Statistical short-term earthquake prediction.

摘要: A statistical procedure, derived from a theoretical model of fracture growth, is used to identify foreshock sequence while it in progress. As predictor, the procedure reduces average uncertainty rate occurrence for future strong earthquake by factor more than 1000 when compared with Poisson occurrence. About one-third all main shocks local magnitude greater or equal 4.0 central California can be predicted this way, starting 7-year database that has lower cut off 1.5. The time scale such predictions order few hours days foreshocks range 2.0 5.0.