AI-Powered Earthquake Nowcasting
24-48 hour advance earthquake alerts powered by artificial intelligence
View California ForecastNatural Time
Event-based time progression rather than clock time for improved pattern recognition
LSTM Neural Networks
Long Short-Term Memory networks capture spatio-temporal earthquake dependencies
Earthquake Potential Score
Quantitative risk assessment based on seismic cycle progression
High Precision Forecasting
≥98% accuracy for latitude, longitude, depth, and magnitude predictions
Research Foundation
Seismikon is co-founded by Dr. Anastasios N. Bikos (University of Patras, Greece) and implements his peer-reviewed research on earthquake nowcasting methodology. The approach uses natural time - counting seismic events rather than clock time - combined with LSTM neural networks and Earthquake Potential Scores (EPS) to forecast significant earthquakes. The method addresses two fundamental questions: Will there be a major event (M≥6.0) in the next year? What are the predicted coordinates (latitude, longitude), focal depth, and magnitude? Currently demonstrated with California seismic data, this research framework is adaptable to any seismically active region.