ECMWF–ESA Workshop on Machine Learning for Earth Observation and Prediction

Name: ECMWF–ESA Workshop on Machine Learning for Earth Observation and Prediction
Start: 2022-11-14T09:00:00+00:00
End: 2022-11-17T17:00:00+00:00
Location: ECMWF

Overview
Presentations and recordings
Posters
Organisers
Workshop photos
Contact

Material Package

Added Since

Include only attachments uploaded after this date

Include*

Everything

Specific sessions

Specific contributions

Specific days

Sessions*

Registration and welcome

Keynotes: Machine Learning for the ESOP - setting the scene

Session 1.1 Machine Learning for Earth Observations

Session 1.2 Machine Learning for Earth Observations

Session 1.3 Machine Learning for Earth Observations

Session 2.1 Hybrid Data Assimilation - Machine Learning

Session 2.2 Hybrid Data Assimilation - Machine Learning

Session 2.3 Hybrid Data Assimilation - ML and ML at the edge

Workshop dinner

Session 3.1 Machine Learning for Model emulation and Model discovery

Session 3.2 Machine Learning for Model emulation and Model discovery

Session 4.2 Machine Learning for user-oriented Earth Science applications

Session 4.1 Machine Learning for user-oriented Earth Science applications

Working Groups

Session 5: Working Groups plenary discussion and close

Include materials from selected sessions

Contributions*

A Deep Feed-Forward Neural Network to reconstruct the Mediterranean 3D chlorophyll-a and temperature fields from satellite measurements (in session "Session 1.1 Machine Learning for Earth Observations")

A Gentle Introduction to Creating and Evaluating Uncertainty Estimates with Neural Networks for Earth Science Applications (in session "Session 2.2 Hybrid Data Assimilation - Machine Learning")

A physically informed recurrent neural network approach for emulating radiative transfer (in session "Session 3.2 Machine Learning for Model emulation and Model discovery")

A reproducible ensemble machine learning approach to forecast dengue outbreaks (in session "Session 4.2 Machine Learning for user-oriented Earth Science applications")

An iterative data-driven model of an Ocean GCM (in session "Session 3.1 Machine Learning for Model emulation and Model discovery")

An overview of AI/ML and Data Assimilation: current state of the art, challenges, and opportunities (in session "Keynotes: Machine Learning for the ESOP - setting the scene")

Application of machine learning to globally model tropospheric parameters (in session "Session 1.3 Machine Learning for Earth Observations")

Beyond one iteration of machine learning and data assimilation steps for learning meteorological models (in session "Session 2.1 Hybrid Data Assimilation - Machine Learning")

Bringing the power of quantum machine learning to Earth observation (in session "Session 2.3 Hybrid Data Assimilation - ML and ML at the edge")

Burned area prediction and model parameter identification for wildfire events using machine learning techniques (in session "Session 2.1 Hybrid Data Assimilation - Machine Learning")

Can we Design NWP Data Assimilation Based Entirely on AI Techniques? Assessing Advantages and Challenges (in session "Session 2.2 Hybrid Data Assimilation - Machine Learning")

Causal Inference for Sustainable and Resilient Agriculture (in session "Session 4.1 Machine Learning for user-oriented Earth Science applications")

Data Assimilation with a Deep Convolutional Variational AutoEncoder (in session "Session 2.3 Hybrid Data Assimilation - ML and ML at the edge")

Data Learning for more reliable digital twins (in session "Session 2.1 Hybrid Data Assimilation - Machine Learning")

Deep Learning for Empirical Downscaling of Earth Science gridded data (in session "Session 4.2 Machine Learning for user-oriented Earth Science applications")

Drought monitoring with Earth Observation Data and Machine Learning methods (in session "Session 4.2 Machine Learning for user-oriented Earth Science applications")

Explainable Uncertainty in Machine Learning Weather Prediction (in session "Session 4.2 Machine Learning for user-oriented Earth Science applications")

Extended Elman Network for Bayesian Data Assimilation (in session "Session 2.1 Hybrid Data Assimilation - Machine Learning")

Generative machine learning methods for multivariate ensemble post-processing (in session "Session 4.1 Machine Learning for user-oriented Earth Science applications")

Ice breaker (in session "Session 1.2 Machine Learning for Earth Observations")

Inferring High-resolution Near-surface NO2 Concentrations over Belgium through Convolutional Neural Networks (in session "Session 1.3 Machine Learning for Earth Observations")

Latent Space Data Assimilation by Deep Learning (in session "Session 2.3 Hybrid Data Assimilation - ML and ML at the edge")

Lunch and Poster session (in session "Keynotes: Machine Learning for the ESOP - setting the scene")

Lunch and Poster session (in session "Session 2.1 Hybrid Data Assimilation - Machine Learning")

Machine Learning for NASA Advanced Information Systems (in session "Session 1.1 Machine Learning for Earth Observations")

Machine learning in (and beyond) the ESA Climate Change Initiative (in session "Session 1.1 Machine Learning for Earth Observations")

Meta-learning data-efficient Machine Learning Models for Diverse Earth Observation Problems (in session "Session 1.2 Machine Learning for Earth Observations")

ML and EO for (global) mapping of the environment: discussing challenges of model extrapolation and accuracy assessment (in session "Session 1.2 Machine Learning for Earth Observations")

NWP Models with ML Components: Challenges and Perspectives (in session "Session 3.1 Machine Learning for Model emulation and Model discovery")

Occam's machete: Data-driven discovery with parsimony and causal invariance (in session "Session 3.1 Machine Learning for Model emulation and Model discovery")

Online model error correction with neural networks -- towards an implementation in the ECMWF data assimilation system (in session "Session 2.2 Hybrid Data Assimilation - Machine Learning")

Open Environmental Data Cube Europe: Analysis-ready datasets produced through ensemble machine learning (in session "Session 1.2 Machine Learning for Earth Observations")

Opening (in session "Registration and welcome")

Quantum Technologies in Earth Observation (in session "Session 2.3 Hybrid Data Assimilation - ML and ML at the edge")

Recurrent Neural Network Emulation for High Resolution Forecasting (in session "Session 3.2 Machine Learning for Model emulation and Model discovery")

Registration (in session "Registration and welcome")

Remote Sensing of Floods (in session "Session 1.1 Machine Learning for Earth Observations")

Representation Learning for Earth Observation and Remote Sensing (in session "Keynotes: Machine Learning for the ESOP - setting the scene")

Seamless prediction of multiple thunderstorm hazards with deep learning (in session "Session 2.2 Hybrid Data Assimilation - Machine Learning")

Segmentation of wildfires, smoke plumes, and burn scars using multi-sensor input and unsupervised and supervised machine learning for improved spatiotemporal coverage and facilitation of automated tracking (in session "Session 1.3 Machine Learning for Earth Observations")

STARCOP: ML models for onboard detection of methane leaks in multispectral and hyperspectral sensors (in session "Session 1.3 Machine Learning for Earth Observations")

Stochastic Downscaling of Precipitation Forecasts with GANs (in session "Session 4.1 Machine Learning for user-oriented Earth Science applications")

SyntEO: Synthetic training data generation for offshore wind energy infrastructure detection in Sentinel-1 imagery (in session "Session 1.2 Machine Learning for Earth Observations")

Temporal downscaling wind climate data using machine learning (in session "Session 3.2 Machine Learning for Model emulation and Model discovery")

Towards flood warnings everywhere - data-driven rainfall-runoff modeling at global scale (in session "Session 4.2 Machine Learning for user-oriented Earth Science applications")

Transforming ocean modeling and forecasting through AI. (in session "Session 4.1 Machine Learning for user-oriented Earth Science applications")

Using GANs as high-resolution, multivariate weather generators (in session "Session 3.1 Machine Learning for Model emulation and Model discovery")

WeatherBench 2.0: A benchmark for the next generation of data-driven weather models (in session "Session 3.2 Machine Learning for Model emulation and Model discovery")

Welcome and introduction - ECMWF (in session "Registration and welcome")

Welcome and introduction - ESA (in session "Registration and welcome")

Working Group 1 - Machine Learning for Earth Observations (in session "Working Groups")

Working Group 2 - Hybrid Data Assimilation - Machine Learning (in session "Working Groups")

Working Group 3 - Machine Learning for Model emulation and Model discovery (in session "Working Groups")

Working Group 4 - Machine Learning for user-oriented Earth Science application (in session "Working Groups")

Working Group 4 - Machine Learning for user-oriented Earth Science applications (in session "Working Groups")

Include materials from selected contributions

Events scheduled on*

14/11/2022

15/11/2022

16/11/2022

17/11/2022

Include materials from sessions/contributions scheduled on the selected dates

ECMWF–ESA Workshop on Machine Learning for Earth Observation and Prediction

Material Package

#AIforEOWS

#ml4esop