******************************EXCERPT********************************
A moderate quake struck southern California right where it was supposed to
A moderate earthquake (magnitude 5.4) in southern California on February 22, 2003 occurred in an area of increased underground stress. It's a fresh example of a promising direction in seismic forecasting.The stress-triggering theory makes intuitive sense: large earthquakes cause stress release on the faults where they occur, but they also add and subtract stresses on the surrounding region. A seismic event, by changing the stresses around it, can make one quake less likely and another one more likely. Ross Stein, a U.S. Geological Survey scientist who is the main driver of this line of research, calls this interaction "earthquake conversations."
An example is the San Francisco Bay area, where the great earthquake of 1906 reduced stresses across the bay, causing a lull in moderate-sized quakes that lasted the rest of the 20th century. The opposite occurred in Turkey along the great North Anatolian fault. A large quake in 1939 sent stresses farther down the fault, triggering a 60-year series of quakes whose latest installment was the deadly Izmit earthquake of 1999. The stresses have risen in the crust near the city of Istanbul, and a quake there is now considered more likely.
Let's see how it works in southern California. A decade ago on the morning of 28 June 1992, the magnitude-7.4 Landers quake changed stresses over a wide region shown on the map below. Red is higher, purple is lower stress.
"Coulomb" stress changes in southern California after Landers earthquake. Pacific Ocean at lower left, Salton Sea at lower right. San Andreas fault zone runs northeast from Salton Sea; Pinto Mountain fault runs east-west in center. White dots are significant aftershocks. All images courtesy U.S. Geological Survey.
Three hours after this earthquake, the magnitude-6.5 Big Bear quake occurred where the stress rise was the greatest (the gray dots are aftershocks and other unrelated earthquakes):
In the nine years afterward, other earthquakes including the Hector Mine quake struck in different parts of the post-Landers stress pattern, but as of the end of 2001 little had changed in the part near Big Bear. If you were looking at this map below, where might you forecast the most likely quake in that part?
Now see where the February 22 event took place:
So this earthquake fits Stein's theory quite well. The only strike against it is that it's not a large quake and won't change the stresses around Big Bear much. But another quake this size happened in the same place on 26 October 1998, and a 5.1 quake occurred nearby in February 2001. A few more of these will start to add up.
Ross Stein's research group at the USGS has some great supporting information online. The stress images come from a fascinating animation there. You can also read a very lucid version of his thinking in the January 2003 Scientific American in the article "Earthquake Conversations."
And I was up close to the Hector Mine Quake.... |