Quasars behind M31 from PTF survey.

This slides is used for my final presentation on 2017 Caltech SURF (Summer Undergraduate Research Fellowship) seminar day. I searched for quasars behind M31 using light curves from the PTF (Palomar Transient Factory) survey.

Transcript

1. Searching for Quasars Behind the Andromeda Galaxy Yuhan Yao Peking University Mentor: Shri Kulkarni Aug, 24, 2017
2. Searching for Quasars Behind the Andromeda Galaxy Artist’s rendering of a quasar in the center of a distant galaxy. Yuhan Yao Peking University Mentor: Shri Kulkarni Aug, 24, 2017
3. Quasars: powerful beacons to probe the gas clouds between galaxies Distant Galaxies Milky Way Quasar Quasar Spectrum 10 billion years ago
4. Absorption lines Quasars: powerful beacons to probe the gas clouds between galaxies Distant Galaxies Milky Way Quasar 8 billion years ago
5. 0 billion years ago Quasars: powerful beacons to probe the gas clouds between galaxies
6. The Andromeda Galaxy Quasars: powerful beacons to probe the gas clouds in Andromeda
7. Data collection: Variability Palomar Transient Factory (PTF) survey Star Quasar 1.2-m telescope
8. Data collection: Variability - 32 features 28 Basic Statistics: Mean magnitude, Root Mean Squared deviation, …… 4 Model Parameters: Damped Random Walk, Structure Function. (Kelly 2009, Schimidt2010, Macleod 2012, Graham 2014) Star Quasar
9. Data collection: Color Optical: Infrared: LGGS survey
10. Data collection: Color - 8 features Optical: Infrared: LGGS: U-B,B-V W1-W2 SDSS: u-g Pan-STARRS: g-r, r-i, i-z, z-y 3.4 µm 4.6 µm
11. Data analysis: machine learning (ML) method • 32 (variability) + 8 (color) = 40 features • Find quasars from 190,000 objects • Training set: 59 known quasars + 500 known stars ( Known objects from: Flesch, 2015; Massey, 2016 )
12. Data analysis: machine learning (ML) method • Training set: 59 known quasars + 500 known stars • Distribution in two most informative features:
13. Data analysis: machine learning (ML) method • Find quasars from 190,000 objects • Use 10 features to do classification – 122 quasar candidates.
14. Results 56 spectra from: 50 confirmed to be quasars! à (1-6/56)=89% of the remaining 66 candidates are likely real quasars.
15. Before: 59 quasars in this field Results
16. After: a large increase! 50 new quasars! including 9 behind disk. Results
17. • Variability selection suffers less from selection effect. Results R<20.2mag: 56 + 37 + 42*0.89 = 130 Quasar areal density: 17 deg-2 for B<20.0mag (Boyle, 1987) brighter fainter PTF FoV 17*7.2=122 20.2
18. Results • Variability selection suffers less from selection effect. • Variability selection can fill in the gaps in color selection methods.
19. Conclusion 33% 28% 3% 36% Quasars Behind the Andromeda Galaxy Known (59) New (50) False (6) Candidate (66) • Good sample to study the ISM of Andromeda. • A pilot study of quasars in current and future synoptic surveys.
20. Acknowledgement • I want to express my immense gratitude to my mentors, professor Shri Kurkarni and Dr. Thomas Kupfer. • I would also acknowledge all members in the ZTF group for their constant help and support. • Special thanks to Yuguang Chen, Tianshu Wang, Anno Ho, and Matthew Graham for valuable discussion. • I would extend my appreciation to SFP, and people ever contributed to the SURF program. • Thanks for listening.