An improved time delay from VLA and ATCA monitoring of the gravitational lens system PKS 1830-211

TL;DR

We address the problem of measuring the time delay in the gravitational lens PKS 1830-211 with archival radio monitoring. We apply two robust time-delay techniques—chi-squared minimization and dispersion analysis—to combined ATCA+VLA light curves spanning 1997–2004, obtaining a precision improvement over previous work. The best-fit delay is $\tau_{B-A} = 25.3 \pm 2.0$ days, and the flux-density ratio $r(t)$ varies smoothly on year timescales, suggesting millilensing by massive objects in the lensing galaxy; VLBI reveals a counter-jet in image A and explains the polarization behavior, while ALMA imaging of the third image offers new constraints for lens models. Overall, the study improves time-delay precision and highlights structured subgalactic mass in PKS 1830-211, with implications for lens modeling and multi-frequency emission region comparisons.

Abstract

We have measured a time delay of 25.3 +/- 2.0 d (1-sigma confidence) in the Einstein ring gravitational lens system PKS 1830-211 from an analysis of archival VLA and ATCA monitoring data observed between 1997 and 2004. A small portion of the ATCA data was previously used to determine a time delay and our result is consistent with the previous value, but with an uncertainty that is smaller by more than a factor of two. The long time-baseline of the monitoring reveals that the flux density ratio is smoothly varying on a time-scale of years, an effect which we attribute to millilensing by massive objects (>>1 M_sun) in the lensing galaxy. Image A is unpolarized in the VLA monitoring, but VLBI observations show that this is partly due to beam dilution by an unpolarized counter-jet that is only present in that image. Based on the identification of this feature as a counter-jet, we conclude that its unexpected prominence in image A is a consequence of lensing and that more detailed modelling is required in order to reconcile the VLBI morphology of each image.

Figures (7)

• Figure 1: Total intensity maps (contours and grey scale) of PKS 1830$-$211 made using the AH593 data in which images A and B have been subtracted. Left: 8.4 GHz, Right: 15-GHz. The positions of the subtracted radio cores are marked with crosses, as is the position of the third image (C) as measured by muller20 which at 15 GHz can be seen to coincide with a feature labelled 'E' by subrahmanyan90. Also plotted are sticks indicating the orientation and relative magnitude of linear polarization. The restoring beams (bottom-left of each image) have dimensions of $360 \times 187$ and $194 \times 107$ mas$^2$ at 8.4 and 15 GHz respectively and contours are plotted at $-1$, 1, 2, 4. 8, etc. multiples of three times the rms noise ($\sigma_{\mathrm{8.4}} = 323$ and $\sigma_{\mathrm{15}} = 242~\mu$Jy beam$^{-1}$). The grey scale gives the image intensity in units of mJy beam$^{-1}$. Positions are offset from $18^{\mathrm{h}} 33^{\mathrm{m}} 39\fs9310, -21\degr 03\arcmin 39\farcs750$ (J2000) and have an astrometric accuracy of $\sim$10 per cent of the synthesized beam-size -- this could account for the offset between image C and the peak of 'E'.
• Figure 2: VLA and ATCA radio light curves of PKS 1830$-$211 at 8.4-8.6 GHz. The error bars are in many cases smaller than the plotting symbol.
• Figure 3: VLA 8.4-GHz percentage polarization (left) and EVPA (right) plots of PKS 1830$-$211 from Season 1. The polarization of image B declines by about a factor of two over the course of the monitoring while the EVPA shows a rotation of about 90. The data for image A, in contrast, have a very low SNR and show no sign of variability.
• Figure 4: Results of the CSM time-delay analysis. Top panel: Combined total flux density radio light curve after removal of the time delay (25.3 d) and flux density ratio. The solid line is a cubic spline fit to the data. Middle panel: polynomials representing the time-varying flux density ratio between A and B. Bottom panel: residuals for each image around the spline.
• Figure 5: Comparison of the time delays measured by previous radio or mm monitoring campaigns.