Flux and color variations of the doubly imaged quasar UM673
D. Ricci, A. Elyiv, F. Finet, O. Wertz, K. Alsubai, T. Anguita, V. Bozza, P. Browne, M. Burgdorf, S. Calchi Novati, P. Dodds, M. Dominik, S. Dreizler, T. Gerner, M. Glitrup, F. Grundahl, S. Hardis, K. Harpsøe, T. C. Hinse, A. Hornstrup, M. Hundertmark, U. G. Jørgensen, N. Kains, E. Kerins, C. Liebig, G. Maier, L. Mancini, G. Masi, M. Mathiasen, M. Penny, S. Proft, S. Rahvar, G. Scarpetta, K. Sahu, S. Schäfer, F. Schönebeck, R. Schmidt, J. Skottfelt, C. Snodgrass, J. Southworth, C. C. Thöne, J. Wambsganss, F. Zimmer, M. Zub, J. Surdej
aa r X i v : . [ a s t r o - ph . C O ] M a y Astronomy&Astrophysicsmanuscript no. 18755 c (cid:13)
ESO 2018November 3, 2018
Flux and color variations of the doubly imaged quasar UM673 ⋆ D. Ricci , , , A. Elyiv , , F. Finet , O. Wertz , K. Alsubai , T. Anguita , , V. Bozza , , P. Browne , M. Burgdorf , ,S. Calchi Novati , , P. Dodds , M. Dominik ,⋆⋆ , S. Dreizler , T. Gerner , M. Glitrup , F. Grundahl , S. Hardis ,K. Harpsøe , , T. C. Hinse , , A. Hornstrup , M. Hundertmark , U. G. Jørgensen , , N. Kains , E. Kerins ,C. Liebig , , G. Maier , L. Mancini , , G. Masi , M. Mathiasen , M. Penny , S. Proft , S. Rahvar , ,G. Scarpetta , , K. Sahu , S. Sch¨afer , F. Sch¨onebeck , R. Schmidt , J. Skottfelt , C. Snodgrass , ,J. Southworth , C. C. Th¨one , , J. Wambsganss , F. Zimmer , M. Zub , and J. Surdej ,⋆⋆⋆ D´epartement d’Astrophysique, G´eophysique et Oc´eanographie, Bˆat. B5C, Sart Tilman, Universit´e de Li`ege, 4000 Li`ege 1,Belgique; e-mail: [email protected] Main Astronomical Observatory, Academy of Sciences of Ukraine, Zabolotnoho 27, 03680 Kyiv, Ukraine Centro de Astro-Ingenier´ıa, Departamento de Astronom´ıa y Astrof´ısica, Pontificia Universidad Cat´olica de Chile, Casilla 306,Santiago, Chile. Max-Planck-Institut f¨ur Astronomie, K¨onigstuhl 17, 69117 Heidelberg, Germany Dipartimento di Fisica “E.R. Caianiello”, Universit`a degli Studi di Salerno, Via Ponte Don Melillo, 84085 Fisciano (SA), Italy Instituto Nazionale di Fisica Nucleare, Sezione di Napoli, Italy SUPA, University of St Andrews, School of Physics & Astronomy, North Haugh, St Andrews, KY16 9SS, United Kingdom Deutsches SOFIA Institut, Universitaet Stuttgart, Pfa ff enwaldring 31, 70569 Stuttgart, Germany Istituto Internazionale per gli Alti Studi Scientifici (IIASS), Vietri Sul Mare (SA), Italy Institut f¨ur Astrophysik, Georg-August-Universit¨at G¨ottingen, Friedrich-Hund-Platz 1, 37077 G¨ottingen, Germany Department of Physics & Astronomy, Aarhus University, Ny Munkegade, 8000 Aarhus C, Denmark Niels Bohr Institute, University of Copenhagen, Juliane Maries vej 30, 2100 Copenhagen Ø, Denmark KASI - Korea Astronomy and Space Science Institute, 776 Daedukdae-ro, Yuseong-gu, Daejeon 305-348, Republic of Korea National Space Institute, Technical University of Denmark, 2800 Lyngby, Denmark Astronomisches Rechen-Institut, Zentrum f¨ur Astronomie, Universit¨at Heidelberg, M¨onchhofstraße 12-14, 69120 Heidelberg,Germany Dipartimento di Ingegneria, Universit`a del Sannio, Corso Garibaldi 107, 82100 Benevento, Italy Bellatrix Astronomical Observatory, Center for Backyard Astrophysics, Ceccano (FR), Italy Physics Department, Sharif University of Technology, Tehran, Iran European Southern Observatory, Casilla 19001, Santiago 19, Chile Max Planck Institute for Solar System Research, Max-Planck-Str. 2, 37191 Katlenburg-Lindau, Germany Astrophysics Group, Keele University, Newcastle-under Lyme, ST5 5BG, United Kingdom Dark Cosmology Centre, Niels Bohr Institute, University of Copenhagen, Juliane Maries Vej 30, Copenhagen , 2100 Denmark INAF, Osservatorio Astronomico di Brera, 23807 Merate, Italy SOFIA Science Center, NASA Ames Research Center, Mail Stop N211-3, Mo ff ett Field CA 94035, USA Centre for Star and Planet Formation, Geological Museum, Øster Voldgade 5, 1350 Copenhagen, Denmark. Jodrell Bank Centre for Astrophyics, University of Manchester, United Kingdom Alsubai’s Establishment for Scientific Studies, Qatar Space Telescope Science Institute (STScI), United States of America Perimeter Institute for Theoretical Physics, 31 Caroline Street North, Waterloo, Ontario N2L 2Y5, Canada INAF / Istituto di Astrofisica Spaziale e Fisica Cosmica, Bologna, Via Gobetti 101, 40129 Bologna, Italy. Instituto de Astronom´ıa, Universidad Nacional Aut´onoma de M´exico, Apdo. Postal 877, Ensenada, B.C. 22800, MexicoNovember 3, 2018
ABSTRACT
Aims.
With the aim of characterizing the flux and color variations of the multiple components of the gravitationally lensed quasarUM673 as a function of time, we have performed multi-epoch and multi-band photometric observations with the Danish 1 .
54m tele-scope at the La Silla Observatory.
Methods.
The observations were carried out in the
VRi spectral bands during four seasons (2008–2011). We reduced the data usingthe PSF (Point Spread Function) photometric technique as well as aperture photometry.
Results.
Our results show for the brightest lensed component some significant decrease in flux between the first two seasons( + . /+ . /+ .
05 mag) and a subsequent increase during the following ones ( − . / − . / − .
10 mag) in the V / R / i spectral bands,respectively. Comparing our results with previous studies, we find smaller color variations between these seasons as compared withprevious ones. We also separate the contribution of the lensing galaxy from that of the fainter and close lensed component. Key words. quasar – lensing – photometric variability 1 . Introduction
Multiply imaged quasars are of great interest in astrophysicsdue to the possibility, from observed flux and color varia-tions between the lensed components, to distinguish betweenintrinsic quasar variations caused by the accretion mechanism,and microlensing e ff ects induced by stars in the lens galaxy(Wambsganss 2006).In a previous paper (Ricci et al. 2011b,a), we have studiedsuch variations for the quadruply imaged quasar HE 0435-1223,observed in the framework of a VRi multi-epoch monitoringof five lensed quasars , a parallel project of the MiNDSTEp(Microlensing Network for the Detection of Small TerrestrialExoplanets) campaign (Dominik et al. 2010).In the current paper, we focus on UM673 / Q0142–100 (seeFig. 1), a doubly imaged quasar discovered by Surdej et al.(1987) during a high resolution imaging survey of HLQs (HighlyLuminous Quasars) and subsequently studied by our team(Smette et al. 1990, 1992; Daulie et al. 1993; Nakos et al. 2005).Surdej et al. (1988) reported a separation of 2 . ′′ betweenthe components “A” (brighter) and “B” (fainter), and found their V magnitudes to be 16 . . z = . R = .
2) lensinggalaxy, located very close to the “B” component, was derived tobe z = .
49, and the time delay between the two lensed compo-nents was estimated around 7 weeks.A photometric monitoring of UM673 was performed duringthe years 1987–1993 (Daulie et al. 1993), but the photometry didnot show any clear evidence for relative variations over the con-sidered period.In the framework of the castles (CfA Arizona SpaceTelescope LEns Survey) project, precise astrometry of the com-ponents and of the lens galaxy “G” was obtained . The colors ofthe lens galaxy were found to match those of a passively evolv-ing early-type galaxy at z ≈ . R filter for six seasons (1995–2000), detecting a significantincrease by 0 . . ⋆ Based on data collected by MiNDSTEp with the Danish 1.54m tele-scope at the ESO La Silla Observatory. Light curves are available viahttp: // cdsweb.u-strasbg.fr / cgi-bin / qcat?J / A + A / ⋆⋆ Royal Society University Research Fellow ⋆⋆⋆ also Directeur de Recherche honoraire du FRS-FNRS UM673 / Q0142-100, HE0435-1223, Q2237 + Table 1.
Number of CCD images collected for each filter andeach year of observation of UM673. The corresponding numberof nights for each filter is also shown. images nightsseason
V R i total
V R i total2008 42 45 43 130 15 15 15 452009 34 35 26 95 12 13 9 342010 72 78 0 150 23 26 0 492011 51 53 9 113 15 16 1 32total 199 211 78 488 65 70 25 160
UM673 R S TU N E Fig. 1.
DFOSC V filter image, taken on 2008-08-03, showingthe position of UM673 and the stars “R”, “S”, “T”, “U”, and“V” used to search for a suitable reference star. The “V” star wasfinally chosen. “G1” and “G2” are field galaxies. The inset zoomshows the two components “A” and “B” of the lensed quasar.Space Telescope) data taken in the R filter, and obtained magni-tudes of 16 .
67, 18 .
96, and 19 .
35 for the “A”, “B” componentsand the lens galaxy, respectively.After spectrophotometric observations performed in 2002 byWisotzki et al. (2004), which did not show any evidence of mi-crolensing, the first multi-filter monitoring of UM673 was car-ried out by Nakos et al. (2005) between 1998 and 1999, in theCousins V and Gunn i filters. Analysis of the light curves wasmade using three di ff erent photometric methods: image decon-volution (Magain et al. 2007), PSF (Point Spread Function) fit-ting, and image subtraction. Nakos et al. (2005) found that com-ponent “A” displayed possible evidence for microlensing.Koptelova et al. (2008, 2010a,b) andKoptelova & Oknyanskij (2010) observed the object in the . Ricci et al.: Flux and color variations of the doubly imaged quasar UM673 N E 2arcsecAB G
Fig. 2.
Composite image of UM673 obtained by superposing the44 best quality CCD images in the R filter, resampled by dividingeach pixel in a grid of 20 ×
20 subpixels and recentering theimages with an accuracy of one new subpixel. The positions ofthe components, provided by HST, are also shown.
VRI bands and succeeded for the first time in determining atime delay: 150 + + − − days (at 68% and 95% confidence levels).Furthermore, Fadely & Keeton (2011) examined the wave-length dependence of the flux ratios for several gravitationallylensed quasars using K and L ′ -band images obtained with theGemini North 8m telescope, detecting no di ff erence betweenthe two flux ratios for the specific case of UM673 (“B” / “A” = . ± .
002 in the K -band and 0 . ± .
006 in the L ′ -band).Finally, in a recent paper, Koptelova et al. (2012) re-estimated the determination of the time delay to a value of89 ±
11 days using 2001–2011 VRI observations, and suggestedthe brightness variations to be mainly due to intrinsic variationsof the quasar.We present multi-epoch photometric monitoring data ofUM673 over four seasons (2008–2011), carried out in three fil-ters (
VRi ) with the DFOSC (Danish Faint Object Spectrographand Camera) instrument of the Danish 1 .
54m telescope at the LaSilla Observatory.The observations and the pre-processing of the images arepresented in Sect. 2. Sect. 3 presents the reduction techniquesand the results are shown in Sect. 4. Finally, Sect. 5 contains themain conclusions.
Table 2.
Maximum di ff erences of the R filter magnitudes be-tween seasons and in σ units for the stars “R”, “S”, “T”, “U”,and “V” in Fig. 1. star ∆ m R ∆ m R /σ R “R” 0.014 0.67“S” 0.030 1.29“T” 0.058 2.69“U” 0.037 0.84“V” 0.020 0.90 Fig. 3.
Reconstructed image of the lens galaxy of UM673.Orientation, pixel scale and marks are the same as those shownin Fig. 2. See the text for the details relative to the reconstructiontechnique.
2. Observations and pre-processing
We monitored UM673 during four seasons (2008–2011) us-ing the Danish 1 .
54m telescope at the La Silla Observatory,equipped with the DFOSC instrument, providing 2147 × . ′ × . ′ with adeclared resolution of 0 . ′′ / pixel. The RON (read-out-noise)of the CCD camera in high-mode (gain g = .
74 elec-tron / ADU) is 3 . / hardware of the telescope did not change over the fourseasons of observation. The data were collected in the Bessel V ,Bessel R , and Gunn i filters .We obtained a total number of 488 VRi images correspond-ing to 160 nights over the four seasons. The details are given inTable 1.In 2010, no i filter image was taken, as the monitoring wasforeseen since the beginning in the only VR filters, and the i filter images were taken depending on the remaining telescopetime with respect to the other MiNDSTEp parallel projects. Allthe frames were acquired with a 180s exposure.We treated the images following the same procedure asthose relative to HE 0435-1223 described in a previous paper(Ricci et al. 2011b), with the exception that we used the imagesalready de-biased and flat-fielded in loco by the IDL (InteractiveData Language) automatic pipeline used at the Danish Telescopefor the daily monitoring of the bulge microlenses.
3. Data reduction
We carefully checked the history of the scale of the images be-tween the various seasons, and we found a constant value of0 . ′′ / pixel. We froze this angular scale in the data reduction.We also checked the evolution of the position angle between theCCD pixel grid and the equatorial coordinate system, findinga change in angle between the seasons: 4 . ′ between 2008 and More details are available at
3. Ricci et al.: Flux and color variations of the doubly imaged quasar UM673 . ′ between 2008 and 2010, and 4 . ′ between 2008 and2011. We took this e ff ect into account in our data reduction.Finally, we checked the seeing values for all the observa-tions. We decided to fit the “U” star (see Fig. 1) with a two-dimensional Gaussian function, and we found that the R filterimages had the best seeing. We then decided to search for a suit-able reference star in that filter.We disregarded all those images for which the two lensedcomponents were unresolved (seeing > ′′ ). Independently wemeasured the flux ratio between the two bright galaxies “G1”and “G2” (see Fig. 1) using aperture photometry (we integrateda square area of 40 ×
40 pixels centered on each galaxy). In theanalysis we only used those images for which this flux ratio wasstable, corresponding to a total of 9–18 images per season, de-pending on the filter.The reference candidates are the stars “R”, “S”, “T”, “U”and “V” in Fig. 1: we compared the fluxes of these stars with thetotal flux of the bright galaxies “G1” and “G2” using aperturephotometry. For this test we decided to use galaxies because wecan be sure of their stability. Table 2 contains the maximum dif-ferences of the magnitudes between seasons and in sigma unitsfor the five concerned stars.On the basis of this analysis, we conclude that star “R” andstar “V” are comparably stable. However, star “V” is closer tothe lens system, and it is therefore better to use its shape asa reference PSF for the lens fitting. Also, it had been foundto be photometrically stable by Sinachopoulos et al. (2001) andNakos et al. (2003); finally it was already used by Nakos et al.(2005) as a reference for the PSF fitting of UM673.From all these considerations, we decided to use star “V”as the reference star for the PSF fitting of the lens system. Tocalibrate the magnitudes in the
VRi filters, we used the valuesof the star “V” provided by Nakos et al. (2003): m V = . ± . m R = . ± .
01, and m I = . ± .
01. Moreover, wecalculated the R magnitude of the “G1” and “G2” galaxies withaperture photometry, using “V” as the reference star. We foundvalues of m R = . ± .
03 for “G1”, and m R = . ± . R band images had better quality, andwe proceeded using these images. Each image was interpolatedwith a bicubic spline and every pixel was divided in a grid of20 ×
20 new sub-pixels. Then we superposed these oversampledimages and we summed them up to obtain an oversampled im-age with a high signal-to-noise ratio (see Fig. 2). We used the“V” reference star as reference PSF. We fitted the gravitationallens system with two PSFs for the “A” and “B” lensed compo-nents, fixing their relative astrometry. We then adjusted the scalefactors of those two PSFs to retrieve the uncontaminated imageof the background lens galaxy. We used aperture photometry toderive its magnitude relatively to the “V” reference star.To improve the accuracy of the photometry, we added twofactors to scale the fluxes of the “A” and “B” lensed components.We varied the factor of the “A” component from 0 .
94 to 1 . . . . . . ′′ , which is ≈
20 new sub-pixels. So we assumed that thedistance between the obtained and expected light center of “G”should be less than half the distance between the galaxy “G” andthe “B” lensed component ( <
10 new sub-pixels).We applied the same criterion between the expected and ob-served position of the maximum of light of the galaxy “G”.Indeed, the light center of “G” may be slightly o ff est from itsmaximum of light.We considered that the overlap between the regions wherethese two conditions are satisfied fixes the region of allowed val-ues for the two scale factors, and the minimum and maximummagnitudes of galaxy “G” which are 19 .
02 and 19 .
56, respec-tively. From that, we then independently calculated the magni-tude of the lensing galaxy “G” in the R band as 19 . ± .
27. If wecalculate this value as an average magnitude over all the allowedvalues for the two scale factors, we obtain 19 . ± .
15. Boththese values, within the error bars, are in good agreement withthe HST data. An image of the reconstructed galaxy is shown inFig. 3.Therefore, in the following analysis we considered the mag-nitudes of the lens galaxy “G” as being those previously mea-sured with HST. HST results (named G HS T ), obtained using HSTfilters, were converted to the ground-based photometric systemby Leh´ar et al. (2000) using Holtzman et al. (1995) calibrations.The V , R and i magnitudes that they derived for the galaxy are: G V = . ± . G R = . ± .
01, and G I = . ± . VRI filters, and derived the photometry without separately tak-ing into account the magnitude of “B” and that of the lens galaxy“G”. As the lens galaxy is located very close to the “B” compo-nent, and for comparison with other works, we also calculatedthe magnitudes of the “B” + “G” components as a simple super-position of their fluxes. Let us label “B + G” the results obtainedin this way.
Table 3.
Average magnitudes for the gravitationally lensed com-ponents of UM673 in the
VRi bands. component season
V R i
A 2008 16 . ± .
02 16 . ± .
03 16 . ± . . ± .
04 16 . ± .
04 16 . ± . . ± .
01 16 . ± . . ± .
03 16 . ± .
03 16 . ± .
01B 2008 19 . ± .
06 19 . ± .
05 18 . ± . . ± .
04 19 . ± .
05 18 . ± . . ± .
06 19 . ± . . ± .
05 19 . ± .
10 18 . ± . + G 2008 18 . ± .
07 18 . ± .
05 18 . ± . . ± .
04 18 . ± .
05 18 . ± . . ± .
06 18 . ± . . ± .
06 18 . ± .
10 18 . ± . m ag MJDUM673 light curveA componentB-1.5 component(B+G)+0.5 component2008 2009 2010 2011iRV
Fig. 4.
Light curves in the
VRi filters of the lensed components“A” and “B” of the gravitationally lensed quasar UM673. Theplot also shows the “B + G” values. The “B” and “B + G” lightcurves have been shifted by − . . .
02 and 0 . .
08 mag for the “A” and “B” components, respectively.
4. Results
The separate light curves of the two lensed components “A” and“B” of UM673 and the “B + G” light curve are shown in Fig. 4.For a robust measurement of variability, we calculated theaverage and the standard deviation over each season. Then, wealso measured the photometry of the whole system ( A + B + G ) aperture using aperture photometry. The aperture photometrywas calculated using two independent routines: a custom rou-tine set up by our team, and the IRAF daophot package. Asthe results were robust and coherent between each other, we de-cided to normalize the averaged PSF fitting results to aperturephotometry. We then calculated for each year a normalizationparameter k = [( A + B + G ) aperture − G HS T ] / ( A PS F + B PS F ), and
Table 4.
Average R − i and V − R color indices for the gravita-tionally lensed components of UM673. component season R − i V − R V − i A 2008 0 . ± .
04 0 . ± .
04 0 . ± . . ± .
05 0 . ± .
06 0 . ± . . ± . . ± .
03 0 . ± .
04 0 . ± .
03B 2008 0 . ± .
11 0 . ± .
08 0 . ± . . ± .
06 0 . ± .
06 0 . ± . . ± . . ± . − . ± .
11 0 . ± . + G 2008 0 . ± .
11 0 . ± .
08 0 . ± . . ± .
07 0 . ± .
06 0 . ± . . ± . . ± .
12 0 . ± .
11 0 . ± . Fig. 5.
Average light curves over the four seasons of observationfor the two lensed components “A” and “B”. The “B + G” averagelight curve is also shown (see the text for details). The error barsindicate the standard deviation over the epoch. The larger back-ground symbols show recent results independently obtained byKoptelova et al. (2012).we corrected all PSF magnitudes for k . These averaged resultsare in agreement with the non normalized results, and are shownin Fig. 5 and in Table 3.Furthermore, in Table 3 we list the magnitudes of “B + G” asa mere superposition of their fluxes. The contribution due to thegalaxy “G” in the total flux of the unresolved component “B + G”is quite important: near 18%, 45% and 52% in the V , R and i bands, respectively.We see an initial common behavior for the di ff erent filtersand components: the flux slightly decreases between the 2008and 2009 seasons, and increases between the 2009 and 2010seasons. Then, during the 2011 season, the flux of the “A” com-ponent keeps increasing, while the “B” component slightly de-creases.In particular, in the V filter we notice a decrease in flux by0 .
09 mag between the 2008 and 2009 seasons for the “A” compo-nent (corresponding to a decrease of 3 σ ), and an increase in fluxby 0 .
10 mag between the two successive seasons (2009–2011).The flux of the “B” lensed component, as well as of “B + G”,slightly decreases in this filter over the four seasons, but not sig-nificantly.In the R filter the behavior is the same: for the “A” compo-nent the flux decreases by 0 .
11 mag (above 3 σ ) between the firsttwo seasons and successively increases by 0 .
11 mag between the2009 and 2011 seasons. The flux of the “B” lensed componentslightly decreases, as well as the flux of “B + G”, with a less sig-nificant amplitude.Finally, in the i filter we notice less evident trends than de-tected in the other filters, excepted for the brighter “A” lensedcomponent which presents a smaller decrease in flux betweenthe first two seasons and a further increase by 0 .
10 mag between2009 and 2011.
5. Ricci et al.: Flux and color variations of the doubly imaged quasar UM673 -0.2-0.100.10.20.30.40.50.6-0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 V - R R-icolor-color diagram of UM673A B B+G
Fig. 6.
Color-color diagram for the 2008, 2009 and 2011 seasons(black bold dots) of the two lensed components “A” and “B” ofUM673. The “B + G” values are relative to the color indices of the“B” component that includes the contribution of the lens galaxy,as in the approach of Koptelova et al. (2010b). The diagramalso includes HST (Mu˜noz et al. 1998) data (black light points)and Koptelova et al. (2010b) data (little gray points). The largerbackground symbols refer to the data from Koptelova et al.(2012).Our results are in good agreement with Koptelova et al.(2012) recent results for the same epochs (see the larger back-ground symbols in Fig. 5). We obtain for the “A” lensed com-ponent a magnitude ≈ . .
08 larger for all the filters. Themagnitudes of “B + G” are slightly smaller: within 2 σ in the R and i bands. These di ff erences might derive from using di ff erenttechniques for PSF fitting and / or setting the photometric zero-points. From the data collected during the 2008, 2009, and 2011 sea-sons, we were able to build a color-color diagram to search forcolor variations of the two lensed components and of “B + G”with time. The results are shown in Fig. 6 and in Table 4.All color variations over each epoch are found to be withinthe error bars. Our results also show that within these error barsthe color indices of the “A” component and of “B + G” are coher-ent with the work of Koptelova et al. (2010b, 2012) data, and wefind small variations with respect to HST data which are relativeto 1994.Moreover, the temporal evolution of the color index betweenthe observations of Koptelova et al. (2010b, 2012) and the cur-rent data shows a weak trend indicating that the quasar becomesredder as its flux decreases, as already observed in our multi-color study of the gravitationally lensed quasar HE 0435-1223(Ricci et al. 2011b).Galaxy “G” a ff ects quite strongly the color of “B + G”. Wefind a di ff erence of 0 . .
13 mag between the R − i color indexof “B + G” and “B”, a di ff erence of 0 . .
47 mag in V − R and of0 . .
60 mag in V − i . On the basis of HST data, corresponding V - i yearUM673 V-i for the A(+1mag), B and B+G componentsHST nakos koptelova this workA+1B+G B+GB B Fig. 7.
Evolution of the V − i color index of “A” (trian-gles), “B” and “B + G” (rhombi) with time, including the datafrom HST (Mu˜noz et al. 1998), from Nakos et al. (2005), fromKoptelova et al. (2010b) and from the present work. Adaptationof recently published data by Koptelova et al. (2012) is alsoshown (larger background symbols). The “A” component isshifted by + ff erences in colors between the “B + G” and “B” componentsare 0 .
17, 0 .
37 and 0 .
54 (Leh´ar et al. 2000). This supports theview that the contribution of galaxy “G” cannot be neglected inany considered band.Nakos et al. (2005) reduced UM673 data in the V and i filtersby using three di ff erent techniques: MCS deconvolution, di ff er-ence imaging, and PSF fitting. The first two techniques allowin principle to get rid of the contribution of “G”, while resultsobtained with PSF fitting are contaminated by the lens galaxy.Despite this, Nakos et al. (2005) results are coherent with eachother. We compared their V − i color index obtained by the di ff er-ent methods. Nakos et al. (2005) obtained di ff erences on the V − i color index between “B + G” and the “B” components smallerthan 0 .
04 mag, which is comparable with their photometric er-rors. This is in contradiction with Leh´ar et al. (2000) results andour results, which lead to 0 .
54 and 0 . − .
60 mag, respectively.As it was shown above, the brightness of galaxy “G” cannot beneglected in the V , R or i bands. The contribution of galaxy “G”significantly changes the color of “B + G”.In Fig. 7 we compare the evolution of the V − i color indexwith time, by using the data collected from HST (Mu˜noz et al.1998), Nakos et al. (2005), and Koptelova et al. (2010b). Anadaptation of recently published data by Koptelova et al. (2012)is also shown. We find good agreement with their data. i ” lightcurve A “global i ” light curve which also includes the results ofMu˜noz et al. (1998), Nakos et al. (2005) and Koptelova et al.(2010b, 2012) is shown in Fig. 8. To construct this figure, firstwe shifted in time the light curve of “B + G” using the value ofthe time delay (89 days) provided by Koptelova et al. (2012).
6. Ricci et al.: Flux and color variations of the doubly imaged quasar UM673 m ag yearUM673 "global" light curveHST nakos koptelova this workiRV Fig. 8. “Global i ” light curve of UM673 built by includ-ing data from HST (Mu˜noz et al. 1998), Nakos et al. (2005),Koptelova et al. (2010b) and the present work. An adaptation ofrecently published data by Koptelova et al. (2012) is also shown(larger background symbols). The technique used to build thiscurve is explained in detail in Sect. 4.3. In particular, filled andopen symbols are used for the “A” and the “B + G” lensed com-ponents, respectively.Then we calculated for each filter the average 2008 di ff erence inmagnitude between the two components, and we corrected the“B + G” light curve for these values. Finally, we corrected the V and R light curves of both components by their average 2008 V − i and R − i color indices, respectively. We chose the 2008season as a reference only because it represents the beginningof our observations. Fig. 8 shows that the flux of the quasar in-trinsically varied over the di ff erent seasons, with an amplitudeof ≈ .
5. Conclusions
We have presented a photometric monitoring, carried out duringfour epochs in three di ff erent filters, of the doubly imaged quasarUM673.The results show a significant decrease in flux of the “A”lensed component between the first two seasons (2008–2009),and a smaller increase between the successive three seasons(2009–2011). This behavior is mostly significant in the V and R bands.Moreover, our observations are in good agreement withthe previous works carried out by Mu˜noz et al. (1998),Koptelova et al. (2010b), and Koptelova et al. (2012) in termsof flux variations and color index. We also separated the con-tribution of the lens galaxy from the fainter lensed component,showing the e ff ects of this operation on the color index of thelatter. We conclude that the contribution of the lens galaxy inthe photometry of UM673 cannot be neglected and we give anindependent estimation of the magnitude of the lens galaxy. Further observations could help in corroborating the sepa-rate color variations of the components, and the slight flux trendobserved between the seasons. Acknowledgements.
This research was supported by ARC – Action de rechercheconcert´ee (Communaut´e Franc¸aise de Belgique – Acad´emie Wallonie-Europe).DR (boursier FRIA) acknlowledges GLObal Robotic telescopes IntelligentArray for e-Science (GLORIA), a project funded by the European UnionSeventh Framework Programme (FP7 / ffi ce. AE is also grateful for partial support in the frameworkof the NASU Target Program “CosmoMicroPhysics”. NK received funding fromthe European Community’s Seventh Framework Programme ( / FP7 / / )under grant agreement No 229517. MD, MH and CL acknowledge the QatarFoundation for support from QNRF grant NPRP-09-476-1-078. Operation of theDanish 1 .
54m telescope is supported by the Danish National Science ResearchCouncil (FNU).
References