Historical Weather

Best practices for retrieving large amounts of historical weather data from the API.

All plans that have access to the Historical Weather API's must break up large requests into several smaller requests. For example, if you are enrolled in the "Business" plan, and want to retrieve hourly historical data for 2017 for a single location, you simply need make

Our database receives data from over 20 different data sources. These data sources feed into our system with varying lag times ranging from 15 minutes (MADIS station data) to 7-10 days (Reanalysis data). This can result in historical data being corrected with higher quality observations as the lagging (higher quality)

Our historical API receives data from the following data sources (including but not limited to): * Weather station data (hourly / sub-hourly reporting) * NOAA Real-Time Mesoscale Analysis (RTMA) * Ground based weather radars * NASA IMERG (Global precipitation estimations) * ERA-5 Re-analysis * GLDAS Re-analysis * CAMS Re-analysis * GHCN (daily reporting stations) * Global satellite data * Air quality

Our API dynamically selects the nearest available station data, and combines it with additional datasets to compensate for sources of error such as: Distance from point, quality of station data, availability of data for time of interest. Ground truth (station data) is always preferred in a response. As the distance

Our API sources data from many station networks. Citizen maintained stations can change their location with time (the station identifier will not change if this occurs), and are considered "Unstable". We recommend avoiding usage these station ID's by either using another query method (lat/lon, postal

Spatial resolution ranges from 1-15 KM for historical data. Spatial resolution varies based on proximity of stations, and data sources available in the region of interest.

Download large amounts of historical data via our easy to use historical report system. No paid subscription required.

This report system is for downloading large amounts of historical forecast data. Or making small one-off requests. Price varies based on the report request properties. Report requests are invoiced separately from subscriptions, and a user's subscription level has no impact on pricing.