资源环境科技发展态势分析平台

Fifty scientists met in the Italian city of Trieste from 5 to 9 June 2017 to explore seasonal predictability and teleconnections in the Earth system using OpenIFS, a version of ECMWF’s Integrated Forecasting System (IFS).

The workshop was a mix of presentations and hands-on sessions with the OpenIFS model, where teams investigated what the weather would have been had the strong El Niño of 2015/16 been different.

The participants included existing OpenIFS users who showed the variety of teaching and research activities under way with the model. These ranged from super-parametrization of convection and idealised climate change studies to new software tools for creating ensemble forecasts.  

Seasonal forecast experiments

ECMWF scientist Franco Molteni devised an experiment based on the El Niño conditions in late 2015 and early 2016. Participants were given ECMWF analyses of the anomalous sea-surface temperatures (SST) in different parts of the globe at that time.

They were able to alter the SST fields using Metview tools developed by ECMWF analyst Sándor Kértesz. They then ran their own seasonal forecasts with the OpenIFS model to explore the effects of their changes.

Some turned the El Niño conditions into a La Niña event, others removed the Indian Ocean SST anomaly. Discussions of the impact on weather through atmospheric teleconnections were instructive and may be followed up in research papers.

For the first time in an OpenIFS workshop, the experiments were run remotely on ECMWF's supercomputer in the UK before the results were transferred to the classroom computers. This enabled participants to learn first-hand how to create an experiment, run the OpenIFS model and process the model output.

Some of the experiments revolved around different El Niño scenarios and their potential consequences for global weather patterns.

User networking

The user meeting, held once every two years, was also an opportunity to exchange information on using OpenIFS and engage with ECMWF scientists.

“The presentations by Director of Research Erland Källén on current developments at ECMWF and by Franco Molteni on seasonal forecasting and predictability research were well received,” said ECMWF scientist Filip Váňa, who is part of the OpenIFS team.

“These workshops are also valuable in allowing existing users to meet with interested groups,” he added.

One of the earliest adopters of the OpenIFS model, Dr Victoria Sinclair from the University of Helsinki, presented a study of storms in a warmer climate.

She described OpenIFS simulations on an 'aqua planet' (no land) in which sea-surface temperatures and carbon dioxide were adjusted in the model to reflect a warmer climate, and explained how this produced more extreme cyclones.

 ‘Most successful yet’

Glenn Carver, ECMWF’s OpenIFS coordinator, described the event as “the most successful OpenIFS workshop yet”.

“It was the first workshop to take place over five days instead of three for previous meetings and was over-subscribed. Feedback was very positive,” he said.

Participants presented and discussed the results of their OpenIFS experiments.

There were plenty of opportunities to present recent work with the IFS and exchange information with other users.

The Trieste meeting took place at the International Centre for Theoretical Physics (ICTP).

It was organised locally by Fred Kucharski with Paolo Ruggieri from the University of L’Aquila, Susanna Corti from ISAC-CNR, Bologna, and Glenn Carver from ECMWF. It was supported financially by ICTP, the EU-funded ESiWACE project and the CETEMPS Environmental Centre of Excellence in L'Aquila, Italy.

During the workshop, a Young Scientist Travel Award from the European Meteorological Society was presented by Professor Erland Källén to Lenka Novak from the University of Reading.

All workshop presentations were recorded and are available on the ICTP website.

OpenIFS provides a supported, easy-to-use version of the IFS model under licence to national meteorological and hydrological services, research institutes and universities.

It is currently used by over 50 licensed institutes in 20 countries worldwide, for both research and education, in fields ranging from weather prediction to climate modelling.

Workshop participants worked in groups to devise their own experiments.