Sloan Extended Quasar, ELG, and LRG Survey (SEQUELS)

Summary

Covering several hundred square degrees within the BOSS footprint, SEQUELS is both a stand-alone ancillary science program and a pilot program for the upcoming eBOSS survey of SDSS-IV. The program targets Luminous Red Galaxies (LRGs), quasars (QSOs), and emission line galaxies (ELGs). A description of this survey can be found in the eBOSS overview paper.

Description

SEQUELS covers several hundred square degrees within the BOSS footprint, targeting LRGs, QSOs, and emission line galaxies (ELGs). SEQUELS serves both as a pilot program for the eBOSS survey of SDSS-IV and as a stand-alone science program.

SEQUELS targets represent a superset of targets selected in the eBOSS program. SEQUELS also encompasses two sub-programs to obtain spectra of variability-selected objects (TDSS) and X-ray detected objects (SPIDERS) that will be described in more detail in the Target Selection section below.

The main SEQUELS footprint lies in the NGC. Targets were selected over the region covering 120° < RA < 210° and 45° < DEC < 60° within the nominal BOSS footprint, although only ~ 1/3 of this area was covered by DR12 spectroscopy. Plates that were drilled but not observed in DR12 will be observed as part of eBOSS. Plates that were tiled but not drilled will be re-tiled with the final eBOSS target selection. SEQUELS LRGs are selected to extend the BOSS LRG redshift coverage, ielding a median redshift of ~ 0.7. SEQUELS QSOs represent a combination of targets selected as Lyman-α forest QSOs and lower-redshift objects that serve as direct tracers of the density field at redshifts 1 < z < 2. Lyman-α forest QSOs are themselves a compilation of known BOSS targets that have been selected for re-observation, and new targets selected from variability methods. SEQUELS ELGs lie within 9 dedicated plates in the NGC, observed separately from the rest of the program. The details of each target class are listed in Target Selection below.

Target Selection

LRGs in SEQUELS were selected using a combination of SDSS imaging and WISE photometry. These two selections are designed to target massive red galaxies at z ~ 0.6. The two selections have substantial overlap, with roughly 50% of objects in one selection also found in the other.

All SDSS imaging comes from v5b reductions. WISE magnitudes come from application of the tractor code to the WISE full-sky maps. All magnitudes are model magnitudes, and all magnitudes have been corrected for extinction. WISE magnitudes have been converted to the AB system.

The individual target bits in the table above are defined as follows:

LRG_IZW: These targets have i – z > 0.7 and i – W1 > 2.143*(i – z) – 0.2. Here W1 is the [X] WISE band. This sample has a limiting magnitude of z < 19.95 and i > 19.9.

LRG_RIW: This target class requires r – i > 0.98 and r – W1 > 2*(r-i) to select red galaxies in the region where stars do not dominate. Additionally, all targets have i – z > 0.625 to weight towards high z. This sample has a limiting magnitude of z < 19.95 and i > 19.9.

QSO_EBOSS_CORE: Quasars targeted as part of the SEQUELS CORE sample are assigned the QSO_EBOSS_CORE bit. The CORE sample is designed to obtain close to the desired target density of 68.8 per square degree of 0.9 < z < 2.2 quasars that comprise the main goal of eBOSS (assuming an exactly 7500 deg2 footprint for eBOSS).

We make no attempt to limit the upper end of the CORE redshift range, meaning that the CORE also selects z > 2.15 quasars that have utility for Lyman- α Forest studies. Quasars in the CORE are selected by a combination of XDQSOz in the optical and a WISE-optical color cut, as detailed in Myers et al. (2015). The basic cuts used to defi ne QSO_EBOSS_CORE targets (and almost all other quasars targeted in SEQUELS) are PRIMARY, objc_type = 6, and faint-end magnitude cuts of g < 22 OR r < 22. QSO_PTF: Quasars intended for Lyman-α Forest studies typically do not have to be selected in a uniform manner. This allows variability selection to be applied to inhomogeneous imaging in order to target additional z > 2.15 quasars, which are an intended eBOSS targeting class.

The QSO PTF bit indicates such quasars, which have been selected using multi-epoch imaging from the Palomar Transient Factory (PTF). Note that PTF targets undergo slightly di fferent initial cuts to other quasar target classes; they are limited in magnitude to r > 19 and g < 22.5. QSO_REOBS: Quasars previously con firmed in BOSS that are of reduced (but not prohibitively low) signal-to-noise have decreased utility for Lyman- α Forest studies. In addition, high-probability BOSS quasar targets that turned out to have zero spectral signal-to-noise in BOSS are likely to have been spectroscopic glitches. The QSO_REOBS bit signifi es quasars that were measured to have 0.75 <= SNR/pixel < 3 OR SNR/pixel = 0 in BOSS, which have been reobserved in SEQUELS. QSO EBOSS KDE: Targets that have the QSO_EBOSS_KDE bit set in SEQUELS were drawn from the Kernel Density Estimation catalog of Richards et al. (2009) and had uvxts = 1 set within that catalog.

QSO EBOSS FIRST: Powerful radio-selected quasars can be at z > 2.15, and can therefore have utility for Lyman-α Forest studies. The QSO_EBOSS_FIRST bit indicates quasars that are targeted in SEQUELS because they have an astrometric match within 1″ of a radio detection in the 13 June 05 version of the FIRST point source catalog (Becker, White, & Helfand 1995).

QSO_BAD_BOSS: Some likely quasars with spectroscopy obtained as part of BOSS have uncertain classifications or redshifts upon visual inspection. Such objects are designated as QSO? or QSO_Z? in the DR12 QSO catalog (Paris et. al., 2017).

The QSO_BAD_BOSS bit signifies such objects, to ensure that ambiguous BOSS quasars are always reobserved, regardless of which other targeting bits are set. A close-to-final but preliminary version of the DR12 catalog was used to define this sample for SEQUELS.

QSO BOSS TARGET: To study the effect of reducing the overall target density for eBOSS, SEQUELS quasar targeting does not retarget any objects with good spectra from BOSS unless otherwise specified. The QSO_BOSS_TARGET bit is set to indicate such objects.

We define an object as having good BOSS spectroscopy if it appears in the file of all spectra that have been observed by BOSS (specifically, the combination of v5_7_0 and v5_7_1 of the BOSS spAll file circa May 30, 2014), and if it does not have either LITTLE_COVERAGE or UNPLUGGED set in the ZWARNING bitmask (see Table 3 of Bolton et al. 2012).

QSO_SDSS_TARGET: Similar to the QSO_BOSS_TARGET bit, SEQUELS does not retarget objects with good pre-BOSS spectra from the SDSS (i.e. spectra from prior to SDSSdre). The QSO_SDSS_TARGET bit is set to indicate such objects.

We define an object as having good SDSS spectroscopy if it appears in the final DR8 spectroscopy files (specifically, the line-by-line parallel spectroscopy and imaging catalogs at
https://data.sdss.org/sas/dr8/sdss/spectro/redux/photoPosPlate-dr8.fits and https://data.sdss.org/sas/dr8/sdss/spectro/redux/specObj-dr8.fits), and if it does not have either LITTLE_COVERAGE or UNPLUGGED set in the ZWARNING bitmask (see Table 3 of Bolton et al. 2012).

QSO_KNOWN: SEQUELS quasar targeting does not re-observe objects with previous good spectra (defined by the QSO_BOSS_TARGET and QSO_SDSS_TARGET bits). The purpose of the QSO_KNOWN bit is to track which previously known objects have a good visually inspected (or otherwise highly confident) redshift and classification from prior spectroscopy.

Objects classified as having excellent prior spectroscopy are those that are of SDSS provenance and match the sample used to define known objects in BOSS (see Ross et al. 2012), or those that match a preliminary version of the DR12 BOSS quasar catalog of Paris et. al. (2017).

DO_NOT_OBSERVE: Which previously known quasars are targeted?

The parameter space for SEQUELS quasar targeting overlaps that of earlier incarnations of the SDSS. The bits QSO_BAD_BOSS, QSO_BOSS_TARGET, QSO_SDSS_TARGET and QSO_KNOWN work together to determine a sample of objects for which SEQUELS does not need to obtain an additional spectrum, because a good classification and redshift should already exist (if the object is a quasar).

Thus, targets are not observed by SEQUELS if any of QSO_BOSS_TARGET, QSO_SDSS_TARGET and QSO_KNOWN are set, unless QSO_BAD_BOSS is set also. In addition, QSO_REOBS always forces a re-observation of an earlier BOSS quasar.

In Boolean notation, DO_NOT_OBSERVE is then set (according to quasar target bits) if:

(QSO_KNOWN || QSO_BOSS_TARGET || QSO_SDSS_TARGET) && !(QSO_BAD_BOSS || QSO REOBS)

DR9_CALIB_TARGET: Which version of the SDSS imaging was used?

SEQUELS targeted quasars selected in both the DR9 imaging used for BOSS and in an updated DR12 imaging calibration intended for use in eBOSS targeting. The DR9_CALIB TARGET bit signifies quasars that were selected for SEQUELS using the DR9 imaging calibrations instead of (or as well as) the updated DR12 imaging.

SEQUELS_COLLIDED: Galaxies from the main BOSS target selection, both LOWZ and CMASS, that were not assigned fibers due to fiber collisions. Observing these galaxies in SEQUELS creates large contiguous areas that have 100% spectroscopic completeness in the final BOSS data sample.

SEQUELS_PTF_VAR: Targets that have the SEQUELS_PTF_VAR bit set are selected from the Palomar Transient Factory (PTF) survey. They are obtained from the union of three classes: hosts of supernovae detected in the PTF supernova program, RRLyrae, and additional sources whose light-curve built from PTF data show variations by 0.4 magnitude or more.

SEQUELS_ELG and SEQUELS_ELG_LOWP: These targets are part of the SEQUELS program, but are listed with the EBOSS_TARGET0 bitmask (bit 34 for SEQUELS_ELG and bit 39 for SEQUELS_ELG_LOWP). This is due to the fact that they were tiled before the main NGC SEQUELS program.

This bit represents galaxies with strong emission lines (ELG) in the redshift range 0.4 < z < 1.6 to test the target selection function of emission line galaxies for eBOSS. The target selection was applied to a combination of SCUSS u-band survey (Zhou et al., 2016) and SDSS gri photometry on a region of sky of 25.7 deg2 around αJ2000 ~ 23° and δJ2000 ~ 20°.

The brightest and bluest galaxy population is selected by: 0.5 < u – r < 0.7*(g – i) + 0.1 AND 20 < u < 22.5 (SEQUELS ELG).

We also observed a low priority ELG selection criterion (SEQUELS_ELG_LOWP) to fill the remaining fibers. It is the same criterion stretched in magnitude and color to investigate the properties of galaxies located around the selection: (20 < u < 22.7 and 0.9 < u – r) AND (u – r < 0.7(em>g – i) + 0.2 or u – r < 0.7).

SPIDERS targets within the SEQUELS program

SPIDERS (SPectroscopic IDentifcation of ERosita Sources) is an SDSS-IV program that will run in parallel to eBOSS and TDSS. The goal of SPIDERS is to obtain SDSS spectroscopy for large samples of i) X-ray selected AGN and ii) member galaxies of X-ray selected clusters. Two SPIDERS pilot programs have been executed within SEQUELS using pre-eROSITA X-ray survey data.

SPIDERS_RASS_AGN: These targets represent a pilot study exploring the feasibility of obtaining near-complete spectroscopic coverage for AGN detected in the ROSAT All Sky Survey (RASS).

A parent sample of X-ray sources is formed from the concatenation of all RASS (both Bright and Faint catalogue; Voges et al. 1999, 2000) detections lying within the SEQUELS footprint.

The RASS positional uncertainties are typically large, and so all SDSS-DR9 photometric detections within 1′ of each RASS detection were initially considered as potential counterparts. The most probable optical counterpart for each RASS source was determined using a novel Bayesian algorithm (Salvato et al. 2017), an extension of the method introduced by Budavári & Szalay (2008). The algorithm calculates the probability of each possible X-ray to optical association, taking account of the distance
between them and the positional uncertainty (dominated by the X-ray positional uncertainty). The probability of the association was then modified by priors on the expected distribution of u- and r-band magnitudes of true counterparts to X-ray bright sources relative to field stars and galaxies.

These priors are derived from a reference sample of X-ray bright serendipitous XMM-Newton sources that can be unambiguously matched to SDSS counterparts (Georgakakis & Nandra 2011). The catalogue of most probable optical counterparts for each RASS source was filtered to remove all objects that had already been observed in previous SDSS spectroscopic programs, that were associated with objects in the Véron-Cetty & Véron (2010) catalogue of known AGN associated with bright stars from the Tycho-II catalogue (Høg et al. 2000), and that lay outside of the the range 17 < r < 22. The remaining targets are submitted for tiling with target bit SPIDERS_RASS_AGN.

SPIDERS_RASS_CLUS: These targets represent follow-up spectroscopy of CODEX and XCLASS galaxy clusters.

The use of galaxy clusters for cosmology requires obtaining redshifts of spectroscopic precision (i.e., Δz <~ 10-3) in order to derive precise mass estimates – for mass function analyses – and precise locations in space – for large-scale structure studies. Cluster redshifts are obtained by collecting a number of individual spectra of their galaxy members.

Performing this task in a controlled and reproducible way therefore enables systematic identification of the most massive structures in the Universe, and removes biases and uncertainties typically associated to photometric redshifts.

The SEQUELS pilot program shares similar objectives as Spectroscopy of a Thousand Massive Galaxy Clusters namely:

  1. confirming RASS-faint sources associated to photometric red-sequences (the CODEX sample)
  2. obtaining velocity dispersions for the most massive of them, a quantity whose systematics differ from other mass proxies (e.g. lensing, X-rays, richness)

Objects of type SPIDERS RASS CLUS are selected from the RedMapper catalogue (Rykoff et al. 2014b) of
et al. 2014b) of cluster members with 17.0 < ifiber2 < 21.0 that lie in the SEQUELS footprint. A prioritization scheme penalizes lower-richness clusters and favors highly-ranked members in the photometric red-sequences. In addition to the RASS-RedMapper clusters, we have added 22 clusters selected in XMM-Newton observations by the XCLASS-RedMapper survey (Sadibekova et al. 2014; Clerc et al. 2012) with richness (i.e. number of candidate members) greater than 20. The better X-ray data allows more detailed characterization of the cluster mass once redshift is known (via e.g. derivation of intra-cluster gas temperatures). No cut on richness is made for these XMM-Newton clusters due to the high reliability of their cluster nature from X-rays.

TDSS targets within the SEQUELS program

TDSS (Time Domain Spectroscopic Survey) is a subclass within SEQUELS targeting variable objects matched between imaging in both PanSTARRS and SDSS. There are two classes of TDSS targets: single-epoch spectroscopy (SES) and few-epoch spectroscopy (FES).

TDSS_A: Comprises the main body of TDSS targets, as SES objects.

Targets are selected by seeking PS1 photometric objects from the PS1 “uberCal” DB of Sep 2013 that match objects cataloged in SDSS DR9. We restrict our search to point sources (SDSS Type = 6) with 16 < iPSF < 21 and more than 10 good detections across the PS1 griz bands. To remove sources that may display false variability because of deblending issues, we also eliminate sources with a g < 22 neighbor within 5" or an i < 12 neighbor within 30". To identify variables within this subsample, we use a 3 Dimensional Kernel Density Estimate. We train our algorithm on known variables, using the Stripe82 variable catalog from Ivezić et al. (2007) and require that the amplitude of variation in the g, r and i bands be greater than 0.1 magnitudes. Our catalog of non-variables is taken from the standards catalog from Ivezić et al. (2007). We improve the purity of this catalog by requiring that our non-variables have at least 8 SDSS observations in Stripe 82 and a constant magnitude-reduced Χ2 of less than 2 in the g, r and i bands. We require that variables, standards, and candidates have SDSS and PS1 magnitude errors of less than 0.1 mag and at least 2 PS1 detections in 3 of the 4 common PS1 and SDSS bands (g,r,i and z).

Across the 3-4 qualified bands (as described above), we use the median PS1-SDSS magnitude difference (corrected photometrically so that it is zero for a typical star), median PS1-only variability (essentially the variance minus the average error squared) and median SDSS magnitude as the three dimensions of our KDE. We bin and convolve both our variable and standard population within this space and define “efficiency” as the fraction of variables divided by the fraction of standards in every region of that space. We then use the PS1-SDSS difference, PS1 variability and median magnitude to assign an efficiency to every source in our sample.

Further cuts serve to effectively remove objects with saturated photometry, yet insure a quality spectrum. We take only objects with SDSS 17 < ifib2 < 21. The further bright-end cuts remove mainly objects with saturated photometry: rfib2 > 17 for u,g, and r. After the above candidate selection, we remove targets for which we have found existing SDSS spectra.

TDSS_FES: These target bits represent FES programs that explicitly seek repeat spectra for objects of interest in order to monitor spectroscopic variability. The TDSS FES program targets are:

  • TDSS_FES_DE: Quasar disk emitters. These targets are select quasars with i < 18.9 and broad, double-peaked, or asymmetric Balmer emission line profiles such as those in Strateva et al. (2006) (z < 0.33 for Hα and Hβ) and higher-redshift analogs from Luo et al. (2013) (z ~ 0:6 for Hβ and MgII). This program characterizes the variability of the broad emission line profiles, especially changes in asymmetry and velocity profiles, for comparison to models of accretion disk emission in the presence of asymmetries and/or perturbations.
  • TDSS_FES_DWARFC: Dwarf carbon stars (dCs). Most targets were chosen from Green (2013), a compilation of 1220 high galactic latitude C stars, selected by having strong cross-correlation peaks with SDSS carbon star spectral templates. To ensure that only dwarfs are targeted, we also require a significant (more than 3-sigma) proper motions. Repeat spectroscopy of this sample seeks to test the hypothesis that all dwarf carbon stars must be explained by past episodes of mass transfer from an AGB star, and to constrain the modes of transfer (e.g. likely wind accretion versus Roche lobe overflow from the distribution of radial velocity variations in the sample).
  • TDSS_FES_NQHISN: Normal, high S/N quasars. These objects are selected for re-observation from low redshift (z < 0.8) normal DR7 quasars with Hβ coverage and a high S/N (median S/N> 20 per pix in the restframe 4750-4950Å region. The purpose of this program is to study the general broad line variability (in particular line shape changes and line centroid shifts) of quasars on multi-year timescales.
  • TDSS_FES_MGII: MgII quasars. These targets represent a third epoch of spectroscopy from a sample of quasars with 0.36 < z < 2 in Ju et al. (2013) showing probable temporal velocity shifts in the MgII broad emission lines. These new spectra may provide direct evidence for super-massive black hole (SMBH) binaries within these quasar host galaxies.
  • TDSS_FES_VARBAL: Broad absorption line quasars. These objects are selected from the Gibson, Brandt, and Schneider (2008) BAL catalog, matched to SDSS known QSOs, with i < 19.28 (uncorrected for reddening), with 'balnicity' BI0 > 100 in one of their BAL troughs, and SN1700 > 6. Further description of this program can be found in Filiz Ak et al. (2012).

Finding Targets

An object whose EBOSS_TARGET0 value includes one or more of the bitmasks in the following table was targeted for spectroscopy as part of this ancillary target program. See SDSS bitmasks to learn how to use these values to identify objects in this ancillary target program.

Program (bit name) Bit number Target Description Number of Fibers
DO_NOT_OBSERVE 0 “Don’t put a fiber on this object” 0
LRG_IZW 1 LRG selection in i/z/W plane” 21,271
LRG_RIW 2 “LRG selection in r/i/W plane with (i-z) cut” 20,967
QSO EBOSS CORE 10 “QSOs in XDQSOz+WISE selection for clustering” 33,667
QSO_PTF 11 “QSOs with variability in PTF imaging” 22,609
QSO_REOBS 12 “QSOs from BOSS to be reobserved” 2,238
QSO_EBOSS_KDE 13 “KDE-selected QSOs (SEQUELS only)” 20,474
QSO_EBOSS_FIRST 14 “Objects with FIRST radio matches” 519
QSO_BAD_BOSS 15 “QSOs from BOSS with bad spectra” 95
QSO_BOSS_TARGET 16 “Known TARGETS from BOSS with spectra” 95
QSO_SDSS_TARGET 17 “Known TARGETS from SDSS with spectra” 1
QSO_KNOWN 18 “Known QSOs from previous surveys” 0
DR9_CALIB_TARGET 19 “Target found in DR9-calibrated imaging” 49,765
SPIDERS_RASS_AGN 20 “RASS AGN sources” 275
SPIDERS_RASS_CLUS 21 “RASS cluster sources” 2,844
TDSS_A 30 “Main PanSTARRS selection for TDSS” 17,394
TDSS_FES_DE 31 “TDSS Few epoch spectroscopy” 70
TDSS_FES_DWARFC 32 “TDSS Few epoch spectroscopy” 29
TDSS_FES_NQHISN 33 “TDSS Few epoch spectroscopy” 148
TDSS_FES_MGII 34 “TDSS Few epoch spectroscopy” 2
TDSS_FES_VARBAL 35 “TDSS Few epoch spectroscopy” 103
SEQUELS_PTF_VARIABLE 40 “Variability objects from PTF” 1,153
SEQUELS_COLLIDED 41 “Collided galaxies from BOSS” 0

REFERENCES

Becker, R. H., White, R. L., & Helfand, D. J. 1995, ApJ, 450, 559

Bolton, A. S., et al. 2012, AJ, 144, 144

Budavári, T., & Szalay, A. S. 2008, ApJ, 679, 301

Clerc, N., et al. 2012, MNRAS, 423, 3561

Filiz Ak, N., et al. 2012, ApJ, 757, 114

Georgakakis, A., & Nandra, K. 2011, MNRAS, 414, 992

Gibson, R. R., Brandt, W. N., & Schneider, D. P. 2008, ApJ, 685, 773

Green, P. 2013, ApJ, 765, 12

Høg, E., et al. 2000, A&A, 355, L27

Ivezić, Ž., et al. 2007, AJ, 134, 973

Ju, W., Greene, J. E., Rafikov, R. R., Bickerton, S. J., & Badenes, C. 2013, ApJ, 777, 44

Luo, B., et al. 2013, MNRAS, 429, 1479

Myers, A.D., et al., 2015, ApJS, 221, 27

Paris, I., et al., 2017, A&A, 597, 79

Richards, G. T., et al. 2009, ApJS, 180, 67

Ross, N. P., et al. 2012, ApJS, 199, 3

Rykoff, E. S., et al., 2014, ApJ, 785, 104

Sadibekova, T., et al. 2014, A&A, 571, A87

Salvato M., et al., 2017, submitted

Strateva, I. V., Brandt, W. N., Eracleous, M., Schneider, D. P., & Chartas, G. 2006, ApJ, 651, 749

Véron-Cetty, M.-P. & Véron, P., 2010, A&A, 518, A10

Voges, W., et al. 1999, A&A, 349, 389

Voges,.W., et al. 2000, VizieR Online Data Catalog, 9029, 0

Zhou, X., et al., 2016, RAA, 16, 69