PaperlayerView exact revision on arXiv

Source-linked AI summary

The First Hyper-Luminous Infrared Galaxy Discovered by WISE

Peter R. M. Eisenhardt, Jingwen Wu, Chao-Wei Tsai, Roberto Assef, Dominic Benford, Andrew Blain, Carrie Bridge, J. J. Condon, Michael C. Cushing, Roc Cutri, Neal J. Evans II, Chris Gelino, Roger L. Griffith, Carl J. Grillmair, Tom Jarrett, Carol J. Lonsdale, Frank J. Masci, Brian S. Mason, Sara Petty, Jack Sayers, S. Adam Stanford, Daniel Stern, Edward L. Wright, Lin Yan

arXiv:1208.5517v1astro-ph.CO

TL;DR

WISE 1814+3412 is an exceptionally luminous dusty galaxy whose energy sources and relation to star formation are uncertain. The authors combine WISE selection with optical, infrared, millimeter, and radio follow-up to characterize its components and emission. The LBG’s measured star formation accounts for less than 10% of the bolometric luminosity, while the inferred Eddington ratio is at least about 4 under the stated black-hole-mass assumption.

  • Problem

    The authors investigate the energy sources of a WISE-selected, exceptionally luminous dusty galaxy whose observed emission must be interpreted across multiple components and wavelengths.

    Source: §1 · p. 2, §2 · p. 3

  • Method

    The study selects W1W2-dropouts and combines imaging and spectroscopy with near-infrared, millimeter, and radio observations to characterize WISE 1814+3412.

    Source: §2 · p. 3, §2.1.2 · p. 5, §2.1.5 · p. 7

  • Results

    The LBG’s extinction-corrected UV star formation rate is ∼300 M⊙yr−1 and accounts for < 10% of the bolometric luminosity; assuming the adjusted local relation applies, the implied Eddington ratio is ∼> 4.

    Source: §4 · pp. 15–16

  • Takeaways & Limitations

    The authors consider WISE 1814+3412 likely to lie well off the local black-hole–bulge mass relation, possibly before building most of its stellar mass.

    Source: §4 · p. 16

    The inferred AGN-powered black-hole mass and dust scale depend on the assumed AGN luminosity and torus model; larger-scale dust would be needed to obscure a kiloparsec-scale narrow-line region.

    Source: §3.7 · p. 15

Abstract

from arXiv · show

We report the discovery by the Wide-field Infrared Survey Explorer of the z = 2.452 source WISE J181417.29+341224.9, the first hyperluminous source found in the WISE survey. WISE 1814+3412 is also the prototype for an all-sky sample of ~1000 extremely luminous "W1W2-dropouts" (sources faint or undetected by WISE at 3.4 and 4.6 microns and well detected at 12 or 22 microns). The WISE data and a 350 micron detection give a minimum bolometric luminosity of 3.7 x 10^13 Lsun, with ~10^14 Lsun plausible. Followup images reveal four nearby sources: a QSO and two Lyman Break Galaxies (LBGs) at z = 2.45, and an M dwarf star. The brighter LBG dominates the bolometric emission. Gravitational lensing is unlikely given the source locations and their different spectra and colors. The dominant LBG spectrum indicates a star formation rate ~300 Msun/yr, accounting for < 10% of the bolometric luminosity. Strong 22 micron emission relative to 350 microns implies that warm dust contributes significantly to the luminosity, while cooler dust normally associated with starbursts is constrained by an upper limit at 1.1 mm. Radio emission is ~10x above the far-infrared/radio correlation, indicating an active galactic nucleus is present. An obscured AGN combined with starburst and evolved stellar components can account for the observations. If the black hole mass follows the local M_BH-bulge mass relation, the implied Eddington ratio is >~4. WISE 1814+3412 may be a heavily obscured object where the peak AGN activity occurred prior to the peak era of star formation.

1. Introduction

WISE was designed to find both nearby brown dwarfs and the most luminous dusty forming galaxies, whose infrared emission can trace dust-obscured galaxy and black-hole formation. The paper reports WISE 1814+3412 as WISE’s first discovered hyper-luminous infrared galaxy.

1. Introduction

  • WISE began sky surveys in January 2010 and completed its first full coverage in July 2010.It later completed two full coverages in W1 and W2 before the all-sky data release in March 2012.Source: §1 · p. 2
  • WISE’s primary science objectives include finding the coldest nearby brown dwarfs and the most luminous dusty forming galaxies.The paper describes the latter as ULIRGs, with dust absorbing UV/optical luminosity and reradiating it in the infrared.Source: §1 · p. 2
  • WISE 1814+3412 is the first hyper-luminous infrared galaxy discovered by WISE.The authors note that such extreme objects may be missed by substantially smaller-area surveys.Source: §1 · pp. 2–3

2. Selection Criteria and Followup Observations

The study selects extremely red WISE sources that are faint in W1 and W2 but bright in W3 or W4, then follows up WISE 1814+3412 as a particularly bright example. The selection also applies exposure, sky-position, artifact, asteroid, and visual-inspection filters.

2. Selection Criteria and Followup Observations

  • W1W2-dropouts are selected by faint W1 and either bright W4 with red W2−W4 color or bright W3 with red W2−W3 color.The criteria also require adequate W3/W4 exposures, separation from the Galactic center and plane, no flagged artifacts or known asteroid association, and visual checks.Source: §2 · p. 3
  • WISE 1814+3412 was the brightest W4 source among 37 objects meeting the selection criteria at the time.It has W1 > 18.67 (< 11 µJy) and a marginal W2 detection of 23 ± 11 µJy.Source: §2 · p. 3

2.1. Images at Other Wavelengths

Multiwavelength imaging and radio observations resolve nearby components and characterize WISE 1814+3412 from optical through millimeter and radio wavelengths. The data include detections at 350 µm and 4.494 GHz, alongside upper limits at other far-infrared and millimeter wavelengths.

2.1.1. Optical Imaging

  • The optical images show components A, B, C, and D near WISE 1814+3412, with A extended and B and C unresolved.Optical photometry for A used an aperture after PSF subtraction of B and C, while B was measured by PSF fitting.Source: §2.1.1 · p. 4

2.1.2. Near-IR Imaging

  • The NIRC2 images detect A, B, C, and D in Ks, while A is faint and D undetected in J; A is extended relative to B and C.WIRC images provided the calibration link between the deeper NIRC2 images and 2MASS.Source: §2.1.2 · p. 5

2.1.3. Spitzer

  • Component A becomes more prominent at longer Spitzer wavelengths, and its [3.6] detection indicates that the WISE W1 2σ limit underestimates its flux.The [4.5] detection is consistent with the WISE W2 measurement.Source: §2.1.3 · pp. 5–6

2.1.4. Far-Infrared Data

  • WISE 1814+3412 is undetected by IRAS, with upper limits of < 40 mJy at 12 and 25 µm, < 100 mJy at 60 µm, and < 600 mJy at 100 µm.These limits are consistent with the WISE W3 and W4 detections.Source: §2.1.4 · p. 6

2.1.5. Caltech Submillimeter Observatory Data

  • The source is detected at 350 µm with 33 ± 9 mJy, but not significantly at 450 µm; the 450 µm 95% confidence upper limit is 35 mJy.At 1.1 mm, the reported 95% confidence upper limit is 2.4 mJy.Source: §2.1.5 · pp. 6–7

2.1.6. Radio

  • The radio observations used EVLA C-band measurements to estimate a spectral index, while GBT Ka-band observations yielded a 2σ upper limit of 175 µJy.The EVLA observations were made simultaneously at 4.494 and 7.93 GHz; the GBT limit covers 26–40 GHz.Source: §2.1.6 · pp. 7–8
  • At 4.494 GHz, an unresolved source coincident with WISE 1814+3412 has flux density 560 ± 70 µJy; its 7.93 GHz measurement is 350 ± 130 µJy.The observations also detected a second unresolved radio source to the southeast; the summed extrapolated 1.4 GHz flux agrees with the NVSS measurement of 3.4 ± 0.5 mJy.Source: §2.1.6 · pp. 7–8

2.2. Optical Spectroscopy

Keck LRIS spectroscopy established WISE 1814+3412’s redshift and examined the spectra of neighboring components. The spectra identify two components as Lyman-break galaxies at z = 2.452, another as a same-redshift quasar, and the remaining component as an M dwarf.

2.2. Optical Spectroscopy

  • Component A has a typical Lyman-break galaxy spectrum at z = 2.452, with Lyα emission, a continuum break, and UV interstellar absorption lines.Its UV spectrum shows no high-ionization, high-equivalent-width emission lines typical of AGN activity; Lyα emission extends more than 30 kpc southeast, making it a Lyα blob.Source: §2.2 · pp. 8–9
  • Component B’s red optical spectrum and colors support its identification as an early-type M-dwarf star.The spectra were obtained with Keck LRIS using multiple slit positions to observe the neighboring components.Source: §2.2 · pp. 8–9
  • Component C is a broad-lined, unobscured Type 1 quasar at the same redshift as WISE 1814+3412.Self-absorption is evident in the Lyα, Si IV, and C IV transitions.Source: §2.2 · p. 9
  • Component D’s continuum break and interstellar absorption lines indicate that it is at the same redshift as WISE 1814+3412.Its continuum is relatively flat in Fν, with a broad absorption and break near the Lyα wavelength at z ∼ 2.45.Source: §2.2 · p. 9

2.3. Astrometry

Astrometry associates the longer-wavelength emission primarily with component A at z = 2.452, while its optical centroid is offset from its Ks and Spitzer positions. The WISE source is unresolved, and the positional shift is considered likely to reflect extinction.

2.3. Astrometry

  • The coordinate measurements assume a minimum 0.2-arcsec uncertainty to account for systematic reference-frame errors.Source: §2.3 · p. 9
  • The WISE source is unresolved, and the longer-wavelength detections align with component A within positional uncertainties.The W3 and W4 profile fits have reduced chi-squared values of 0.98 and 1.09, respectively.Source: §2.3 · p. 9
  • Component A at z = 2.452 is judged to dominate the flux from the single component detected at longer wavelengths.Source: §2.3 · p. 10
  • Component A’s optical centroid lies about 0.5 arcsec east of its Ks and Spitzer positions, a shift considered real and likely related to appreciable extinction.The offset corresponds to 4 kpc at z = 2.452.Source: §2.3 · p. 10

3. Analysis

WISE 1814+3412 has unusually extreme W2-to-W3 colors, and its observed photometry points to warm dust, an AGN, and active star formation. The authors fit the photometry with stellar and reddened AGN templates alongside simple dust-emission components.

3. Analysis

  • Compared at z = 2.452 and normalized at 22 µm, WISE 1814+3412 is more extreme in W2-to-W3 color than the other unusually red objects shown.The passage notes that W1W2-dropouts satisfy the DOG color criterion and that AGN may play an important role in them.Source: §3 · p. 10

3.1. SED Modeling

  • The W3, W4, and 350 µm detections, together with the 450 µm and 1.1 mm upper limits, indicate warm dust dominates the luminosity; radio data show an AGN is present.The LBG-like optical spectrum also implies active star formation.Source: §3.1 · p. 10
  • The example SED fit combines starburst and evolved stellar templates with a reddened Type 1 AGN template, alongside 490 K and 45 K blackbody fits to mid-IR and submillimeter data.The AGN template fit gives rest-frame E(B − V) = 15.6 ± 1.4, or A_V = 48 ± 4.Source: §3.1 · p. 11

3.2. Luminosity Estimates

WISE 1814+3412 has a bolometric luminosity above 3.7 × 10^13 L⊙, with a template-based estimate near 9.2 × 10^13 L⊙. The authors consider significant gravitational lensing unlikely and conclude that the large luminosity appears intrinsic.

3.2. Luminosity Estimates

  • L_bol > 3.7 × 10^13 L⊙ is the lower limit, while integration of the best-fit SED gives 9.2 ± 1.0 × 10^13 L⊙.The template estimate assigns 7.2 ± 0.8 × 10^11 L⊙, less than 1%, to stellar components; the luminosity is unlikely to be twice as high.Source: §3.2 · p. 11

3.2.1. Lensing

  • Significant gravitational lensing is considered unlikely because the same-redshift components A, C, and D have different SEDs and spectra.Source: §3.2.1 · p. 11
  • Component B appears to be a Galactic M dwarf, and extended Ks emission coincides with z = 2.452 Lyα emission, supporting the conclusion that the large luminosity is intrinsic.Source: §3.2.1 · p. 12

3.3. Starburst Component

The UV spectrum indicates active star formation, with extinction-corrected rates estimated at hundreds of solar masses per year. The stellar light also requires an evolved population, and the inferred stellar mass depends on the assumed population model.

3.3. Starburst Component

  • Without correcting for UV extinction, the inferred star-formation rate is about 50 M⊙ yr−1 for a Salpeter IMF or 30 M⊙ yr−1 for a Chabrier IMF.These estimates use the rest-frame 1500–2800 Å starburst-component luminosity.Source: §3.3 · p. 12
  • Extinction-corrected star-formation rates range from about 300 to 600 M⊙ yr−1 for a Salpeter IMF and 180 to 350 M⊙ yr−1 for a Chabrier IMF.The color-based extinction estimates span 2.00–2.75 mag at 0.16 µm; higher rates are possible if star formation is completely obscured in the UV.Source: §3.3 · p. 12

3.4. Stellar Component

  • The starburst template falls below the observed Ks, [3.6], and [4.5] fluxes, requiring a significant contribution from older stars.Source: §3.4 · p. 12
  • Allowing 0.34 mag of extinction at [4.5], the upper end inferred from observed-versus-model g′ − r′ colors, gives an upper stellar-mass limit of 5.8 × 10^11 M⊙, many times a present-day field L* galaxy’s ∼5 × 10^10 M⊙ stellar mass.Source: §3.4 · p. 13
  • A plausible model with equal [4.5] contributions from starburst and evolved stars yields a stellar mass of 3.6 × 10^11 M⊙ for a Salpeter IMF or 2.0 × 10^11 M⊙ for a Chabrier IMF.The lower-bound model, assigning all [4.5] light to a starburst, gives 4.1 × 10^10 M⊙.Source: §3.4 · p. 13

3.5. Dust Component

The infrared colors indicate substantial warm dust, while the 1.1 mm limit constrains cooler dust. A cooler component associated with the observed star formation could still contribute substantially to the 350 µm flux.

3.5. Dust Component

  • The W3-to-W4 and W4-to-350 µm flux ratios imply warm dust at approximately 490 K and 170 K, respectively.The estimated warm-dust radius is 30–250 pc, with a dust mass of at least 10^7 M⊙.Source: §3.5 · p. 13
  • A 350 µm-to-1.1 mm flux ratio greater than 13.75 and the 1.1 mm upper limit constrain dust cooler than 45 K.Cool dust at those temperatures, normalized to the observed 350 µm flux, would exceed the 2.4 mJy 95% confidence limit.Source: §3.5 · p. 13
  • A roughly 35 K dust component associated with the approximately 300 M⊙ yr−1 star formation rate could contribute over half the 350 µm flux while satisfying the 1.1 mm limit.This component would have an estimated dust mass of approximately 2.5 × 10^8 M⊙.Source: §3.5 · p. 13

3.6. Radio Source

Radio measurements identify WISE 1814+3412 as radio-loud and indicate that its radio emission is powered by an AGN rather than star formation.

3.6. Radio Source

  • The rest-frame 5 GHz-to-4400 Å flux-density ratio is approximately 200, qualifying WISE 1814+3412 as radio-loud.The 1.4 GHz rest-frame luminosity is approximately 5 × 10^25 W Hz−1.Source: §3.6 · p. 14
  • The 175 µJy Ka-band upper limit is consistent with the 115 µJy expected by extrapolating the EVLA measurements, arguing against star formation dominating the energetics.Free-free emission at the level seen in M82 would have exceeded the upper limit.Source: §3.6 · p. 14
  • The far-infrared/radio parameter q is 1.36, about 10× more radio-loud than the normal value of 2.4 ± 0.24.The authors conclude that the radio emission is due to AGN activity rather than star formation.Source: §3.6 · p. 14

3.7. AGN Component

The SED fit supports a heavily obscured AGN, although the optical spectrum lacks AGN features. The inferred dust geometry may extend beyond the torus if it obscures a kpc-scale narrow-line region.

3.7. AGN Component

  • The SED fit assigns over 99% of the bolometric luminosity to a reddened Type 1 AGN component, despite no AGN features in component A’s UV spectrum.The fitted extinction is AV = 48 mag, corresponding to approximately 200 mag in the rest UV.Source: §3.7 · p. 14
  • The absence of typical high-ionization narrow emission lines suggests either a more compact narrow-line region or obscuration on scales larger than the torus.The approximately 4 kpc shift between rest-UV and optical continuum supports the larger-scale obscuration possibility.Source: §3.7 · p. 14
  • If accretion onto a supermassive black hole supplies most of the luminosity, the dust torus is expected to extend no more than 30 times the approximately 7 pc sublimation radius.This scale is consistent with the discussed warm-dust size range, but larger-scale dust would be needed to obscure a kpc-scale narrow-line region.Source: §3.7 · p. 15

4. Summary and Discussion

WISE 1814+3412 is the first hyper-luminous infrared galaxy identified by WISE, with star formation unable to account for most of its luminosity under the reported estimates. The authors favor an obscured, dominant AGN and discuss a potentially unusual evolutionary stage.

4. Summary and Discussion

  • WISE 1814+3412 has a minimum bolometric luminosity of 3.7 × 10^13 L⊙, with a more likely estimate of approximately 9 × 10^13 L⊙, making it the first hyper-luminous infrared galaxy identified by WISE.Its luminosity approaches that of the most luminous known quasars, while some W1W2-dropouts reported by Wu et al. (2012) are more luminous.Source: §4 · p. 15
  • The LBG has an extinction-corrected UV star formation rate of approximately 300 M⊙ yr−1, but even the upper estimate accounts for less than 10% of the bolometric luminosity.A cooler star-forming dust component would need to be significantly warmer than 35 K to avoid exceeding the 1.1 mm upper limit.Source: §4 · p. 15
  • The modest radio source indicates an AGN, and the lack of submillimeter emission with predominant hot dust is consistent with an obscured AGN supplying the luminosity.The authors note that this SED may resemble the intrinsic dusty central AGN region more than that of many quasars.Source: §4 · pp. 15–16
  • The fitted AGN luminosity implies an Eddington ratio of at least approximately 4 if the adjusted local black-hole-to-bulge relation applies.The authors note that if the source instead has a typical quasar Eddington ratio, at least one of the assumptions does not hold.Source: §4 · p. 16
  • Alternative explanations include star formation with warm dust scaled up from low-metallicity starburst dwarfs, underestimated stellar mass, or super-Eddington accretion.The authors note that scaling the starburst scenario by a factor of a thousand would be required.Source: §4 · p. 16
  • The authors favor an early evolutionary stage in which AGN activity dominates before the object has built most of its stellar mass.This proposed timing would contradict the expectation that AGN activity lags the peak of star formation.Source: §4 · p. 16
  • Three systems at the same redshift within 50 kpc suggest that significant stellar-mass assembly lies in WISE 1814+3412’s future.Source: §4 · p. 17
Loading 1208.5517v1…