Collection
Artificial Intelligence and Meteorology
Progress in Numerical Weather Prediction, and more generally in meteorology, has traditional stemmed from increased availability of Earth observations, improved knowledge of the bio-geo-physical processes represented in numerical models, and an ever growing computational capacity to realistically simulate weather and environmental phenomena.
As typical of scientific and technological developments, periods of continuous and gradual developments alternate pivotal moments in which cognitive and technological advancements permit more rapid and disruptive innovations. These phases call for different approaches and methodologies to take advantage of the new opportunities and aim at real breakthroughs.
This is the case for Meteorology and Climate sectors, thanks to a generational leap in the HPC infrastructure, combined with unprecedented data availability for Earth Observations, which truly belong to Big Data. This confluence of data and computational resources has been calling for new approaches to optimally extract the potentially available information.
Artificial Intelligence, and Machine Learning methods in particular, have been identified as a key innovative methodology to leverage these opportunities and it is now necessary to proceed with development plans that can progressive integrate traditional model development, based on physical parameterisation, with AI-based approaches, that are extremely powerful and can be complementary. This is well explained in the comprehensive Technical Memo “Machine learning at ECMWF: A roadmap for the next 10 years”, by Peter Dueben et al. published in January 2021 (n. 878) as complement to the new ECMWF 10-year strategy.
The attention on Big Data, AI and Machine Learning methodologies is currently in a growing phase, as demonstrated by the numbers of scientific publications and applications. This special topic issue of BAST is therefore dedicated to "Artificial Intelligence and Meteorology" to foster an exchange of the ongoing scientific efforts and experiences.
Editors
-
Tiziana Paccagnella
Tiziana Paccagnella cooperates with ARPAE-SIMC (Regional Hydro-Meteo-Climate Service of the Emilia-Romagna Environmental Agency, Italy) where she has been working for 35 years, the last three as Director. She is a member of the Steering Committee for Meteorology and Climatology of the Italian National Agency for Meteorology and Climatology, and of the Scientific Council of the IFAB foundation for Big Data and Artificial Intelligence. She always worked in Atmospheric Physics and Meteorology with a focus on Numerical Weather Prediction and Data Assimilation.
-
Marco Arpagaus
Marco Arpagaus is the Leader of the Development Team of the Numerical Prediction Division at the Federal Office of Meteorology and Climatology at MeteoSwiss, Switzerland. He is a member of C2SM and of the TEAMx Coordination and Implementation Group.
-
Gianpaolo Balsamo
Gianpaolo Balsamo is a researcher and a team leader at the European Centre for Medium-range Weather Forecasts,UK (ECMWF) headquarters in Reading. He joined ECMWF since 2006 and is currently working as a principal scientist focusing on the Earth system coupled process at interface of atmosphere-ocean-land. He had previously worked at Environment & Climate Change Canada in Montréal, and Météo-France in Toulouse, specialising on data assimilation of Earth Observations for the land surface. He has a Ph.D. in Meteorology and a Physics degree and he is an invited professor for climate change at a Politecnico di Torino.
-
Pierpaolo Alberoni
Pier Paolo Alberoni has been working since 1987 at ARPAE-SIMC, Italy (Regional Hydro-Meteo-Climate Service of the Agency for the Environment of Emilia-Romagna) becoming responsible for the management of weather radars and the development of radar and nowcasting products. Currently he is also responsible for the meteorological numerical modeling team. President of the scientific council of CETEMPS.
