Scientific Research
The peer-reviewed work that Seismikon's earthquake nowcasting is built on
A methodology grounded in the scientific literature
Seismikon's published research describes natural-time analysis combined with LSTM neural networks. The peer-reviewed paper and its earlier preprint are available below so that the published work can be read in its original form.
The principal Greek research dataset was the National Observatory of Athens (NOA) catalogue: earthquakes in the Greek region with magnitude greater than 2.0 from 1950 through 2024. USGS ANSS Comprehensive Earthquake Catalog (ComCat) data were co-utilized as a supplementary global public catalogue for proof-of-concept work and to support catalogue completeness.
Seismikon's current S.A.F. (Seismikon Algorithmic Framework) is an operational evolution of the published research. It uses California seismic catalogue data with historical coverage extending back to 1932, magnitude-specific catalogue branches, and continuous live regional monitoring. Seismikon's seismic research is active and ongoing, with further research and scientific dissemination in development.
Research Publications
Seismic Nowcasting: A Systemic Artificial Neural Network Predictive Model
This peer-reviewed paper introduces an earthquake nowcasting framework combining natural-time analysis and Long Short-Term Memory (LSTM) neural networks. Its Greek research dataset uses National Observatory of Athens (NOA) events of magnitude greater than 2.0 from 1950 through 2024, with USGS ANSS ComCat co-utilized as a supplementary global public catalogue. The publication describes a historical four-parameter output—latitude, longitude, focal depth, and magnitude—with occurrence time-frame estimation investigated separately.
Note on the ≥98% figure: This retrospective result concerns approximate latitude, longitude, focal depth, and magnitude parameters in Bikos (2024); time-frame estimation was also investigated, but was not one of those four reported parameters. It is not a live prospective performance claim. For forward-issued prediction outcomes, see the Track Record page.
Seismic Nowcasting: A Holistic Artificial Neural Network Predictive Model
This is the preprint version of the research. It is provided alongside the peer-reviewed publication for transparency and is not presented as peer-reviewed research.
About the Researcher
Research Areas
- Natural-time seismology
- LSTM neural networks for geophysical prediction
- Earthquake Potential Score (EPS) methodology
- Prospective prediction evaluation frameworks
Dr. Anastasios N. Bikos is a seismologist and data scientist at the University of Patras, Greece. His research focuses on earthquake nowcasting - the application of natural-time statistical mechanics and machine learning to prospective seismic event prediction.
Unlike retrospective pattern-matching or early warning systems, the nowcasting methodology Dr. Bikos developed produces predictions before the fact: a specific location, magnitude class, focal depth range, and time window, issued prospectively with defined tolerances. This allows independent evaluation against real seismic catalogue data, treating each prediction as a testable scientific hypothesis.
The published methodology originated in work on Greek seismic sequences, using 74 years of National Observatory of Athens (NOA) catalogue data for magnitude greater than 2.0 events from 1950 through 2024; USGS ANSS ComCat was co-utilized as a supplementary global public catalogue. The current S.A.F. uses California catalogue data with historical coverage since 1932, supplemented by magnitude-specific branches and continuous live regional monitoring.
Dr. Bikos serves as Co-Founder and Chief Scientist of Seismikon and is directly involved in the operational prediction pipeline - the same pipeline that produces the predictions published on this website.
Publications
Peer-reviewed research. Geoinformatics & Geostatistics: An Overview, 12:4. Download PDF
Preprint / non-peer-reviewed version. Download PDF
Methodology at a Glance
The 2024 paper describes four interconnected components of its historical research pipeline. The current S.A.F. has evolved beyond that configuration; its additions are identified separately below. The Research Methodology page and the How It Works guide provide plain-language explanations for non-specialist readers.
Natural Time
Seismic event count replaces clock time as the unit of analysis. This event-based representation reveals patterns in the seismic cycle that are invisible in ordinary calendar time - specifically, the accumulation of stress and strain between successive large earthquakes in a defined geographic region.
Originally introduced by Varotsos et al. in the context of complex system physics. Applied to earthquake nowcasting in Bikos (2024).
Earthquake Potential Score (EPS)
A 0-100 index that quantifies how far through the seismic cycle a given region currently is. Derived from natural-time entropy statistics (specifically the variance of the natural-time probability distribution weighted by seismic energy release). A high EPS signals that the region is approaching conditions historically associated with a significant event.
Formulated in Bikos (2024). Methodology detailed in full in the paper.
Sliding Window Feature Extraction
A dynamic sliding window technique divides seismic catalogues into overlapping natural-time segments and extracts statistical features from each: earthquake count, mean and maximum magnitude, seismic energy release, spatial clustering metrics, and temporal derivatives of the EPS. These features form the input to the LSTM network.
Described as the stochastic filter component in Bikos (2024).
LSTM Neural Network
A Long Short-Term Memory recurrent network trained on historical seismic sequences learns the temporal and spatial dependencies that precede significant events. Training uses Connectionist Temporal Classification (CTC) loss to handle variable-length seismic sequences. Output is a historical four-parameter forecast: latitude, longitude, focal depth, and magnitude.
Network architecture and training protocol described in full in Bikos (2024).
- Can we nowcast the exact GCS (Geographic Coordinate System), focal depth, and magnitude within the next 48-hour time window for any applicable global region under monitoring?
- Can we nowcast the exact EPS (Earthquake Potential Score) within the next 48-hour time window for the same applicable global region under monitoring?
Validation Results
From Research Paper to Live System
The Bikos (2024) methodology informs a live, continuously running operational system. Dr. Bikos is directly involved in the pipeline that generates each prediction. The current S.A.F. is an evolution beyond the exact configuration described in the paper; the distinctions are:
Paper (Bikos 2024)
- Retrospective back-test on historical data
- Greece: National Observatory of Athens catalogue, magnitude greater than 2.0 (1950-2024)
- USGS ANSS ComCat co-utilized as a supplementary global public catalogue
- Results evaluated after the fact against known earthquake sequences
- Fixed training and test splits
- ≥98% retrospective precision across approximate latitude, longitude, focal depth, and magnitude parameters
Live System (Seismikon)
- Prospective predictions - issued before the evaluation window opens
- California seismic catalogue data with historical coverage extending back to 1932, magnitude-specific branches, and live regional tracking
- Every prediction is timestamped before the event window
- Results evaluated by independent FDSN catalogue in your browser
- 24-hour live monitoring and a continuously evolving model
- 5-tuple predictive output: longitude, latitude, focal depth, magnitude, and time
- LSTM neural networks and Graph Neural Networks (GNNs)
- Dual Sliding Window technique: two windows move inversely along the same time axis to increase dynamic convolutional sensitivity
- Live performance visible on the Track Record page
Scientific Enquiries
We welcome critical engagement from the seismological community. If you are a researcher interested in the methodology, would like to discuss the evaluation framework, or want to propose independent validation studies, please get in touch.