[huishao-r]

Description

Update the PathSum TEC test configuration (and associated test data in ufo-data repo) to use SI units consistently. This includes YAML configurations, reference observations, and variable naming to better follow CF conventions.

Changes include:

Updated GeoVaLs files (ufo-data)
electron density from cm⁻³ to m⁻³
height-related quantities from km to m
Removed unused fields from some of the testing files (ufo-data)
Updated new reference values (better precision) (ufo-data, yaml)
Renamed height_wrt_height to height_above_reference_ellipsoid to follow CF conventions. (ufo-data, yaml)

Issue(s) addressed

Resolves #609
build-group=https://github.com/JCSDA-internal/ufo/pull/4428

Dependencies

These data are used by https://github.com/JCSDA-internal/ufo/pull/4428

• https://github.com/JCSDA-internal/ufo/pull/4416
• https://github.com/JCSDA-internal/ufo/pull/4428
• https://github.com/JCSDA-internal/pyiri-jedi/pull/206

Checklist

• I have performed a self-review of my own code
• I have made corresponding changes to the documentation
• I have run the unit tests before creating the PR

[BenjaminRuston]

this file is almost 6Mb can it be reduced?
testinput_tier_1/geovals_madirgal_slant-tec_202010232345_100locs.nc4

[jhaiduce]

this file is almost 6Mb can it be reduced?

testinput_tier_1/geovals_madirgal_slant-tec_202010232345_100locs.nc4

Should be doable. This file was created by randomly sampling data from a larger source file (which was a couple gigabytes in size of I remember right). Assuming none of the tests this is used in actually require 100 obs we can just use a smaller subset of this data. @huishao-r do you think 5 obs will be enough?

[huishao-r]

this file is almost 6Mb can it be reduced?
testinput_tier_1/geovals_madirgal_slant-tec_202010232345_100locs.nc4

Should be doable. This file was created by randomly sampling data from a larger source file (which was a couple gigabytes in size of I remember right). Assuming none of the tests this is used in actually require 100 obs we can just use a smaller subset of this data. @huishao-r do you think 5 obs will be enough?

@BenjaminRuston @jhaiduce I reduced the file size using nccopy. I hope it is ok now.

[huishao-r]

this file is almost 6Mb can it be reduced?
testinput_tier_1/geovals_madirgal_slant-tec_202010232345_100locs.nc4

Should be doable. This file was created by randomly sampling data from a larger source file (which was a couple gigabytes in size of I remember right). Assuming none of the tests this is used in actually require 100 obs we can just use a smaller subset of this data. @huishao-r do you think 5 obs will be enough?

@BenjaminRuston @jhaiduce I reduced the file size using nccopy. I hope it is ok now.

@BenjaminRuston I compressed the other two geoval files. Now the sizes are quite small.

[fcvdb]

Thanks Hui!

[BenjaminRuston]

excellent thanks for reducing the file size @huishao-r