Machine learning for seamless thunderstorm nowcasting from multiple data sources
Online: 6 October 2021, 11:00 UTC
Photo credit: Carsten Schaefer
Webinar with Jussi Leinonen (MeteoSwiss), moderator: Mark Higgins (EUMETSAT).
29 September 2021
16 September 2021
We are working to develop thunderstorm nowcasting machine learning methods that make use of multiple sources of data simultaneously and can be trained to create products for different hazards according to the needs of the end user.
This session will present the results of two related nowcasting studies:
- Analysis of the usefulness of different data sources — this analysis was carried out using the gradient boosting method, which allows the value of different predictors to be quantified in a straightforward fashion.
- Current focus of our efforts, building on the results of the above-mentioned work — the goal of this work is to create a framework based on deep learning for probabilistic nowcasts of the occurrence of hazardous events.
How to participate
Register to join, at least 10 minutes before the session starts at 11:00 UTC.
During the webinar, we will be using Slido to answer your questions. Go to the event on Slido.com (no log in required).
Note: The webinar will be recorded. After the session, the recording will be made available from the Machine Learning for Seamless Thunderstorm Nowcasting from Multiple Data Sources course page.
For further information about training email our Training team.
Other online courses
Gap filling ocean data (statistical and machine learning methods)
Online webinar, 19 October 2011, 11:00 UTC.
Application of machine learning approaches to ocean colour topics
Online webinar, 4 November 2011, 12:00 UTC.