trained personnel, according to a new AAAS report on the quality of fire investigation in the United States.

The report, the first of two forensic science assessments organized by the AAAS Scientific Responsibility, Human Rights and Law Program, examines current practice in fire investigation, identifies gaps in scientific knowledge and makes recommendations for further research. It is particularly timely given the deadly fire that engulfed West London’s residential housing apartment building – the Grenfell Tower – in the early morning hours of June 14, taking at least 80 lives.

The report covers the two phases of fire investigation – the on-scene effort to determine the origin and cause of a fire and the laboratory analysis of debris from the fire, which can provide important information contributing to a conclusion about the cause of the fire. It was written by a working group that included a fire investigator, two analytical chemists, a human factors expert and a fire engineer.

The wide-ranging report explores inaccuracies in the existing literature about fire investigation that can affect the beliefs and behavior of investigators; assesses laboratory analytical methods and the state of computer modeling of fires; weighs the value of dogs over current electronic “sniffing” devices for finding residues of ignitable liquids at fire scenes; and stresses the need for more use of controlled test fires under a wide variety of conditions to produce “ground truth” for fire investigators.

The AAAS project is also evaluating the status of latent fingerprint analysis. Valid forensic analysis is an essential tool in apprehending criminals, but the reliability of many of the forensic sciences has been challenged in trials, expert reviews and congressional hearings. Reports and testimony based on substandard science have contributed in some cases to convictions of individuals later proved innocent through DNA testing.

“As advances in fire science are being adopted, investigators and attorneys are revisiting old cases,” the AAAS report notes. “And what they are finding is chilling.” John Lentini, a forensic scientist who co-authored the AAAS report, estimates that “a couple hundred” people have been wrongly imprisoned based on the use of invalid fire science. According to the National Registry of Exonerations, 63 individuals have been exonerated since 1991.

“We will never know for sure how many people have been wrongfully convicted” based on invalid practices, the AAAS report says. “Likewise, we will never know how many criminal arsonists have not been indicted because of inadequate forensic tools.”

The new report, aimed at prosecutors and defense lawyers, law enforcement officers, academic scientists, fire investigators, funding agencies and policymakers, is meant to help change that situation and provide improved confidence in the criminal justice system.

Fire has been extensively studied, but the complex chemical and physical processes involved still are not fully understood, the report says. That is particularly true for fire scene investigation, where interpretation of the evidence too often comes down to subjective evaluation by fire investigators who may lack adequate scientific training and familiarity with the cues that can help truly pinpoint the origin and cause of a fire.

Burn patterns can be confusing and misleading, particularly when an enclosure fire grows to the point of flashover and the room or enclosure becomes fully involved. Flashover, the near-simultaneous ignition of directly exposed combustible material in an enclosed area, is considered the transition from “a fire in a room” to “a room on fire.”

Modern furnishings, particularly upholstered furniture made with polyurethane or polyester fiberfill cushions, can burn so quickly that a room can be fully engulfed in flames in less than five minutes, the report says. That can mimic the rapid spread of an arsonist-set fire that is fueled by an accelerant such as gasoline. Accordingly, the speed with which a fire spreads in modern residences cannot be used alone to classify a fire as accidental or incendiary, according to the report.

Post-flashover fires can create new ventilation-generated burn patterns while obscuring existing burn patterns, the report says, adding that “the longer a fire burns in a fully involved condition, the more difficult is the determination of the correct area of origin.” One study cited in the report found that fires burning for even a few minutes beyond flashover produced fire patterns that caused error rates among practitioners in excess of 75% in determining the origin of a fire.  In one experiment, only 3 of 53 participants were able to identify the room quadrant of origin for a test fire that burned for two minutes beyond flashover.

“The most useful information sometimes comes from eyewitnesses or security cameras, not from fire patterns in compartments that burned for more than a few minutes beyond flashover,” the report notes. Research during the past decade has shown that, in many cases, the ability of a fire investigator to use burn patterns to correctly identify the origin of a fire in a fully involved room is no better than chance.

While on-site fire investigation remains challenging, the laboratory study of debris taken from a fire scene has become one of the most fully standardized fields within forensics, according to the report. Nevertheless, there is more work to be done, given the changing nature of consumer products and ignitable liquids, to better understand the residues left after a fire. The report calls for more research on sample identification and collection at the fire scene. Laboratory analysis is irrelevant, it says, if a sample is incorrectly identified or collected. And when it comes to identifying samples for collection, the report says, “accelerant detection” dogs still perform better than available electronic “sniffing” instruments in locating residues of interest.

Perhaps most importantly, the report stresses the human factor in fire investigation. The determination of the origin and cause of a fire necessarily involves some subjective judgments and investigators might be “biased” by information not relevant to their task. To minimize and mitigate such biases, the report recommends separating the fire scene investigation from other components of the fire investigation. Further, given the amount of “widespread, persistent and inaccurate literature affecting the beliefs and the behavior of many investigators,” it says, there is a crucial need for improved education and proficiency testing for fire scene investigators.

“Over the past three decades, much has been learned about fire behavior and investigation, but the distribution of the knowledge among field investigators, many of whom have only limited scientific education, has not been uniform,” the report notes. More research is needed on the effects of education and training on fire investigators’ ability to determine fire origin and cause. Meanwhile, “there is no scientific basis for concluding that the accuracy of certified fire investigators, in particular, is better than the accuracy of non-certified fire investigators,” the report says.

The use of test fires will help establish a “ground truth” against which the validity of investigators’ conclusions can be assessed, the report adds, while acknowledging that broader use of test fires will be costly. The authors note that reasonable substitutes are available, with many municipalities in the process of razing buildings that could be used instead for controlled fires under a variety of conditions.

The AAAS report was not intended to reach a conclusion regarding the acceptability of a fire scene investigation as evidence in a court of law. It does call for rigorous scientific research to better understand the strengths and weaknesses of fire investigation. And it recommends that investigator-initiated research be “from a broader range of disciplines and institutions than have traditionally been engaged in the study or work of the forensic sciences.” It cites the value of disciplines such as physics, psychology, statistics and engineering and expresses the hope that a culture of science will become more engrained among forensic practitioners.

[Associated image: State Farm/Flickr (CC BY 2.0)]