Scientists search for death’s aroma

Mary Cablk and her dog Inca spent a week last September searching the ashes of remote California towns ravaged by forest fires. They were looking for people who had died as their homes and everything in them burned to the ground.

Cablk had just finished training Inca, an 18-month-old Malinois, to alert her to the presence of odors wafting from human remains. The weeklong trip from their home base in Reno, Nev., to help a California disaster response team was one of Inca’s first major assignments, literally a trial-by-fire.

When they arrived at one home reduced almost entirely to ash, instead of the odor of burnt wood and smoke, Cablk and Inca were enveloped by the scent of a propane tank leak. Cablk’s heart sank. All she could smell was the sulfurous stink of ethyl mercaptan, a compound added to propane to warn humans of a tank leak. A dog’s nose is much more sensitive than a human’s. If Cablk was overwhelmed with ethyl mercaptan, how was Inca going to smell anything else?

“But not only did Inca locate the human remains—just tiny bone fragments—in the background of all of the stuff that had burned in the house—clothes, food, the walls, the insulation—but she did so with a face full of propane odorant,” says Cablk, an olfaction scientist at Nevada’s Desert Research Institute in Reno. “It was incredible.”

Just 15 mg of human tissue, blood, or bone can be enough to alert cadaver dogs to the location of a corpse, says Arpad A. Vass, a forensic anthropologist at the University of Tennessee’s Law Enforcement Innovation Center. Although they aren’t perfectly accurate, properly trained cadaver dogs can focus on human remains while ignoring the corpses of other dead animals nearby, a fact that’s given scientists hope that they might identify a unique scent for decomposing human flesh—a human odor mortis, so to speak.

Vass and Cablk are among a small community of researchers trying to find the macabre cocktail of chemicals that form that unique scent. With this cocktail in hand, trainers could teach cadaver dogs to find human remains without having to use decomposing human flesh, which can be illegal to possess in some countries, including the U.K. In those countries, trainers often use the commonly touted “death scent” molecules putrescine and cadaverine or decomposing animals, with mixed results. “If you train a cadaver dog on pig remains, that dog is going to be really good at finding dead pigs—but not humans,” Cablk says.

But a human cadaver aroma would be more than just a training tool. It could also be used to build an electronic nose, a detector that could be used routinely by law enforcement during searches for missing homicide victims buried in shallow graves or could be deployed on a drone into a disaster zone. “Think about earthquakes or tsunamis, where hundreds or thousands of people suddenly disappear,” says Jan Tytgat, a forensic toxicologist at the University of Leuven. Having a cocktail of chemicals that pinpoint decomposing human bodies would have worldwide benefit, he says.

This goal of finding the stripped-down, unique odors of human corpses may sound straightforward—just a matter of measuring the molecules wafting off dead people and distinguishing them from the volatiles emanating from dead animals. In practice though, the aroma of decomposing human cadavers can be just as complex as the personalities that once inhabited those bodies.

For starters, the chemicals wafting off a dead body are not stable over time because cadavers transition through many stages of decomposition. Moments after death, carbon dioxide levels rise in a corpse’s blood, and the body’s pH begins to drop. The brain’s neurotransmitters begin to pool, and enzymes that normally break down materials in the body are no longer held in check.

Amazingly, Vass says, iridescent blowflies are attracted to corpses for egg laying within minutes of death by some still-unknown chemical, even when there have barely been any noticeable changes to the body.

Soon, a corpse’s cells begin to spill out their nutrient-packed interiors, and the trillions of microbes inhabiting a person’s body take note that the immune system is no longer operational. Thus begins a cadaver’s bloat phase, when the body’s own microbiome feasts on its proteins, carbohydrates, and lipids, breaking down these biological building blocks into component parts and creating internal gas that cannot escape through the usual exit valves. Insects, bacteria, fungi, and animals from the environment or soil join the feeding frenzy as the body loses its human form. Depending on the burial conditions, after weeks, months, or years, just a skeleton, or sometimes a mummy, remains.

Through all these stages, a wide variety of volatile chemicals wax and wane: Hydrocarbons, aldehydes, ketones, nitrogen-heavy compounds, sulfides, and organic acids are produced by the decomposing flesh. Among the most commonly measured chemicals in the air above a cadaver are dimethyl disulfide, a disagreeable garlicky odor; toluene; and p-xylene. Curiously, the commonly touted corpse chemicals putrescine and cadaverine aren’t always measured in the air around a dead body. They may be present within the corpse, Vass says, but they don’t always become volatile and escape.

The incredible diversity of chemicals emanating from human remains isn’t just a function of time and how far a corpse has decomposed. “So much depends on what’s in the environment of a corpse,” Cablk says. For example, decomposition in the presence or absence of oxygen produces different odor compounds. That means a body buried in 1 meter of sand, which has access to air, isn’t going to emit the same chemical signature as a body buried in 1 meter of clay, which is sequestered from oxygen.

Temperature is also a consideration: Not only can it accelerate or decelerate decay, temperature can also change a corpse’s odor profile because different decomposing microbes and insects thrive in different climates.

Even a person’s diet during life can impact the odors emanating from bodies postmortem. Vass and his colleagues have measured a variety of fluorinated compounds wafting off corpses, including trichlorofluoromethane and carbon tetrachloride—molecules, he says, that are likely a consequence of drinking fluoridated water.

For all these reasons, it should come as no surprise that different researchers report different lists of candidate chemicals for a human odor mortis.

Part of the problem is that human decomposition aroma is inherently variable, as is the aroma of the decomposing animals being used as references: To isolate a human-cadaver-specific chemical cocktail, researchers often compare the human decomposition odor with that of dead animals, hoping to find key molecular differences. They also look for other trends. For instance, it could be ratios of volatiles, instead of specific compounds, that distinguish human and animal remains from one another.

But which animal to choose in these comparison experiments? Odors of decomposing pig, dog, chicken, sparrow, sturgeon, frog, turtle, deer—even ground beef—have all been used for comparison with human decay. The more animals used in comparison, the better the likelihood that human-specific markers or trends will emerge—but researchers can’t test every animal in nature.

With all these complicating variables, it might seem that researchers will never pinpoint the precise cocktail of human chemicals that canines recognize in multiple landscapes.

“It is very complicated. But I don’t think we should just throw up our hands and walk away,” Vass says. “The dogs do it somehow. Given enough people interested in doing systematic research, we’ll come up with a more universal profile of what the human signature is.

“Unfortunately there are very few people who do this type of work,” Vass continues, and that’s partly because “a lot of people don’t have access to human remains.”

To do a systematic study in enough different environments and with enough repetition to get statistical certainty would require a lot of dead cadavers. In addition to the extensive ethical and legal permissions required to work with human corpses, not a lot of people donate their bodies specifically to the study of postmortem aromas. And there’s a lot of competition for cadavers donated to science.

Vass is affiliated with one of the few research institutes in the U.S. focused on human decomposition. The Forensic Anthropology Center at the University of Tennessee, Knoxville, has a two-acre facility for studying all manner of decomposition science, from the insect populations that feast on human corpses to what happens to a body decomposing in cement.

Vass and his collaborators have carried out some of the most systematic corpse odor studies to date in realistic conditions at the institute: They sampled four individuals over a four-year period in outdoor graves that ranged from about 0.5 meters to 1 meter deep (J. Forensic Sci. 2008, DOI: 10.1111/j.1556-4029.2008.00680.x). Sifting through the hundreds of molecules that wafted from these corpses, they detected 30 molecules with mass spectrometry, including tetrachloroethene, important for human decomposition.

Ideally, Vass says, researchers should be using human remains, they should be using the entire body, and they should be doing systematic comparative studies under a variety of burial conditions. But not everyone has a facility like the one in Tennessee at their disposal.

That’s why some in the field have opted to use dead pigs as proxies for human cadavers. Cablk takes issue with using pigs. In particular, when she and her colleagues compared the odor of decomposing pigs, chickens, and cows with that of humans, they found that the pig aroma was most different from humans, while the chicken aroma was most similar (Forensic Sci. Int. 2012, DOI: 10.1016/j.forsciint.2012.02.007).

Others in the field wanting to stick with human decay have opted to measure volatiles emanating from cadavers in morgues and crematoriums, or they’re studying the decomposition of human parts left over from other scientific endeavors.

When Tytgat and his colleagues at the University of Leuven measured volatiles emerging from decomposing human body parts in glass jars in a lab, they added intestinal bacteria to more realistically seed the early stages of the decomposition process (PLOS One 2015, DOI: 10.1371/journal.pone.0137341). But the lab experiments don’t offer the full spectrum of soil bacteria, insects, and carnivores that would normally participate in decomposition. In defense of the more simplistic decomposition setup, Tytgat argues that “to be systematic, we have to start somewhere.” He says there’s value in starting simply and then progressing to more complex outdoor experiments, such as those his team is now planning.

As researchers struggle for access to cadavers and to obtain permission to do decomposition experiments in realistic environments, many searching for human remains continue to rely on the mysterious nasal biology of cadaver dogs. Although their record is not perfect, these trained canines have an ability to locate tiny amounts of human material, even when masked by other competing scents that defies explanation, Vass notes. Their successes “still exceed the ability of our best instrumentation.”