Skip to content

Compare datasets

Pick up to four datasets to see their period, resolution, update status, key DOIs and successors in one table. Type a variable such as "precipitation", "2 m temperature" or tp into Find a variable (or use Find by variable) to list every dataset that has it — no selection needed — with its name, units and whether it is instantaneous, mean or accumulated. Different short names for the same quantity (tp, PRECTOT, prate …) are grouped by CF standard name (see Standard variables).

Choose two or more datasets above, or type a variable to find the datasets that have it. Your selection is kept in the page address, so you can share the link.

About resolution values — please read
  • Provided resolution is the grid spacing you download from the main provider, not the model's native resolution. The same dataset can come on different grids from different providers or products, so always check the grid of the file you actually downloaded.
  • km values are approximations at the equator (1° ≈ 111 km). Longitude spacing shrinks with cos(latitude): at 60° it is half (0.25° ≈ 14 km), so "28 km" is not 28 km everywhere.
  • Gaussian grids have uneven latitude spacing, and ocean model grids (tripolar, curvilinear) vary from place to place. Grids with different latitude and longitude spacing are written as 0.5°×0.625° (lat × lon).
  • Polar-stereographic and EASE2 grids of sea-ice satellite products are defined in km; the degree value in brackets is a simple conversion. The NSIDC polar stereographic grid matches its nominal spacing only at 70° latitude.
  • Grid spacing is not the size of the smallest feature actually resolved (effective resolution). Features are typically resolved only at several times the grid spacing, and interpolating to a finer grid adds no information.
  • Temporal resolution is the shortest interval provided. Intervals and whether a value is instantaneous, mean or accumulated differ between variables — see 'Available variables' on each dataset page.

The table is a summary

Check the exact definitions (start and end of accumulation windows, units, grids) on each dataset page and in the official documentation. Always double-check against the official source before using anything in a paper.