The Global Earthquake-Tsunami Risk Assessment is a global dataset for earthquake and tsunami risk assessment, designed to provide a standardized and computable data foundation for tsunami risk prediction, earthquake event analysis, and earthquake hazard assessment.

This dataset contains detailed observational information on 782 major earthquakes worldwide between 2001 and 2022, covering a region from latitude –61.85° to 71.63° and longitude –179.97° to 179.66°. The data is published in CSV format and is complete and without missing data. Its core objective is to conduct a binary classification prediction study of earthquake-triggered tsunamis to assess the potential tsunami risk of seismic events. The data includes 478 events that did not trigger tsunamis (61.11 TP3T) and 304 events with potential tsunami risk (38.91 TP3T), with a balanced distribution. The magnitude ranges from 6.5 to 9.1, with an average magnitude of 6.94. It includes 28 mega-earthquakes (≥8.0), such as the two 9.1 magnitude super-earthquakes in 2004 and 2011. This dataset is representative and balanced in terms of magnitude, depth, and geographical coverage.

Main fields:

• Magnitude: Earthquake magnitude (Richter scale), ranging from 6.5 to 9.1, is a major tsunami predictor.
• CDI (Community Decimal Intensity): ranging from 0 to 9, reflects the degree to which earthquakes are perceived by the population.
• mmi (Modified Mercalli Intensity), ranging from 1 to 9, measures the degree of damage to buildings caused by earthquakes.
• sig (Significance Score): The score ranges from 650 to 2910. It is used to comprehensively assess the overall seismic hazard.
• nst (Number of Seismic Stations): Range 0–934, reflects the coverage quality of seismic observation data.
• dmin (Distance to Nearest Station): Range 0.0–17.7°, used to evaluate positioning accuracy.
• Azimuthal gap (range 0.0–239.0°) represents the difference in azimuth coverage between observation stations.
• Focal depth: ranging from 2.7 to 670.8 km. Shallow earthquakes are usually accompanied by a higher risk of tsunami.
• Latitude: Epicenter Latitude, ranging from –61.85° to 71.63°, indicates the north-south distribution of earthquake locations.
• Epicenter Longitude, ranging from –179.97° to 179.66°, indicates the east-west distribution of earthquake locations.
• Year: The year in which the earthquake occurred, ranging from 2001 to 2022, used for time distribution analysis.
• Month: The month in which the earthquake occurred, ranging from 1 to 12, which can be used for seasonal analysis.
• tsunami (tsunami potential): Tsunami Potential (target variable), takes the value 0 or 1, indicating whether there is a potential risk of triggering a tsunami.