BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Statistics and AI for Weather and Climate Science
DTSTART:20270309T080000Z
DTEND:20270311T170000Z
DTSTAMP:20261119T224600Z
UID:indico-event-5779@indico.kit.edu
DESCRIPTION:AI is transforming weather and climate science as AI-based wea
 ther and climate models now rival their physics-based counterparts on vari
 ous standard benchmarks while producing forecasts at a fraction of the com
 putational cost. However\, many important questions remain beyond overall 
 prediction skills. How reliable are their forecasts for extreme events\, p
 articularly for those outside their training range? Can they capture spati
 o-temporal and inter-variable dependence\, and are their forecasts physica
 lly consistent? How well do they quantify uncertainty? What strengths of p
 hysics-based models do they still lack?\nAnswering these questions is esse
 ntial for understanding how AI-based and physics-based numerical models di
 ffer\, identifying where AI models could be further improved\, and buildin
 g trust in using AI weather and climate models for scientific discovery an
 d operational forecasting. However\, the mathematical and statistical tool
 s to address these questions remain underdeveloped.\nThis workshop brings 
 together researchers from applied mathematics\, statistics\, atmospheric a
 nd Earth sciences\, and computer science to foster interdisciplinary discu
 ssions on these challenges and deepen our understanding of AI weather and 
 climate models.\nMore details about the workshop can be found at: https://
 zhongwei-zhang.github.io/Workshop_Heidelberg_2027/\n\nhttps://indico.kit.e
 du/event/5779/
URL:https://indico.kit.edu/event/5779/
END:VEVENT
END:VCALENDAR
