New near-infrared observations and lens-model constraints for UM673

Aims: We performed a detailed photometric analysis of the lensed system UM673 (Q0142-100) and an analysis of the tentative lens models. Methods: High-resolution adaptive optics images of UM673 taken with the Subaru telescope in the H band were examined. We also analysed the J, H and K-band observational data of UM673 obtained with the 1.3m telescope at the CTIO observatory. Results: We present photometry of quasar components A and B of UM673, the lens, and the nearby bright galaxy using H-band observational data obtained with the Subaru telescope. Based on the CTIO observations of UM673, we also present J- and H-band photometry and estimates of the J, H and K-band flux ratios between the two UM673 components in recent epochs. The near-infrared fluxes of the A and B components of UM673 and their published optical fluxes are analysed to measure extinction properties of the lensing galaxy. We estimate the extinction-corrected flux ratio between components A and B to be about 2.14 mag. We discuss lens models for the UM673 system constrained with the positions of the UM673 components, their flux ratio, and the previously measured time delay

Time delay between images of the lensed quasar UM673

We study brightness variations in the double lensed quasar UM673 (Q0142-100) with the aim of measuring the time delay between its two images. In the paper we combine our previously published observational data of UM673 obtained during the 2003 - 2005 seasons at the Maidanak Observatory with archival and recently observed Maidanak and CTIO UM673 data. We analyze the V, R and I-band light curves of the A and B images of UM673, which cover ten observational seasons from August 2001 to November 2010. We also analyze the time evolution of the difference in magnitudes between images A and B of UM673 over more than ten years. We find that the quasar exhibits both short-term (with amplitude of sim 0.1 mag in the R band) and high-amplitude (sim 0.3 mag) long-term variability on timescales of about several months and several years, respectively. These brightness variations are used to constrain the time delay between the images of UM673. From cross-correlation analysis of the A and B quasar light curves and error analysis we measure the mean time delay and its error of 89 pm11 days. Given the input time delay of 88 days, the most probable value of the delay that can be recovered from light curves with the same statistical properties as the observed R-band light curves of UM673 is 95{+5/-16}{+14/-29} days (68 and 95 % confidence intervals). Analysis of the V - I color variations and V, R and I-band magnitude differences of the quasar images does not show clear evidence of the microlensing variations between 1998 and 2010.