Discovery of the Lensed Quasar System DES J0408-5354

Strong gravitational lensing systems provide valuable tools for studying the properties and evolution of galaxies and quasars, for measuring the distribution of dark matter, and for constraining cosmological parameters. In particular, lensed quasar systems can provide information on, for example, intervening absorption line systems (e.g., Smette et al. 1995), properties of quasar host galaxies (e.g., Peng et al. 2006), dark matter substructure in the lens (e.g., Dalal & Kochanek 2002), and the stellar content of the lensing galaxy (e.g., Schechter et al. 2014). Moreover, when combined with time delay measurements and careful lens modeling, lensed quasar systems can provide powerful cosmological constraints that are complementary to those of other techniques (e.g., Suyu et al. 2016; Treu & Marshall 2016; Bonvin et al. 2017).

Previously, large samples of lensed quasars have been found by surveys such as the Cosmic Lens All-Sky Survey (Browne et al. 2003; Myers et al. 2003) in the radio and the Sloan Digital Sky Survey Quasar Lens Search (SQLS; Oguri et al. 2006; Inada et al. 2012) and the SDSS-III BOSS quasar lens survey (More et al. 2016) in the optical. The Dark Energy Survey (DES; Dark Energy Survey Collaboration 2005, 2016) is an ongoing imaging survey covering 5000 deg² of the Southern Galactic Cap in the grizY filters using the Dark Energy Camera (Flaugher et al. 2015), and it holds the promise of significantly increasing the numbers of lensed quasars. In particular, based on the forecasts of Oguri & Marshall (2010), we expect to find in DES about 120 lensed quasar systems brighter than i = 21 (magnitude limit applies to the fainter image for pairs and third brightest image for quadruple systems, or quads; see Figure 1 of Ostrovski et al. 2017). DES, specifically with the external collaboration project STRIDES⁵⁵ (PI: T. Treu), aims to use the resulting large lensed quasar sample for the primary science goal of improving constraints on cosmological parameters. This lensed quasar sample is predicted to include about 20 of the very rare quad systems that will provide additional valuable information compared to pair systems for constraining lens models (specifically extra constraints from two more positions and two more time delays), in particular for cosmology purposes (e.g., Suyu et al. 2013) and substructure studies (e.g., Dalal & Kochanek 2002).

To date, we have discovered and spectroscopically confirmed three lensed quasar pair systems (Agnello et al. 2015b; Ostrovski et al. 2017) in DES. Here in this Letter we report the discovery and confirmation of the first lensed quasar quad (or quad-like) system, DES J0408-5354, in the DES. We first describe our lensed quasar search procedure, system discovery, and photometric data in Section 2. We then describe our spectroscopic observations and present our data in Section 3. We summarize and conclude in Section 4. Detailed photometry analysis and lens modeling for this system are presented in a modeling paper, Agnello et al. (2017), to which we will refer the reader where relevant below.

The system DES J0408-5354 was discovered during a systematic visual search for candidate strong lensing systems in the DES Year 1 (Y1) data (Diehl et al. 2014). A number of different search methods have been applied to the DES Y1 data, but the specific method involved in this case was a "blue-near-red" technique, where we first automatically identified candidate systems of red galaxies with multiple neighboring blue objects within some radius, and then used visual inspections of these systems to rate them and to select the best candidates for subsequent spectroscopic follow-up. This method has been used previously in the Sloan Digital Sky Survey (SDSS) data to search for bright strong lensing systems and then to confirm them spectroscopically with good success (e.g., Belokurov et al. 2009; Diehl et al. 2009).

Here, we started with a DES Y1 red galaxy sample selected using the redMaGiC technique (Rozo et al. 2016). To minimize stellar contamination of this sample, we rejected objects with SExtractor (Bertin & Arnouts 1996) spread_model(Desai et al. 2012; Bouy et al. 2013) values $\leqslant 0.01$. We then identified our initial set of candidates as those redMaGiC galaxies with three or more blue objects within a radius $\lt 10^{\prime\prime}$, where we defined a blue object as one with colors $-1\lt =\,g-r\lt 1$ and $-1\lt =\,r-i\lt 1$. We did not apply any star/galaxy separation cut to the blue objects, but did reject objects that are saturated in any of the $g,r,i$ filters using a SExtractor flags $\lt \,=\,3$ cut. We also applied a magnitude cut $r\lt 22$ on the blue objects to keep the number of candidates manageable for the visual inspection step and to have relatively brighter candidates to ease follow-up spectroscopic redshift measurements.

Applying the above criteria resulted in a list of 6526 systems that one of us (HL) then inspected visually by examining their DES gri color composite images. The discovery image of DES J0408-5354 is shown in Figure 1, where we see a central red galaxy (G1) surrounded by three blue objects (A, B, and D) and a fourth red object (G2) (objects labeled as in Agnello et al. 2017). This system stood out as a potential quadruply lensed quasar system, except that the fourth putative lensed quasar image has a conspicuously redder color than those of the other three images. Regardless, the quad-like configuration of DES J0408-5354 made it a very good candidate for subsequent spectroscopy, described in the next section. We present a detailed lensing model for this system in Agnello et al. (2017), which concludes that the system is indeed a quad where the reddened image is a blend of the fourth lensed quasar image with a perturbing galaxy.

Zoom In Zoom Out Reset image size

In the DES Y1 catalog, DES J0408-5354 is composed of three objects due to the fact that components D and G1 are blended and that B and G2 are also blended. As such, the catalog contains entries from A, D+G1, and B+G2. The SExtractor segmentation map for the system is shown in Figure 1, overplotted on the DES i-band image of the system. We need to point out here that all three catalog objects actually meet our blue object color criteria given before (see Figure 2, leftmost panel), so that the redMaGiC galaxy for this system is not G1 but rather the red galaxy G0 marked in Figure 1. G0 is about $6^{\prime\prime}$ away from G1 and has a redMaGiC photometric redshift of 0.66 ± 0.03, close to G1's spectroscopic redshift of 0.597 (see Section 3). Therefore, in this case, our blue-near-red method found the system via another red galaxy likely associated with the lensing galaxy, rather than directly via the red lensing galaxy itself. The discovery was thus likely not coincidental, but was also not by the direct route as intended by the method. In similar DES Y1 blue-near-red and related searches, we also see other cases in galaxy group and cluster environments where the lensing system is being indirectly found this way (H. T. Diehl et al. 2017, in preparation).

Zoom In Zoom Out Reset image size

The coordinates and photometry of the three DES catalog objects are given in Table 1. The DES SExtractor auto magnitudes of the three catalog objects are listed in Table 1, as are the near-infrared Kron magnitudes (Kron 1980) from the VISTA Hemisphere Survey (VHS; McMahon et al. 2013) and the mid-infrared magnitudes from the Wide-field Infrared Survey Explorer (WISE; Wright et al. 2010). For the near-infrared catalog, magnitudes were obtained by cross-matching the DES and VHS catalogs using a 15 search radius. Component D+G1 does not have a J-band magnitude value due to blending with B+G2 in this band. For the mid-infrared magnitudes, the DES and WISE catalogs were cross-matched using a 40 search radius. All components of the system are blended into a single WISE source, so that the W1 and W2 magnitudes displayed in the table apply to A+D+G1+B+G2 as a whole.

Table 1.  DES J0408-5354 Positions (J2000 Coordinates) and Photometry (AB Magnitudes)

	A	D+G1	B+G2	A	B	D	G1	G2	G0
R.A.	4h8m2192	4h8m2171	4h8m2150	4h8m2192	4h8m2149	4h8m2171	4h8m2170	4h8m2149	4h8m2183
Decl.	−53°54'10	−53°53'587	−53°53'592	−53°54'10	−53°53'592	−53°53'581	−53°53'596	−53°54'03	−53°54'59
g	19.99 ± 0.01	20.45 ± 0.01	19.74 ± 0.01	20.07 ± 0.07	19.98 ± 0.07	20.90 ± 0.07	22.18 ± 0.20	22.68 ± 0.20	23.46 ± 0.19
r	19.94 ± 0.01	20.08 ± 0.01	19.51 ± 0.01	20.16 ± 0.07	19.95 ± 0.07	20.94 ± 0.10	20.65 ± 0.03	21.98 ± 0.15	21.68 ± 0.05
i	19.75 ± 0.01	19.57 ± 0.01	19.18 ± 0.01	20.16 ± 0.07	19.74 ± 0.10	20.73 ± 0.12	19.77 ± 0.04	21.46 ± 0.15	20.37 ± 0.03
z	19.51 ± 0.02	19.13 ± 0.02	18.82 ± 0.01	19.96 ± 0.10	19.28 ± 0.08	20.42 ± 0.10	19.31 ± 0.03	20.91 ± 0.12	19.90 ± 0.04
Y	19.48 ± 0.07	19.09 ± 0.05	18.69 ± 0.03	20.04 ± 0.10	19.34 ± 0.10	20.77 ± 0.13	19.12 ± 0.05	20.56 ± 0.16	19.79 ± 0.12
J	19.58 ± 0.06	⋯	18.77 ± 0.03
H	19.62 ± 0.08	18.90 ± 0.04	18.65 ± 0.03
K	19.06 ± 0.07	18.51 ± 0.04	18.15 ± 0.03
	A + D+G1 + B+G2
W1		16.78 ± 0.02
W2		16.51 ± 0.02

Download table as:  ASCIITypeset image

We present in Agnello et al. (2017) a detailed analysis that provides the deblended positions and grizY magnitudes of all five components A, B, D, G1, and G2 in this system. For completeness we also provide this information in Table 1, as well as include the DES catalog position and photometry for galaxy G0.

After DES J0408-5354 was first announced to the STRIDES Collaboration as a lensed quasar candidate from the blue-near-red search, a number of other search methods within STRIDES were examined and also seen to identify the system as a candidate. These other search methods are described in more detail in Agnello et al. (2017). We provide here a brief summary: (1) the Gaussian Mixture Model (GMM) method of Ostrovski et al. (2017), which uses supervised machine learning in a five-dimensional optical plus infrared color space and identified DES J0408-5354 as a candidate pair (D+G1 not found separately in J band due to blending noted above); (2) CHITAH (Chan et al. 2015), which uses pixel-based automatic recognition on grizY cutout images and identified the system as a candidate pair (not flagged as a quad because the fourth image G2 is too red); and (3) the Artificial Neural Network method of Agnello et al. (2015a, 2015b), which uses griz and $W1W2$ magnitudes and identified DES J0408-5354 as a candidate extended quasar. Figure 2 shows how the system was flagged by the GMM method as a candidate, due to the quasar-like colors of its components. We note that none of the search methods mentioned in this Letter selected DES J0408-5354 as a quad candidate, due to blending issues and/or reddended colors. The unusual color and configuration of this system and potential others like it argue for incorporating more flexibility into our search techniques (Agnello et al. 2017).

Spectroscopic observations of DES J0408-5354 were carried out using the Gemini Multi-Object Spectrograph (GMOS-S) on the Gemini South telescope, as part of a larger Gemini Large and Long Program (LLP; PI: E. Buckley-Geer; program IDs GS-2015B-LP-5, GS-2016A-LP-5) of spectroscopic follow-up for DES strong lensing systems and for DES photometric redshift (photo-z) calibrations. GMOS-S was used in multi-object spectroscopy (MOS) mode on 2015 December 9 UT to take spectra of the three blue objects A, B, and D. The i-band GMOS-S acquisition image is shown in Figure 1 (top right). A single slit mask was used, with one slit for A and a second slit for B and D together; see blue slits in Figure 1 (bottom right). (Another 38 slits were assigned to unrelated DES photo-z calibration targets.) Two sets of spectra were taken: blue data using the B600 grating (dispersion $\approx 1.0$ Å pixel⁻¹) and red data using the R400 grating (dispersion $\approx 1.5$ Å pixel⁻¹). We used 4 × 900 s exposures for cosmic ray rejection and processed the data using the IRAF Gemini reduction package. The seeing was 08, measured from spectra of mask-alignment stars in the science data.

From the extracted (but not flux calibrated) 1D spectra shown in Figure 3 (top panels), we clearly see that all three blue images A, B, and D show strong quasar emission lines at the same redshift, specifically Lyα 1216 Å and C iv 1549 Å in the blue spectra, and C iii] 1909 Å and Mg ii 2800 Å in the red spectra. From the C iii] line, which shows the cleanest, most symmetric line profile among these four broad emission features, we use the emsao task in the IRAF external package rvsao (Kurtz & Mink 1998) to report a redshift z = 2.375 for the quasar.

Zoom In Zoom Out Reset image size

To compare the spectra of the three blue images in more detail, we also show in Figure 3 (bottom panels) the fractional differences between the blue and red spectra of images A and D relative to those of image B, which has the spectra with the most counts. We first rescale the spectra of A and D so that they each have the same median counts as those of B over the respective blue and red wavelength ranges plotted. The results in Figure 3 show generally good agreement among the spectra, especially for the red spectra. For the blue spectra, we do see an overall slope for the spectrum of A relative to that of B; we attribute this slope to extinction differences between the lines of sight to these two images, and/or to differences in atmospheric differential refraction between the two differently oriented slits (Figure 1) used to observe images A and B. In addition, we also see conspicuous differences between the spectra of D and B at the locations of the strong Lyα and C iv emission lines, and to a lesser extent at the C iii] line. We attribute these differences to the effects of microlensing by stars in the lensing galaxy, manifested as flux ratio differences, in pairs of lensed quasar images, for the continuum versus the emission line regions in the spectra (e.g., Motta et al. 2012). Microlensing may also cause the spectral slope differences mentioned above.

Subsequent to these observations that confirmed the three blue images as having the same quasar redshift, we designed a second MOS slit mask targeting the central lensing galaxy G1 and the fourth image G2, each allocated to a single slit. The slits for G1 and G2 are shown in red in Figure 1 (bottom right). Red R400 observations were obtained in the Gemini South semester 2016A observing queue, on 2016 April 9, under 11 seeing conditions and using the same observing setup as above. Blue B600 observations were also put into the queue, but no data were obtained before the system set in 2016A. We show the R400 spectra in Figure 4, where we see that G2 shows broad C iii] and Mg ii emission at the same redshift as found for the three blue images, while G1 shows Ca H+K absorption lines, indicating that G1 is an early-type lensing galaxy with redshift z = 0.597.

Zoom In Zoom Out Reset image size

However, we also see that the G1 spectrum unexpectedly shows the same C iii] emission line as in the lensed quasar spectra, leading us to suspect there is contaminating light from the neighboring lensed quasar image D, which is only about 13 away (cf. the 11 seeing). Likewise, object B, the brightest quasar image, is only about 11 away from the substantially fainter object G2, suggesting that the C iii] line seen in the G2 spectrum is also contamination from object B. Using the image modeling code GALFIT (Peng et al. 2010), we simulate image B as a Moffat profile with an FWHM of 11 and find that about 12% of its light falls within the 1''-wide slit and 2''-long spectral extraction aperture used for G2. Though seemingly small, this 12% fraction actually amounts to about 80% of the r-band light from G2 itself, given that $r({\rm{B}})=19.95$ and $r({\rm{G}}2)=21.98$ from Table 1 (Agnello et al. 2017). (We use r band as the C iii] line falls within that filter.) This estimated contamination is thus substantial and unfortunately we are unable to correct for it as image B lies off the slit for G2 (Figure 1, bottom right) and we do not have a concurrent spectrum of B. We thus cannot rule out contamination from B as the source of the C iii] line seen in the G2 spectrum (likewise for D and G1) and will need future observations to verify whether G2 indeed shows quasar emission features or not.

DES J0408-5354 was identified as a candidate lensed quasar quad system from the DES first-year data. Subsequent follow-up spectroscopy confirmed the three bright blue objects in this system to be the images of a quasar at redshift z = 2.375, lensed by an early-type red galaxy with redshift z = 0.597. Another reddened object in the system, possibly a blend of a perturbing galaxy and the fourth lensed quasar image, was also observed spectroscopically, though without conclusive results.

Our modeling paper, Agnello et al. (2017), presents a detailed model of the system, from which we expect the longest time delay in the system to be about 80 days, making DES J0408-5354 well suited (Treu & Marshall 2016) for time delay measurements, which we are undertaking via a monitoring campaign within the STRIDES Collaboration. The lensing model also constrains the mass of the possible perturbing galaxy and thus provides information about substructure in the lensing mass distribution. Further spectroscopic follow-up and high-resolution imaging data should provide more needed details about the main lensing galaxy and its environment. We thus expect this quad system to be particularly useful for the application of time delay cosmography (e.g., Treu & Marshall 2016) and substructure studies (e.g., Dalal & Kochanek 2002). DES J0408-5354 heralds a much larger sample of some 20 of these very rare and valuable quad lensed quasar systems anticipated to be discovered by the DES.

Funding for the DES Projects has been provided by the U.S. Department of Energy, the U.S. National Science Foundation, the Ministry of Science and Education of Spain, the Science and Technology Facilities Council of the United Kingdom, the Higher Education Funding Council for England, the National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign, the Kavli Institute of Cosmological Physics at the University of Chicago, the Center for Cosmology and Astro-Particle Physics at the Ohio State University, the Mitchell Institute for Fundamental Physics and Astronomy at Texas A&M University, Financiadora de Estudos e Projetos, Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro, Conselho Nacional de Desenvolvimento Científico e Tecnológico and the Ministério da Ciência, Tecnologia e Inovação, the Deutsche Forschungsgemeinschaft and the Collaborating Institutions in the Dark Energy Survey.

The Collaborating Institutions are Argonne National Laboratory, the University of California at Santa Cruz, the University of Cambridge, Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas-Madrid, the University of Chicago, University College London, the DES-Brazil Consortium, the University of Edinburgh, the Eidgenössische Technische Hochschule (ETH) Zürich, Fermi National Accelerator Laboratory, the University of Illinois at Urbana-Champaign, the Institut de Ciències de l'Espai (IEEC/CSIC), the Institut de Física d'Altes Energies, Lawrence Berkeley National Laboratory, the Ludwig-Maximilians Universität München and the associated Excellence Cluster Universe, the University of Michigan, the National Optical Astronomy Observatory, the University of Nottingham, The Ohio State University, the University of Pennsylvania, the University of Portsmouth, SLAC National Accelerator Laboratory, Stanford University, the University of Sussex, Texas A&M University, and the OzDES Membership Consortium.

The DES data management system is supported by the National Science Foundation under grant number AST-1138766. The DES participants from Spanish institutions are partially supported by MINECO under grants AYA2012-39559, ESP2013-48274, FPA2013-47986, and Centro de Excelencia Severo Ochoa SEV-2012-0234. Research leading to these results has received funding from the European Research Council under the European Union's Seventh Framework Programme FP7/2007-2013) including ERC grant agreements 240672, 291329, and 306478.

Based on observations obtained at the Gemini Observatory (acquired through the Gemini Observatory Archive and processed using the Gemini IRAF package), which is operated by the Association of Universities for Research in Astronomy, Inc., under a cooperative agreement with the NSF on behalf of the Gemini partnership: the National Science Foundation (United States), the National Research Council (Canada), CONICYT (Chile), Ministerio de Ciencia, Tecnología e Innovación Productiva (Argentina), and Ministério da Ciência, Tecnologia e Inovação (Brazil).

The analysis presented here is based on observations obtained as part of the VISTA Hemisphere Survey, ESO Programme, 179.A-2010 (PI: McMahon).