Geoscientific Model Development | 2021
OpenIFS@home version 1: a citizen science project for ensemble weather and climate forecasting
Abstract
Abstract. Weather forecasts rely heavily on general circulation models of\nthe atmosphere and other components of the Earth system. National\nmeteorological and hydrological services and intergovernmental\norganizations, such as the European Centre for Medium-Range Weather\nForecasts (ECMWF), provide routine operational forecasts on a range of\nspatio-temporal scales by running these models at high resolution on\nstate-of-the-art high-performance computing systems. Such operational\nforecasts are very demanding in terms of computing resources. To facilitate\nthe use of a weather forecast model for research and training purposes\noutside the operational environment, ECMWF provides a portable version of\nits numerical weather forecast model, OpenIFS, for use by universities and\nother research institutes on their own computing systems. In this paper, we describe a new project (OpenIFS@home) that combines\nOpenIFS with a citizen science approach to involve the general public in\nhelping conduct scientific experiments. Volunteers from across the world can\nrun OpenIFS@home on their computers at home, and the results of these\nsimulations can be combined into large forecast ensembles. The\ninfrastructure of such distributed computing experiments is based on our\nexperience and expertise with the climateprediction.net ( https://www.climateprediction.net/ , last access: 1\xa0June\xa02021) and\nweather@home systems. In order to validate this first use of OpenIFS in a volunteer computing\nframework, we present results from ensembles of forecast simulations of\nTropical Cyclone Karl from September\xa02016 studied during the NAWDEX field\ncampaign. This cyclone underwent extratropical transition and intensified in\nmid-latitudes to give rise to an intense jet streak near Scotland and heavy\nrainfall over Norway. For the validation we use a 2000-member\nensemble of OpenIFS run on the OpenIFS@home volunteer framework and a\nsmaller ensemble of the size of operational forecasts using ECMWF s forecast\nmodel in 2016 run on the ECMWF supercomputer with the same horizontal\nresolution as OpenIFS@home. We present ensemble statistics that illustrate\nthe reliability and accuracy of the OpenIFS@home forecasts and\ndiscuss the use of large ensembles in the context of forecasting extreme\nevents.