Periodograms for Multiband Timeseries

The Lomb-Scargle periodogram has long been a workhorse in astronomy for detecting periodicity from irregularly sampled observations of variable sources. The classic periodogram cannot, however, make use of the multi-band photometry of many current and future surveys. While past studies have used K-corrections and/or ad hoc methods to address this, a more principled approach will become necessary for future surveys. Our solution is the multiband periodogram, a generalization of the classic Lomb-Scargle method which flexibly accounts for multi-band observations within a single model, without any physical assumptions about corrections between bands. Experiments on simulated data indicate that this new method vastly improves on previous approaches, pushing the effective limit of LSST RR Lyrae studies 1-2 magnitudes deeper.

Transcript

1. Jake VanderPlas Periodograms for Multiband Time Series Jake VanderPlas #LGAstats; June 3rd 2015
2. Jake VanderPlas Periodograms for Multiband Time Series Jake VanderPlas #LGAstats; June 3rd 2015 The Bomb-Scargle Periodogram!!!
3. Jake VanderPlas Mapping the MW with RR Lyrae Sesar et al. 2010 SDSS II Stripe 82: - 483 RR Lyrae to r~22 - 300 deg2 - d ~ 100 kpc Supports the idea of an early-forming smooth inner halo, and late-forming accreted outer halo.
4. Jake VanderPlas RR Lyrae in LSST SDSS II LSST 300 deg2 ~20,000 deg2 r ~ 22 mags r ~ 24 mags d ~ 100 kpc d ~ 300 kpc 483 RR Lyrae > 30,000 RR Lyrae? ? ? ?
5. Jake VanderPlas RR Lyrae in LSST SDSS II LSST 300 deg2 ~20,000 deg2 r ~ 22 mags r ~ 24 mags d ~ 100 kpc d ~ 300 kpc 483 RR Lyrae ~107 RR Lyrae ? ? ? Potential for vastly increased understanding of structure of our halo & detailed constraints on models of Milky Way formation history!
6. Jake VanderPlas But it is not entirely straightforward... Naive single-band approach for LSST: - 1 year: 50% completeness at g ~ 22 - 10 years: 50% completeness at g~24.5 Can we do better?
7. Jake VanderPlas The issue: For LSST-style data (1 band each visit), single band approaches fail! We need a method that utilizes all bands at once.
8. Jake VanderPlas Let’s think about a periodogram which can utilize multiple bands simultaneously . . .
9. Jake VanderPlas The Lomb-Scargle Periodogram If you’ve ever come across the Lomb-Scargle Periodogram, you’ve probably seen something like this... But this obfuscates the beauty of the algorithm: the classical periodogram is essentially the 2 of a single sinusoidal model-fit to the data:
10. Jake VanderPlas Standard Lomb-Scargle cf. Lomb (1976), Scargle (1982) Figure: VanderPlas & Ivezic 2015 Periodogram peaks are frequencies where a sinusoid fits the data well:
11. Jake VanderPlas Connection between Fourier periodogram and least squares allows us begin generalizing the periodogram . . .
12. Jake VanderPlas Floating Mean Model cf. Ferraz-Mello (1981); Cumming et al (1999); Zechmeister & Kurster (2009) Figure: VanderPlas & Ivezic 2015 . . . in which we simultaneously fit the mean
13. Jake VanderPlas Truncated Fourier Model cf. Bretthorst (1988) Figure: VanderPlas & Ivezic 2015 . . . in which we fit for higher-order periodicity
14. Jake VanderPlas Regularized Model . . . in which we penalize coefficients to simplify an overly-complex model. The “trick” is adding a strong prior which pushes coefficients to zero: higher terms are only used if actually needed!
15. Jake VanderPlas Putting it all together: The Multiband Periodogram
16. Jake VanderPlas Putting it all together: The Multiband Periodogram - define a truncated Fourier base component which contributes equally to all bands.
17. Jake VanderPlas Putting it all together: The Multiband Periodogram - for each band, add a truncated Fourier band component to describe deviation from base model
18. Jake VanderPlas Putting it all together: The Multiband Periodogram - Regularize the band component to drive common variation to the base model. + =
19. Jake VanderPlas Putting it all together: The Multiband Periodogram - Regularize the band component to drive common variation to the base model. + = Key: Regularization reduces added model complexity & pushes common variation into the base model.
20. Jake VanderPlas Multiband Periodogram on sparse, LSST-style data . . . Detects period with high significance when single-band approaches fail!
21. Jake VanderPlas The Money Plot: Prospects for LSST Based on simulated LSST cadence & photometric errors; see VanderPlas & Ivezic (2015, in prep) Fraction Recovered 6 months 1 year 2 years 5 years multiband model Oluseyi (2012) approach g-band mag
22. Jake VanderPlas The Money Plot: Prospects for LSST Based on simulated LSST cadence & photometric errors; see VanderPlas & Ivezic (2015, in prep) Fraction Recovered 6 months 1 year 2 years 5 years multiband model Oluseyi (2012) approach g-band mag e.g. after 2 years: ~0% →~75% completeness at survey limit!
23. Jake VanderPlas The Money Plot: Prospects for LSST Based on simulated LSST cadence & photometric errors; see VanderPlas & Ivezic (2015, in prep) Fraction Recovered 6 months 1 year 2 years 5 years multiband model Oluseyi (2012) approach g-band mag ~2 mag improvement in effective depth of LSST!
24. Jake VanderPlas Code to reproduce the study & figures: http://github.com/jakevdp/multiband_LS/ Python periodogram implementation: http://github.com/jakevdp/gatspy/ “If it’s not reproducible, it’s not science.”
25. Jake VanderPlas ~ Thank You! ~ Email: jakevdp@uw.edu Twitter: @jakevdp Github: jakevdp Web: http://vanderplas.com/ Blog: http://jakevdp.github.io/
26. Jake VanderPlas Extras . . .
27. Jake VanderPlas Oluseyi 2012 Simulated LSST Measurements: Faintest RR- Lyrae: Pessimistic period recovery even with 5-10 years of LSST data!
28. Jake VanderPlas
29. Jake VanderPlas