[AstroPy] astropy stability for time-series differential photometry pipeline

[Samuel Harrold]

Hello,

My research group at UT Austin is considering migrating our data reduction
pipeline from IRAF/PyRAF to astropy. We do time-series differential
photometry for asteroseismology, and because we use our data for many years
after we've processed it, we'd like our pipeline to be stable for 5-10
years. We're unsure if we should wait several years for the astropy
collaboration to release a paper demonstrating comparisons with
IRAF/PyRAF/ESO-MIDAS/other, or if we should migrate now.

We have a few questions about the stability of astropy before we attempt to
migrate. We greatly appreciate any advice:

- In the long term, is STSCI replacing IRAF/PyRAF with astropy?

- If we migrate to astropy now, how often would we need to incorporate
updates to astropy, use non-astropy packages, and/or make our own
work-arounds? Our key pipeline needs are:
- - Operations are done on ~1000 - 5000 images at a time. (We take
sequences of ~10-second exposures for many hours at a time, following the
target from horizon to horizon.)
- - Bias, dark, flat calibration
- - Align calibrated images (to correct for bad tracking)
- - Circular aperture photometry of target star and comparison stars
- - Correct timestamps to BJD with accuracy to better than 0.01s. (Our
instrument's timestamps are controlled by GPS. Light curves are combined
with others taken over ~20 years.)

Thanks for all your help, and thanks for making such a useful tool!
Sam Harrold
----------
Samuel Harrold
PhD Candidate
Astronomy Department
University of Texas at Austin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/astropy/attachments/20140211/2a69dfd1/attachment.html>