How to Scrape USGS Earthquake Data

USGS publishes rich earthquake events, but raw GeoJSON adds mapping work when a team wants a flat table or scheduled feed. The USGS Earthquake Scraper normalizes event properties and coordinates into one dataset row per event.

Run a small recent query

Begin with a meaningful magnitude threshold and bounded result count.

{"minMagnitude":5,"orderBy":"time","maxResults":20}

Review event ID, magnitude, place, time, longitude, latitude, depth, significance, alert, status, and source URL. Open several event pages and compare their update timestamps. Use the USGS event ID as the durable key and update existing rows when the source revision time changes.

Dates can be ISO dates or timestamps. If both are empty, the Actor uses the last seven days rather than searching the full catalog. Historical work should always set explicit bounds. The upstream service limits very large result sets, so split broad studies by time or increase the minimum magnitude.

Coordinates use longitude, latitude, and depth in kilometers. Preserve that order when creating geometry. Null intensity or alert fields are valid; they do not prove that no effects occurred.

Once validation passes, save the input as an Apify Task and schedule it. Route new IDs and material revisions separately. This produces a traceable catalog feed, but it is not an early-warning system. Emergency and tsunami decisions must rely on the responsible official authorities.

Record the query thresholds with each dataset because changing magnitude or depth filters can alter trend lines even when the underlying catalog is unchanged.