[Paper Review] The variability analysis of PKS 2155-304

J. H. Fan, R. G. Lin|arXiv (Cornell University)|Jan 4, 2000

Atmospheric Ozone and Climate3 citations

TL;DR

This study analyzes post-1977 photometric light curves of the BL Lac object PKS 2155-304 using the Jurkevich and Discrete Correlation Function (DCF) methods to investigate long-term periodic variability. It identifies tentative periodicities of 4.16 ± 0.2 years and 7.0 ± 0.16 years in the V-band light curve, with statistical significance levels of 93.8% and 96.2%, respectively, suggesting possible quasi-periodic oscillations in this highly variable AGN.

ABSTRACT

In this paper, the post-1977 photometric observations of PKS 2155-304 are compiled and used to discuss the variation periodicity. Largest amplitude variations ($ΔU = 1^{m}.5$; $ΔB = 1^{m}.65$; $ΔV = 1^{m}.85 $; $ΔR = 1^{m}.25$; $ΔI = 1^{m}.14 $) and color indices ($(B-V) = 0.30\pm 0.06$; $(U-B) = -0.72\pm 0.08$; $(B-R) = 0.62\pm 0.07$; $(V-R) = 0.32\pm 0.04$) are found. The Jurkevich's method and DCF (Discrete Correlation Function) method indicate possible periods of 4.16-year and 7.0-year in the V light curve.

Motivation & Objective

• To compile and analyze post-1977 UBVRI photometric data of PKS 2155-304 for variability and spectral index behavior.
• To investigate the presence of long-term periodicity in the optical light curve of this highly variable BL Lac object.
• To assess the significance and reliability of detected periodic signals using robust statistical methods.
• To explore the implications of periodic variability for models of AGN activity and jet dynamics.

Proposed method

• The Jurkevich method is applied to the V-band light curve by folding data at trial periods and minimizing the mean square deviation to detect periodicity.
• The F-test is used to evaluate the statistical significance of the minima in the variance profile, assessing whether observed periodicities are real or spurious.
• The Discrete Correlation Function (DCF) method is employed to detect self-correlation in the light curve at various time lags, identifying potential periodic signals.
• Data from 140 U-B, 105 B-V, 90 B-R, and 98 V-R color index pairs are used to compute color index dispersions and their uncertainties.
• The light curve is simulated using two sinusoidal components with periods of 4.16 and 7.0 years to test the fit quality.
• Time coverage of approximately 16 years is used to assess the reliability of periodic signals, considering the risk of false positives due to long-term trends.

Experimental results

Research questions

• RQ1 Does the optical light curve of PKS 2155-304 exhibit significant long-term periodic variability?
• RQ2 What are the most probable periods of variability in the V-band light curve, and how significant are they?
• RQ3 How do the results from the Jurkevich method compare with those from the DCF method in detecting periodicity?
• RQ4 To what extent do the observed amplitudes and color indices suggest intrinsic variability or contamination from host galaxy light?
• RQ5 Can the observed variability be explained by a simple two-periodic model, or are additional components required?

Key findings

• The largest amplitude variations observed are ΔU = 1.5 mag, ΔB = 1.65 mag, ΔV = 1.85 mag, ΔR = 1.25 mag, and ΔI = 1.14 mag across the UBVRI bands.
• The color indices are (B-V) = 0.30 ± 0.06, (U-B) = -0.72 ± 0.08, (B-R) = 0.62 ± 0.07, and (V-R) = 0.32 ± 0.04, based on 140, 105, 90, and 98 pairs respectively.
• The Jurkevich method identifies two broad minima in the variance profile at 4.16 ± 0.2 years and 7.0 ± 0.16 years, with F-test significance levels of 93.8% and 96.2% respectively.
• The DCF method confirms the presence of self-correlation at time lags of 4.20 ± 0.2 years and 7.31 ± 0.16 years, consistent with the Jurkevich results.
• The simulated light curve using both periods does not fit the observations well, suggesting the presence of more than two periodic components or non-sinusoidal behavior.
• The study concludes that the detected periodicities are tentative and require confirmation by independent observations due to the limited time span (~16 years), which is only about three times the period length.