The number of large destructive earthquakes in 2010, plus a flurry of medium magnitude quakes in California, led many people to ask, Are we in a period of heightened temblor activity, and is it likely to continue? It's also raised questions among both scientists and laypeople about whether these events are related—and if so, how. The eruption of an Icelandic volcano, which disrupted air traffic in Europe for weeks, serves as an additional reminder that we live on a volatile planet. What, if anything, does this apparent uptick in geological activity portend, and how does it compare to events in Earth's past history? E&S sat down with Hiroo Kanamori, the Smits Professor of Geophysics, Emeritus, and Joe Kirschvink, the Van Wingen Professor of Geobiology, to hear their thoughts.
Between February and September 2010, earthquakes ranging from magnitude 6.8 to 8.8 occurred in regions as far-flung as Sumatra, China, Chile, New Zealand, and Baja, California. Are we in fact seeing more large quakes than usual?
Hiroo Kanamori: There are a couple of ways to answer this question. If you look at very major earthquakes, we are not seeing as much activity as between 1950 and 1965, when there were three events of magnitude 9 or greater in which an enormous amount of energy was released.
However, if we total up the number of quakes over magnitude 8 that have occurred since the first great Sumatran quake of 2004, we do find that these numbers really have increased. On average about one quake per year is magnitude 8 or larger. Since 2004, on average we have had two quakes of that size or more annually.
Is this statistically significant?
HK: We don't really know! Thanks to a study that's been going on for about the last 18 years, we do know a great deal more than we used to about triggering events in earthquakes. We now know that every large earthquake sends out seismic waves that can travel some distance and potentially activate seismic activity elsewhere.
How well do scientists understand the physical mechanisms that might touch off a quake cascade like this?
HK: We have several different models and theories. The most straightforward mechanism would be one in which the seismic waves increase stress on other faults that they're passing through. If those faults are already close to rupture, this seismic impact may be enough to push things over the edge.
There are also cases in which this activity is delayed. This appears to be what happened this summer when the magnitude 7.2 quake that had occurred in Baja California in April touched off two moderate quakes in June and July on the San Jacinto fault in Southern California.
Joe, you've made in-depth studies of ancient geological upheavals. Can you put these recent events in perspective for us?
JK: Just to take volcanoes, the Icelandic eruption that we saw this spring was tiny compared to eruptions that have happened previously in Earth's history. In California alone, about three-quarters of a million years ago—which geologically is nothing—the Long Valley Caldera, between Mono Lake and Mammoth, blew its top. The eruption covered the southwestern United States with a blanket of ash that extended all the way to the Mississippi. The sediments that washed off from the Mississippi delta produced deposits that in some places were hundreds of meters thick. That episode was far, far worse than anything in human memory. There was a similar eruption about two million years ago in what is today Yellowstone.
One question we often hear from both the public and the media is, will we ever be able to predict earthquakes the way we can—more or less—forecast the weather? What are your views?
HK: There are such fundamental differences between weather forecasting and earthquake prediction. With weather, the situation basically changes on an almost daily basis. With quakes, we are dealing with long-term processes in which the timescale for stress buildup and release is very long—100 to 1,000 years or more—while the length of time in which quakes occur is very short.
As I said earlier, we have made major advances in our understanding of how these seismic processes operate over these lengthy timescales. But to be able to say there's a strong likelihood that a magnitude 8 earthquake will occur in some specific area within the next hundred years or so is not necessarily very useful for the average layperson. You simply can't handle it like a weather forecast. If the forecast says, "rain tomorrow," you may take your umbrella, and either it rains or it doesn't. However, in the case of earthquakes, if you say that something big is going to happen tomorrow and nothing happens, that can be a problem. And that's really a key difference between climatology and seismology.
JK: I agree with Hiroo. If you want to see how completely distinct the two areas are, just turn the analogy around. Certainly, meteorology averaged over a very long period of time gives you climate. Or, to put it another way, climate is just long-term weather. But I certainly wouldn't advocate analyzing ancient climates to determine whether you'll have a thunderstorm next Tuesday.
To read the full interview, go to Engineering and Science online.