Credit: AFP
In the mid-1990s, Terry Wallace, now at America's Los Alamos National Laboratory, was part of a team studying earth movements in the Andes Mountains along the border between Bolivia and Chile. Occasionally, one of the seismometers would record strange micro-earthquakes, less than 1.0 on the Richter scale. Small temblors are common, but these occurred only late at night in a remote plain, 4,000 metres above sea level.
Most people associate seismology with earthquakes, but Wallace knew that seismometers aren't picky about the source of vibrations. Given the remote location, he wondered, could he be detecting trucks, carrying contraband though the desert?
Poring over the seismograph traces, he came to believe he could even determine in which direction the smugglers were going, based on the fact that from one side, they would be approaching across flat land, whereas from the other, they would be going uphill. On the upgrade, the trucks would have to gear down, producing different vibrations.
So, one night, he hid and watched as trucks, running without lights, rolled across the plain. Staking out South American smugglers in a remote desert wasn't, he would later admit, the smartest thing he'd ever done. "But it was fun," he said. More significantly, it was an early exercise in what has come to be known as 'forensic seismology.'
Forensic seismology originated during the Cold War as a means of monitoring the enemy's underground nuclear tests. But it is a field whose uses are as broad as the imagination.
In 2000, Emile Okal of Northwestern University in Ohio used it to track a gigantic iceberg as it floated away from Antarctica, scraping across the sea bottom as it went. Vera Schulte-Pelkum, from the University of Colorado, in Boulder, has used it to monitor storm waves crashing into Canada's Labrador coast so hard that they show up on seismometers in California.
But the most interesting investigations involve airline crashes, mine disasters, and explosions.
The key to such studies lies in the fact that seismometers are common. By one estimate, a few years ago there were 16,000 of them permanently installed around the globe, many continuously uploading to the Internet.
Once a seismometer is installed, it's always tuned in. "You may be trying to listen to one thing, but the 'noise' may tell you something unexpected," Wallace has said. "We're just beginning to eke out every wiggle on the seismogram and what it's telling us."
Canadian seismologist David McCormack set the groundwork for this type of analysis in the investigation of Pan Am Flight 103, brought down in 1988 over Lockerbie, Scotland, by a terrorist's bomb.
Nearby seismometers recorded six impacts, indicating that the plane had broken into that many pieces before it hit.
This was already evident from the pattern of debris at the crash site, but the seismic data allowed McCormack to calculate the kinetic energy of each piece.
Comparing this to the weight of the recovered fragments allowed him to determine each piece's speed at the time of impact – information that could be used to reconstruct the explosion that ripped the jetliner apart. Such analyses are now done for other major crashes, at least when they occur on land.
On 19 August 2000, a dozen people were on a family reunion, camping beside a river in southern New Mexico, USA, 200 m from an underground natural gas pipeline. What they saw and heard isn't entirely known, but most likely, there was a noise like a clap of thunder followed by a shower of rocks and stones.
Seconds ticked by and nothing else happened ... until the earth erupted in a gout of flame, accompanied by an apocalyptic roar that went on and on and on. At the pipeline, the temperature hit 1,200 °C. At the campsite, some of the campers lived long enough to be rescued an hour later, but all subsequently died.
One of the people called in for the ensuing investigation was Keith Koper, a seismologist at Saint Louis University in St. Louis, Missouri, USA.
Drawing on records from seismometers as far as 180 km away, Koper was able to determine that there had actually been three explosions. First came the pipeline blowout – no fire, just a blast as the pipeline ruptured. It appeared on the seismograms as an explosion-style jolt, followed by a continuous high-frequency vibration from gas jetting out of the hole. The fire had not yet started.
Then, 24 seconds later, there was a second bang, followed by a roaring-style vibration that must have been created by flames gouting into the air. That was followed 17.8 seconds later by a third explosion, probably from a pool of gas not ignited by the initial spark. At this time, the roaring of the flames stepped up another notch, as the pipeline was converted into a giant torch.
Koper's most important finding was the 24-second delay between the blowout and the initial ignition. That's about the length of time needed for a breeze to waft gas from the pipeline to the campsite. Koper suspected that the gas was then ignited by something at the campsite, perhaps a campfire or cook stove.
Nobody argues that the campers were to blame for the fire. But a quirk of New Mexico law allows punitive damages to be assessed on the basis of how long people are placed in life-threatening danger before they are injured — in this case, the length of time between the pipeline rupture and the start of the fire.
The amount of the eventual legal settlement was not made public, but Koper has said that the 24-second figure played a significant role in determining its size.
In another study, Wallace and Koper collaborated to use seismic data from Norway, Sweden, Finland, and Spitsbergen to reconstruct the sinking of the Russian submarine Kursk on 12 August 2000.
When the sub sank, the Scandinavian seismometers showed two events a little more than two minutes apart. Underwater explosions produce bubbles of hot gases, which oscillate as they rise.
The frequency of these oscillations depends on the size of the explosion and the depth at which it occurs. This allowed the scientists to determine that the Kursk was sunk by an explosion, not a collision, and that it had occurred when the sub was at a depth of 83 m.
The first blast carried power equivalent to 250 kg of TNT; the second was several times larger. Wallace and Koper therefore suspected that the first was a torpedo misfire during live-fire exercises, which the Kursk was known to be undertaking at the time. The second, larger explosion, they believed, probably occurred when fire from the initial accident detonated additional warheads.
Shortly after the sinking, the seismograms revealed many other small seismic events, with magnitudes between 1.25 and 1.86. These appear to have been depth charges dropped by the Russians to discourage other nations from sending scuba divers to spy out the secrets of the stricken submarine. The seismic readings were so precise that Wallace and Koper could even track the speed and course of the naval vessel used to lay down the underwater barrage.
Ever-vigilant seismometers were also on duty during the 9-11 terrorist attacks on the World Trade Centre. Because tall buildings are built on deep footings, energy from the impact shivered down the spine of these structures, directly into bedrock. The impacts were then recorded by seismometers many kilometres away.
The collapses of the buildings also produced seismic traces. These traces were so detailed that it is possible to identify the crash of each floor as the buildings pancaked downward in a series of rat-a-tat-tat impacts, each stronger than the one before as collapsing floors added more mass to the falling debris.
Scientifically, it is fascinating that these records are so incredibly detailed. But looking at them is also disturbing: in its way, just as disturbing as watching the famous videos of the collapsing towers.
Perhaps all forms of forensic science are inherently disturbing because they are tools for reconstructing events that are often quite terrible. But forensics carries the fascination of solving a good mystery, which, undoubtedly, is what fuels the appeal of television shows such as "CSI."
It is also probably what keeps the best forensic seismology stories circulating at scientific meetings. In 2007, the Geological Society of America scheduled a symposium on forensic seismology, and the stories of the Kursk, the New Mexico pipeline, and the World Trade Centre collapse were still among the stars of the show.
Richard A. Lovett is a contributing editor of Cosmos based in Portland, Oregon.
