BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260624T171341Z
LOCATION:Bldg. 6 - 002
DTSTART;TZID=Europe/Stockholm:20260701T140000
DTEND;TZID=Europe/Stockholm:20260701T160000
UID:submissions.pasc-conference.org_PASC26_sess122@linklings.com
SUMMARY:MS5A - Data-Driven Regional Weather Modeling: Towards Trustworthy 
 Convection-Resolving Forecasts
DESCRIPTION:Organizer(s): Oliver Fuhrer (MeteoSwiss, ETH Zurich), and Laur
 e Raynaud (Météo-France)\n\nData-driven weather prediction has advanced ra
 pidly in recent years, with machine-learning-based models now complementin
 g traditional numerical weather prediction. While global data-driven model
 s have demonstrated impressive skill, extending these approaches to region
 al, convection-resolving forecasting introduces new scientific and computa
 tional challenges. At kilometer and sub-kilometer scales, models must repr
 esent complex physical processes, integrate high-frequency observations, a
 nd provide trustworthy uncertainty estimates, particularly for extremes. T
 his minisymposium focuses on recent progress and open challenges in data-d
 riven regional weather modeling. Topics include generative diffusion model
 s for convective-scale downscaling, graph-based neural network architectur
 es for high-resolution domains, training strategies that improve generaliz
 ation across scales and regions, and operational perspectives from nationa
 l meteorological services. The session emphasizes scientific trustworthine
 ss, evaluation, and physical consistency, and discusses how these requirem
 ents interact with high-performance computing workflows and model design. 
 By bringing together experts from academia and operational forecasting, th
 e minisymposium provides a forum to assess the state of the art and explor
 e pathways toward reliable, convection-resolving data-driven forecasts, wi
 th relevance to a wide range of computational science domains.\n\nTowards 
 Operational Data-Driven Regional Forecasting at Convection-Resolving Scale
 s\n\nTranslating global data-driven weather models to regional, convection
 -resolving prediction remains a key scientific challenge. In Alpine enviro
 nments, orographic precipitation, convective initiation, and valley flows 
 demand kilometer-scale resolution and hourly to sub-hourly output to provi
 de physic...\n\n\nCarlos Osuna, Claire Merker, Alberto Pennino, Andreas Pa
 uling, Daniele Nerini, Francesco Zanetta, Hugues de Laroussilhe, Jonas Bhe
 nd, Katrin Ehlert, and Mary McGlohon (MeteoSwiss); Michele Cattaneo (Swiss
  data science center); and Oliver Fuhrer and Radi Radev (MeteoSwiss)\n----
 -----------------\nComputational and Verification Challenges in Data-Drive
 n Atmospheric Downscaling\n\nUsing generative machine learning for perform
 ing atmospheric downscaling (super-resolution for meteorological data) is 
 of growing interest, as the methods used for data-driven downscaling are c
 omputationally inexpensive compared with statistical downscaling methods o
 r with data-driven models for for...\n\n\nMary McGlohon and Petar Stamenko
 vic (MeteoSwiss, ETH Zurich); David Leutwyler, Xavier Lapillonne, and Oliv
 er Fuhrer (MeteoSwiss); Fabian Bösch, Lukas Drescher, and Henrique Mendonç
 a (ETH Zurich / CSCS); Sebastian Schemm (University of Cambridge); and Sid
 dhartha Mishra (ETH Zurich)\n---------------------\nFrom Numerical to Data
 -Driven Regional Forecasting: Challenges from a Scientific and Operational
  Perspective\n\nAI has opened a new path for atmospheric modeling, with ga
 ins in both quality and computational efficiency. At Météo-France, like at
  other national weather services, the topic of AI for weather prediction h
 as developed rapidly and is being explored from various angles, with appli
 cations to both glob...\n\n\nLaure Raynaud (Météo-France)\n---------------
 ------\nMulti-Domain: Improving Generalization Across Scales and Regions\n
 \nThe domain of weather forecasting is currently undergoing a significant 
 transformation driven by advances in machine learning. Following these dev
 elopments, high-resolution regional models have emerged. Among these regio
 nal models is the stretched-grid model (SGM), a global model with an incre
 ased s...\n\n\nSophie Buurman (KNMI), Aram Farhad Shafiq Salihi and Even M
 arius Nordhagen (Norwegian Meteorological Institute), Mario Santa Cruz (EC
 MWF), Michiel van Ginderachteren (RMI Belgium), and Thomas Nils Nipen (Nor
 wegian Meteorological Institute)\n\nDomain: Climate, Weather, and Earth Sc
 iences, Physics, Computational Methods and Applied Mathematics\n\nSession 
 Chair: Oliver Fuhrer (MeteoSwiss, ETH Zurich)
END:VEVENT
END:VCALENDAR