Closed jameswilburlewis closed 2 months ago
See https://github.com/tsssss/geopack/issues/18 : someone else also reported issues with T96, and proposed some patches. I forked the repo and implemented the fixes, and that resolved the worst of the differences I was seeing, but it's still not quite right. See the third panel in this plot for the IDL vs. Python differences in the T96 output:
Updates merged into master geopack repo, waiting for new version to be uploaded to pypi.
Comparing IDL vs. Python results for the T96 model (using THEMIS-A positions on 2007-03-23), for the time intervals approx. 15:00-16:00 and 20:00-21:00, the Python results seem to be disrupted, diverging from the IDL results, then snapping back suddenly. Agreement looks pretty good outside these intervals.