Methodology

How Climate Stats aggregates daily weather measurements into monthly climate normals and summary statistics.

Data Source

All raw data comes from the Open-Meteo Archive API (https://archive-api.open-meteo.com/v1/archive), which provides daily historical weather data derived from reanalysis and station observations. We request the following daily variables for each city:

Averaging Period

We fetch daily data for a five-year window: 2020-01-01 to 2024-12-31. A five-year window is recent enough to reflect current conditions while long enough to smooth out single-year anomalies.

Monthly Aggregation

For each calendar month (January through December), we pool all daily values across the five years and compute:

Summary Statistics

Each city also gets an annual summary: hottest/coldest month (by mean temperature), wettest/driest month (by monthly precipitation total), annual mean temperature, total annual precipitation, total annual rainy days, and annual averages for humidity, wind speed, and UV index.

Best Time to Visit

Each month is scored out of 100 as a weighted blend: temperature comfort (50%, peaking near 23°C), low precipitation (30%), and sunshine (20%). The top three months are presented as the recommended visiting window. This is a heuristic, not a forecast — personal preferences for heat, cold, or rain may differ.

Rounding and Units

Temperatures are rounded to one decimal place. Precipitation and snowfall to one decimal place or whole millimetres for annual totals. Coordinates use the city's reference latitude and longitude; the API returns data in the city's local timezone (timezone=auto).

Limitations

Five-year averages are not full 30-year climate normals (WMO standard). They reflect recent conditions and may include short-term variability. Snowfall, UV, and humidity availability can vary by station density. Always check a local forecast before traveling.

Last updated: 2026-07-30.