We are invited to present at the Using ECMWF’s Forecasts' (UEF 2023) in Reading (UK) on June 5-8th, 2023, the challenges we face in developing an AI-Enhanced operational demonstrator of sub-seasonal and seasonal forecasting of detection and attribution of heatwaves and warm nights we develop in in the H2020 CLINT project
The 'Using ECMWF’s Forecasts' (UEF 2023) forum serves as a global platform for ECMWF forecast users to engage in discussions about the utilization and performance of ECMWF's forecasts and associated products. Held at ECMWF's headquarters in Reading, UK, the event provides an opportunity for users worldwide to offer feedback on forecast performance, explore available products, learn about recent developments in ECMWF's forecasting system, and foster connections within the ECMWF community.
This year's theme is 'Ensemble Forecasting,' chosen based on suggestions from previous UEF meetings, recent events, and discussions with ECMWF staff and users. The event features livestreamed oral presentations, facilitating global participation.
The focus of the forum includes an overview of ECMWF's ensemble predictions, tracing their evolution over 30 years. Beginning with the first operational ensemble predictions in 1992, the Ensemble Prediction System (EPS) has evolved to run twice a day at 18 km resolution with 137 levels and 51 members, extending predictions up to 15 days. The Extended Range ensemble (ENS-extended) now runs twice a week at 36 km, with 51 members forecasting up to 46 days ahead.
Ensemble forecasts play a crucial role across various fields, including meteorology, hydrology, air quality, and climate. The upcoming model cycle upgrade (48r1) in 2023 will bring significant changes. These include increased resolution to 9 km for ENS and daily runs with 101 members for ENS-extended.
While ensemble data contributes to various ECMWF forecast products, UEF2023 also aims to address challenges in daily ensemble usage. Despite their potential benefits, some users lean towards more deterministic approaches. The forum seeks to explore these challenges and understand the reasons behind the underutilization of ensembles, fostering a deeper appreciation for their role in forecasting.
