BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Training course: Numerical methods for weather prediction
DTSTART:20241111T090000Z
DTEND:20241115T174500Z
DTSTAMP:20260720T162700Z
UID:indico-event-381@events.ecmwf.int
CONTACT:training@ecmwf.int
DESCRIPTION:This course describes contemporary methods\, recent advancemen
 ts\, and the challenges in numerical techniques which are used across a wi
 de spectrum of weather prediction models for solving the governing equatio
 ns of atmospheric dynamics on High-Performance Computing (HPC) systems. Th
 e covered methods have relevance in short-term and medium-term forecasting
 \, seasonal forecasting\, climate prediction\, as well as global and limit
 ed-area modelling.TopicsGoverning equations and hydrostatic/non-hydrostati
 c dynamicsThe spectral transform\, semi-implicit\, semi-Lagrangian integra
 tion method of the ECMWF model dynamical coreHorizontal and vertical discr
 etizationsAdvection schemes\, time-integration methods\, semi-implicit met
 hods and elliptic solversFinite volume and finite element methods\, high-o
 rder discontinuous spectral element discretization methodsStructured and q
 uasi-uniform grids\, unstructured meshes and mesh generationMassively para
 llel computingContrasts and synergies between established numerical weathe
 r prediction models and currently explored machine learning models for wea
 ther predictionRequirementsParticipants should have a good mathematical ba
 ckground and are expected to be familiar with the contents of standard met
 eorological and mathematical textbooks. A familiarity with python is usefu
 l for some of the practical sessions.Other complementary material can be f
 ound in our lecture note series.Some general knowledge or experience in n
 umerical weather prediction is an advantage.All lectures are given in Engl
 ish.\n\nhttps://events.ecmwf.int/event/381/
IMAGE;VALUE=URI:https://events.ecmwf.int/event/381/logo-3547453490.png
LOCATION:ECMWF 
URL:https://events.ecmwf.int/event/381/
END:VEVENT
END:VCALENDAR