How to Scrape NASA Near-Earth Object Data

NASA's Near Earth Object Web Service provides structured asteroid data, but production use still needs date validation, deduplication, unit conversion, and stable output fields. The NASA Near-Earth Objects Scraper packages the feed and exact-object lookup into one bounded Actor.

Query a date window

Start with a one-day sample and inspect objects that have multiple close approaches, missing optional fields, or broad diameter ranges. This confirms the desired row grain before scheduling overlapping windows and prevents approach-level data from being mistaken for timeless object properties.

{"startDate":"2026-07-20","endDate":"2026-07-21","maxResults":25}

Each row includes NASA reference ID, name, absolute magnitude, estimated diameter range in kilometers, potentially hazardous and sentry flags, approach date, velocity, miss distance, orbiting body, full close-approach records, and JPL links. Objects are deduplicated by NASA ID when they appear in the requested feed.

NASA limits feed windows to seven days. Split longer research into daily or weekly Tasks, then deduplicate downstream. Blank dates use the current day, which is useful for a scheduled monitor. Exact asteroid IDs can be supplied for objects already on a watchlist.

Treat diameter as an estimate range, not an exact size. Hazard and sentry fields should be quoted as NASA classifications and never reframed as a prediction. For historical storage, key objects by NASA ID and approaches by object ID plus approach timestamp. Retain source URLs so readers can validate an observation against NASA and JPL. This model supports education, data journalism, and research without overstating what the feed says.