The CSPP Geo team is pleased to announce the release of the Geo2Grid Version 1.0.0 software package. The Geo2Grid software provides the capability to create high quality single band and combined Red/Green/Blue (RGB) image files in an easy and efficient manner from geostationary imager data files in mission-standard format. Images can optionally be reprojected to a pre-defined or user-defined grid.
Development of this software was funded by the NOAA GOES-R Program.
Output images are written in GeoTIFF format, which can be displayed in typical image viewers but contains extra georeferencing information that can be used in geo-spatially aware applications.
Creation of the images is carried out by simple bash shell script commands, which wrap underlying Python code. By default, images of all Imager Bands plus true and false color output image files will be generated, however, users can choose to create one or more band images using optional arguments.
The software supports the following satellite/instrument combinations:
- •GOES-16 and GOES-17 Advanced Baseline Imager (ABI)
•Himawari-8 Advanced Himawari Imager (AHI) (both HSD and HimawariCast formats)
• True and natural color imagery creation includes:
- • Upsample and sharpen composite bands to the highest spatial resolution (500m).
• Creation of pseudo “green” band for the ABI instruments.
• Reflectance adjustment (dividing by cosine of the solar zenith angle).
• Removal of atmospheric Rayleigh scattering (atmospheric correction).
• Command line option supports latitude/longitude subsets.
• Support for combining daytime RGB/nighttime RGB in one image.
• Option to reproject image onto a user defined grid.
• Utility script provided for overlaying borders, coastlines, etc.
• Utility script provided for creating animations from the output GeoTIFF files.
• Thorough documentation, examples, and verification data provided.
Minimum System Requirements
•Intel or AMD CPU with 64-bit instruction support (2+ cores - 2.4 GHz),
•16 GB RAM (minimum),
•CentOS 6 (or other compatible 64-bit Linux distribution),
•5 GB disk space (minimum).
System Requirements for Multi-Band Real-time Processing
•Intel or AMD CPU with 64-bit instruction support (20+ cores - 2.4 GHz),
•46 GB RAM (minimum),
•CentOS 6 (or other compatible 64-bit Linux distribution),
•1 TB disk space (minimum for ~1 week of images, does not include long-term storage)
The software uses the number of specified CPU threads to create high quality reprojections in the lowest latency possible thanks to the use of the Dask python library. Dask splits data arrays into multiple “chunks” and processes them in parallel.
Geo2Grid software development is led at the University of Wisconsin Space Science and Engineering Center by David Hoese and Kathy Strabala. The software, test dataset and documentation can be obtained from the download page on the CSPP Geo website (login required).
http://download.ssec.wisc.edu/files/csppgeo/
Please contact csppgeo.issues@ssec.wisc.edu with any questions, or to report any issues encountered when installing or running the software.