Those of us who have spent a lot of time looking at seismographs (and hey, who hasn’t?) know that they can present quite a riddle for the inexperienced. There was a time when even seismologists had trouble figuring them out. Understanding what you see on one is largely a matter of experience over time. This is especially true in the context of volcanic activity.
That was the case in in 1985 when Nevado Del Ruiz erupted generating lahars that wiped out the town of Armero killing 24,000 people. This tragedy is regarded by many as volcanology’s biggest failure. There was no reliable way to predict an eruption, and you couldn’t just tell thousands of people to get out of town when that volcano has not bothered anyone in living memory and it’s more than 40 miles away in the first place. Back then it was just a matter of guesswork for the most part. Even if you saw harmonic tremor it might stop at any time as quickly as it started with no eruption.
Then a really smart guy got to work on the problem. Bernard Chouet and his colleague got to looking over the seismic records of the Ruiz eruption while modeling magma filled fractures in volcanoes, and Chouet had an epiphany. Up until then volcanologists had been looking only at one particular seismic signature, designated the A type event. It looks like this:
That’s rock breaking. Snap! On the seismographs from Ruiz there were thousands of them, and the scientists did their best to find a pattern in them that would give some clue as to when an eruption might occur, but try as they might they could establish no clear pattern because every volcano is different, and rock under pressure breaks at different times and pressures in each instance. So they seemed to be without a clue until Bernard Chouet’s epiphany.
What Chouet had seen in the seismographs was a second type of signal that others had also seen but just didn’t know what to make of it. It came to be called the B type or the Long Period Event, and Chouet instinctively realized what it was. Here is what he saw:
These were from the seismographs on Galeras, a volcano in southern Columbia. They were recorded in 1993, and Chouet recognized them as resonance signatures, like those of a water hammer, vibrations of liquid rock surging through the fissures of a volcano.
Chouet noted that in previous seismic records these had increased in frequency until the point of eruption. They were in fact the signals of a pressurizing volcano, and he would use them to successfully predict two future eruptions, those of the Redoubt volcano in the Aleutians, and that of Popocatepetl in Mexico, where timely evacuation saved many lives.
In 1993 not everyone was aware of Chouet’s discovery, or preferred to rely on more traditional methods of predicting when a volcano might erupt, and one such person was Stanley Williams, who was leading a team of volcanologists and tourists into the crater of the Galeras volcano after it had already erupted once not long before. Now to my mind that’s a pretty foolish thing to do, but for Williams it was the only way, and he managed to talk 11 other people into going with him.
What followed is known as the Galeras disaster. At about 9:30 am on January 14th 1993 Stanley Williams lead his team into the crater of Galeras, where it’s lava dome smoldered and fumaroles gushed volcanic gases. It was the gases that Williams was there for, because he believed (rather obsessively) that measuring the gas output of a volcano was the only real way to predict when it might erupt, and for that you had to be there, right in the center of things, in the crater, even though there had never been a successful prediction of an eruption from gas data.
Williams had in fact been told of Chouet’s discovery, but chose to ignore it, even when he was notified while his team was in the crater that the same long period events were showing up on the seismographs. Personally I think Stanley Williams was very much driven by professional hubris and arrogance, but that’s just my opinion.
The previous eruption of Galeras had occurred only four days after the appearance of the long period events on the seismographs, and they were again appearing during the volcanology conference Williams was at there in Pasto, but he ignored them because to his mind the gas output of the volcano was at safe levels, even though there as a lava dome in the crater, an ominous sign for any volcanologist. Not only did Williams ignore the seismic warnings, he kept them from his team as well. In a later interview Andy Adams, one of the three survivors, stated “I had no prior knowledge of any seismic precursors before entering the volcano. ”
Down into the crater Williams led his team, and at 9:47 am an ominous sound was heard, that of a rockfall in the crater. Two more followed in the space of a minute. Dr. Andrew MacFarlane turned to Williams and asked him about the rockfalls, and immediately after that the volcano blew, killing nine of the team and severely injuring MacFarlane and Williams. It would be two years before Williams could even walk again. It was a flat-out miracle that anyone survived in the crater of an erupting volcano.
After the disaster Stanley Williams claimed to be the “sole survivor” of Galeras on the “NBC Nightly News” and in a number of publications. He kept on with this quite shameful self-aggrandizing lie for a full seven years after the disaster, until pressured to retract his statement by the two other survivors.
While seismology might be a relatively inexact science the significance of Bernard Chouet’s long period events has been well proven and documented. Monitoring these seismic signatures in active volcanoes has served to predict other eruptions, including that of Mt. Pinatubo in the Philipines, and it has saved many lives. For my part I would certainly rather rely on them than go mucking about in the crater of an active volcano.