Add Barycentric Correction Tool by jlothringer · Pull Request #185 · spacetelescope/stistools

jlothringer · 2025-11-19T17:53:07Z

We've developed a new script, barycentric_correction.py, to replace the old functionality of the odelaytime IRAF routine. This new tool uses modern tools (astropy, JPL Horizons) for coordinate transformations and positional queries. Users can still use older HST orbital files, which will agree with the JPL Horizons calculations to within about 1 ms.

The script works with any STIS file (tag, raw, flt, x1d) and inherits its file modification logic from the original odelaytime port. The script has been tested against the old Python port of odelaytime and has been found to agree within about 1 ms. Standalone functions are also available to calculate barycentric corrections without file modifications.

A couple tests have also been added to tests/ with the test files available at this box link. A file in docs/ has been added and init.py has also been updated.

jlothringer · 2025-11-19T18:34:46Z

I also have this testing script and notebook that may be helpful. I also attach the odelaytime.py port from before, which was tested against.
odelaytime.py
Odelay_Verification.ipynb
final_odelay_testing.py

jkcarlberg · 2025-11-20T15:43:42Z

I assume the decision to define unit transformations were adopted from the original "odelaytime" python conversion, but it seems that these should really rely on astropy units and constants to avoid limiting the precision unnecessarily. (E.g., LYPERPC = (1 * u.pc).to('lightyear').value)

Of note, LYPERPC (defined as 3.261633 in the code) disagrees with the astropy calculation beginning in the 4th decimal place
In [21]: (1. *u.pc).to('lightyear')
Out[21]: <Quantity 3.26156378 lyr>

sean-lockwood · 2025-11-24T18:14:17Z

Will need to add it to the sphinx index:
https://github.com/spacetelescope/stistools/blob/master/doc/source/index.rst

sean-lockwood · 2025-11-24T18:15:50Z

+
+    def test_raw_JPL(self):
+        """Compare output for a x1d file."""
+        self.get_data("input", "od9m97020_raw.fits".fits")


Looks like some typos here in the filename. (And below on L26)

Good catch- done

sean-lockwood · 2025-11-24T18:18:16Z

+
+        bary_corr("od9m97020_raw.fits", hst_orb="p2o0000r_TIME.fit")
+
+        outputs = [(output, "od9m97020_raw_orbfile_out.fits")]


@jlothringer -
Let me know where this file lives and I can add it along with the input data to Artifactory.

Linked from the PR description here

sean-lockwood · 2025-11-24T18:22:12Z

+
+#should be done on the tag file, right? cuz changes will propogate to, like, inttag
+#nah, make it generic, may not TIME-TAG data!
+def bary_corr(table_names, verbose=True, distance=1e9, hst_orb = None, 


For consistency with other functions in stistools, I think we should either rename your main function (and associated references) to barycentric_correction or rename the module (file name) to bary_corr/bary_corr.py.

I'd vote for the former.

I agree with former- there is already another tool called barycorr: https://github.com/tronsgaard/barycorr

sean-lockwood · 2025-11-24T18:28:59Z

+    def test_raw_JPL(self):
+        """Compare output for a x1d file."""
+        self.get_data("input", "od9m97020_raw.fits".fits")
+        #output = "od9m97020_barycorr_raw_out.fits"


The output variable is not defined but used in the comparison. Fix in both tests.

sean-lockwood · 2025-11-24T18:30:51Z

+
+        bary_corr("od9m97020_raw.fits")
+
+        outputs = [(output, "od9m97020_raw_JPL_out.fits")]


The second filename should be the reference we upload to Artifactory. You should probably rename out to ref in both tests.

Fixed and renamed.

sean-lockwood · 2025-11-24T18:59:13Z

+        including barycorr, astroutils, and pintbary.
+
+        HST's changing location around the Earth can lead to time-delay
+        differences of up to ~23 seconds. HST's location can be determined


23 s or ms?

ms, though that's relative to the geocenter, so true maximum error is more like 42.5 ms: https://www.wolframalpha.com/input?i=%28Earth+diameter%2B2*550+km%29+%2F+speed+of+light. Fixed.

…sue with interpolation, added HST orbfile column flexibility

jlothringer · 2025-11-24T20:46:19Z

I assume the decision to define unit transformations were adopted from the original "odelaytime" python conversion, but it seems that these should really rely on astropy units and constants to avoid limiting the precision unnecessarily. (E.g., LYPERPC = (1 * u.pc).to('lightyear').value)

Of note, LYPERPC (defined as 3.261633 in the code) disagrees with the astropy calculation beginning in the 4th decimal place In [21]: (1. *u.pc).to('lightyear') Out[21]: <Quantity 3.26156378 lyr>

I've updated the script to use values derived from the Astropy units where appropriate.

…tance calculation (now fixed)

… BJD_TDB with option for BJD_UTC

Authored-by: jlothringer <jlothringer@stsci.edu> Contributions from: Sean Lockwood, Chris Hayes, Robert Jedrzejewski

sean-lockwood · 2026-05-28T15:21:47Z

Code merged in #189. Closing this PR.

jlothringer added 2 commits November 19, 2025 12:39

initial commit

74e72c6

initial commit

7b9a2f0

sean-lockwood added the enhancement label Nov 24, 2025

sean-lockwood self-requested a review November 24, 2025 16:49

sean-lockwood assigned jlothringer Nov 24, 2025

sean-lockwood reviewed Nov 24, 2025

View reviewed changes

jlothringer added 2 commits November 24, 2025 15:22

fixed test class names, fixed units to be astropy, fixed numpy>2.0 is…

a417dec

…sue with interpolation, added HST orbfile column flexibility

typo fixes in test and main barycentric correction script.

60bcc83

jlothringer added 2 commits November 25, 2025 15:16

fixed tag file re-stuff issue which uncovered a bug in the finite-dis…

e3d9ddc

…tance calculation (now fixed)

added UTC to TDB option for final time output. Times should now be in…

a1279a0

… BJD_TDB with option for BJD_UTC

sean-lockwood mentioned this pull request Mar 23, 2026

refactor of barycentric correction code from #185 #189

Merged

sean-lockwood added a commit that referenced this pull request May 28, 2026

Added barycentric timing correction code (originally from #185; #189)

a1f2345

Authored-by: jlothringer <jlothringer@stsci.edu> Contributions from: Sean Lockwood, Chris Hayes, Robert Jedrzejewski

sean-lockwood closed this May 28, 2026


		bary_corr("od9m97020_raw.fits", hst_orb="p2o0000r_TIME.fit")

		outputs = [(output, "od9m97020_raw_orbfile_out.fits")]


		bary_corr("od9m97020_raw.fits")

		outputs = [(output, "od9m97020_raw_JPL_out.fits")]

Uh oh!

Conversation

jlothringer commented Nov 19, 2025

Uh oh!

jlothringer commented Nov 19, 2025

Uh oh!

jkcarlberg commented Nov 20, 2025

Uh oh!

sean-lockwood commented Nov 24, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jlothringer commented Nov 24, 2025

Uh oh!

sean-lockwood commented May 28, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants