Tag - Computing services and tools

Machine Learning for Earth Systems Modelling

Foundations and New Frontiers

Online training course 1
From foundations to hands-on applications!

16 March 2026
Online, self-paced

Level: Fundamentals

Certification course: Yes

Full description:

Overview

Discover how machine learning is transforming extreme weather and climate simulations. Under the Destination Earth (DestinE) initiative, ECMWF is developing a new series of three online courses on Machine Learning for Earth Systems Modelling.

The courses offer a structured learning pathway, from foundations and context to advanced architectures and real-world applications, showing how machine learning (ML) complements physical modelling and supports next-generation digital twins of the Earth.

This first training course of the series introduces the foundations of machine learning in Earth Systems modelling, connecting ML concepts with physical modelling and real applications. Participants explore how AI is used within the DestinE initiative, including the digital twins and next-generation AI simulation systems. The course also addresses ethics, regulation, and future skills, bringing together expert perspectives to prepare learners for advanced ML applications.

Registration Start Date: 11 February, 2026
Registration End Date: March 2026
Course Start Date: 16 March 2026
Course End Date: 17 April 2026

Objectives of the training

After this course, you will be able to:

Objectives

Target audience

This course is the first in a three-part series of DestinE online training courses on Machine Learning for Earth Systems Modelling, targeting researchers, practitioners, or technical specialists from meteorological services, climate centres, academic institutes, or industries.

This first course, Foundations and New Frontiers, specifically targets a broad audience of high-level information users and policy/decision-makers from both European public and private sectors, academia, and industry. This course will also be of interest to a technical audience of non-ML meteorologists who want to be onboarded into the world of ML.

Prerequisites

A background in meteorology, climate systems, or Earth Systems Sciences.

Familiarity with statistics and data interpretation.

Some prior knowledge of machine learning concepts and some coding experience is helpful, but not essential.

Course structure and content

The course consists of a series of modules that combine a range of learning activities and applied examples to support a progressive understanding of ML in Earth System Sciences (ESS). The modules guide learners from core concepts to practical applications, illustrating how ML methods are developed, evaluated, and integrated into weather and climate workflows.

Learning materials include video lectures with visual explanations, interactive webinars, panel discussions featuring experts, and hands-on Python notebooks providing practical demonstrations using open climate datasets and models. Additional resources include curated readings and DestinE documentation, podcasts, auto-graded quizzes to check understanding, and forum discussions that encourage reflection and exchange after selected modules.

Modules

Course lectures and contributors

Logos

This course was created by ECMWF under the DestinE initiative and contracted to Wageningen University (WU) in collaboration with the Karlsruhe Institute of Technology (KIT) and Wageningen Research (WR). Scientific content is provided by European experts from ECMWF, academia, national meteorological services, and private industry (contributor details will be published in the following update).

Lectures

William Becker, ECMWF

Christian Lessig, ECMWF

Matthew Chantry, ECMWF

Stephan Siemens, ECMWF

Peter Dueben, ECMWF

Sara Hahner, ECMWF

Rachel Furner, ECMWF

Peer, Nowack, KIT

Tobias Selz, KIT

Sebastian Schmidt, Fraunhofer IAIS

Tom Beucler, University of Lausanne

Fernando Iglesias Suárez, Predictia

Martin Schultz, Forschungszentrum Jülich

Phillip Stier, University of Oxford

Sophie Buurman, Royal Netherlands Meteorological Institute

Stefanie Holborn, Deutscher Wetterdienst

Contributors

William Becker, ECMWF

Samuel Sutanto, Wageningen University

Wouter Smolenaars, Wageningen University

Natalia Gomez Solano, Wageningen University

Imme Benedict, Wageningen University

Martin Janssens, Wageningen University

Bouke Hefting, Wageningen University

Dwaipayan Chatterjee, KIT

Janine Quist, Wageningen University

Estimated duration: 10 hours

Hide last updated info: Yes

Hide number of enrollments: Yes

Hide full description label: Yes

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Machine Learning

MLWC MOOC 1: Introduction to Machine Learning in Weather and Climate

Six modules introducing the main topics in machine learning in the context of weather and climate.

Level: Fundamentals

Certification course: No

What you'll learn:

An overview of machine learning in weather and climate
Unpack, at a conceptual level, key concepts and topics in ML
Applications and recent advances in the field

Prerequisites:

A basic knowledge of weather and climate, statistics and computing.

Full description:

In this first tier of our MOOC on Machine Learning (ML) in Weather and Climate, we give a broad overview of the key concepts ML and its applications in recent years to topics from forecasting and data assimilation to post-processing, observations and computing. This course is aimed as an introduction, which is accessible to those with an interest in ML, weather and climate, but without necessarily requiring a very technical background.

This course features a mixture of interactive modules, webinar recordings, quizzes, podcasts and code examples.

Please note that this course ran live in early 2023 and reflected the state of the art at that point in time.

Estimated duration: 14 hours

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: Yes

Computing services and tools: Yes

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Machine Learning

Introduction to the Scalable Acquisition and Pre-Processing (SAPP) system

The SAPP system is the ECMWF's operational acquisition and pre-processing system for observations and other input data.

Level: Fundamentals

Certification course: No

What you'll learn:

Raw skills to use the SAPP Jupyter notebook to get you started
How to modify the SAPP system for your requirements
Better understanding of the three-stage data workflow: Acquisition stage, Processing stage and Extraction stage

Full description:

The SAPP system is the ECMWF's operational acquisition and pre-processing system for observations and other input data. This lesson is aimed at users with little or no experience of SAPP and will provide the raw skills needed to use SAPP.

Estimated duration: 40 minutes

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Computing services and tools

Using ECMWF computing facilities: the batch system

This lesson focuses on ECGATE - ECMWF's server allocated for users' tasks, from submitting jobs to correcting errors.

Level: Others

Certification course: No

What you'll learn:

How to run tasks in batch
How to submit, query and cancel jobs
How to correct common errors
How to check the accounting database

Full description:

This lesson will focus on ECGATE - ECMWF's server dedicated to the users' work. You will learn how to run tasks in batch, submit, query and cancel jobs, correct common errors and check the accounting database.

Estimated duration: 40 minutes

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Computing services and tools

A starter guide to ecFlow

ecFlow is a workflow package that enables users to run a large number of programmes behind a firewall.

Level: Fundamentals

Certification course: No

What you'll learn:

Understand ecFlow vocabulary
Start and stop the server
Use the command line and Python clients
Design a simple suite with its related task headers and script templates

Full description:

ecFlow is a workflow package that enables users to run programmes behind a firewall. During this lesson, you will learn ecFlow vocabulary and how to run a simple suite.

Estimated duration: 40 minutes

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Computing services and tools

ecCodes – Manipulating GRIB data with tools and API

This lesson is focused on how to gain flexibility and control when handling GRIB data using advanced ecCodes tools.

Level: Others

Certification course: No

What you'll learn:

How to handle and manipulate GRIB data with the advanced ecCodes GRIB tools, such as: grib_copy, grib_set, grib_-lter
Basic features of the ecCodes Application Programming Interfaces (API), which will give you the highest flexibility and level of control when handling GRIB data

Full description:

ecCodes is software developed by ECMWF to decode and encode in WMO GRIB and BUFR formats. This lesson focuses on handling GRIB data with ecCodes tools.

Estimated duration: 40 minutes

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Computing services and tools

MARS ECMWF's meteorological archive

The Meteorological Archival and Retrieval System (MARS) enables users to access and retrieve ECMWF’s historical data.

Level: Fundamentals

Certification course: No

What you'll learn:

What the MARS catalogue is
How the data is organised in the archive
How to create a customised data retrieval
How to improve the efficiency of your MARS requests

Full description:

The Meteorological Archival and Retrieval System (MARS) enables users to access ECMWF’s data. This lesson will teach you how to create a customised data retrieval.

Estimated duration: 30 minutes

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Computing services and tools

Metview for the Single-Column Model (SCM)

This lesson provides an overview of Metview's main features to analyse and edit input data for the single-column model.

Level: Others

Certification course: No

What you'll learn:

A quick overview of Metview's main features
How to use Metview to analyse and edit input data for the single-column model
How to run the model and visualise its output

Full description:

This lesson will provide a quick overview of Metview's main features and enable you to use Metview to analyse and edit input data for the single-column model, run the model and visualise its output.

Estimated duration: 30 minutes

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Computing services and tools

Exploring meteorological data through OGC web services

This lesson describes the web services used to visualise geographical data and outlines what OGC and INSPIRE are.

Level: Others

Certification course: No

What you'll learn:

Define what web services are
Explain the typical challenges of meteorological data
Describe how the meteorological community can benefit from web services
Explain why we need data standards
Outline what OGC and INSPIRE are

Full description:

Web services are used to visualise geographical data. This lesson describes web services, data standards and outlines what OGC and INSPIRE are.

Estimated duration: 40 minutes

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Computing services and tools

Introduction to Metview

Metview is a meteorological workstation application that enables you to access and visualise meteorological data.

Level: Fundamentals

Certification course: No

What you'll learn:

How to use Metview to handle meteorological data files
How to inspect metadata
How to perform computations
How to visualise the results
Hands-on experience in creating sophisticated layouts comprising multiple plots on the same page
Generate and run Macro programs to obtain plots via an automated process

Full description:

Metview is a powerful meteorological workstation application that enables you to access, process and visualise meteorological data. In this lesson you will learn how to use MetView.

Estimated duration: 40 minutes

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Computing services and tools

ecCodes: Decoding with GRIB tools

ecCodes is an application programming interface developed by ECMWF to decode and encode in WMO GRIB format.

Level: Others

Certification course: No

What you'll learn:

What ecCodes is
What GRIB format is
Inspect the content of GRIB messages using the following GRIB tools: grib_ls, grib_dump, grib_get and grib_get_data

Full description:

ecCodes is software developed by ECMWF to decode and encode in WMO GRIB and BUFR formats. This lesson will teach you how to inspect GRIB messages using GRIB tools.

Estimated duration: 30 minutes

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Computing services and tools

Introduction to BUFR decoding with ecCodes

ecCodes is an application programming interface developed by ECMWF to decode and encode in WMO GRIB and BUFR formats.

Level: Fundamentals

Certification course: No

What you'll learn:

Explain what BUFR is
Describe its structure
Identify data descriptors
Use ecCodes tools for decoding BUFR data

Full description:

ecCodes is software developed by ECMWF to decode and encode in WMO GRIB and BUFR formats. This lesson will introduce you to the BUFR format for decoding of BUFR data.

Estimated duration: 30 minutes

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Computing services and tools

MARS – advanced retrievals, data manipulation and computations

The Meteorological Archival and Retrieval System (MARS) enables access to ECMWF data. Explore its computing capability

Level: Others

Certification course: No

What you'll learn:

How to combine multiple requests into one call to MARS
How to write fields into multiple files
How to manipulate already retrieved data
Explain the concept of fieldsets
MARS computations

Full description:

The Meteorological Archival and Retrieval System (MARS) enables users to access ECMWF’s data. This lesson will look at MARS requests and explore its compute capability.

Estimated duration: 40 minutes

Hide last updated info: No

Hide number of enrollments: No

Hide full description label: No

Atmospheric composition: No

Climate: No

Computing services and tools: No

Data applications: No

Forecasting: No

Machine learning: No

Numerical weather prediction: No

Click to enter this course

Category: Computing services and tools

Computing services and tools

Courses tagged with "Computing services and tools"

Foundations and New Frontiers

Overview

Objectives of the training

Target audience

Prerequisites

Course structure and content

Course lectures and contributors

Lectures

Contributors