Python for variable star astronomy: A status report
- 1. Python for Variable Star Astronomy a status update Matt Craig Department of Physics and Astronomy 1
- 2. Acknowledgements •Current students •Isobel Snellenberger •Madelyn Madsen 2 •Former students •Adam Kline •Erin Aadland •Andy Block •Jane Glanzer •Elias Holte •Laura Maixner •Stefan Nelson •Elizabeth Dougherty •Nathan Walker •Laura Herzog •Michael Meraz •Connor Stotts •Nathan Heidt •Colleagues (MSUM) •Juan Cabanela •Linda Winkler
- 3. Outline •Motivation •Educational materials •Image viewer •Source detection and photometry •Data reduction •VSP 3
- 4. Motivation •Prepare undergraduate students for career •Include some programming •Choose accessible language •Re-use as much community software as possible •Do science: TESS/exoplanet follow-up 4
- 5. Goals •Software interoperability •Read and write AstroImageJ photometry files •Generate/read AIJ source lists •Data “interoperability” • TESS-style for reporting TESS data •AAVSO-style for reporting variable star data •Let others do the hard work •Use the community-based astro software large institutions are developing •Make user interaction browser-based, built on tools widely used outside astro 5
- 6. Guide to Data Reduction •http://bit.ly/ccd-guide 7
- 7. Image viewer •Browser-based; can run •on cloud server •on local machine •Interaction with browser and Python uses •Jupiter framework •widely support by data science and finance 8
- 9. Image viewer •Viewer design •Python designed to allow multiple implementations in future •STScI considering its use internally •LSST considering plugging their viewer into this framework 10
- 10. Aperture photometry •Source detection using sample image •Perform aperture photometry on all images in folder for those sources 11
- 11. Photometry 12
- 12. Aperture photometry •Produces (per source per image): •Net counts •Instrumental magnitude •Filter •Sky background •RA/Dec •Error (from CCD equation) 13
- 13. Differential photometry •Determine color correction and zero point for each frame •Use APASS stars in frame as standard stars •Prefer those with small error •Apply corrections to all sources 14
- 14. It works 😀 15
- 15. Sometimes 😬 16
- 16. Differential photometry •Need to do differential photometry in addition •TESS/AIJ style •flux (count) ratio •target flux / (sum of comp fluxes) •AAVSO style: •target magnitude from • Difference between instrumental target and comp • Catalog magnitude of comp •average over all comp stars 17
- 18. VSP •Relatively isolated code • Clearly defined result (single plot) 19
- 19. VSP: Chart for EY UMa 20
- 20. VSP: First pass in Python 21
- 21. VSP: Status •Start: 3,500 lines of well-written Perl •Now: 190 lines of code •Really poorly written Python… •…that uses current VSP API to get comp stars •Eventually: 500-600 lines of code 22
- 22. Challenges •Local computer •Installation is…painful •Relatively easy to break working install •Launch from a terminal •two or three platforms to support •Server: •Setup is…painful •Authentication is…more painful •Storage and CPU cost money 23
- 23. Advantages •Local computer •You control the compute •You have already paid for storage and CPU •Server •Software for users easier to manage •Can provide large data files without downloads 24
- 24. Questions? 26 Slides/links/how to try it out at: http://bit.ly/aavso-2019