Travel seasons methodology

How Sfrtna rates each month for each destination, and where the numbers come from.

Data sources

Contains modified Copernicus Climate Change Service information (CC-BY 4.0). Neither the European Commission nor ECMWF is responsible for any use that may be made of it.

What we show

For every destination and month we show the typical average high and low, rain, humidity, the season, and one of five ratings: Best time to visit, Excellent time to visit, Good time to visit, Less favorable time to visit, Not recommended climatically. The rating is Sfrtna's own, produced by the method below — it is not issued by any weather agency.

Where the numbers come from

Temperatures, rain and humidity are computed from ERA5-Land, published by the Copernicus Climate Change Service (C3S) at ECMWF, for each city's own coordinates — never a country average. We use 1991–2020, the World Meteorological Organization's standard climate-normal period. The average high is the mean of each day's highest temperature in the city's local time, the average low the mean of each day's lowest, and a rainy day is one with 1 mm or more.

Normals, not forecasts

The figures describe the usual weather in that month over thirty years. They are not a forecast for the days of your trip, and any given year can differ.

How a month is rated

First, how well the climate suits a visit: daytime and night-time temperatures, humidity, how often and how much it rains, wind, and snow for ski destinations — judged by the kind of trip, since 31 °C suits a beach better than a city walk. A month is only as good as its weakest essential part: perfect temperatures do not make up for rain on half its days. Some limits no month can pass whatever else it has going for it: extreme heat, heat with oppressive humidity (the "feels like" temperature from the heat-index formula used by the US National Weather Service, NOAA), very heavy rain, and deep cold except for ski destinations — and "Not recommended climatically" is kept for months where one of these limits applies. Then the month is compared with the destination's other months: "Best time to visit" has to be close to the best month of its year, and the least bad of twelve poor months never becomes the best time.

What tourism sources do

When a destination's official tourism board — or a well-known travel guide when there is none — names a month among the best times to visit, that lifts its rating a little, and the source is named and linked on the city. No source can lift a month out of extreme heat or rain. Where we found no source, the rating is climate-based only; we never invent a recommendation.

Confidence and review

Each destination carries an internal confidence score from how complete its climate data is, how close the data point is to the city, and whether an official tourism source exists. The home page's best destinations never show low-confidence ones. Coastal cities are measured at the nearest land cell no more than 25 km away. Because a data cell is about 9 km across, we compare its height with the city's own: where they differ a lot (a valley town whose cell sits on the mountain) we use a neighbouring cell closer to the city's height, and where a large difference remains we lower the confidence — and beyond 400 m we do not publish the city's figures until official station normals are available. No temperature is ever adjusted by hand.

Known limits of the data

Reanalyses such as ERA5-Land tend to count more light-rain days than weather stations record, especially in the humid tropics and on wet coasts, so our rating leans on how much it rains more than on how many days. On some coasts temperatures can read slightly lower than the station's, because the cell includes some sea or inland ground.

Last updated: 30 September 2026 · Methodology version 1.0