Searcharxiv⌕ Search

arXiv subjects

G. H. Rieke

Publications and source records attributed to G. H. Rieke.

At least 127 records · Page 7Linked to original sources
arXiv2006

Why X-ray--Selected AGN Appear Optically Dull

We investigate why half of X-ray--selected active galactic nuclei (AGN) in deep surveys lack signs of accretion in their optical spectra. The majority of these ``optically--dull'' AGN are no more than 6 times fainter than their host galaxies in rest-frame R-band; as such, AGN lines are unlikely to be overwhelmed by stellar continuum in at least half the sample. We find that optically--dull AGN have the mid--infrared emission and L(x)/L(IR) ratios characteristic of local Seyferts, suggesting that the cause of optical dullness is not missing UV--optical continua. We compare the morphologies of 22 optically--dull and 9 optically--active AGN at 0.5<z<0.8, and find that optically--dull AGN show a wide range of axis ratio, but optically--active AGN have only very round axis ratios. We conclude that hard X-rays select AGN in host galaxies with a wide range of inclination angle, but only those AGN in the most face-on or spheroidal host galaxies show optical emission lines. Thus, extranuclear dust in the host galaxy plays an important role in hiding the emission lines of optically--dull AGN.

astro-ph↗
arXiv2006

The Bolocam 1.1 mm Lockman Hole Galaxy Survey: SHARC II 350 micron Photometry and Implications for Spectral Models, Dust Temperatures, and Redshift Estimation

We present 350 micron photometry of all 17 galaxy candidates in the Lockman Hole detected in a 1.1 mm Bolocam survey. Several of the galaxies were previously detected at 850 microns, at 1.2 mm, in the infrared by Spitzer, and in the radio. Nine of the Bolocam galaxy candidates were detected at 350 microns and two new candidates were serendipitously detected at 350 microns (bringing the total in the literature detected in this way to three). Five of the galaxies have published spectroscopic redshifts, enabling investigation of the implied temperature ranges and a comparison of photometric redshift techniques. Lambda = 350 microns lies near the spectral energy distribution peak for z = 2.5 thermally emitting galaxies. Thus, luminosities can be measured without extrapolating to the peak from detection wavelengths of lambda > 850 microns. Characteristically, the galaxy luminosities lie in the range 1.0 - 1.2 x 10^13 L_solar, with dust temperatures in the range of 40 K to 70 K, depending on the choice of spectral index and wavelength of unit optical depth. The implied dust masses are 3 - 5 x 10^8 M_solar. We find that the far-infrared to radio relation for star-forming ULIRGs systematically overpredicts the radio luminosities and overestimates redshifts on the order of Delta z ~ 1, whereas redshifts based on either on submillimeter data alone or the 1.6 micron stellar bump and PAH features are more accurate.

astro-ph↗
arXiv2006

The Albedo, Size, and Density of Binary Kuiper Belt Object (47171) 1999 TC36

We measured the system-integrated thermal emission of the binary Kuiper Belt Object 1999 TC36 at wavelengths near 24 and 70 microns using the Spitzer space telescope. We fit these data and the visual magnitude using both the Standard Thermal Model and thermophysical models. We find that the effective diameter of the binary is 405 km, with a range of 350 -- 470 km, and the effective visible geometric albedo for the system is 0.079 with a range of 0.055 -- 0.11. The binary orbit, magnitude contrast between the components, and system mass have been determined from HST data (Margot et al., 2004; 2005a; 2005b). Our effective diameter, combined with that system mass, indicate an average density for the objects of 0.5 g/cm3, with a range 0.3 -- 0.8 g/cm3. This density is low compared to that of materials expected to be abundant in solid bodies in the trans-Neptunian region, requiring 50 -- 75% of the interior of 1999 TC36 be taken up by void space. This conclusion is not greatly affected if 1999 TC36 is ``differentiated'' (in the sense of having either a rocky or just a non-porous core). If the primary is itself a binary, the average density of that (hypothetical) triple system would be in the range 0.4 -- 1.1 g/cm3, with a porosity in the range 15 -- 70%.

astro-ph↗
arXiv2006

IRS Spectra of Solar-Type Stars: A Search for Asteroid Belt Analogs

We report the results of a spectroscopic search for debris disks surrounding 41 nearby solar type stars, including 8 planet-bearing stars, using the {\it Spitzer Space Telescope}. With accurate relative photometry using the Infrared Spectrometer (IRS) between 7-34 $\micron$ we are able to look for excesses as small as $\sim$2% of photospheric levels with particular sensitivity to weak spectral features. For stars with no excess, the $3σ$ upper limit in a band at 30-34 $μ$m corresponds to $\sim$ 75 times the brightness of our zodiacal dust cloud. Comparable limits at 8.5-13 $μ$m correspond to $\sim$ 1,400 times the brightness of our zodiacal dust cloud. These limits correspond to material located within the $<$1 to $\sim$5 AU region that, in our solar system, originates from debris associated with the asteroid belt. We find excess emission longward of $\sim$25 $μ$m from five stars of which four also show excess emission at 70 $μ$m. This emitting dust must be located around 5-10 AU. One star has 70 micron emission but no IRS excess. In this case, the emitting region must begin outside 10 AU; this star has a known radial velocity planet. Only two stars of the five show emission shortward of 25 $\micron$ where spectral features reveal the presence of a population of small, hot dust grains emitting in the 7-20 $μ$m band. The data presented here strengthen the results of previous studies to show that excesses at 25 $\micron$ and shorter are rare: only 1 star out of 40 stars older than 1 Gyr or $\sim 2.5$% shows an excess. Asteroid belts 10-30 times more massive than our own appear are rare among mature, solar-type stars.

astro-ph↗
arXiv2006

IRS Spectra of Solar-Type Stars: \break A Search for Asteroid Belt Analogs

We report the results of a spectroscopic search for debris disks surrounding 41 nearby solar type stars, including 8 planet-bearing stars, using the {\it Spitzer Space Telescope}. With accurate relative photometry using the Infrared Spectrometer (IRS) between 7-34 $\micron$ we are able to look for excesses as small as $\sim$2% of photospheric levels with particular sensitivity to weak spectral features. For stars with no excess, the $3σ$ upper limit in a band at 30-34 $μ$m corresponds to $\sim$ 75 times the brightness of our zodiacal dust cloud. Comparable limits at 8.5-13 $μ$m correspond to $\sim$ 1,400 times the brightness of our zodiacal dust cloud. These limits correspond to material located within the $<$1 to $\sim$5 AU region that, in our solar system, originates from debris associated with the asteroid belt. We find excess emission longward of $\sim$25 $μ$m from five stars of which four also show excess emission at 70 $μ$m. This emitting dust must be located around 5-10 AU. One star has 70 micron emission but no IRS excess. In this case, the emitting region must begin outside 10 AU; this star has a known radial velocity planet. Only two stars of the five show emission shortward of 25 $\micron$ where spectral features reveal the presence of a population of small, hot dust grains emitting in the 7-20 $μ$m band. The data presented here strengthen the results of previous studies to show that excesses at 25 $\micron$ and shorter are rare: only 1 star out of 40 stars older than 1 Gyr or $\sim 2.5$% shows an excess. Asteroid belts 10-30 times more massive than our own appear are rare among mature, solar-type stars.

astro-ph↗
arXiv2006

Spitzer/MIPS Infrared Imaging of M31: Further Evidence for a Spiral/Ring Composite Structure

New images of M31 at 24, 70, and 160 micron taken with the Multiband Imaging Photometer for Spitzer (MIPS) reveal the morphology of the dust in this galaxy. This morphology is well represented by a composite of two logarithmic spiral arms and a circular ring (radius ~10 kpc) of star formation offset from the nucleus. The two spiral arms appear to start at the ends of a bar in the nuclear region and extend beyond the star forming ring. As has been found in previous work, the spiral arms are not continuous but composed of spiral segments. The star forming ring is very circular except for a region near M32 where it splits. The lack of well defined spiral arms and the prominence of the nearly circular ring suggests that M31 has been distorted by interactions with its satellite galaxies. Using new dynamical simulations of M31 interacting with M32 and NGC 205 we find that, qualitatively, such interactions can produce an offset, split ring like that seen in the MIPS images.

astro-ph↗
arXiv2006

Ultraviolet-to-Far Infrared Properties of Lyman Break Galaxies and Luminous Infrared Galaxies at z ~ 1

We present the first large, unbiased sample of Lyman Break Galaxies (LBGs) at z ~ 1. Far ultraviolet-dropout (1530 A) galaxies in the Chandra Deep Field South have been selected using GALEX data. This first large sample in the z ~ 1 universe provides us with a high quality reference sample of LBGs. We analyzed the sample from the UV to the IR using GALEX, SPITZER, ESO and HST data. The morphology (obtained from GOODS data) of 75 % of our LBGs is consistent with a disk. The vast majority of LBGs with an IR detection are also Luminous Infrared Galaxies (LIRGs). As a class, the galaxies not detected at 24 microns are an order of magnitude fainter relative to the UV compared with those detected individually, suggesting that there may be two types of behavior within the sample. For the IR-bright galaxies, there is an apparent upper limit for the UV dust attenuation and this upper limit is anti-correlated with the observed UV luminosity. Previous estimates of dust attenuations based on the ultraviolet slope are compared to new ones based on the FIR/UV ratio (for LBGs detected at 24 microns), which is usually a more reliable estimator. Depending on the calibration we use to estimate the total IR luminosity, beta-based attenuations A_{FUV} are larger by 0.2 to 0.6 mag. than the ones estimated from FIR/UV ratio. Finally, for IR-bright LBGs, median estimated beta-based SFRs are 2-3 times larger than the total SFRs estimated as SFR_{TOT} = SFR_{UV} + SFR_{IR} while IR-based SFRs provide values below SFR_{TOT} by 15 - 20 %. We use a stacking method to statistically constrain the 24 microns flux of LBGs non individually detected. The results suggest that these LBGs do not contain large amounts of dust.

astro-ph↗
arXiv2006

Mid-infrared properties of X-ray sources in the Extended Groth Strip

Mid-infrared observations of Active Galactic Nuclei (AGN) are important for understanding of the physical conditions around the central accretion engines. Chandra and XMM-Newton X-ray observations of a 300 arcmin^2 region in the Extended Groth Strip are used to select a sample of~150 AGN. The Spitzer instruments IRAC and MIPS detect 68-80% of these sources, which show a wide range of mid-infrared properties. About 40% of the sources have red power-law spectral energy distributions (f_nu ~ nu^alpha, alpha<0) in the 3.6-8 um IRAC bands. In these sources the central engine dominates the emission at both X-ray and IR wavelengths. Another 40% of the sources have blue mid-IR spectral energy distributions (alpha>0) with their infrared emission dominated by the host galaxy; the remaining 20% are not well-fit by a power law. Published IRAC color criteria for AGN select most of the red sources, but only some of the blue sources. As with all other known methods, selecting AGN with mid-IR colors will not produce a sample that is simultaneously complete and reliable. The IRAC SED type does not directly correspond to X-ray spectral type (hard/soft). The mid-IR properties of X-ray-detected Lyman-break, radio, submillimeter, and optically-faint sources vary widely and, for the most part, are not distinct from those of the general X-ray/infrared source population. X-ray sources emit 6-11% of the integrated mid--IR light, making them significant contributors to the cosmic infrared background.

astro-ph↗
arXiv2005

Detecting Faint Galaxies by Stacking at 24 micron

We stack Spitzer 24 micron images for ~7000 galaxies with 0.1<z<1 in the Chandra Deep Field South to probe the thermal dust emission in low-luminosity galaxies over this redshift range. Through stacking, we can detect mean 24 micron fluxes that are more than an order of magnitude below the individual detection limit. We find that the correlations for low and moderate luminosity galaxies between the average L_IR/L_UV and rest-frame B-band luminosity, and between the star formation rate (SFR) and L_IR/L_UV, are similar to those in the local Universe. This verifies that oft-used assumption in deep UV/optical surveys that the dust obscuration-SFR relation for galaxies with SFR < 20 solar mass per year varies little with epoch. We have used this relation to derive the cosmic IR luminosity density from z=1 to z=0.1. The results also demonstrate directly that little of the bolometric luminosity of the galaxy population arises from the faint end of the luminosity function, indicating a relatively flat faint-end slope of the IR luminosity function with a power law index of 1.2+-0.3.

astro-ph↗
arXiv2005

An Initial Look at the Far Infrared-Radio Correlation within Nearby Star-forming Galaxies using the Spitzer Space Telescope

(Abridged) We present an initial look at the far infrared-radio correlation within the star-forming disks of four nearby, nearly face-on galaxies (NGC 2403, NGC 3031, NGC 5194, and NGC 6946). Using Spitzer MIPS imaging and WSRT radio continuum data, observed as part of the Spitzer Infrared Nearby Galaxies Survey (SINGS), we are able to probe variations in the logarithmic 24mu/22cm (q_24) and 70mu/22cm (q_70) surface brightness ratios across each disk at sub-kpc scales. We find general trends of decreasing q_24 and q_70 with declining surface brightness and with increasing radius. The residual dispersion around the trend of q_24 and q_70 versus surface brightness is smaller than the residual dispersion around the trend of q_24 and q_70 versus radius, on average by ~0.1 dex, indicating that the distribution of star formation sites is more important in determining the infrared/radio disk appearance than the exponential profiles of disks. We have also performed preliminary phenomenological modeling of cosmic ray electron (CRe^-) diffusion using an image-smearing technique, and find that smoothing the infrared maps improves their correlation with the radio maps. Exponential kernels tend to work better than Gaussian kernels which suggests that additional processes besides simple random-walk diffusion in three dimensions must affect the evolution of CRe^-s. The best fit smoothing kernels for the two less active star-forming galaxies (NGC 2403 and NGC 3031) have much larger scale-lengths than those of the more active star-forming galaxies (NGC 5194 and NGC 6946). This difference may be due to the relative deficit of recent CRe^- injection into the interstellar medium (ISM) for the galaxies having largely quiescent disks.

astro-ph↗
arXiv2005

Infrared power-law galaxies in the Chandra Deep Field South: AGN and ULIRGs

We investigate the nature of a sample of 92 Spitzer/MIPS 24 micron selected galaxies in the CDFS, showing power law-like emission in the Spitzer/IRAC 3.6-8 micron bands. The main goal is to determine whether the galaxies not detected in X-rays (47% of the sample) are part of the hypothetical population of obscured AGN not detected even in deep X-ray surveys. The majority of the IR power-law galaxies are ULIRGs at z>1, and those with LIRG-like IR luminosities are usually detected in X-rays. The optical to IR spectral energy distributions (SEDs) of the X-ray detected galaxies are almost equally divided between a BLAGN SED class (similar to an optically selected QSO) and a NLAGN SED (similar to the BLAGN SED but with an obscured UV/optical continuum). A small fraction of SEDs resemble warm ULIRG galaxies (e.g., Mrk231). Most galaxies not detected in X-rays have SEDs in the NLAGN+ULIRG class as they tend to be optically fainter, and possibly more obscured. Moreover, the IR power-law galaxies have SEDs significantly different from those of high-z (z_sp>1) IR (24 micron) selected and optically bright (VVDS I_AB<=24) star-forming galaxies whose SEDs show a very prominent stellar bump at 1.6 micron. The galaxies detected in X-rays have 2-8 keV rest-frame luminosities typical of AGN. The galaxies not detected in X-rays have global X-ray to mid-IR SED properties that make them good candidates to contain IR bright X-ray absorbed AGN. If all these sources are actually obscured AGN, we would observe a ratio of obscured to unobscured 24 micron detected AGN of 2:1, whereas models predict a ratio of up to 3:1. Additional studies using Spitzer to detect X-ray-quiet AGN are likely to find more such obscured sources.

astro-ph↗
arXiv2005

Spitzer Observations of Massive Red Galaxies at High Redshift

We investigate the properties of massive galaxies at z=1-3.5 using HST observations, ground-based near-IR imaging, and Spitzer Space Telescope observations at 3-24 micron. We identify 153 distant red galaxies (DRGs) with J-K > 2.3 mag (Vega) in the southern GOODS field. This sample is approximately complete in stellar mass for passively evolving galaxies above 10^11 solar masses and z < 3. The galaxies identified by this selection are roughly split between objects whose optical and near-IR rest-frame light is dominated by evolved stars combined with ongoing star formation, and galaxies whose light is dominated by heavily reddened starbursts. Very few of the galaxies (< 10%) have no indication of current star formation. Using SFR estimates that include the reradiated IR emission, the DRGs at z=1.5-3 with stellar masses > 10^11 solar masses have specific SFRs (SFRs per unit stellar mass) ranging from 0.2 to 10 Gyr^-1, with a mean value of ~2.4 Gyr^-1. The DRGs with stellar masses > 10^11 solar masses and 1.5 < z < 3 have integrated specific SFRs greater the global value over all galaxies. In contrast, we find that galaxies at z = 0.3-0.75 with these stellar masses have integrated specific SFRs less than the global value, and more than an order of magnitude lower than that for massive DRGs at z = 1.5-3. At z < 1, lower-mass galaxies dominate the overall cosmic mass assembly. This suggests that the bulk of star formation in massive galaxies occurs at early cosmic epochs and is largely complete by z~1.5. [Abridged]

astro-ph↗
arXiv2005

Linking stellar mass and star formation in Spitzer/MIPS 24 micron galaxies

We present deep Ks<21.5 (Vega) identifications, redshifts and stellar masses for most of the sources composing the bulk of the 24 micron background in the GOODS/CDFS. Our identified sample consists of 747 Spitzer/MIPS 24 micron objects, and includes ~94% of all the 24 micron sources in the GOODS-South field which have fluxes Snu(24)>83 microJy (the 80% completeness limit of the Spitzer/GTO 24 micron catalog). 36% of our galaxies have spectroscopic redshifts (mostly at z<1.5) and the remaining ones have photometric redshifts of very good quality, with a median of |dz|=|zspec-zphot|/(1+zspec)=0.02. We find that MIPS 24 micron galaxies span the redshift range z~0-4, and that a substantial fraction (28%) lie at high redshifts z>1.5. We determine the existence of a bump in the redshift distribution at z~1.9, indicating the presence of a significant population of galaxies with PAH emission at these redshifts. Massive (M>10^11 Msun) star-forming galaxies at redshifts 2 500 Msun/yr), and some of them are able to construct a mass of 10^10-10^11 Msun in a single burst lifetime (~0.01-0.1 Gyr). At lower redshifts z<2, massive star-forming galaxies are also present, but appear to be building their stars on long timescales, either quiescently or in multiple modest burst-like episodes. At redshifts z~1-2, the ability of the burst-like mode to produce entire galaxies in a single event is limited to some lower (M<7x10^10 Msun) mass systems, and it is basically negligible at z<1. Our results support a scenario where star-formation activity is differential with assembled stellar mass and redshift, and where the relative importance of the burst-like mode proceeds in a down-sizing way from high to low redshifts. (abridged)

astro-ph↗
arXiv2005

Frequency of Debris Disks around Solar-Type Stars: First Results from a Spitzer/MIPS Survey

We have searched for infrared excesses around a well defined sample of 69 FGK main-sequence field stars. These stars were selected without regard to their age, metallicity, or any previous detection of IR excess; they have a median age of ~4 Gyr. We have detected 70 um excesses around 7 stars at the 3-sigma confidence level. This extra emission is produced by cool material (< 100 K) located beyond 10 AU, well outside the ``habitable zones'' of these systems and consistent with the presence of Kuiper Belt analogs with ~100 times more emitting surface area than in our own planetary system. Only one star, HD 69830, shows excess emission at 24 um, corresponding to dust with temperatures > 300 K located inside of 1 AU. While debris disks with Ld/L* > 10^-3 are rare around old FGK stars, we find that the disk frequency increases from 2+-2% for Ld/L* > 10^-4 to 12+-5% for Ld/L* > 10^-5. This trend in the disk luminosity distribution is consistent with the estimated dust in our solar system being within an order of magnitude, greater or less, than the typical level around similar nearby stars.

astro-ph↗
arXiv2005

Infrared Spectral Energy Distributions of Nearby Galaxies

The Spitzer Infrared Nearby Galaxies Survey (SINGS) is carrying out a comprehensive multi-wavelength survey on a sample of 75 nearby galaxies. The 1-850um spectral energy distributions are presented using broadband imaging data from Spitzer, 2MASS, ISO, IRAS, and SCUBA. The infrared colors derived from the globally-integrated Spitzer data are generally consistent with the previous generation of models that were developed based on global data for normal star-forming galaxies, though significant deviations are observed. Spitzer's excellent sensitivity and resolution also allow a detailed investigation of the infrared spectral energy distributions for various locations within the three large, nearby galaxies NGC3031 (M81), NGC5194 (M51), and NGC7331. Strong correlations exist between the local star formation rate and the infrared colors f_nu(70um)/f_nu(160um) and f_nu(24um)/f_nu(160um), suggesting that the 24 and 70um emission are useful tracers of the local star formation activity level. Preliminary evidence indicates that variations in the 24um emission, and not variations in the emission from polycyclic aromatic hydrocarbons at 8um, drive the variations in the f_nu(8.0um)/f_nu(24um) colors within NGC3031, NGC5194, and NGC7331. If the galaxy-to-galaxy variations in spectral energy distributions seen in our sample are representative of the range present at high redshift then extrapolations of total infrared luminosities and star formation rates from the observed 24um flux will be uncertain at the factor-of-five level (total range). The corresponding uncertainties using the redshifted 8.0um flux (e.g. observed 24um flux for a z=2 source) are factors of 10-20. Considerable caution should be used when interpreting such extrapolated infrared luminosities.

astro-ph↗
arXiv2005

Unveiling a Population of X-ray Non-Detected AGN

We define a sample of 27 radio-excess AGN in the Chandra Deep Field North by selecting galaxies that do not obey the radio/infrared correlation for radio-quiet AGN and star-forming galaxies. Approximately 60% of these radio-excess AGN are X-ray undetected in the 2 Ms Chandra catalog, even at exposures of > 1 Ms; 25% lack even 2-sigma X-ray detections. The absorbing columns to the faint X-ray-detected objects are 10^22 cm^-2 < N_H < 10^24 cm^-2, i.e., they are obscured but unlikely to be Compton thick. Using a local sample of radio-selected AGN, we show that a low ratio of X-ray to radio emission, as seen in the X-ray weakly- and non-detected samples, is correlated with the viewing angle of the central engine, and therefore with obscuration. Our technique can explore the proportion of obscured AGN in the distant Universe; the results reported here for radio-excess objects are consistent with but at the low end of the overall theoretical predictions for Compton-thick objects.

astro-ph↗
arXiv2005

Star Formation in NGC5194 (M51a): The Panchromatic View from GALEX to Spitzer

(Abridged) Far ultraviolet to far infrared images of the nearby galaxy NGC5194, from Spitzer, GALEX, Hubble Space Telescope and ground--based data, are used to investigate local and global star formation, and the impact of dust extinction in HII-emitting knots. In the IR/UV-UV color plane, the NGC5194 HII knots show the same trend observed for normal star-forming galaxies, having a much larger dispersion than starburst galaxies. We identify the dispersion as due to the UV emission predominantly tracing the evolved, non-ionizing stellar population, up to ages 50-100 Myr. While in starbursts the UV light traces the current SFR, in NGC5194 it traces a combination of current and recent-past SFR. Unlike the UV emission, the monochromatic 24 micron luminosity is an accurate local SFR tracer for the HII knots in NGC5194; this suggests that the 24 micron emission carriers are mainly heated by the young, ionizing stars. However, preliminary results show that the ratio of the 24 micron emission to the SFR varies by a factor of a few from galaxy to galaxy. While also correlated with star formation, the 8 micron emission is not directly proportional to the number of ionizing photons. This confirms earlier suggestions that the carriers of the 8 micron emission are heated by more than one mechanism.

astro-ph↗
arXiv2005

Metallicity Effects on Mid-Infrared Colors and the 8 micron PAH Emission in Galaxies

We examine colors from 3.6 micron to 24 micron as a function of metallicity (O/H) for a sample of 34 galaxies. The galaxies range over 2 orders of magnitude in metallicity. They display an abrupt shift in the 8 micron to 24 micron color between metallicities 1/3 to 1/5 of the solar value. The mean 8 micron to 24 micron flux density ratio below and above 12 + log (O/H) = 8.2 is 0.08 +/- 0.04 and 0.70 +/- 0.53, respectively. We use mid-infrared colors and spectroscopy to demonstrate that the shift is primarily due to a decrease in the 8 micron flux density as opposed to an increase in the 24 micron flux density. This result is most simply interpreted as due to a weakening at low metallicity of the mid-infrared emission bands usually attributed to PAHs (polycyclic aromatic hydrocarbons) relative to the small-grain dust emission. However, existing empirical spectral energy distribution models cannot account for the observed short-wavelength (i.e., below 8 micron) colors of the low-metallicity galaxies merely by reducing the strength of the PAH features; some other emission source (e.g., hot dust) is required.

astro-ph↗