Earthquake Prediction and Forecasting
Abstract
Prospects for earthquake prediction and forecasting, and even their definitions, are actively debated. Here, "forecasting" means estimating the future earthquake rate as a function of location, time, and magnitude. Forecasting becomes "prediction" when we identify special conditions that make the immediate probability much higher than usual and high enough to justify exceptional action. Proposed precursors run from aeronomy to zoology, but no identified phenomenon consistently precedes earthquakes. The reported prediction of the 1975 Haicheng, China earthquake is often proclaimed as the most successful, but the success is questionable. An earthquake predicted to occur near Parkfield, California in 1988±5 years has not happened. Why is prediction so hard? Earthquakes start in a tiny volume deep within an opaque medium; we do not know their boundary conditions, initial conditions, or material properties well; and earthquake precursors, if any, hide amongst unrelated anomalies. Earthquakes cluster in space and time, and following a quake earthquake probability spikes. Aftershocks illustrate this clustering, and later earthquakes may even surpass earlier ones in size. However, the main shock in a cluster usually comes first and causes the most damage. Specific models help reveal the physics and allow intelligent disaster response. Modeling stresses from past earthquakes may improve forecasts, but this approach has not yet been validated prospectively. Reliable prediction of individual quakes is not realistic in the foreseeable future, but probabilistic forecasting provides valuable information for reducing risk. Recent studies are also leading to exciting discoveries about earthquakes.
- Publication:
-
Geophysical Monograph Series
- Pub Date:
- 2004
- DOI:
- 10.1029/150GM26
- Bibcode:
- 2004GMS...150..335J