KNMI: Predict Lightning

94 views

Skip to first unread message

Bas van Stein

unread,

Feb 1, 2022, 4:56:46 AM2/1/22

to LIACS thesis projects

contact: Cor Veenman <c.j.v...@liacs.leidenuniv.nl>

This project aims to develop and implement a machine-learning/artificial intelligence applications for near-real-time prediction of chance of lightning.

Lightning is a potential weather hazard that is important for many economic sectors, where the chance of lightning or the occurrence of lightning is a trigger for changing standard operations. Knowing and understanding where and when lightning may occur is therefore important information, even on time scales as short as 15-30 minutes in advance.

A few years ago, the first satellite designed to monitor lightning from space has been launched on the American GOES satellites, the so-called Global Lightning Mapper (GLM). This high-speed space camera provides lightning detection information every 20 seconds, day-in-day-out.

Scientists from NOAA/CIMSS (University Wisconsin) have recently developed a ML/AI algorithm to predict the chance of lightning up to a few hours ahead of time based on geostationary satellite data. This tool is now operational for the United States as part of the NOAA/CIMSS ProbSevere algorithm and provides valuable information about chances of lightning.

https://cimss.ssec.wisc.edu/satellite-blog/archives/38136

Within a few years an instrument similar to GLM will be launched as part of the new series of Meteosat Third Generation satellites. KNMI is currently preparing for use of MTG data, and including the use of planned novel satellite instruments like the MTG Lightning Imager (LI).

At the same time, KNMI has a responsibility for providing weather data and forecast for a number of Dutch overseas islands in the Caribbean, the so-called BES islands (Bonaire, st. Eustatius, Saba), and recently KNMI has also installed instrumentation for monitoring the volcanoes on st. Eustatius and Saba.

Because the Caribbean fall within the observational domain of the GOES GLM satellite, KNMI has started exploring the GLM data in order to learn how to use satellite measurements of lightning in KNMI operations. Plans are to implement the GOES data into the KNMI web-based online weather visualization tool GEOWEB via a dedicated BES Natural Hazard Monitoring platform.

https://journals.ametsoc.org/view/journals/wefo/aop/WAF-D-20-0028.1/WAF-D-20-0028.1.xml