Experts sift the ashes, seeking clues about what causes raging firestorms

A trained firefighter, an educator and an ecologist, Halsey was hardly a wide-eyed bystander to the fires that had begun erupting across the region. Still, even to his trained eye, it did not appear that the flames were a threat to his home in this San Diego County city.

The next morning, they were in his backyard.

"The whole world was in flames," he said. "It was just: boom, boom, boom, boom, boom. And you're thinking: What is going on?"

The result, by the end of the week, was no mere wildfire. It was a full-fledged firestorm of 35 simultaneous fires, 20 of them sizable blazes requiring a significant dedication of firefighting resources.

It stretched from north to south along more than 200 miles of coastline and so far east that it encompassed the full spectrum of Southern California's ecological diversity: the Mediterranean waterfront of Malibu, the piney woods of Running Springs, the rugged peaks of the San Diego mountains.

Such events are rare enough that firefighters, ecologists and meteorologists don't really have a name for them, but they have happened before, and no one is entirely sure how to predict their behavior.

"If you've been in fire long enough, it makes the hair on the back of your neck stand up," said Halsey, founder of the California Chaparral Institute, which promotes understanding of the region's natural habitat. "You don't know where it's going to start. But it's going to start."

Scientists are looking to delve into the issue. Some are planning to fly over recently scorched hills in an effort to develop more detailed maps of the fires.

In doing so, they may be able to determine whether fires burned "clean," temporarily eliminating the brush inside the perimeter, or whether there are still unburned "islands" of volatile brush. The information could play into a divisive debate among ecologists about whether communities near recently burned patches are immunized, to some degree, from the next round of fires.

Others are hoping to assemble detailed accounts of regionwide fire "events" -- not just the October fires but past conflagrations too. They'll soon begin overlaying fire maps with additional data, including moisture levels in plants and trees, sources of ignition and maps of smaller fires that burned out or were stopped before they grew into significant blazes.