Back to the deformation and Stress Change Modeling home page

Back to the team's online recent papers pages

Summary. We make calculations of earthquake probability based on what we know about past earthquake activity. If, for example, earthquakes were known to strike some fault every 200 years exactly, we could predict the next one to happen exactly 200 years after the last one. Unfortunately, while we do recognize some regularity with earthquake occurrence, they also demonstrate a lot of variability. Thus instead of saying exactly when the next event will occur, we give the odds of it occurring over some given period. We use the long-term record of earthquakes and an estimate of their regularity to calculate those odds.

We also know that if an earthquake happens on one fault, it might advance or delay events on a second, nearby fault. The purpose of this paper is to investigate whether the interaction between earthquakes is important enough to alter our probability calculations in a statistically significant way. It turns out that even when there is a lot of uncertainty, if a very large earthquake happens, we should revise our probability calculations on other, nearby faults to account for the changes in stress caused by the earthquake.