New open dataset enables the research of ensemble prediction

Dataset enables deeper and more exact understanding of the uncertainties and errors in weather prediction. Through this, it is possible to further improve the skill of the weather prediction models, which leads to more precise weather forecasts.

Open dataset called ”OpenEnsemble 1.0” provides unparalleled possibility for the academic research community to participate in the research of ensemble and probabilistic weather forecasts. Newly opened dataset includes a set of perturbed initial conditions that enables ensemble forecasting and helps to acknowledge also the alternative but equally plausible future evolutions of the atmospheric state.

Ensemble forecasting provides researchers a key to e.g. understand, how are the uncertainties and errors in weather forecasts depend on the errors in the initial conditions of the atmospheric state used in the weather prediction models. Dataset has been compiled in 2019–2020 by the Finnish Meteorological Institute, University of Helsinki and European Centre for Medium-Range Weather Forecasts (ECMWF).

Further information:

Postdoctoral Researcher Pirkka Ollinaho, pirkka.ollinaho@fmi.fi, Finnish Meteorological Institute

Scientific article: Ollinaho, P., Carver, G. D., Lang, S. T. K., Tuppi, L., Ekblom, M., and Järvinen, H.: Ensemble prediction using a new dataset of ECMWF initial states – OpenEnsemble 1.0, Geosci. Model Dev., 14, 2143–2160, https://doi.org/10.5194/gmd-14-2143-2021, 2021.

You can read the scientific article here: https://doi.org/10.5194/gmd-14-2143-2021

Science newsWeatherWeather forecastWeather model