The ensemble models are a type of weather forecasting technique that use multiple members or versions of a model to produce a range of possible outcomes for a given forecast. Each member is initialized with slightly different initial conditions and/or model parameters to account for uncertainties and variations in the atmosphere, resulting in a set of perturbed forecasts. By combining the perturbed forecasts, the ensemble model generates a probability distribution of possible outcomes, indicating not only the most likely forecast but also the range of possible outcomes and their likelihoods. This probabilistic approach provides more comprehensive and accurate forecast guidance, especially for high-impact weather events where uncertainties are high. Different national weather services calculate ensemble models, each with varying resolutions of weather variables and forecast time-range. For instance, the German weather service DWD's ICON model provides exceptionally high resolution for Europe but only forecasts up to 7 days. Meanwhile, the GFS model can forecast up to 35 days, albeit at a lower resolution of 50 km. The appropriate ensemble model to use would depend on the forecast horizon and region of interest.

• JSON

  Open-Meteo Ensemble API
  Download
• HTML

  Open-Meteo Ensemble API Documentation
  Download