← All datasets

Major Earthquakes

A CodeSights flagship for the geoscience and disaster-data crowd: 173 major earthquakes of the instrumental era, the deadliest and the strongest of each year, with magnitude, focal depth, felt intensity on the Modified Mercalli scale, death toll, coordinates and country. Built from two tables of Wikipedia's lists (deadliest by year since 1920, strongest by magnitude each year) deduped by article, with In_Deadliest_List and In_Strongest_List flags recording which list each came from; coordinates and a Wikidata QID are fetched per quake. The famous finding is that magnitude is not death toll: across the set the correlation between magnitude and the log of deaths is essentially zero, slightly negative by test, because energy is logarithmic while deaths track depth, shaking and where people live. The strongest quake ever recorded, Valdivia 1960 at magnitude 9.5, killed about 2,000; the 2010 Haiti quake at magnitude 7.0 killed over 100,000; the deadliest row, Haiyuan 1920, was a shallow magnitude 7.9 that killed 273,000. Depth and MMI explain far more than magnitude does. Magnitudes span 5.3 to 9.5, dates run from 1907 to the present, coordinates cover most rows for mapping, and 38 remote strong quakes without a Wikipedia article are out of scope. Ten acceptance tests anchored on Haiti, Valdivia and Tohoku.
173 rows 1 file last pulled 2026-10-15
Earthquakes 173 rows × 16 columns
EarthquakeEarthquake_IDDateYearCountryLocationLatitudeLongitudeMagnitudeDepth_kmMMIDeaths_LowDeaths_HighIn_Deadliest_ListIn_Strongest_ListEarthquake_URL

Questions and answers

What is the Major Earthquakes dataset?

A CodeSights flagship for the geoscience and disaster-data crowd: 173 major earthquakes of the instrumental era, the deadliest and the strongest of each year, with magnitude, focal depth, felt intensity on the Modified Mercalli scale, death toll, coordinates and country. Built from two tables of Wikipedia's lists (deadliest by year since 1920, strongest by magnitude each year) deduped by article, with In_Deadliest_List and In_Strongest_List flags recording which list each came from; coordinates and a Wikidata QID are fetched per quake. The famous finding is that magnitude is not death toll: across the set the correlation between magnitude and the log of deaths is essentially zero, slightly negative by test, because energy is logarithmic while deaths track depth, shaking and where people live. The strongest quake ever recorded, Valdivia 1960 at magnitude 9.5, killed about 2,000; the 2010 Haiti quake at magnitude 7.0 killed over 100,000; the deadliest row, Haiyuan 1920, was a shallow magnitude 7.9 that killed 273,000. Depth and MMI explain far more than magnitude does. Magnitudes span 5.3 to 9.5, dates run from 1907 to the present, coordinates cover most rows for mapping, and 38 remote strong quakes without a Wikipedia article are out of scope. Ten acceptance tests anchored on Haiti, Valdivia and Tohoku.

How big is the Major Earthquakes dataset?

173 rows in one CSV file, with 16 columns in total: Earthquake, Earthquake_ID, Date, Year, Country, Location, Latitude, Longitude, Magnitude, Depth_km, MMI, Deaths_Low, Deaths_High, In_Deadliest_List, In_Strongest_List, Earthquake_URL.

Where does the data come from?

It is built by programmatically scraping en.wikipedia.org, last pulled on 2026-10-15. Datasets are versioned; older versions stay downloadable.

Is the dataset free to download?

Yes. Every CodeSights dataset is completely free as a CSV download; a free account is all it takes. Anyone can preview the data without signing in.

Can I see the code that built this dataset?

Yes. The exact Python scraper that built it is viewable on the dataset page by any signed-in member, so every number is reproducible.

Can the data contain errors?

Automated scraping leaves room for error and the underlying sources change over time, so no version is guaranteed accurate or complete. If a number matters, verify it against the original source.

Browse every free dataset on CodeSights