G. Wilson

The Infrared Array Camera (IRAC) for the Spitzer Space Telescope

The Infrared Array Camera (IRAC) is one of three focal plane instruments on the Spitzer Space Telescope. IRAC is a four-channel camera that obtains simultaneous broadband images at 3.6, 4.5, 5.8, and 8.0 � m. Two nearly adjacent 5A2 ; 5A2 fields of view in the focal plane are viewed by the four channels in pairs (3.6 and 5.8 � m; 4.5 and 8 � m). All four detector arrays in the camera are 256 ; 256 pixels in size, with the two shorter wavelength channels using InSb and the two longer wavelength channels using Si:As IBC detectors. IRAC is a powerful survey instrument because of its high sensitivity, large field of view, and four-color imaging. This paper summarizes the in-flight scientific, technical, and operational performance of IRAC.

Obscured and unobscured active galactic nuclei in the Spitzer Space Telescope First Look Survey

Selection of active galactic nuclei (AGNs) in the infrared facilitates the discovery of AGNs whose optical emission is extinguished by dust. In this paper, we use the Spitzer Space Telescope First Look Survey (FLS) to assess the fraction of AGNs with mid-infrared (MIR) luminosities that are comparable to quasars and that are missed in optical quasar surveys because of dust obscuration. We begin by using the Sloan Digital Sky Survey (SDSS) database to identify 54 quasars within the 4 deg^2 extragalactic FLS. These quasars occupy a distinct region in MIR color space by virtue of their strong, red continua. This has allowed us to define an MIR color criterion for selecting AGN candidates. About 2000 FLS objects have colors that are consistent with them being AGNs, but most are much fainter in the MIR than the SDSS quasars, which typically have 8 μm flux densities S_(8.0) ~ 1 mJy. We have investigated the properties of 43 objects with S_(8.0) ≥ 1 mJy that satisfy our AGN color selection. This sample should contain both unobscured quasars as well as AGNs that are absent from the SDSS survey because of extinction in the optical. After removing 16 known quasars, three probable normal quasars, and eight spurious or confused objects from the initial sample of 43, we are left with 16 objects that are likely to be obscured quasars or luminous Seyfert 2 galaxies. This suggests that the numbers of obscured and unobscured AGNs are similar in samples selected in the MIR at S_(8.0) ~ 1 mJy.

The All-wavelength Extended Groth Strip International Survey (AEGIS) Data Sets

In this the first of a series of Letters, we present a panchromatic data set in the Extended Groth Strip region of the sky. Our survey, the All-Wavelength Extended Groth Strip International Survey (AEGIS), aims to study the physical properties and evolutionary processes of galaxies at z ~ 1. It includes the following deep, wide-field imaging data sets: Chandra/ACIS X-ray, GALEX ultraviolet, CFHT/MegaCam Legacy Survey optical, CFHT/CFH12K optical, Hubble Space Telescope/ACS optical and NICMOS near-infrared, Palomar/WIRC near-infrared, Spitzer/IRAC mid-infrared, Spitzer/MIPS far-infrared, and VLA radio continuum. In addition, this region of the sky has been targeted for extensive spectroscopy using the Deep Imaging Multi-Object Spectrograph (DEIMOS) on the Keck II 10 m telescope. Our survey is compared to other large multiwavelength surveys in terms of depth and sky coverage.

The Far- and Mid-Infrared/Radio Correlations in the Spitzer Extragalactic First Look Survey

Using the Spitzer Space Telescope and the Very Large Array (VLA), we present the first direct evidence that the well-known far-infrared/radio correlation is valid to cosmologically significant redshift. We also confirm, with improved statistics compared with previous surveys, a similar result for the mid-IR/radio correlation. We explore the dependence of monochromatic q_(24) and q_(70) on z. The results were obtained by matching Spitzer sources at 24 and 70 μm with VLA 1.4 GHz microjansky radio sources obtained for the Spitzer First Look Survey (FLS). Spectroscopic redshifts have been obtained for over 500 matched IR/radio sources using observations at WIYN and Keck, and archival Sloan Digital Sky Survey (SDSS) data extending out to z > 2. We find that q_(24) shows significantly more dispersion than q_(70). By comparing the observed fluxes at 70, 24, and 4.5 μm with a library of SED templates, we find that the larger dispersion in q_(24) is predictable in terms of systematic variations in spectral energy distribution (SED) shape throughout the population. Although the models are not able to encompass the full range of observed behavior (both the presence of either extremely flat or extremely steep IR SEDs), the fitting parameters were used to k-correct the higher z galaxies, which resulted in a reduced scatter in q. For comparison, we also corrected these data using the SED for M82. The results for 24 and 70 μm provide strong consistent evidence for the universality of the mid- and far-IR/radio correlations out to redshifts of at least z = 1.

Spitzer 70 and 160 μm Observations of the Extragalactic First Look Survey

We present 70 and 160 μm observations from the Spitzer extragalactic First Look Survey (xFLS). The data reduction techniques and the methods for producing co-added mosaics and source catalogs are discussed. Currently, 26% of the 70 μm sample and 49% of the 160 μm–selected sources have redshifts. The majority of sources with redshifts are star-forming galaxies at z < 0.5, while about 5% have infrared colors consistent with active galactic nuclei. The observed infrared colors agree with the spectral energy distributions (SEDs) of local galaxies previously determined from IRAS and Infrared Space Observatory data. The average 160 μm/70 μm color temperature for the dust is T_d ≃ 30 ± 5 K, and the average 70 μm/24 μm spectral index is α ≃ 2.4 ± 0.4. The observed infrared-to-radio correlation varies with redshift as expected out to z ~ 1 based on the SEDs of local galaxies. The xFLS number counts at 70 and 160 μm are consistent within uncertainties with the models of galaxy evolution, but there are indications that the current models may require slight modifications. Deeper 70 μm observations are needed to constrain the models, and redshifts for the faint sources are required to measure the evolution of the infrared luminosity function.

Spitzer Observations of MAMBO Galaxies: Weeding Out Active Nuclei in Starbursting Protoellipticals

We present 3.6-24 μm Spitzer observations of an unbiased sample of nine luminous, dusty galaxies selected at 1200 μm by MAMBO on the IRAM 30 m telescope, a population akin to the well-known submillimeter or SCUBA galaxies (hereafter SMGs). Owing to the coarse resolution of submillimeter/millimeter cameras, SMGs have traditionally been difficult to identify at other wavelengths. We compare our multiwavelength catalogs to show that the overlap between 24 and 1200 μm must be close to complete at these flux levels. We find that all (4/4) of the most secure ≥4 σ SMGs have ≥4 σ counterparts at 1.4 GHz, while the fraction drops to 7/9 using all ≥3 σ SMGs. We show that combining mid-infrared (MIR) and marginal (≥3 σ) radio detections provides plausible identifications in the remaining cases, enabling us to identify the complete sample. Accretion onto an obscured central engine is betrayed by the shape of the MIR continuum emission for several sources, confirming Spitzers potential to weed out active galaxies. We demonstrate the power of an S24 μm/S8 μm versus S8 μm/S4.5 μm color-color plot as a diagnostic for this purpose. However, we conclude that the majority (~75%) of SMGs have rest-frame mid/far-IR spectral energy distributions commensurate with obscured starbursts. Sensitive 24 μm observations are clearly a useful route to identify and characterize reliable counterparts to high-redshift far-IR-bright galaxies, complementing what is possible via deep radio imaging.

Spitzer Observations of the SCUBA/VLA Sources in the Lockman Hole: Star Formation History of Infrared-Luminous Galaxies

We present Spitzer Space Telescope imaging observations at 3.6, 4.5, 5.8, 8.0, and 24 ?m of the SCUBA submillimeter sources and ?Jy VLA radio sources in a 5 ? 5 area in the Lockman Hole East region. Out of the ~40 SCUBA/VLA sources in the field, Spitzer counterparts were detected for nearly all except for the few low-weight SCUBA detections. We show that the majority (80%-90%) of the detected sources are cold (i.e., starburst-like) infrared-luminous galaxies (LIR > 1011 L?) at redshift 0.5 < z < 3.5, whose star-formation rate density (SFRD) is comparable to that of the optically selected star-forming galaxies.

The Spitzer Space Telescope Extragalactic First Look Survey: 24 μm Data Reduction, Catalog, and Source Identification

We present the reduction of the 24 μm data obtained during the first cosmological survey performed by the Spitzer Space Telescope. Images of a region of sky at moderately high Galactic latitude (l = 88.3°, b = +34.9°) were obtained on 2003 December 9-11. The survey consists of a shallow observation of 2.5° × 2° centered at 17h18m, +59°30 (main survey) and a deeper observation of 1° × 0.5° centered at 17h17m, +59°45 (verification survey). Issues with the reduction of the 24 μm MIPS data are discussed and solutions to attenuate instrumental effects are proposed and applied to the data. Approximately 17,000 sources are extracted with a signal-to-noise ratio (S/N) greater than 5. The photometry of the point sources is evaluated through point-spread function (PSF) fitting using an empirical PSF derived from the data. Aperture corrections and the absolute calibration have been checked using stars in the field. Astrometric and photometric errors depend on the S/N of the source varying between 0farcs35^-1 and 5%-15%, respectively, for sources detected at 20-5 σ. The fluxes of the 123 extended sources have been estimated through aperture photometry. The extended sources cover less than 0.3% of the total area of the survey. Based on simulations, the main and verification surveys are 50% complete at 0.3 and 0.15 mJy, respectively. Counterparts have been searched for in optical and radio catalogs. More than 80% of the 24 μm sources have a reliable optical counterpart down to R = 25.5; 16% of the sources have a 20 cm counterpart down to 0.1 mJy and ~80% of the radio-IR associations have a reliable optical counterpart. A residual map is obtained by subtracting point sources detected at the 3 σ level and interpolating the regions occupied by extended sources. Several Galactic clouds with low and intermediate velocities are identified by comparison with neutral hydrogen data from this field.

Star Formation Rates and Extinction Properties of IR-luminous Galaxies in the Spitzer First Look Survey

We investigate the instantaneous star formation rates (SFRs) and extinction properties for a large (N = 274), near-infrared (NIR: 2.2 μm) + mid-infrared (MIR: 24 μm)-selected sample of normal to ultraluminous infrared galaxies (ULIRGs; 10^9 < L_(IR)/L_☉ < 10^(12.5)) with〈 z〉~ 0.8 in the Spitzer Extragalactic First Look Survey (FLS). We combine 24 μm observations with high-resolution Keck DEIMOS spectroscopy to derive optical emission-line (Hα, Hβ, and [O II]) and infrared star formation rates (SFR_(opt) and SFR_(IR), respectively). Comparison of SFR diagnostics reveals a wide extinction range (1.0 < A_V < 4.0 mag) for this sample, even after removing spectroscopic and IRAC color-selected AGN candidates (≈12% of the sample). Objects with SFRs of a few M_☉ yr^(-1) have extinction values consistent with normal spirals (A_V ≈ 1.0 mag). By contrast, LIRGs at z ≳ 1, which comprise a fraction of our sample, have SFR ≈ 100 M_☉ yr^(-1) and a mean A_V ≈ 2.5 mag. This translates to a 97% mean [O II] λλ3727 attentuation and in extreme cases is as high as 99.7%. We derive an IR-luminosity-dependent A^(IR)_V function [A^(IR)_V = 0.75 log(L_(IR)/L_☉) - 6.35 mag] that we use to extinction correct our line luminosities. The resulting correlation between SFR_(IR) and SFR_(opt) has a dispersion of ~0.2 dex (semi-interquartile range). Comparison of the A_V dependence on redshift and LIR reveals that for a fixed L_(IR), there is no significant AV evolution. Comparison to previous studies reveals a mean attenuation that is intermediate between that of local optical/UV- and radio-selected samples with a marginally stronger L_(IR) dependence.

AEGIS20: A radio survey of the Extended Groth Strip

We describe AEGIS20-a radio survey of the Extended Groth Strip (EGS) conducted with the Very Large Array (VLA) at 1.4 GHz. The resulting catalog contains 1123 emitters and is sensitive to ultraluminous (10^12 L⊙) starbursts to z ≤ 1.3, well matched to the redshift range of the DEEP2 spectroscopic survey in this region. We use stacking techniques to explore the microjansky-level emission from a variety of galaxy populations selected via conventional criteria-Lyman break galaxies (LBGs), distant red galaxies (DRGs), UV-selected galaxies, and extremely red objects (EROs)-determining their properties as a function of color, magnitude, and redshift and their extinction-free contributions to the history of star formation. We confirm the familiar pattern that the star formation rate (SFR) density, ρ_*, increases by at least a factor of ~5 from z = 0 to 1, although we note highly discrepant UV- and radio-based SFR estimates. Our radio-based SFRs become more difficult to interpret at z > 1 where correcting for contamination by radio-loud active galactic nuclei (AGNs) comes at the price of rejecting luminous starbursts. While stacking radio images is a useful technique, accurate radio-based SFRs for z » 1 galaxies require precise redshifts and extraordinarily high fidelity radio data to identify and remove accretion-related emission.