on reverberation mapping lag uncertainties
play

On Reverberation Mapping Lag Uncertainties Zhefu Yu, Department of - PowerPoint PPT Presentation

On Reverberation Mapping Lag Uncertainties Zhefu Yu, Department of Astronomy, The Ohio State University Advisor: Christopher Kochanek, Bradley Peterson Continuum Time lag Between continuum and lines or Ly different continuum


  1. On Reverberation Mapping Lag Uncertainties Zhefu Yu, Department of Astronomy, The Ohio State University Advisor: Christopher Kochanek, Bradley Peterson

  2. Continuum Time lag • Between continuum and lines or Ly 𝛽 different continuum wavelengths Si ¡IV • Critical for: • BH mass estimates C IV • R – L relation • Accretion physics (continuum RM) • …… He ¡II 2 (De Rosa et al. 2015)

  3. Lag measurement: ICCF • Linearly interpolate lightcurves • Lag: centroid / peak of the cross-correlation function • Uncertainty: flux randomization + random subsampling CCF ¡r Lag ¡(days) (Yu et al. 2019b) 3

  4. Lag measurement: JAVELIN • Assumptions: • Correct, Gaussian errors Flux • DRW stochastic process for interpolation • Line lightcurve is a shifted, scaled, and top-hat smoothed Flux version of the continuum • Uncertainty: MCMC based JD ¡-­‑ 2400000 (Yu et al. 2019b) 4

  5. Discrepancy of lag uncertainties JAVELIN ICCF • JAVELIN generally gives much smaller lag uncertainties than ICCF • Widely noticed, but few systematic studies • We use simulations to study: • Which uncertainty is more reliable? • How do the two algorithms behave with various systematic errors? • What happens to JAVELIN if its assumptions break down? 5 Lag ¡(days) (Fausnaugh et ¡al. ¡2016) ¡

  6. Observed Lightcurve of NGC 5548 Flux Simulated Lightcurve Simulated Lightcurve (Observed Cadence) Input lag: 2 – 4 days 6 JD ¡-­‑ 2400000 (Yu et al. 2019b)

  7. Parameterization & Baseline results • 𝜏obs : width of the (𝑢fit−𝑢 , ) distribution (“true” uncertainty) • 𝜏est : uncertainty from the algorithms • 𝜃 = 𝜏est/𝜏obs ( 𝜃 > 1 : Overestimate | 𝜃 < 1 : Underestimate) • Result: JAVELIN gets closest to correct uncertainty; ICCF overestimates the uncertainty Estimated ¡ Uncertainty 𝑢fit − 𝑢 , (days) (Yu et al. 2019b) 7

  8. Violating JAVELIN assumptions • Correct, Gaussian errors • DRW stochastic process • Line lightcurve is a shifted, scaled and top-hat smoothed version of the continuum 8

  9. Results: incorrect lightcurve errors • JAVELIN is more sensitive than ICCF 9 𝑢fit − 𝑢 , (days) (Yu et al. 2019b)

  10. Violating JAVELIN assumptions • Correct, Gaussian errors • DRW stochastic process • Line lightcurve is a shifted, scaled and top-hat smoothed version of the continuum 10

  11. Stochastic process: “Kepler” process • Less variability at short time scales Flux Flux 11 MJD ¡-­‑ 56000 (Yu et al. 2019b)

  12. Results: “Kepler” process • No significant effect 𝑢fit − 𝑢 , (days) (Yu et al. 2019b) 12

  13. Violating JAVELIN assumptions • Correct, Gaussian errors • DRW stochastic process • Line lightcurve is a shifted, scaled and top-hat smoothed version of the continuum 13

  14. Transfer functions • No significant effect Input ¡lag JA VELIN Assumption t ¡(days) 14 (Yu et al. 2019b)

  15. Violating JAVELIN assumptions • Correct, Gaussian errors • DRW stochastic process • Line lightcurve is a shifted, scaled and top-hat smoothed version of the continuum 15

  16. Varying background • Additional long time scale variability MJD ¡-­‑ 56000 (Yu et al. 2019b) 16

  17. Results: varying background • Strong deviation from input 𝑢fit − 𝑢 , (days) 17 (Yu et al. 2019b)

  18. Cadence and SNR (previous work) • Yu et al. 2019a: effect of cadence on LSST Deep Drilling Fields 3-day cadence, 23 epochs 2-day cadence, 31 epochs 1-day cadence, 54 epochs Lag ¡(days) 3750 3800 3850 3900 18 MJD ¡-­‑ 56000

  19. Summary • Systematic study on lag uncertainties with simulated lightcurves • JAVELIN gets closest to correct lag uncertainties in most circumstances, while ICCF tends to overestimate lag uncertainties. JAVELIN is more sensitive to incorrect single-epoch errors. • Underlying stochastic processes and transfer functions do not significantly affect lag measurements. • Both methods are significantly biased by additional sources of variability (Related papers: Yu et al. 2019a: arxiv 1811.03638 Yu et al. 2019b: arxiv 1909.03072) 19

  20. Correlated Errors 20 (Yu et al. 2019b)

  21. Result: Correlated Errors • No effect for the same sign errors • Declination of 𝜃 ¡ for the Matern 3/2 model 𝑢fit − 𝑢 , (days) (Yu et al. 2019b) 21

  22. Effect of Outliers 𝑢fit − 𝑢 , (days) (Yu et al. 2019b) 22

  23. Transfer functions: results 23 𝑢fit − 𝑢 , (days) (Yu et al. 2019b)

  24. 24

Recommend


More recommend