A need for studying the science of fire
Dan Drollette
Tuesday, 06 October 2009 08:52 UTC
A few issues ago, iSGTW did a story, Burning down the house, about how supercomputers are used to provide hard, scientific, fact-based data on what happens when a burning building “explodes” from a heat tolerable to humans into a super-hot “flashover” of 700 degrees C or more. (To do this work, researchers hooked up gauges, sensors and videocameras to a mock apartment inside a controlled environment, and then set the “apartment” on fire. The data is then analyzed via computer to see what happens from one split-second to the next.)
(Photo of warehouse fire, courtesy the London Fire Brigade)
A fascinating, if rather academic exercise, perhaps?
Now it turns out that getting hard, scientific facts about fire has very serious real-world implications.
According to a lengthy feature story in the September 7 issue of The New Yorker, a man in a Texas prison may have been wrongly executed by authorities on a charge of criminal arson in the case of a housefire in which his two children died. Much of fire investigation has been more art than science, says the article; the “evidence” for arson is based largely upon rules of thumb and old wives’ tales, with little scientific study behind it all.
Superhot fires were nearly always thought to be the result of arson, with the classic evidence being things such as a “brown stain” on the floor (supposedly indicating the use of flammable liquids) and crazed window glass. But independent research has found that such stains occur whenever rust or gunk from charred debris has mixed with water, such as that from a fire hose. As for crazed window glass, it happens any time that very hot glass encounters cold water — such as that from a firefighthers’ hose.
(A similar article on the emerging trend towards the science-based study of arson and fire appeared in New Scientist about a year or two ago.)
A thought to ponder.
Updated 06 October 2009 11:45 UTC
-
Replies
-
I also read the New Yorker article, and was surprised by how little experiment had gone into science forensics a decade ago. Its nice to see that this is changing. Given the consequences of convicting the wrong person, more work on this seems vital. And it doesn’t cost that much compared to lawyers fees for court cases and appeals.
-
Anonymous
Apparently this topic is getting more interest, as a forensics investigator condemns the “junk science” used in the case of a Texas man executed on a charge of arson, says an article in American Prospect. More and more it looks like he may have been innocent.
Getting good, hard facts on the science behind fire is long overdue. Three cheers for the Edinburgh Supercomputing Center and their data-collection and modeling of fire.
-
Just ran across this October 1 article in The Guardian, which said that a "51-page report, written by a nationally-recognised expert on fire safety, Craig Beyler, tore apart the original case against Willingham (the Texas man executed on a charge of criminal arson) on virtually every count. It found that the key evidence upon which the conviction was based had no basis in modern fire science and that ‘a finding of arson could not be sustained.’
The report was particularly critical of one of the fire inspectors, who has since died, saying his findings were ‘nothing more than a collection of personal beliefs’ and more ‘characteristic of mystics or psychics.’ "
. . . the specific burn patterns on the floors and skirting board of the house upon which the initial fire inspectors had largely based their opinion were based on assumptions that had been overruled by modern scientific experiments. He said these assumptions had no basis in good investigative work even back in 1991, and suggested the fire could have been the result of natural causes, pointing to a space heater in the children’s room.’ "
The rest of the story is here.
-
