ECMWF launches Massive Open Online Course on Machine Learning in Weather and Climate

MOOC Machine Learning in Weather & Climate illustration

Registration is now open for a new Massive Open Online Course (MOOC) on Machine Learning in Weather and Climate. The MOOC will be launched on 9 January 2023 by ECMWF in partnership with the International Foundation on Big Data and Artificial Intelligence for Human Development (IFAB).

The objective is to train a wider community on the impact and use of machine learning in numerical weather and climate predictions. This is a fully online free training course, accessible to a global audience, with different elements suitable for various levels of expertise.

“This is a great opportunity for a wide audience to find out about recent developments concerning machine learning in weather and climate, which has seen a dramatic rise in recent years,” says ECMWF machine learning specialist Peter Düben. “It will include contributions from many experts in our Member and Co-operating States and beyond.”

Who is the MOOC for?

The MOOC is aimed at anyone interested in the topic, be they academics, from operational services, policymakers or the general public. Only a basic knowledge of weather and climate science, statistics and computing is assumed.

Learners may choose to follow one or more of the three tiers of the MOOC. Within each tier, the programme is modular. Participants may select one or more modules within a particular topic subset, according to their interests.

The first tier is an introduction to machine learning in weather and climate. Tier two takes a deeper look at the concepts of machine learning, and tier three demonstrates practical machine learning applications in weather and climate.

The last two tiers are more hands-on and include coding assignments, in which learners have the opportunity to apply machine learning to real-world problems. These tiers are therefore more suited to technical data users from our Member and Co-operating States, academia or industry.

To allow for self-paced learning, the MOOC runs over ten weeks, with around three to four hours of study per week, for a total of 36 hours of training. The end of the MOOC coincides with the application phase of ECMWF’s Summer of Weather Code (ESoWC). Learners will be encouraged to submit proposals for ESoWC coding projects. Participation in ESoWC will come with ECMWF mentoring and the chance to obtain a cash stipend.

What does the MOOC offer?

The MOOC mixes interactive e‑learning with webinars. Learning will be consolidated through a wide variety of activities, from videos and commuter-friendly podcasts to discussion forums and e‑learning modules.

Guided by a host of domain specialists from around the world, learners will be able to explore the subject from different angles. The webinars, for instance, are perfect occasions to interact with leading experts from the field.

The acquired knowledge will then be honed in practical exercises. Using interactive Jupyter notebooks, learners will be able to gain hands-on practical experience with machine learning algorithms.

Throughout their learning journey, participants will be accompanied by a dedicated science communicator, Lisa Burke, who will break down the technical content into plain language whenever needed. Each completed tier will be rewarded with a certificate of completion.

At the end of the live run, the material will continue to be freely accessible from the ECMWF website.

Registration for the MOOC is now open.