[daip] Question about DECONV.FOR {External}

[Eric Greisen]

If the convolved source is about the size of the beam along the major axis but slightly fatter along the minor axis then the deconvolution could move the PA by 90.  Try IMFIT and/or JMFIT on the images.  They do a correct deconvolution (of Gaussians) and they try the fit values and plus/minus 1 sigma to make a matrix of deconvolved values.  This should help to determine the reliability of the deconvolution.

Eric Greisen
________________________________
From: Daip <daip-bounces at listmgr.nrao.edu> on behalf of Hanno Spreeuw via Daip <daip at listmgr.nrao.edu>
Sent: Thursday, October 16, 2025 7:16 AM
To: daip <daip at nrao.edu>
Subject: [daip] Question about DECONV.FOR {External}

Hello,

I have been running DECONV.FOR for some time and its output always seemed fine.
I guess the current version - 31DEC25/APL/SUB/DECONV.FOR - does not differ from the previous ones.

Yet, I am a bit confused about its output when the convolved source is aligned with the restoring beam.
Then the returned, i.e. deconvolved, position angle is 90° larger than those position angles.

E.g.
deconv(7.1941910149411346, 6.826481869094818, 86.02929942633546, 5.699093590220353, 4.181319049159876, 86.7612722265415)
returns
(5.396174907684326, 4.390378952026367, 177.1446533203125, 0)
So rpa=177.1446533203125 is 90° larger than one would expect.
Could you please help me to explain this?

Kind regards,
Hanno Spreeuw

PS To reproduce using Python, see the attached source file, which is the original DECONV.FOR with a few additions for f2py.
python -m numpy.f2py -c DECONV.FOR -m deconv_python
ipython
import numpy as np
import deconv_python
# print(deconv_python.deconv.__doc__)
deconv_python.deconv(7.1941910149411346, 6.826481869094818, 86.02929942633546, 5.699093590220353, 4.181319049159876, 86.7612722265415)
---> (5.396174907684326, 4.390378952026367, 177.1446533203125, 0)

--
[https://esciencecenter.nl/wp-content/themes/raadhuis/dist/assets/img/e-signature.png]
Hanno Spreeuw
eScience Research Engineer
+31 6 21 537 257
Netherlands eScience Center
www.esciencecenter.nl <https://www.esciencecenter.nl/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listmgr.nrao.edu/pipermail/daip/attachments/20251017/02127b9e/attachment-0001.html>