About the data
World Climates publishes long-term monthly climate averages for 4,250 places. This page explains where the numbers come from and what they can and cannot tell you.
Climate data: NASA POWER
The data were obtained from the National Aeronautics and Space Administration (NASA) Langley Research Center’s Prediction Of Worldwide Energy Resources (POWER) project, funded through the NASA Earth Science Division. We use the POWER Climatology API (version 2.10, accessed 2026-10-16), which provides pre-computed monthly averages for January 2001 – December 2025.
- Temperature, humidity, wind, precipitation and snow depth come from NASA’s MERRA-2 reanalysis — a weather model constrained by observations — on a grid of 0.5° × 0.625° (roughly 50–70 km). Precipitation is the bias-corrected MERRA-2 product (PRECTOTCORR).
- Cloud amount and solar radiation come from the CERES satellite products (1° grid).
- Average high and low are calculated as the monthly mean temperature plus/minus half of the average daily temperature range.
- Elevation adjustment: temperatures are adjusted for the difference between a city’s elevation (from GeoNames) and the grid cell’s average elevation, using the standard lapse rate of 6.5 °C per 1,000 m. Where the difference is 50 m or more, the city’s page says so.
- Day length is calculated astronomically for the 15th of each month.
- Climate type is the Köppen–Geiger classification computed from the monthly averages shown on each page (0 °C threshold between temperate and continental climates).
Because the values describe a grid cell rather than a weather station, they can differ from the official station normals published by national weather services — especially in mountains, on coasts and for rainfall in areas with strong local contrasts. Rainy-day counts and sunshine hours are not part of this dataset, so we do not show them.
Places
Place names, coordinates, elevations and populations come from the GeoNames geographical database, licensed under CC BY 4.0. We include every city with at least 100,000 inhabitants in the GeoNames “cities15000” list, the largest city of every other country, and the places that had pages on the earlier version of this website (some of them former weather-station locations). Cities that fall into the same NASA POWER grid cell as a larger city are listed on that city’s page instead of getting a separate page with identical numbers.
Comfort and packing tips
“Best time to visit” ranks the months of each place by a comfort score applied to the averages. The ideal is an average high between 21 and 28 °C, an average low above 5 °C, little precipitation and moderate humidity; months that miss it by less score better. The best-ranked months of a place are recommended even when none of them meets every condition — in the tropics or at high altitude that is often the case. Packing tips are general suggestions derived from the average highs, lows and precipitation. Real weather on a given day can be very different from the long-term average.
Frequently asked questions
Why can the numbers differ from other climate websites?
Many sites show records from one weather station, often at an airport. We show averages for a grid cell of about 50 km. The period is also different: we use 2001 to 2025, while many tables still use 1961 to 1990 or 1991 to 2020.
How often is the data updated?
NASA updates the POWER climatology every few years. When a new period is published, we rebuild every page. The date of the last data access is written at the bottom of each page.
Why is rainfall in the mountains sometimes too high?
Reanalysis models spread rain over the whole grid cell. In mountains and in the tropics this can overestimate the rain at a single town. For example, Bogotá shows more rain than its weather station records. Treat these values as a regional guide.
Can I use the data?
The climate values come from NASA POWER, which is free to use, including for commercial purposes, with a citation. Place data come from GeoNames under CC BY 4.0. Please cite both sources if you reuse the numbers.