Probing the broad line region geometry and size of the gravitationally lensed quasar Q2237+0305 with microlensing time series

TL;DR

This study uses microlensing time series data from the gravitationally lensed quasar Q2237+0305 to estimate the size and geometry of its broad line region, favoring a Keplerian disk model and providing new size measurements.

Contribution

It introduces an independent microlensing method analyzing index time series to determine BLR structure and size, validated with archival spectrophotometric data.

Findings

The BLR is best modeled as a Keplerian disk.

Estimated CIV BLR half-light radius is 51±23 light days.

System viewed at an intermediate inclination angle.

Abstract

Lensed quasars are powerful cosmic laboratories; they are used to simultaneously probe various astrophysical phenomena. Microlensing by stars within distant galaxies acts as strong gravitational lenses of multiply imaged quasars, and provides a unique and direct measurement of the lensed quasar internal structure. Microlensing of the continuum emitting region as well as the broad-line region (BLR) is well characterized by four observable indices, $\mu^{cont}$, $\mu^{BLR}$, $WCI$ (wing-core), and $RBI$ (red-blue), measured directly from the spectra. During the 2004-2007 monitoring period, image A of the quadruply lensed system Q2237+0305 underwent a strong microlensing amplification, while image D remained unaffected. We used 35 epochs of archival spectrophotometric data of Q2237+0305 obtained with the Very Large Telescope of the European Southern Observatory to develop an independent microlensing method for estimating the geometry and size of the BLR. We measured the index time series for the CIV line and the continuum emission at $1450\,\unicode{x212B}$. We built a library of the simulated microlensing index time series that reproduce the observed times series based on three representative BLR models: Keplerian disk (KD), polar wind (PW), and equatorial wind (EW). After sampling the model parameter space, we find that KD is the predominant model, while PW and EW are less likely. We infer that the system is viewed at an intermediate viewing angle $i\sim 35^\circ$, and we estimate the most likely CIV BLR half-light radius $r_\mathrm{1/2}=51\pm 23$ light days. Our results are in good agreement with previous findings in the literature and extend the validity of the index-based approach to a temporal domain.