ECMWF Users

🎉𝗪𝗲’𝗿𝗲 𝗴𝗼𝗶𝗻𝗴 𝗢𝗣𝗘𝗡! 🎉 From 1 October 2025 on, ECMWF forecast data will be more accessible than ever - under a 𝗖𝗖-𝗕𝗬-𝟰.𝟬 𝗹𝗶𝗰𝗲𝗻𝗰𝗲. From now until our Open Data is here, we will be providing you with with monthly info posts around a specific topic to help you wrap your head around the changes. 𝗢𝘂𝗿 1️⃣𝘀𝘁 𝘁𝗼𝗽𝗶𝗰: “𝗢𝗽𝗲𝗻” 𝗱𝗼𝗲𝘀𝗻’𝘁 𝗺𝗲𝗮𝗻 𝗲𝘃𝗲𝗿𝘆𝘁𝗵𝗶𝗻𝗴 𝗶𝘀 𝗳𝗿𝗲𝗲! ECMWF produces ~100 TB of data daily, and to ensure sustainable access, we’re introducing a clear split: 𝗢𝗽𝗲𝗻 𝗟𝗶𝗰𝗲𝗻𝗰𝗲 - 𝗙𝗿𝗲𝗲 𝗦𝘂𝗯𝘀𝗲𝘁 ✅ Selected real-time forecast data ✅ Free to use, even commercially ✅ Share, adapt, remix ✅ Built for openness and reuse 𝗢𝗽𝗲𝗻 𝗟𝗶𝗰𝗲𝗻𝗰𝗲 - 𝗪𝗶𝘁𝗵 𝗦𝗲𝗿𝘃𝗶𝗰𝗲 𝗖𝗵𝗮𝗿𝗴𝗲𝘀 💡 Same open terms (commercial use, share and adapt) 💡 Delivery to your chosen platform, user access tools 💡 High-volume push delivery options 💡 Support and tailored services 🎓 𝗚𝗼𝗼𝗱 𝗻𝗲𝘄𝘀 𝗳𝗼𝗿 𝗿𝗲𝘀𝗲𝗮𝗿𝗰𝗵𝗲𝗿𝘀: Approved academic projects will continue to receive free access under our Basic service pack — as long as the work isn’t commercial or service-based (incl. redistribution). We’ve put together a visual below on "what's free & what's not" if you want to share with colleagues. Let us know what your questions are, either in the comments or via our Support Portal! --- Links: 🔗 Open Data FAQs: https://lnkd.in/eZJs-sVz 🔗 User Support: http://support.ecmwf.int/ #ECMWF #OpenData #Forecasting #DataAccess #Licensing #Meteorology #ScienceForSociety

🌐 𝗧𝗵𝗶𝘀 𝘁𝗶𝗺𝗲 𝗻𝗲𝘅𝘁 𝘆𝗲𝗮𝗿 𝗮𝗹𝗹 𝗘𝗖𝗠𝗪𝗙 𝗳𝗼𝗿𝗲𝗰𝗮𝘀𝘁 𝗱𝗮𝘁𝗮 𝘄𝗶𝗹𝗹 𝗯𝗲 𝗽𝗿𝗼𝘃𝗶𝗱𝗲𝗱 𝗶𝗻 𝗚𝗥𝗜𝗕𝟮 𝗳𝗼𝗿𝗺𝗮𝘁🌐 And no time like the present to get ready! GRIB2 will potentially be a big change for some of our users so we have started preparing documentation to help you with the transition. Don't forget that now already, the totality of our AIFS forecast and our Open Data forecast datasets are available in GRIB2 data format, but the rest of our products remain a mix of GRIB1 and GRIB2. 𝗪𝗵𝗮𝘁 𝘀𝘂𝗽𝗽𝗼𝗿𝘁 𝗶𝘀 𝗮𝘃𝗮𝗶𝗹𝗮𝗯𝗹𝗲 𝗻𝗼𝘄? 🔹A documentation web space describing the changes, and also providing reminders about GRIB data. This is being updated continuously on: https://lnkd.in/eKZhfVTM 🔹A recent webinar recording highlighting the GRIB2 migration roadmap and significant changes. This is available on our YouTube channel: https://lnkd.in/eZgRJHVy 𝗪𝗵𝗮𝘁 𝘀𝘂𝗽𝗽𝗼𝗿𝘁 𝗶𝘀 𝗰𝗼𝗺𝗶𝗻𝗴? 🔹We are preparing a static test dataset for users to examine the changes in advance - and of course to give us feedback. 🔹We are working on a version of ecCodes that will be able to handle the new data. 🔹New ecCodes and test datasets are coming soon, and we will inform users when they are ready. ➡️ To keep up to date about all things GRIB2 - updates, user documentation, deadlines, changes - make sure subscribe to our 𝗠𝗶𝗴𝗿𝗮𝘁𝗶𝗼𝗻 𝘁𝗼 𝗚𝗥𝗜𝗕𝟮 mailing list, by sending an email to 𝘴𝘺𝘮𝘱𝘢@𝘭𝘪𝘴𝘵𝘴.𝘦𝘤𝘮𝘸𝘧.𝘪𝘯𝘵, with the subject line "𝘚𝘜𝘉𝘚𝘊𝘙𝘐𝘉𝘌 𝘮𝘵𝘨2@𝘭𝘪𝘴𝘵𝘴.𝘦𝘤𝘮𝘸𝘧.𝘪𝘯𝘵". ⁉️As always, if you need any further information you can contact us via our Support Portal at https://support.ecmwf.int/

🤖 𝗔𝗜 𝗪𝗲𝗮𝘁𝗵𝗲𝗿 𝗤𝘂𝗲𝘀𝘁 𝗺𝗼𝘃𝗲𝘀 𝗶𝗻𝘁𝗼 𝘁𝗲𝘀𝘁𝗶𝗻𝗴 𝗺𝗼𝗱𝗲🤖 The 𝗘𝗖𝗠𝗪𝗙 𝗔𝗜 𝗪𝗲𝗮𝘁𝗵𝗲𝗿 𝗤𝘂𝗲𝘀𝘁 invites the ECMWF user community to explore how AI/ML can support sub-seasonal prediction skill — with a focus on probabilistic forecasts for days 19–25 and 26–31 ahead. 💡 𝗛𝗼𝘄 𝗶𝘀 𝘁𝗵𝗶𝘀 𝗿𝗲𝗹𝗲𝘃𝗮𝗻𝘁 𝘁𝗼 𝗘𝗖𝗠𝗪𝗙 𝘂𝘀𝗲𝗿𝘀? This is a practical opportunity to test AI/ML-driven forecasting workflows, benchmark different methods, and engage with ECMWF’s data and evaluation tools in a structured, transparent setting. 📊 𝗙𝗼𝗿𝗲𝗰𝗮𝘀𝘁 𝘁𝗮𝗿𝗴𝗲𝘁𝘀: 🔹Near-surface air temperature 🔹Precipitation 🔹Mean sea level pressure 🧠 𝗣𝗮𝗿𝘁𝗶𝗰𝗶𝗽𝗮𝗻𝘁𝘀 𝗰𝗮𝗻: 🔹Develop AI/ML models using any observational or forecast datasets — including ECMWF-supported data 🔹Use any programming language 🔹Submit any type of AI/ML model 📅 The Testing JJA Period (May–August) is part of the current Initial Training Phase — a chance to submit forecasts and self-evaluate performance before the competitive scoring begins. Join one of our 𝘄𝗲𝗯𝗶𝗻𝗮𝗿𝘀 to learn more, taking place on 𝗪𝗲𝗱𝗻𝗲𝘀𝗱𝗮𝘆 𝟳 𝗠𝗮𝘆: 📌08:00–09:00 UTC ➡️ https://lnkd.in/duzx7zvc 📌14:00–15:00 UTC ➡️ https://lnkd.in/duHxT5F5 We hope to see you take part in the #AIWeatherQuest! --- Useful links: 🔗 AI Weather Quest website: https://lnkd.in/gEq6rAXn 🔗 Morning webinar slot: https://lnkd.in/duzx7zvc 🔗Afternoon webinar slot: https://lnkd.in/duHxT5F5

 📢 𝗔𝗱𝗱𝗶𝘁𝗶𝗼𝗻𝗮𝗹 𝗿𝗲𝘀𝗼𝘂𝗿𝗰𝗲𝘀 𝗳𝗼𝗿 𝘆𝗼𝘂𝗿 𝗘𝗖𝗠𝗪𝗙 𝗿𝗲𝗮𝗹-𝘁𝗶𝗺𝗲 𝗱𝗮𝘁𝗮 𝗺𝗮𝗻𝗮𝗴𝗲𝗺𝗲𝗻𝘁 ❗ ❗ As a Member & Co-Operating State official or a Gold & Silver licensed user, you can install the forecast data you need and manage its dissemination. Need a hand? We are happy to offer video recordings for self-learning that explore the two ECMWF’s core data applications, the 𝙋𝙧𝙤𝙙𝙪𝙘𝙩 𝙍𝙚𝙦𝙪𝙞𝙧𝙚𝙢𝙚𝙣𝙩𝙨 𝙀𝙙𝙞𝙩𝙤𝙧 (𝙋𝙍𝙀𝙙) and the 𝙀𝘾𝙈𝙒𝙁 𝙋𝙧𝙤𝙙𝙪𝙘𝙩𝙞𝙤𝙣 𝘿𝙖𝙩𝙖 𝙎𝙩𝙤𝙧𝙚 (𝙀𝘾𝙋𝘿𝙎). We recently held these two webinars as part of our ongoing user support, but wanted to make them available for everyone to rewatch and learn from.   𝗛𝗼𝘄 𝗰𝗮𝗻 𝗜 𝘂𝘀𝗲 𝘁𝗵𝗲 𝗣𝗥𝗘𝗱? The PREd is a web application designed for configuring and maintaining real-time dissemination requirements. It allows you to: ⚙️ Modify and manage real-time data dissemination requirements for your organisation ⚙️ Utilize a validation system to check for configuration errors before submitting publication requests ⚙️ Ensure that once requirements are installed in the PREd, the ECPDS will produce and deliver the corresponding data to your system Learn how 📹 https://lnkd.in/d2bgFqDC   𝗛𝗼𝘄 𝗰𝗮𝗻 𝗜 𝘂𝘀𝗲 𝘁𝗵𝗲 𝗘𝗖𝗣𝗗𝗦? The ECPDS is designed to store and disseminate real-time forecast data. It allows you to: ⚙️  Receive and manage real-time forecast data through various transfer protocols such as: FTP, SFTP, Amazon S3 (Native and S3-compatible API), Microsoft Azure Blob Storage, Google Cloud Storage ⚙️  Monitor and control data transmissions, with options to pause, restart, and re-queue transfers, view host configuration and check network throughput ⚙️ Based on requirements set in the PREd system, ECPDS stores and disseminates the requested data to your end point Learn how 📹 https://lnkd.in/dg5dQnFB --- Links: 🔗 ECPDS page: https://lnkd.in/d7myWXZK 🔗 PREd page: https://lnkd.in/dp-MJXHx 🔗 User Support: http://support.ecmwf.int/

📢 𝗖𝗼𝗱𝗲 𝗳𝗼𝗿 𝗘𝗮𝗿𝘁𝗵 𝗯𝗿𝗶𝗻𝗴𝘀 𝗼𝗽𝗲𝗻 𝗶𝗻𝗻𝗼𝘃𝗮𝘁𝗶𝗼𝗻 𝘁𝗼 𝗼𝘂𝗿 𝘂𝘀𝗲𝗿𝘀!  📢 The Call for Participation for the new edition in 2️⃣0️⃣2️⃣5️⃣ is open until Tuesday, 1 April!  𝗪𝗵𝗮𝘁 𝗶𝘀 𝗖𝗼𝗱𝗲 𝗳𝗼𝗿 𝗘𝗮𝗿𝘁𝗵? 🌍It is a European Centre for Medium-Range Weather Forecasts - ECMWF-led innovation programme that supports cutting-edge open source developments in weather forecasting, climate and atmospheric monitoring, Copernicus and Destination Earth. #CodeforEarth challenges the ECMWF user community to create advancements and deliver technical solutions related to the broad scope of activities at ECMWF, including data science, AI/ML applications to weather, climate or other Earth sciences, visualisation and more. 𝗪𝗵𝗮𝘁’𝘀 𝗻𝗲𝘄 𝗶𝗻 𝟮𝟬𝟮𝟱? 🆕This edition has released 15 unique mentor-identified challenges focused on three areas: #datavisualization, #machinelearning and #softwaredevelopment All the challenges are listed on GitHub ➡️ https://lnkd.in/e6Pgy-nr 𝗚𝗲𝘁 𝗶𝗻𝘃𝗼𝗹𝘃𝗲𝗱! ⚙️Developers, data enthusiasts, and experts in Earth sciences (weather, climate, atmosphere), computer science, and software engineering from the ECMWF user community are invited to submit proposals and, if selected, team up with experienced mentors from ECMWF and partner organisations to work on project-driven solutions during the summer. Selected and successfully completed projects will receive a €5,000 stipend❗ ℹ️Learn more about the programme at https://lnkd.in/eWqTb6gN  📩 Don’t forget to 𝘀𝘂𝗯𝗺𝗶𝘁 𝘆𝗼𝘂𝗿 𝗽𝗿𝗼𝗽𝗼𝘀𝗮𝗹 𝗯𝘆 𝟭 𝗔𝗽𝗿𝗶𝗹 𝟮𝟬𝟮𝟱, 𝟮𝟯:𝟱𝟵 𝗖𝗘𝗧

📣 𝗘𝘃𝗲𝗻 𝗺𝗼𝗿𝗲 𝘁𝗿𝗮𝗶𝗻𝗶𝗻𝗴𝘀 𝗳𝗼𝗿 2️⃣0️⃣2️⃣5️⃣ 📣 We are pleased to announce the dates for the our next batch of training courses, taking place at our headquarters in Reading (UK), as well as online:    🎓 Online course: 𝗔𝘁𝗼𝘀 𝗛𝗣𝗖 𝘁𝗿𝗮𝗶𝗻𝗶𝗻𝗴 - 16 to 19 June 2025 - https://lnkd.in/eVHBu5Uj 🎓 In-person course: 𝗨𝘀𝗲 𝗮𝗻𝗱 𝗜𝗻𝘁𝗲𝗿𝗽𝗿𝗲𝘁𝗮𝘁𝗶𝗼𝗻 𝗼𝗳 𝗘𝗖𝗠𝗪𝗙 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝘀 – 6 to 9 October 2025 - https://lnkd.in/efbvcmqv 🎓 In-person course: 𝗠𝗮𝗰𝗵𝗶𝗻𝗲 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗳𝗼𝗿 𝘄𝗲𝗮𝘁𝗵𝗲𝗿 𝗽𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝗼𝗻 – 27 to 31 October 2025 - https://lnkd.in/eGGKk7bi 🎓 In-person course: 𝗔 𝗵𝗮𝗻𝗱𝘀-𝗼𝗻 𝗶𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝘁𝗼 𝗡𝗪𝗣 𝗺𝗼𝗱𝗲𝗹𝘀: 𝘂𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱𝗶𝗻𝗴 𝗮𝗻𝗱 𝗲𝘅𝗽𝗲𝗿𝗶𝗺𝗲𝗻𝘁𝗶𝗻𝗴 – 10 to 14 November 2025 - https://lnkd.in/e2F6VRdA   Applications are open as of today, and the deadline for the online course is 𝟮𝟬 𝗠𝗮𝘆 𝟮𝟬𝟮𝟱, and all in-person courses close on 𝟭 𝗝𝘂𝗹𝘆 𝟮𝟬𝟮𝟱. We look forward to receiving your applications!   For more information, including a course description, please have a look at the individual event webpages. More information, such as the modules to be covered, will be published in due course. We will also be continuing our series of shorter webinar-style trainings on the European Weather Cloud (EWC) throughout the year, they will be announced on our website. --- Links: 🔗 ECMWF trainings website: https://lnkd.in/eaHBJX8 🔗 Application guidelines: https://lnkd.in/eyYZmytd 🔗 Our e-learning platform: https://learning.ecmwf.int

🚀 𝗛𝗲𝗹𝗽 𝘀𝗵𝗮𝗽𝗲 𝗼𝘂𝗿 𝗺𝗮𝗰𝗵𝗶𝗻𝗲 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝘁𝗿𝗮𝗶𝗻𝗶𝗻𝗴 𝗶𝗻 𝘄𝗲𝗮𝘁𝗵𝗲𝗿 & 𝗰𝗹𝗶𝗺𝗮𝘁𝗲! 🌍 We are launching a survey to understand your training needs in machine learning in weather and Earth modelling. Your insights will help us design future courses that truly meet your needs! ➡️Take the survey now: https://lnkd.in/e9P2iSNe 🕒 It only takes five minutes. Let’s build better training together—your input makes a difference! --- 🔗 Training website: https://lnkd.in/ev6vQE9U 🔗 Our training offering: https://lnkd.in/eaHBJX8 🔗 Our e-learning platform: https://learning.ecmwf.int 🔗 Contact us: at training@ecmwf.int #MachineLearning #WeatherScience #ClimateAI #Training #ECMWF

📢 𝗨𝗘𝗙𝟮𝟬𝟮𝟱: 𝗥𝗲𝗴𝗶𝘀𝘁𝗿𝗮𝘁𝗶𝗼𝗻 𝗮𝗻𝗱 𝗮𝗯𝘀𝘁𝗿𝗮𝗰𝘁 𝘀𝘂𝗯𝗺𝗶𝘀𝘀𝗶𝗼𝗻𝘀 𝗡𝗢𝗪 𝗢𝗣𝗘𝗡! 📢 Join us at Using ECMWF Forecasts (UEF2025) which will be held 15 - 18 September 2025 in Bologna, Italy! 💡 𝗪𝗵𝘆 𝗶𝘀 𝘁𝗵𝗶𝘀 𝗶𝗺𝗽𝗼𝗿𝘁𝗮𝗻𝘁 𝗳𝗼𝗿 𝘆𝗼𝘂? The UEF is the largest user-focused event for our meteorological services users and forecasters. It provides a forum to: 🔹 Give feedback on ECMWF forecast performance and available products and services 🔹 Stay informed about ECMWF’s latest updates and plans 🔹 Connect with ECMWF and fellow users to share experiences and network 🔥 𝗪𝗵𝗮𝘁 𝗶𝘀 𝘁𝗵𝗲 𝘁𝗵𝗲𝗺𝗲 𝗳𝗼𝗿 𝟮𝟬𝟮𝟱? This year, the theme is 𝘌𝘹𝘱𝘭𝘰𝘳𝘪𝘯𝘨 𝘵𝘩𝘦 𝘗𝘢𝘴𝘵, 𝘗𝘳𝘦𝘴𝘦𝘯𝘵 𝘢𝘯𝘥 𝘍𝘶𝘵𝘶𝘳𝘦 𝘰𝘧 𝘌𝘊𝘔𝘞𝘍 𝘧𝘰𝘳𝘦𝘤𝘢𝘴𝘵 𝘱𝘳𝘰𝘥𝘶𝘤𝘵𝘴 𝘢𝘯𝘥 𝘴𝘦𝘳𝘷𝘪𝘤𝘦𝘴. As ECMWF celebrates its 50th anniversary, UEF2025 will explore the evolution of ECMWF forecast products and services over the years and into the future. 📌𝗦𝘂𝗯𝗺𝗶𝘁 𝘆𝗼𝘂𝗿 𝗮𝗯𝘀𝘁𝗿𝗮𝗰𝘁 - oral, poster or workshop - now! UEF2025 focuses on the following thematic areas: 🔹Exploring the past - celebrating and examining how ECMWF's products and services have evolved, key milestones achieved, and the impact of collaborative activities 🔹Evaluating the present - showcasing recent advancements in forecasting products and services and their ability to meet the needs of the meteorological community 🔹Envisioning the future - focusing on emerging technologies, novel forecasting methodologies, the future integration of machine learning, evolving user needs, and the ongoing importance of collaboration and communication  🔹Special session - UEF2025 meets HPC - joined with the 21st High Performance Computing workshop, exploring crossover which will appeal to both forecast user and HPC user communities ℹ️ More information, registration and abstract submission: https://lnkd.in/dBKTgVi6

📅𝗜𝗻𝗳𝗼𝗦𝗲𝘀𝘀𝗶𝗼𝗻: 𝗦𝗽𝗮𝗰𝗲 𝗳𝗼𝗿 𝗘𝗮𝗿𝗹𝘆 𝗪𝗮𝗿𝗻𝗶𝗻𝗴 𝗶𝗻 𝗔𝗳𝗿𝗶𝗰𝗮📅 We're pleased to announce our first webinar for our new programme SEWA - Space for Early Warning in Africa. Please join us on 📌 𝗪𝗲𝗱𝗻𝗲𝘀𝗱𝗮𝘆 1️⃣9️⃣ 𝗠𝗮𝗿𝗰𝗵 𝟮𝟬𝟮𝟱 at 𝟭𝟬:𝟬𝟬𝗮𝗺 𝗖𝗘𝗧. In this InfoSession you'll learn about the scope of the programme, the roles of the involved stakeholders, as well as the procurement and grants opportunities of the programme. If you're from the African meteorological communities or from a European National Meteorological and Hydrological Service - or simply want to know more - please register here: 🔗https://lnkd.in/ehRWpUab 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗦𝗘𝗪𝗔? SEWA is funded by the European Commission's DG for International Partnerships (DG INTPA), and ECMWF is contributing to 𝗦𝗘𝗪𝗔 alongside EUMETSAT and the African Union Commission (AUC). The programme is to develop space-based services and applications to strengthen early warning systems for hazardous weather and climate-related events in sub-Saharan Africa. This Action also contributes to 𝗨𝗡 #𝗘𝗪𝟰𝗔𝗹𝗹. We officially started this project in January 2025, you can read more here: https://bit.ly/3ELMHQL We look forward to seeing you! --- 🔗Registration: https://lnkd.in/ebQYJX2t 🔗Agenda: https://lnkd.in/ehRWpUab

💫 𝗗𝗶𝗱 𝘆𝗼𝘂 𝗸𝗻𝗼𝘄 𝘁𝗵𝗮𝘁 𝘆𝗼𝘂 𝗰𝗮𝗻 𝗿𝘂𝗻 𝗔𝗜𝗙𝗦 𝘆𝗼𝘂𝗿𝘀𝗲𝗹𝗳? 💫 We have taken a huge step towards making our machine-learning-driven forecast even more accessible! Besides exploring AIFS Single real-time data and charts, you can now also run the latest version of the AIFS Single model on your own device using ECMWF's 𝘰𝘱𝘦𝘯 𝘥𝘢𝘵𝘢 𝘥𝘢𝘵𝘢𝘴𝘦𝘵 and the 𝘢𝘯𝘦𝘮𝘰𝘪-𝘪𝘯𝘧𝘦𝘳𝘦𝘯𝘤𝘦 package. 𝗛𝗼𝘄 𝘁𝗼 𝗴𝗲𝘁 𝘀𝘁𝗮𝗿𝘁𝗲𝗱? The latest deterministic version of the AIFS (AIFS Single) v1, is not only open source, it is also is available on 𝘏𝘶𝘨𝘨𝘪𝘯𝘨 𝘍𝘢𝘤𝘦, a platform for sharing and collaborating on machine learning models. ➡️Simply head to AIFS Single model card: https://lnkd.in/eY7TMEqw where you will find a Jupyter notebook with step-by-step instructions including: 🔹Installing necessary packages 🔹Retrieving initial conditions from ECMWF Open Data 🔹Generating forecasts using anemoi-inference 🔹Inspecting your generated forecast We’re excited to see your experiments, innovations, and applications using the AIFS! 🌍☁️⚡ Try it out and share your insights with us, either through support portal: http://support.ecmwf.int/ or our forum: https://forum.ecmwf.int/