A Multi-Band Forced-Photometry Catalog in the ELAIS-S1 Field
Fan Zou, W. N. Brandt, Mark Lacy, Qingling Ni, Kristina Nyland, Guang Yang, Franz E. Bauer, Giovanni Covone, Aniello Grado, Nicola R. Napolitano, Maurizio Paolillo, Mario Radovich, Marilena Spavone, Mattia Vaccari
DD raft version F ebruary
25, 2021Typeset using L A TEX default style in AASTeX63
A Multi-Band Forced-Photometry Catalog in the ELAIS-S1 Field F an Z ou , W. N. B randt , M ark L acy , Q ingling N i , K ristina N yland , G uang Y ang , F ranz E. B auer , G iovanni C ovone , A niello G rado , N icola R. N apolitano , M aurizio P aolillo , M ario R adovich , M arilena S pavone , and M attia V accari Department of Astronomy and Astrophysics, 525 Davey Lab, The Pennsylvania State University, University Park, PA 16802, USA Institute for Gravitation and the Cosmos, The Pennsylvania State University, University Park, PA 16802, USA Department of Physics, 104 Davey Laboratory, The Pennsylvania State University, University Park, PA 16802, USA National Radio Astronomy Observatory, 520 Edgemont Road, Charlottesville, VA 22903, USA National Research Council, resident at the U.S. Naval Research Laboratory, 4555 Overlook Ave. SW, Washington, DC 20375, USA Department of Physics and Astronomy, Texas A & M University, College Station, TX, 77843-4242 USA George P. and Cynthia Woods Mitchell Institute for Fundamental Physics and Astronomy, Texas A & M University, College Station, TX, 77843-4242 USA Instituto de Astrof´ısica, Facultad de F´ısica, Pontificia Universidad Cat´olica de Chile Av. Vicu˜na Mackenna 4860, 782-0436 Macul, Santiago, Chile National Radio Astronomy Observatory, Pete V. Domenici Array Science Center, P.O. Box O, Socorro, NM 87801, USA Space Science Institute, 4750 Walnut Street, Suite 205, Boulder, CO 80301, USA INAF - Osservatorio Astronomico di Capodimonte, Salita Moiariello 16, I-80131, Napoli, Italy Dipartimento di Fisica, Universit`a di Napoli “Federico II”, via Cinthia 9, 80126 Napoli, Italy INFN - Sezione di Napoli, via Cinthia 9, 80126 Napoli, Italy INAF - Osservatorio Astronomico di Padova, vicolo Osservatorio, 5 I-35122 Padova, Italy Inter-university Institute for Data Intensive Astronomy, Department of Physics and Astronomy, University of the Western Cape, Robert Sobukwe Road, 7535Bellville, Cape Town, South Africa INAF - Istituto di Radioastronomia, via Gobetti 101, 40129 Bologna, Italy
ABSTRACTThe ELAIS-S1 field will be an LSST Deep Drilling field, and it also has extensive multiwavelength coverage.To improve the utility of the existing data, we use
The Tractor to perform forced-photometry measurements inthis field. We compile data in 16 bands from the DeepDrill, VIDEO, DES, ESIS, and VOICE surveys. Usinga priori information from the high-resolution fiducial images in VIDEO, we model the images in other bandsand generate a forced-photometry catalog. This technique enables consistency throughout di ff erent surveys,deblends sources from low-resolution images, extends photometric measurements to a fainter magnitude regime,and improves photometric-redshift estimates. Our catalog contains over 0.8 million sources covering a 3 . area in the VIDEO footprint and is available at 10.5281 / zenodo.4540178.To aid counterpart characterization for the XMM-SERVS survey (Chen et al. 2018; Ni et al., in preparation) and help preparefor the upcoming Deep Drilling Fields (e.g., Brandt et al. 2018) of the Vera C. Rubin Observatory Legacy Survey of Space andTime (LSST), we generate a multi-band photometric catalog in the European Large-Area ISO Survey-S1 (ELAIS-S1) field fromthe Spitzer
DeepDrill (3.6 and 4 . µ m; Lacy et al. 2021) , VIDEO ( ZY JHK s ; Jarvis et al. 2013) , DES DR2 ( g riz y ; Abbott et al.2021) , ESIS ( BVR ; Berta et al. 2006; Vaccari et al. 2016) , and VOICE ( u ; Vaccari et al. 2016) surveys using The Tractor (Lang et al. 2016) forced-photometry approach (see below). The coverages of these surveys are displayed in Fig. 1. Comparedto the conventional single-band photometric measurements, this forced-photometry technique can distinguish blended sourcesin low-resolution surveys (e.g., DeepDrill) and enables much more reliable photometric-redshift measurements (e.g., Lang et al.2014; Nyland et al. 2017; Wold et al. 2019).We follow the same approach to make
The Tractor photometric measurements as Nyland et al. (2017) for VIDEO-detectedsources in the full VIDEO area. We only briefly outline the procedures here, and interested readers can refer to Nyland et al.
E-mail: [email protected] http: // personal.psu.edu / wnb3 / xmmservs / xmmservs.html https: // irsa.ipac.caltech.edu / data / SPITZER / DeepDrill / http: // horus.roe.ac.uk / vsa / https: // des.ncsa.illinois.edu / releases / dr2 http: // / mattiavaccari.net / esis / http: // / voice / a r X i v : . [ a s t r o - ph . GA ] F e b (2017) for more details. For each source, we adopt the band with the longest wavelength among the VIDEO bands in whichthe source is detected as the fiducial band and then obtain the source position and intrinsic surface-brightness profile (point-like, deVaucouleurs, or exponential profile) from the corresponding VIDEO catalog and image. We fix this prior position andintrinsic surface-brightness profile, only allowing the fluxes to vary, and fit the images in other bands, where the intrinsic profileis convolved with the corresponding PSFs to match the observed profiles. This procedure returns forced photometry. Thephotometric errors are derived using the equations in the Appendix of Nyland et al. (2017). The astrometric errors and imageo ff sets of our data are smaller than the sizes of single pixels (0 . (cid:48)(cid:48) − . (cid:48)(cid:48) ), and thus do not considerably downgrade our photometricaccuracy (Nyland et al. 2017). We flag possibly saturated sources as those with magnitudes smaller than 13.5 (4 . µ m), 14.0(3 . µ m), 14.0 ( K s ), 14.2 ( H ), 14.5 ( J ), 13.6 ( Y ), 14.0 ( Z ), 14.5 ( R ), 14.2 ( V ), and 14.6 ( B ). These thresholds are similar to thosein Nyland et al. (2017). After comparing The Tractor photometry and original photometry for DES, we find that there are somesignificant outliers. These outliers are not a result of
The Tractor fitting; instead, their original DES z - or y -band magnitudes alsoshow significant deviations compared to their original VIDEO Z - or Y -band magnitudes, indicating that their DES images maybe problematic (e.g., saturated), and thus these sources are flagged. We release our catalog online. It contains fiducial bands, intrinsic surface-brightness profile, AB magnitudes, magnitudeerrors, flags, reduced χ values, and nearest-neighbor angular separations. The separations can be used as a basic indication forpossible blending issues in DeepDrill images. A “ readme.txt ” file along with the catalog provides detailed explanations of thecatalog columns. Our released data also include several diagnostic figures to characterize the reliability of our catalog: Fig. S1compares the forced photometry and the original photometry, with flagged sources displayed explicitly. Fig. S2 plots The Tractor magnitude errors versus magnitudes for all sources, based on which the nominal 5 σ magnitude limits are estimated. The limitsare estimated to be 24.3 (4 . µ m), 24.5 (3 . µ m), 24.6 ( K s ), 25.1 ( H ), 25.7 ( J ), 26.0 ( Y ), 26.2 ( Z ), 22.7 ( y ), 23.9 ( z ), 24.5 ( i ), 25.2( r ), 25.5 ( g ), 25.8 ( R ), 25.9 ( V ), 26.5 ( B ), and 25.9 ( u ). Fig. S3 displays the magnitude distributions for both the original catalogsand our catalog after applying a 2 σ cut and demonstrates that we are able to place photometric constraints on a considerablenumber of faint sources not formally detected in single-band images.Our catalog greatly improves the photometric-redshift ( z phot ) results, which was a primary motivation for our work. We useEAZY (Brammer et al. 2008) to estimate z phot , following an iterative method of magnitude zero-point correction (Yang et al.2014), and then compare z phot values and spectroscopic redshifts ( z spec ) for ∼ z spec measurements (typicallywith R magnitudes ≈ − z spec compilation is described in Ni et al. (in preparation). Using publicly availableSERVS, VIDEO, and DES cross-matched catalogs, the normalized median absolute deviation ( σ NMAD ) and the outlier fraction( f outlier ; defined as | z phot − z spec | / (1 + z spec ) > .
15) are 0.071 and 17.4%, respectively. After utilizing our catalog, the values dropto σ NMAD = .
032 and f outlier = . z phot measurements as well as the o ffi cial release of the corresponding z phot catalog.The forced-photometry catalogs in the other parts of XMM-SERVS are or will be publicly available as well; see Nyland et al.(2017) and Nyland et al. (in preparation). Acknowledgements . We acknowledge support from NASA grant 80NSSC19K0961. Basic research in radio astronomy atthe U.S. Naval Research Laboratory is supported by 6.1 Base Funding. FEB acknowledges support from ANID-Chile BasalAFB-170002, FONDECYT Regular 1200495 and 1190818, and Millennium Science Initiative – ICN12 009. MV acknowledgessupport from the Italian Ministry of Foreign A ff airs and International Cooperation (MAECI Grant Number ZA18GR02) and theSouth African Department of Science and Technology’s National Research Foundation (DST-NRF Grant Number 113121) aspart of the ISARP RADIOSKY2020 Joint Research Scheme. See, e.g., part 3 in https: // herschel-vos.phys.sussex.ac.uk / vo-data / dmu6 / dmu6 v ELAIS-S1 / help elais-s1 checks.html, and we follow the same approach todefine outliers. https: // doi.org / / zenodo.4540178 D E C ( d e g ) VIDEOESISVOICEXMM-SERVS
Figure 1.
The boundaries of multiwavelength survey regions in ELAIS-S1. The black circles represent the
XMM-Newton pointings composingthe XMM-SERVS survey in this field. The
Spitzer
DeepDrill and DES surveys cover the whole field, and thus their coverages are not displayed.Our cataloged 3 . region is the VIDEO survey region. REFERENCES
Abbott, T. M. C., Adamow, M., Aguena, M., et al. 2021, arXive-prints, arXiv:2101.05765Berta, S., Rubele, S., Franceschini, A., et al. 2006, A&A, 451, 881Brammer, G. B., van Dokkum, P. G., & Coppi, P. 2008, ApJ, 686,1503Brandt, W. N., Ni, Q., Yang, G., et al. 2018, arXiv e-prints,arXiv:1811.06542Chen, C.-T. J., Brandt, W. N., Luo, B., et al. 2018, MNRAS, 478,2132 Jarvis, M. J., Bonfield, D. G., Bruce, V. A., et al. 2013, MNRAS,428, 1281Lacy, M., Surace, J. A., Farrah, D., et al. 2021, MNRAS, 501, 892Lang, D., Hogg, D. W., & Mykytyn, D. 2016, The Tractor:Probabilistic astronomical source detection and measurementLang, D., Hogg, D. W., & Schlegel, D. J. 2014, arXiv e-prints,arXiv:1410.7397Mauduit, J. C., Lacy, M., Farrah, D., et al. 2012, PASP, 124, 714Nyland, K., Lacy, M., Sajina, A., et al. 2017, ApJS, 230, 9
Vaccari, M., Covone, G., Radovich, M., et al. 2016, in The 4thAnnual Conference on High Energy Astrophysics in SouthernAfrica (HEASA 2016), 26Wold, I. G. B., Kawinwanichakij, L., Stevans, M. L., et al. 2019,ApJS, 240, 5 Yang, G., Xue, Y. Q., Luo, B., et al. 2014, ApJS, 215, 27
APPENDIXHere are some supplementary notes: • Fig. S1. The slightly tilted trends in the DeepDrill panels may be due to calibration di ff erences for extended sources andalso exist in other work (e.g., Nyland et al. 2017). The small “antenna-shaped” structures in the VIDEO panels, wherethe original magnitudes are higher than The Tractor magnitudes for some faint sources, are mainly composed of sourcesaround bright stars, and we attribute much of these source fluxes to the backgrounds contributed by the adjacent stars. Thedegree to which the plots appeal to “funnel” out in the faint regime depends on the detection significance thresholds of theoriginal catalogs as well as the total number of plotted sources (or the plotting settings). For example, the original ESIScatalog has a relatively higher threshold and fewer sources, and thus the noise of the ESIS panels is less. • DES DR2 outliers. DES outliers are selected by comparing VIDEO ZY and DES z y magnitudes (see Footnote 7). Fig. S1further shows that The Tractor magnitudes of these outliers also deviate from the corresponding DES magnitudes in all theDES bands, and thus we suggest applying the DES outlier flags to all the DES bands. Although many of these outlierspresent clearly saturated images (i.e., with trapezoidal brightness profiles), the others do not show abnormal images. Theactual causes for these outliers, however, are still unclear. T r a c t o r D ee p D r ill . m M a g T r a c t o r D ee p D r ill . m M a g K s Mag3025201510 T r a c t o r V I D E O K s M a g H Mag3025201510 T r a c t o r V I D E O H M a g J Mag3025201510 T r a c t o r V I D E O J M a g Y Mag3025201510 T r a c t o r V I D E O Y M a g Z Mag3025201510 T r a c t o r V I D E O Z M a g T r a c t o r D E S y M a g T r a c t o r D E S z M a g T r a c t o r D E S i M a g T r a c t o r D E S r M a g T r a c t o r D E S g M a g R Mag3025201510 T r a c t o r E S I S R M a g V Mag3025201510 T r a c t o r E S I S V M a g B Mag3025201510 T r a c t o r E S I S B M a g Figure S1.
Comparison between
The Tractor photometry and original AB photometry for all the bands except u band, which only has availableimages but does not have an original catalog. The black dashed lines mark one-to-one relations. The red points are flagged sources. ForDeepDrill, the orange points are possibly blended sources (i.e., nearest-neighbor separations < . (cid:48)(cid:48) ), and the green points are other sources,which are less likely to su ff er from blending issues. The original magnitudes used for comparisons here are aperture-corrected magnitudeswithin an aperture of radius = . (cid:48)(cid:48) for DeepDrill, Petrosian magnitudes for VIDEO, and MAG AUTO magnitudes for DES and ESIS.
15 18 21 24 27 30DeepDrill 4.5 m Mag0.00.10.20.30.40.5 D ee p D r ill . m M a g e rr o r
5 limit: 24.3 15 18 21 24 27 30DeepDrill 3.6 m Mag0.00.10.20.30.40.5 D ee p D r ill . m M a g e rr o r
5 limit: 24.5 15 18 21 24 27 30VIDEO K s Mag0.00.10.20.30.40.5 V I D E O K s M a g e rr o r
5 limit: 24.6 15 18 21 24 27 30VIDEO H Mag0.00.10.20.30.40.5 V I D E O H M a g e rr o r
5 limit: 25.115 18 21 24 27 30VIDEO J Mag0.00.10.20.30.40.5 V I D E O J M a g e rr o r
5 limit: 25.7 15 18 21 24 27 30VIDEO Y Mag0.00.10.20.30.40.5 V I D E O Y M a g e rr o r
5 limit: 26.0 15 18 21 24 27 30VIDEO Z Mag0.00.10.20.30.40.5 V I D E O Z M a g e rr o r
5 limit: 26.2 15 18 21 24 27 30DES y Mag0.00.10.20.30.40.5 D E S y M a g e rr o r
5 limit: 22.715 18 21 24 27 30DES z Mag0.00.10.20.30.40.5 D E S z M a g e rr o r
5 limit: 23.9 15 18 21 24 27 30DES i Mag0.00.10.20.30.40.5 D E S i M a g e rr o r
5 limit: 24.5 15 18 21 24 27 30DES r Mag0.00.10.20.30.40.5 D E S r M a g e rr o r
5 limit: 25.2 15 18 21 24 27 30DES g Mag0.00.10.20.30.40.5 D E S g M a g e rr o r
5 limit: 25.515 18 21 24 27 30ESIS R Mag0.00.10.20.30.40.5 E S I S R M a g e rr o r
5 limit: 25.8 15 18 21 24 27 30ESIS V Mag0.00.10.20.30.40.5 E S I S V M a g e rr o r
5 limit: 25.9 15 18 21 24 27 30ESIS B Mag0.00.10.20.30.40.5 E S I S B M a g e rr o r
5 limit: 26.5 15 18 21 24 27 30VOICE u Mag0.00.10.20.30.40.5 V O I C E u M a g e rr o r
5 limit: 25.9
Figure S2.
The Tractor magnitude errors versus AB magnitudes. The red lines mark the 5 σ limits, where the errors are 0.2, and the limitsare displayed as the titles. Note that the actual limits vary among di ff erent regions in the field for non-uniform surveys (e.g., the u band of theVOICE survey, whose exposure time varies across the whole field), and the limits quoted here only refer to the deepest regions.
15 20 25DeepDrill 4.5 m Mag0369 N u m b e r o f s o u r c e s ×10 Tractor photometryOrignial photometry
15 20 25DeepDrill 3.6 m Mag 15 20 25VIDEO K s Mag 15 20 25VIDEO H Mag15 20 25VIDEO J Mag0369 N u m b e r o f s o u r c e s ×10
15 20 25VIDEO Y Mag 15 20 25VIDEO Z Mag 15 20 25DES y Mag15 20 25DES z Mag0369 N u m b e r o f s o u r c e s ×10
15 20 25DES i Mag 15 20 25DES r Mag 15 20 25DES g Mag15 20 25ESIS R Mag0369 N u m b e r o f s o u r c e s ×10
15 20 25ESIS V Mag 15 20 25ESIS B Mag 15 20 25VOICE u Mag
Figure S3.
Magnitude distributions for our catalog from
The Tractor and original catalogs in each band. Only sources with magnitude errorssmaller than 0.54 are shown, where the threshold roughly corresponds to signal-to-noise ratios = uu