The CSPP Geo Team is pleased to announce a second beta release of the LightningCast software package, which predicts the probability of a Geostationary Lightning Mapper (GLM) observation of lightning occurring in the next hour within a region of interest.
The main new feature in this release is support for processing ABI data obtained from the GOES-19 satellite, which is scheduled to become operational as GOES-East on April 4, 2025. Users who are currently running the beta1 version of the LightningCast software to process GOES-East data are advised to update their software to beta2 before that date.
This release is built for Linux variants compatible with Rocky 8 or Rocky 9. CentOS 7 is no longer supported as it reached end of life in June 2024. Otherwise, this software is designed to be a drop-in replacement for the previous beta release, with similar hardware requirements and a backward-compatible user interface.
Capabilities
The underlying algorithm uses a deep learning model trained on data from the GLM instrument on-board the GOES-16 satellite. However, the only input that is required for near real-time processing is imager data from the GOES-16, GOES-18, GOES-19, or Himawari-9 satellite.
The main capabilities offered in this beta release are:
- Generation of lightning predictions in several data formats (GeoJSON, NetCDF, GR placefiles)
Generation of single band and RGB imagery with lightning probability contours overlaid
Optional over-plotting of GLM Flash Extent Density (as generated by the CSPP Geo Gridded GLM package)
User-definable regions of interest
Optional parallax correction
Optional AWIPS-compatible output
What's New
The following changes are included, relative to the previous beta release (v1.0beta1):
- GOES-19 support.
Two new image types: true color and true color / IR cloud phase composite.
Target platform is Rocky 8/9 or compatible Linux variant.
Updated versions of bundled software libraries.
Bug fixes and various minor improvements.
History and Attribution
This package is based on science software that was developed by NOAA and CIMSS scientists John Cintineo, Mike Pavolonis, Justin Sieglaff and Levi Pfantz. For more information on the algorithm, refer to the 2022 paper ProbSevere LightningCast: A Deep-Learning Model for Satellite-Based Lightning Nowcasting. Another useful resource is the LightningCast Quick Guide.
CSPP Geo software is developed at CIMSS at the University of Wisconsin under NOAA / GOES-R funding.
Beta Software
Since this is a beta release, please bear in mind that it has not been as thoroughly tested as a production release. In addition, users should expect that functionality and interfaces may be changed for the v1 production release, which is planned for later this year.
Software Download and Installation
To obtain the software, system requirements, test data and documentation, please visit the CSPP Geo website (free registration required for downloads). The software package is self-contained; installation of additional third-party software is not required. Refer to the CSPP Geo LightningCast Software Users' Guide for instructions on installing and running the software, as well as a description of features and usage examples.
We appreciate the feedback received from users regarding beta1, which has helped to improve this version. Any and all additional feedback is welcome at csppgeo.issues@ssec.wisc.edu.
Best regards,
Graeme, on behalf of the CSPP Geo Team