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:20220812T074358Z LOCATION:Sydney Room DTSTART;TZID=Europe/Stockholm:20220627T160000 DTEND;TZID=Europe/Stockholm:20220627T180000 UID:submissions.pasc-conference.org_PASC22_sess127@linklings.com SUMMARY:MS2C - Nexus of AI and HPC for Weather, Climate, and Earth System Modelling DESCRIPTION:Minisymposium\n\nAccurately and reliably predicting weather an d climate change and associated extreme weather events are critical to pla n for disastrous impacts well in advance and to adapt to sea level rise, e cosystem shifts, and food and water security needs. The ever-growing deman ds of high-resolution weather and climate modelling require exascale syste ms. Simultaneously, petabytes of weather and climate data are produced fro m models and observations each year. Artificial Intelligence (AI) offers n ovel ways to learn predictive models from complex datasets, at scale, that can benefit every step of the workflow in weather and climate modelling: including data assimilation, process emulation, solver acceleration, and e nsemble prediction. Further, how do we make the best use of AI to build Ea rth digital twins for a wide range of applications from extreme weather to renewable energy, including at highly localized scales such as cities? Th e next generation of breakthroughs will require a true nexus of high-perfo rmance computing (HPC) and large-scale AI. This minisymposium will delve i nto the challenges and opportunities at the nexus of HPC and AI. Presenter s will describe scientific and computing challenges and the development of efficient and scalable AI solutions for weather, climate, and Earth syste m modeling.\n\nThe Maelstrom Protocol - Workflow for the Development of AI on HPC\n\nEmmerich\n\nOne of the challenges of today’s machine learning ( ML) workflows lies in the storage and exchange of machine learning models. Apart from these collaborative issues, running machine learning applicati ons for training or prediction on HPC infrastructure is not standardized. With Maelstrom, we un...\n\n---------------------\nExtreme Weather Forecas ting using Fourier Neural Operator on HPC Systems\n\nKurth, Pathak, Subram anian, Harrington, Raja...\n\nWe present FourCastNet, short for Fourier Fo recasting Neural Network, a global data-driven weatherforecasting model th at provides accurate short to medium-range global predictions at 0.25-degr ee resolution. FourCastNet accurately forecasts high-resolution, fast-time scale variables such as the surfac...\n\n---------------------\nDeep Learn ing for Weather Prediction and Ensemble Post-Processing\n\nDryden\n\nRecen t advances in deep learning have shown great promise in weather and climat e tasks. However, the neural network architectures and training infrastruc ture are often adapted from "off-the-shelf" approaches for tasks such as i mage classification. Such approaches need to be scaled and tuned for thes. ..\n\n---------------------\nStatistical Downscaling of Surface Temperatur e and Precipitation with Deep Neural Networks\n\nGong, Langguth, Ji, Mozaf fari, Mache...\n\nIn light of the success of superresolution (SR) applicat ions in computer vision, recent studies have started to develop statistica l downscaling methods for meteorological data based on deep neural network s (DNNs). DNNs are attractive, because they are computationally cheap, onc e they are trained.