Global Grid Stations – High-Quality Data for Over a Million Locations Across the Entire Globe

We offer these grid stations alongside the thousands of real-world weather stations in our system.

For grid stations there is a lag of about 5 days before the latest data becomes available. Degree days are calculated for complete days only, and for real-world weather stations our system can usually provide degree days up to and including yesterday. For grid stations our system can usually provide degree days up to and including the day that is 5 days before yesterday.

Where global grid data comes from

We make use of the excellent ERA5 hourly data that is kindly provided by the Copernicus Climate Change Service (C3S) Climate Data Store (CDS). We maintain our own performance-optimized, multi-terabyte database that enables rapid grid-data access and degree-day calculation for any location on Earth.

The ERA5 hourly data is what's known as reanalysis data. It comes from a sophisticated global weather model based on historical observations made by weather stations, satellites, weather balloons, aircraft, ships, and more. The video below explains further:

Our investigations into the quality of this data have found it to be excellent. And the fact it covers absolutely everywhere on Earth with no gaps makes it ideal for degree-day analysis of historical energy data. (For ongoing monitoring, real-world weather stations may be preferable because they can usually provide degree days up to and including yesterday, whereas grid stations can usually only provide degree days up to and including the day 5 days before yesterday.)

Grid cells and grid "stations"

For the global grid, the Earth is divided into grid cells, each covering 0.25° of longitude and 0.25° of latitude. These grid cells are pretty much square at the equator, but their east/west dimension gets narrower as you move towards the poles.

At the central point of each grid cell is what we call a grid "station". It is the best representation the reanalysis dataset can give us of the weather at that specific location, but it's not tightly localized in the way that a real-world weather station is. The spatial resolution of the reanalysis dataset is such that the data for a grid station is broadly representative of the weather across its grid cell.

When you click a grid station on the map on our website you will see a rectangle appear to show its grid cell. You can think of this as the station's area of influence.

Choosing a grid station to represent your building

If a significant proportion of a grid cell is over sea, or mountainous terrain, that can influence the data for the grid station, even if the grid station itself (the central point of the grid cell) is located on flat land. For example, a grid station with a grid cell dominated by sea will tend to have data that is more sea-like (typically milder nights, cooler days, and smaller day/night temperature swings) than a grid station with a grid cell dominated by land.

When choosing a real-world weather station you would typically look for one in a location with similar weather conditions to your building. For grid stations, instead you should consider the weather of the grid cell as a whole and how representative it is likely to be of the weather at your building.

When you search for stations on our website, our system will automatically aim to list the best matches for your target location first. Similarly, if your software is using our API to fetch data for specified longitude/latitude positions or postal/zip codes, our system will automatically consider factors like distance, data quality, and land/sea proportion when choosing which stations to use.

Specify and Generate Your Degree Days Now