-
Understanding Post-Starburst Quasars and Their Role in Mutual
Black Hole -
Bulge Growth of Galaxies
PI: S. L. Cales, Advisor: M. S. Brotherton
[email protected], [email protected]
Physics & Astronomy Department, University of Wyoming,
Laramie, WY 82071
ABSTRACT
The observational result that massive galaxies harbor
supermassive black holes(BHs) at their centers along with the tight
correlation between black hole mass andstellar velocity dispersion
has led astronomers to hypothesize that supermassive BHsand their
host galaxies have mutual formation and evolutionary histories.
Hierarchi-cal merging, the process/mechanism by which larger
(elliptical) galaxies are formedby mergers of smaller (spiral)
ones, is a common thread among astronomers studyingmutual BH/bulge
growth. Indeed, the ubiquitous presence of ultra-luminous
infraredgalaxies (ULIRGs) involved in mergers lends circumstantial
evidence to this theory.Furthermore, the host galaxies of “normal”
quasars have been found to be disturbed,a telltale sign of
interaction. A new class of objects, post-starburst quasars
(PSQs;quasars showing starburst activity within the last few 100
Myrs), might represent anintermediate phase between merger and the
end product (red elliptical) of hierarchicalmerger theory, in which
feedback from the quasar has recently blown away obcsur-ing/feeding
material. To test this theory I propose a multi-wavelength study of
thisnew class (PSQs) using high-quality data (HST, Keck, KPNO,
IRTF, Spitzer) of asample of 29 PSQs. Specifically, points of study
include: i) morphology and quasar tohost galaxy light fractions via
HST ACS and IRTF imaging; ii) demographics of hostgalaxy and quasar
(starburst mass and age, black hole mass and accretion rate) via
Keckand KPNO spectra; iii) PSQ infrared colors and dust
contributions via Spitzer spectra,and; iv) comparisons with other
classes of objects (ULIRGs, post-starburst galaxies,normal quasars,
elliptical galaxies. . .) and theoretical models. Ultimately, this
projectwill establish the potentially critical role of
post-starburst quasars in evolutionary mod-els for mutual BH/galaxy
evolution, determining fundamental observational constraintson
theoretical models, which are popular and running rampant on this
topic.
-
– 2 –
1. Introduction
This thesis will be testing the theory that the MBH -σ∗
correlation (§ 1.1) arises as a directresult of quasar feedback
(“blowout”) via extragalactic observations of PSQs. Many classes
ofobjects (e.g., ULIRGs, post-starburst galaxies, normal quasars,
elliptical galaxies) are expected toadhere to the MBH -σ∗ relation
(§ 1.3) who’s formation is attributed to hierarchical merging (§
1.2).Furthermore, PSQs are predicted to be a phase in the life of a
quasar/galaxy at or around “blowout”(§ 1.4) and thus, the “smoking
gun” evidence for AGN/quasar feedback showing how the MBH
-σ∗correlation arises. The detailed characterization of the first
discovered PSQ, UN J1025-0040 (§ 1.5),solidifies the proposed
evolutionary sequence as a plausible theory.
1.1. Supermassive Black Hole Mass-Bulge Mass Correlation
Recent research has shown that most galaxies harbor supermassive
black holes at their centers(Kormendy & Richstone 1995).
Additionally, observations have established that there exists
atight correlation between the masses of these black holes and that
of the host galaxy stellar bulgecomponent (MBH ∼ 0.15% Mbulge;
Figure 1; Magorrian et al. 1998; Gebhardt et al. 2000b), ormeasured
from the stellar velocity dispersion with a tighter correlation (M
-σ∗ relation; Gebhardtet al. 2000a; Ferrarese & Merritt 2000;
Tremaine et al. 2002). This MBH -Mbulge relation suggeststhat the
growth of black holes is intimately linked to the formation and
evolution of their hostgalaxies and hints at their originating via
a common physical process (Kauffmann & Haehnelt2000; Granato et
al. 2004; Hopkins et al. 2007).
1.2. Modeling the Starburst-AGN Connection
According to cold dark matter cosmology, hierarchical merging is
responsible for the formationof larger galaxies via mergers of
smaller ones. Additionally, the triggering of major starburst
isattributed to mergers, possibly accounting for the growth of
supermassive black holes and theformation of quasars (i.e., high
luminosity class of active galactic nuclei), and ultimately
predictingthe formation of many of the structures that we see in
the Universe today. Theorists have goneon to model star formation
and black hole growth in galaxy mergers in an effort to explain
theMBH -Mbulge link (e.g., Kauffmann & Haehnelt 2000; Burkert
& Silk 2001; Granato et al. 2004).Many of the models invoke the
hierarchical model of galaxy formation and evolution put forth
byToomre (1977) in which most elliptical galaxies form by
mergers.
Hydrodynamical simulations incorporating star formation,
supernova feedback, BH growth andAGN (active galactic nuclei)
feedback support hierarchical clustering. These simulations follow
starformation and BH growth simultaneously during galaxy-galaxy
collisions and find that mergers leadto strong gaseous inflows that
feed the quasar, starbursts, and, in due course, energy released
by
-
– 3 –
Fig. 1.— BH mass versus (a) bulge luminosity and (b) velocity
dispersion of Gebhardt et al. (2000b, Figure1). There are 33 points
in the dispersion plot: 26 from the compilation of Gebhardt et al.
(2000a, open circles)and seven from reverberation mapping (filled
circles). Solid and dotted lines are the best-fit correlations
andtheir 68% confidence bands from Gebhardt et al. (2000a) fitted
only to the galaxies with spatially resolvedkinematics. The error
bars in the lower right of each plot are representative of the
reverberation mappinguncertainties.
the quasar quenches both star formation and further black hole
growth (Di Matteo, Springel, &Hernquist 2005). In a simulation
of the color evolution of merger remnants (Springel, Di
Matteo,& Hernquist 2005), it was found that reddening (i.e.,
aging of the galaxy) was significantly quickerin remnants
containing black holes, an effect brought on by feedback from the
quasar (“blowout”).Furthermore, the simulations of Cox et al.
(2006) concerning the morphological and kinematicproperties of
merger remnants argue that gas-rich major mergers trigger quasars
and starbursts,fuel the growth of black holes, and produce remnant
galaxies that have the colors, scaling relationsand kinematic
structures in concordance with ellipticals of the current epoch.
Thus, black holeaccretion and AGN feedback is expected to have a
crucial impact on the evolution of its hostgalaxy.
1.3. The Importance of Being Ultraluminous
Ultraluminous infrared galaxies (ULIRGs), strongly interacting
merger systems heated by bothstarburst and AGN power sources, have
been hypothesized to evolve into normal quasars after thecentral
engine clears away the dust associated with the massive star
formation (Sanders et al. 1988;Sanders & Mirabel 1996; Veilleux
2006). These objects are possibly the first observational step
inthe evolution and formation of a larger elliptical galaxy.
Furthermore, the host galaxies of “normal”
-
– 4 –
quasars have been found to be disturbed (Canalizo et al. 2007;
Bennert et al. 2008), a telltale signof interaction. Post-starburst
galaxies are thought to be the evolved galaxies after a few 100
Myrafter the peak of the star formation and AGN activity (Tremonti,
Moustakas, & DiamondStanic2007). Thus, an incorporated
evolutionary schema upon merger could be: ULIRG → ? → quasaror
post-starburst galaxy → late type galaxy, elliptical. There are
other possibilities still, such as;minor mergers, harassment, and
other interactions which could lead to both black hole growth
andbulge growth.
1.4. PSQs, Objects During Feedback?
A new class of objects, post-starburst quasars, might represent
objects during the feedbacktime-frame, (i.e., at or around
“blowout”). Post-starburst quasars (PSQs) are broad-lined AGN
thatalso possess the spectral signatures of massive, moderate-aged
stellar populations (in excess of tenbillion solar masses and ages
of hundreds of Myrs). The typical composite spectra of PSQs
displaybroad emission lines as well as Balmer jumps and strong
Balmer absorption lines characteristic oftype-A stars. The absence
of obscuring dust in the host galaxy along with the presence of
both amassive starburst and BH suggests that they are evolved
ULIRGs at or around “blowout”. Followingrecent merger theory, PSQs
might then represent a phase in the life of a galaxy/quasar where
amerger has activated the central engine, as well as, a burst of
star formation, and feedback from thequasar (and/or supernovae) has
recently blown away the obscuring (feeding) material. WhetherPSQs
represent an evolutionary phase common to all galaxies that shows
how the correlationbetween black hole mass and stellar bulge mass
arises, or they represent a rare but particularlyviolent
interaction that leads to both massive starbursts and nuclear
activity is not yet clear.
1.5. The First PSQ, UN J1025-0040
The post-starburst quasar prototype, UN J1025-0040, upholds many
features of the evolu-tionary theory. UN J1025-0040 is hosted by a
galaxy with a ∼400 Myr old strong starburst(Brotherton et al.
1999), the host appears as a merger remnant (Brotherton et al.
2002), and hasa companion galaxy in a post-starburst phase
(Canalizo et al. 2000). A younger UN J1025-0040(tens of Myrs after
the starburst) would have a more luminous stellar population and
would likelybe dust-enshrouded, placing it in the ULIRG class.
These observations suggest that UN J1025-0040 is a plausible
transition between ULIRGs and quasars. Timescales of the quasar
activityand post-starburst phase likely affect the evolutionary
sequence. If, for example, the timescale ofthe quasar’s duty cycle
is less than that of the post-starburst phase, then after the
quasar’s dutycycle UN J1025-0040 would presumably appear as a
post-starburst galaxy. Alternatively, if thepost-starburst phase is
shorter, we would expect to see a “normal” quasar after the
post-starburstphase had subsided. After both quasar activity and
post-starburst phase timescales, UN J1025-0040 would likely appear
as an elliptical, i.e., the evolutionary schema then becomes: ULIRG
→
-
– 5 –
Fig. 2.— The stacked PSQ and QSO, 70 percent of the normalized
PSF and corresponding host radialbrightness profiles (as
indicated). The differential surface brightness scale is correct,
while the absolute scalemight be arbitrary. The PSQs host galaxy is
more dominant than that of the QSOs (normal quasars). Also,notice
that the shapes of the galactic components differ.
PSQ → quasar or post-starburst galaxy → late type galaxy,
elliptical.
2. Foundational Research
2.1. PSQ Catalog
PSQs are a rare class of objects. Of the over 50,000 quasars
found by Sloan Digital SkySurvey data release 3 (SDSS DR3;
Schneider et al. 2002) ∼600 were spectroscopically selected
andcatalogued by Brotherton et al. (2008a, in preparation;
hereafter B8a) as PSQs. The automatedselection algorithm was
primarily based on that of Zabludoff et al. (1996), an algorithm
used toselect post-starburst galaxies. The selection criteria
required that the spectra display both thebroad emission lines of
luminous AGNs (Seyfert galaxies or quasars) and the Balmer jumps
andhigh-order Balmer absorption lines of massive stellar
populations on order of hundreds of Myrold. SDSS images of the
catalog reveal a 40% near neighbor rate (i.e., close interacting
doublesor companions within 10”). Follow up Keck spectroscopy,
along with SDSS and HST images ona select 25 PSQs, shows similar
results (Brotherton et al. 2008b, in preparation). Of the
singlesources, most show extended fuzz or sometimes tidal tails;
both symptoms of interaction and mergeractivity. This suggests that
PSQs are quasar analogs to post-starburst galaxies; galaxies which
areuniformly interacting or merger systems.
-
– 6 –
2.2. Multiwavelength UW AGN Group Project
The efforts of the UW AGN group (i.e., M. Brotherton, R.
Ganguly, Z. Shang. . . ) are outlinedbriefly to distinguish my
efforts of carefully characterizing a small subsample of objects
(∼29) froma larger broader analysis of the ∼600 PSQs defined by the
SDSS DR3 catalog:
• Catalog Selection - Modified Zabludoff et al. (1996) selection
criterion (§ 2.1).
• Environments of PSQs - Large and small scale environments of
PSQs in comparison withquasars (§ 2.3).
• Basic Properties - Analysis of SDSS spectra, imaging and
archive parameters to determinePSQ colors, starburst age and mass
(via evolutionary stellar population synthesis models),BH mass and
Eddington accretion rate fraction.
• Comparisons with other classes of objects - By analysis of the
data in the SDSS archive andrecent published work on other classes
(e.g., ULIRGs, normal quasars) ask the questions:How do the basic
properties of PSQs compare with recently published work on other
classes?
• Final Catalog - Select final catalog from SDSS Final Release
using identifying properties ofPSQs determined from this
thesis.
2.3. PSQ Environments
Follow up Keck spectroscopy of the B8a catalog, along with SDSS
and HST imaging, wasperformed on a select 25 PSQs with
close/possibly interacting companions. Analysis reveals that∼40% of
these are indeed close interacting doubles or companions within 10”
Brotherton et al.(2008b, in preparation; hereafter B8b). B8b
compares the environments of PSQs, of the B8acatalog, with those of
normal quasars (QSOs, quasi-stellar objects) on a variety of
scales. Usingphotometric redshifts based on SDSS imaging B8b
compared the incidence of companion galaxieson scales of 10 to 100
kiloparsecs, finding only a marginal excess of galaxies around PSQs
(i.e.,they live in denser regions). Different large-scale
environments would suggest that PSQs are notrun-of-the-mill quasars
seen at a special time. Typically, PSQs tend to have more neighbors
thannormal quasars; about 4:3. In addition, more PSQs have
neighbors than quasars; 90%:76%. Amore pronounced difference was
found on smaller scale environments. Differences in
small-scaleenvironments might indicate direct triggers of the
activity that are lost from view at later times. Iutilize image
stacking and point source subtraction to compare the immediate
environments on 10kiloparsec and smaller scales. The images of the
sample are binned by redshift according to 10%differences in size
scale. Here I present the smallest redshift bin (Figure 2). The
difference in thePSQ and quasar host galaxies is apparent. In all,
the large-scale environments of post-starburstquasars are not
dissimilar from those of other quasars, thus it is concluded that
the more immediate
-
– 7 –
small-scale interaction is the primary cause of the
post-starburst activity in most objects of theclass.
3. Project Description
While the work of Brotherton et al. (2008a,b) focuses on the
broad determination of propertiesof PSQs from a catalog, the aim
for this thesis is the careful determination of the properties
andnature of PSQs from a smaller sample ∼29. This can be
accomplished, by obtaining and analyzinghigh-quality data from
various observatories (HST, IRTF, KPNO, Keck and Spitzer ; see
Table 1)otherwise not attainable for a large catalog. Specifically,
knowledge of: starburst properties(age and mass), quasar black hole
mass and accretion rate (relative to Eddington),
PSQ colors, the infrared color of PSQs and dust contribution of
the host galaxy, PSQ
allegiances with other classes and theoretical models will
broaden our understanding
of PSQs and their environments. Ultimately, the goal paramount
to this investigation
is to aid in the understanding of how black hole mass/bulge mass
correlations arise
and their evolutionary role. To gain such knowledge, I propose
the following small projects asmy thesis:
• Imaging of HST Sample: HST and NASA’s IRTF imaging will
determine morphologiesand quasar to host galaxy light contributions
(§ 4).
• Host Component Analysis: Keck and KPNO-4m high quality
spectroscopy obtained inorder to model starburst properties (age,
mass), their black hole mass and accretion rates,and determination
of redshifts and classifications of nearby companion galaxies (§
5).
• Infrared Properties of PSQs: Mid-infrared spectroscopy from
Spitzer will provide infrared(IR) colors (mid-IR slope of spectra)
and PAH (polycyclic aromatic hydrocarbon) featurestrengths
diagnosing the heating source of the dust radiating in the IR (§
6).
• Statistical Analysis: With the three previously mention
studies and further comparisonsto the current AGN-starburst and
feedback modeling a comparative (with other classes of ob-jects,
such as ULIRGs, normal quasars, elliptical galaxies and possibly
post-starburst galaxies)statistical analysis will be performed (§
7).
4. Imaging of HST Sample
4.1. Le’ Crème
HST’s SNAP programs take advantage of gaps in the General
Observer program schedule.During these gaps, nearby SNAP targets
can be imaged in a short visit. There are no ‘guaranteed’
-
– 8 –
observations of targets, thus, observations of a program are
generally not 100 percent completed.As such, chosen targets are
randomized with respect to the target list, but do depend on
certaingap constraints (e.g., current pointing on the sky). We have
received 29 HST/ACS-F606W (broadV-band at 0.05”/pixel) SNAP images
(Figure 3) of a target list totaling 80 post-starburst SDSSDR3
quasars. The PSQs were selected from the B8a catalog to have SDSS r
< 18.6, Hδ absorptionequivalent widths > 1 Å, 0.25 < z
< 0.45 (ensuring high luminosity and similar sizescales,
resolvingstructures a half Kpc across) focusing on the clearest
most luminous (〈Mr〉 ∼ −22.7) examples ofPSQs. The sample of 29 will
hereafter be referred to as the HST sample and is the defining
sampleof this thesis. It is a good sample in the sense that these
29 objects are representative of theoriginal SNAP target list of 80
PSQs. Besides important morphological information (§ 4.2),
theSNAPshot image analysis will involve point source fitting (§
4.3) and/or modeling (§ 5.2) similarto what Brotherton et al.
(2002) accomplished with UN J1025-0040.
4.2. Morphology
The first step in characterizing PSQs naturally comes from the
images themselves. The HSTSNAP images give us a general feel for
the morphologies, companions/neighbors and
merger/interactionindications of PSQs. Some of the PSQs are
obviously quasars living in spiral galaxies, while othersappear to
be disturbed ellipticals. There are even hosts that appear to be
in-between. In two ofthe images we see obvious merging and about 40
percent have neighbors within 10”. The vast ma-jority of systems
show signs of interaction/merger activity, such as, tidal tails,
shells, star-formingknots and asymmetries. Brightness profiles of
the host galaxy and quasar to host galaxy light ratio(discussed in
detail below) will help quantify the nature of PSQ host
morphologies.
4.3. Point Source Fitting and Subtraction
Brotherton et al. (2002) estimate the point-spread function
(PSF; quasar counterpart to thecomposite image) by a combination of
Tiny Tim generated model PSFs and the spectral shape(composite blue
quasar spectrum) of the point source. If there was a suitable star
in the image fieldone would use that for the PSF estimation. Having
obtained good estimates of the PSF, the imageswere then deconvolved
using the Richard-Lucy algorithm. From here extended starlight and
“fuzz”(centrally concentrated light) colors can be evaluated.
PSF-subtracted images separate the hostand quasar components of PSQ
UN J1025-0040. The range of the point source to extended fluxratio
is determined by finding the minimum percent subtraction of the
normalized PSF necessaryto create a smooth surface brightness
profile (one with no central dip) and the maximum
percentsubtraction necessary for a flux of zero to be present at
the center (Figure 4). Knowing the quasarto host galaxy light ratio
constrains the composite starburst population modeling of the
opticalspectra as discussed further in the next section.
-
– 9 –
Fig. 3.— A mosaic of the 29 HST/ACS images. Each image is 10” ×
10” and NE is towards the upperleft. At HST’s high resolution
(∼0.05”) and an average redshift of 〈z〉 ∼ 0.319, these 200 × 200
pixel imagescorrespond to about 46.7 kpc across. We choose a flat
universe in which Ho is 71 km/s/Mpc and ΩM equalto 0.27. The
cosmology is chosen to be a flat universe in which Ho is 71
km/s/Mpc and ΩM equal to 0.27.See also
http://phsics.uwy.edu/agn/psq/index.html for a finer detailed
view.
-
– 10 –
Fig. 4.— PSF subtraction, left, and evolutionary stellar
population synthesis model fit, right, for UNJ1025-0040. Left,
Brotherton et al. (1999) surface brightness profile from
ellipse-fitting, UN J1025-0040(triangles) vs. a normalized point
source (squares). The circles show the difference between the total
lightand 70% of the normalized PSF. Right, the rescaled spectrum of
Brotherton et al. (1999, red spectrum) withthe HST -band photometry
marked (upper points) and the error bars for the non-PSF (stellar)
contribution(lower points). The starburst model (blue spectrum) is
a 400 Myr old population plus a young population(in this case, 10
Myr) and should be compared to the lower points. The quasar
residual (green spectrum) isthe difference between the two spectra
above.
A more novel approach to fitting the point source involves
principle component analysis (PCA).PCA is a statistical tool that
reduces the number of dimensions of a data set by using the
linearcombination of the variables. What does that mean? Well,
simply put, the first component of adata set represents the
direction in which the data has the largest variance, the second
componentmust be normal to the first and represent the direction in
which the next largest variance is present,and so on. Thus, in
terms of the the PSQ images (in which a host galaxy and quasar
point sourceare superposed), the first principle component should
be the ‘nice’ PSF (quasar point source). Thatis, the most
defining/predictable characteristic of the images will be the PSF.
Preliminary resultsof the PCA reveal that the first principle
component does indeed look like a PSF although furtheranalysis is
required. Originally, I centered the images by using a simple
centroiding task in IRAF(Image Reduction and Analysis Facility). I
hope to increase the accuracy of the PCA by re-binning,re-centering
and then re-running the PCA.
-
– 11 –
4.4. IRTF Data and Analysis
We have obtained IRTF/NSFCAM near-IR (K-band) moderately deep
imaging of the HSTsample. These will help us further characterize
the morphology of the host galaxies and possibleevidence for recent
mergers. With a resolution of 0.15”, size scales of about 3 kpc
will be probed.Analysis is in progress to determine the quasar
contributions at K-band. Like the analysis ofthe HST F606W imaging
(§ 4.3), this data will provide constraints on the evolutionary
stellarpopulation synthesis modeling of the Keck and KPNO spectra
(§ 5.3). Infrared properties of PSQwork (§ 6.2) will also be
constrained by the addition of an integrated light data point at
K-band.R-K colors are attainable to compare host galaxies,
companions and fainter features.
5. Host Component Analysis
5.1. KPNO and Keck Spectroscopy
We have obtained high signal-to-noise optical spectroscopy of
all but one object of the HSTsample in order to investigate
fundamental characteristics of the PSQ sample. This includes
eightobjects observed with KPNO-4m RC-Spec and 20 objects with Keck
LRIS. Keck LRIS spectroscopycovers 3100 Å to 1 µm. The 300/5000
grating on the red side and 400/3400 grism on the blue yield2.54
Å/pixel and 1.071 Å/pixel, respectively. The KPNO-4m RC-Spec data
using the KPC-10A316/4000 grating spans 3200 Å to 7200 Å with a
dispersion of 2.75 Å/pixel. The analysis of thespectra will focus
on the determination of the ages and masses of the host stellar
populations(utilizing stellar population synthesis models) and the
black hole masses and their Eddingtonfractions (by making virial
mass estimates). Our approach is to model the spectra as a
combinationof a simple instantaneous starburst (more commonly known
as simple stellar populations, SSPs;Bruzual & Charlot 2003) and
a quasar power-law plus emission lines (including ultraviolet
andoptical Fe II blends (Vestergaard & Wilkes 2001). I will
model the approach after that of Brothertonet al. (2002) as shown
in Figure 4. Host age and mass measurements will be a direct result
of thepopulation synthesis models (§ 5.2; Bruzual & Charlot
2003; Maraston 2005; Charlot & Bruzual2007) while the black
hole mass and Eddingon fraction estimates (§ 5.3 can be determined
from thequasar’s contribution to the composite spectra. The quasar
to host light ratio will help constrainthis modeling (§ 4.3).
5.2. Stellar Population Synthesis Modeling
Because the spectra of PSQs depend strongly on starburst
strength and time elapsed sincethe starburst, I will build a
library of quasar+A type spectra using stellar population
synthesis(or evolutionary population synthesis, EPS) models
(Bruzual & Charlot 2003; Maraston 2005;Charlot & Bruzual
2007). The technique utilizes the physical properties of age, star
formation
-
– 12 –
history, metallicity and dust content as the input parameters to
create model stellar populations.Bruzual & Charlot (2003)
models compute the spectral evolution of stellar populations of
differentmetallicities at ages between ∼1 × 105 and 2 × 1010 yr at
a resolution of 3 Å FWHM across thewavelength range of 3200 to
9500 Å. Maraston (2005) incorporates contributions from
thermallypulsating asymptotic giant branch (TP-AGB) stars into the
models for U to K bands with lowerspectral resolution (∼5-10 Å for
visible and 20-100 Å for near-IR). TP-AGB stars are expectedto
dominate SSPs at intermediate ages (0.3 < t < 2 Gyr). Thus,
the addition of the TP-AGBcontributions will be necessary in
accurately modeling PSQs, which have stellar populations of∼100s
Myrs. The models of Charlot & Bruzual (2007) are similar to
Bruzual & Charlot (2003)with the addition of contributions from
TP-AGB stars.
The simple stellar population (SSP) modeling technique assumes
that all stars are coeval withthe same chemical composition.
Stellar populations with star formation history can be expandedin
series of these SSPs. More complex models incorporate starbursts
that are not instantaneous(composite stellar populations, CSPs) but
begin at a certain time and decay exponentially charac-terized by
some e-folding time. Higher complexity modeling is gained when SSP
and CSP meet inlinear combination. I will model the host stellar
population as a linear combination of SSPs and/orCSPs.
The stellar contribution to the library will be built using EPS
models of different measuresof the input parameters (i.e., age,
star formation history, metallicity and dust content). Addedto each
model in the library will be quasar contributions (AGN power-law
with emission lines) ofvarying strength. It will be necessary to
rebin the library spectra to the same spectral resolutionas the
data in order to perform a least-squares error analysis between
models of the library andthe spectral data. Since we don’t know the
metallicity or dust content of the stellar populationsthere is some
uncertainty. The ability of the model to reproduce the strengths of
prominent stellarabsorption features in PSQ spectra is notable. In
particular, the equivalent width ratio of the Ca IIlines compared
to the Balmer lines is an effective age indicator over the age
range of these objects(∼50-500 Myr). The addition of the TP-AGB
star contribution in the modeling alleviates someof the
uncertainty. Also, as mentioned above the determination of quasar
to host ratio will helpconstrain this aspect of the modeling.
5.3. Black Hole Mass and Eddington Accretion Rate
Estimations of black hole mass MBH of broad-lined AGN are
performed via the reverberationmapping (RM) technique using data
from multiple epochs (Peterson 1993). This technique utilizesthe
variability properties of AGN to estimate the central mass using
the virial theorem, MBH ∼RV 2/G, where R is the size of the broad
line region (BLR; as estimated by the mean time delay, τ ,of
emission line relative to continuum variations, R = cτ), V is the
width of the variable emission-line. The active galaxy RM M -σ∗ (σ∗
is the stellar velocity dispersion of the host bulge
component)relation can then calibrated to the M -σ∗ for quiescent
galaxies assuming an appropriate scaling
-
– 13 –
factor, f (i.e., MBH = fRV 2/G), where f depends on the
geometry, kinematics and orientation ofthe BLR. In most instances,
RM is time-consuming and impractical. Fortunately, the luminosityof
the AGN (L) has been to found to scale with R and can be used as a
proxy for cτ in determiningthe MBH (Kaspi et al. 2000). I will rely
on scaling relations which relate RM results of physicalproperties
(i.e., M , R) to observable quantities (e.g., L, FWHM of C IV, Mg
II, Hβ or Hα andcorresponding continuum luminosities) applied to
single-epoch data. Mg II and/or Hα will beused due to their
location in the optical window for our objects (0.25 < z <
0.4) and the lack ofsignificant contamination from the
post-starburst stellar population (strong Balmer absorption).For
example, Dietrich & Hamann (2004) give:
MBHM�
= 3.37× 106(
14503000
)0.47(1+α) (λLλ(1450 Å)1044 ergs s−1
)0.47 [FWHM Mg II103 km s−1
]2, (1)
where α is determined from the quasar power-law continuum fit
(Fν ∝ να). Estimations of theEddington ratio (Lbol/LEdd) include
our MBH estimate and the conversion of the monochromaticluminosity
into a bolometric luminosity:
Lbol = f × λLλ(3000Å), (2)
where f ∼ 8.3 is the bolometric correction at 3000 Å and LEdd =
1.51 × 1038(MBH/M�) ergs s−1
(Krolik 1999). Determination of MBH is also useful in estimating
a bulge mass, Mbulge. This canbe done using the MBH -Mbulge
relation.
6. Infrared Properties of PSQs
6.1. Spitzer Data
We have obtained Spitzer mid-IR spectra for 16 of the 29 PSQs
using the two low-resolution(∼60-127) IRS modules. The wavelength
range spans 5 to 40 µm. During late Summer of 2007 Iattended the
IRS sections of the Spitzer Data Analysis Workshop #5. This
workshop covered thevarious pipeline processes for extraction (BCD,
Post-BCD, subtraction, extraction). The SPICE(Spitzer IRS Custom
Extraction) or SMART (Spitzer contributed software) packages will
be usedto re-extract and recalibrate the spectra.
6.2. “Warm” and “Cool” Contributions
Sanders et al. (1988) showed that ULIRGs with “warm” (fν(25
µm)/fν(60 µm) > 0.3) IRcolors have mid-IR emission (∼5-40 µm) of
over an order of magnitude stronger than their
“cooler”counterparts. This fraction of ULIRGs spans many types of
objects, including quasars and isthe premise for the evolutionary
connection between ULIRGs and quasars (Sanders et al.
1988).Furthermore, the mid-IR slope has become an AGN/starburst
diagnostic of the heating source of
-
– 14 –
Fig. 5.— Near to far-IR spectral models of a HyperLIRG, a normal
radio-quiet quasar and a star-forminggalaxy (Dale et al. 2006).
the dust radiating at those wavelengths. AGN have powerful UV
and X-ray emission which heatsthe surrounding dust to a range of
temperatures (even up to the sublimation temperature, Tsub ∼1500 K)
resulting in a shallow (warm) mid-IR power-law continuum.
Alternatively, stellar powersources do not heat dust to such high
temperatures resulting in a steeper mid-IR (cooler) continuumslope
(Figure 5). I propose to determine just how warm PSQs are.
6.3. PAH Features of Composite-types (PSQs)
Genzel et al. (1998) showed forbidden mid-IR emission line
ratios versus the 7.7 µm PAHemission equivalent width can be used
as a diagnostic to quantify energetic contributions fromAGN and
stellar sources. This can help constrain the quasar to stellar
light ratio. Recent researchhas improved upon this diagnostic by
utilizing the clean 6.2 µm feature which does not showcontamination
from the 10 µm Si absorption feature (Figure 6). The left panel of
Figure 6 clearlyshows why the distribution of the 6.2 µm feature
for star-forming objects, ULIRGs and broad-linedAGNs is a useful
diagnostic (Dale et al. 2006). To help test the evolutionary theory
of ULIRGsevolving into normal quasars via possible transitions
objects such as PSQs I will further explorethe previously mentioned
diagnostics. Quasars in general do not show PAH features, thus any
PAHfeatures found in the SEDs of PSQs presumably come from the host
galaxy, which could put themon a path between ULIRGs and normal
quasars.
-
– 15 –
Fig. 6.— Left, ratios of mid-IR forbidden lines as a function of
the 6.2 µm PAH feature equivalent widthfrom Dale et al. (2006).
SINGS data are displayed in red with 1 σ error bars based on the
statisticaluncertainties; archival data without error bars are
indicated with black symbols. The dotted lines are linearmixing
models of a “pure” AGN and a “pure” star-forming source. The dashed
line in the top panel is asimilar mixing model first presented by
Genzel et al. (1998). The solid lines and Roman numerals
delineateregions distinguished by Seyfert galaxies, LINERs, star
formation, etc. Right, distributions of the equivalentwidth of the
clean 6.2 µm PAH feature (Spitzer data) for star-forming objects,
ULIRGs and broad-line AGNs(Dale 2006, unpublished). There are more
star-forming objects with EW > 1.4 (not shown).
-
– 16 –
7. How Do PSQs Fit in the Grand Scheme of Things?
7.1. Comparisons of PSQs with Other Classes of Objects
Once characteristics of the class are defined, a comparative
study with other classes (ULIRGs,post-starburst galaxies, normal
quasars, normal galaxies) will be performed. The following is
abrief overview of the proposed science and the comparisons that
will be made:
• Host galaxy demographics - Is there a starburst (SB)
mass/bulge mass relation? ASB mass/BH mass relation? If so, how do
they compare with other classes that show SBs(ULIRGs,
post-starburst galaxies)? Do the host galaxy stellar population
ages place them inan evolutionary timeline between ULIRGs, “normal”
quasars and red ellipticals?
• Quasar demographics - Same MBH -Mbulge-MSB question as above.
How do the BH ac-cretion rate of each class compare?
• Quasar to host galaxy light ratio - Where does the percent
contribution of quasar inPSQs put them with respect to ULIRGs,
starburst galaxies, normal quasars and galaxies?
• IR spectral shape - If stellar dominated classes are cool and
AGN dominated sources arewarm what color should PSQs have, do they
lie there and is that in accord with the percentAGN contribution
found?
• PAH feature strength - In general PAH features are weak/not
found around dominantAGN sources. Given the percent AGN
contribution of PSQs do their PAH strengths placethem in logical
sequence with AGN and starburst classes?
7.2. Inclusion of PSQs and Other Classes in Theoretical
Modeling
It will also be promising to compare results and data from other
classes with current theoret-ical models and simulations. This
thesis will be testing Hopkins et al. (2007) detailed theory
ofhierarchical clustering (Figure 7). As mentioned in § 1.2, Di
Matteo et al. (2005) simulated thesimultaneous star formation and
BH growth in galaxy-galaxy collisions. Results showed that
themerger/interaction triggers inflows which feed (turn-on) the
quasar and starburst. But, the energyreleased by the quasar blows
away the feeding material quenching the star formation and black
holegrowth. Snapshots of the simulation with and without central BH
are shown in Figure 8, whilethe star formation rate, BH mass and
accretion rate as a function of time are shown in Figure 9.If PSQs
are objects at blowout one might expect to see a trend correlating
BH accretion rate,mass or star formation rate with an appropriate
time since merger indicator (say, dominant stellarpopulation age).
If the evolutionary sequence (ULIRG to PSQ to normal quasar or
elliptical) iscorrect, then we should see trends there as well
placing ULIRGs with the most dominant SB and
-
– 17 –
youngest stellar populations/AGN and ellipticals/normal quasars
with the least dominant SB andoldest stellar populations/AGNs.
8. Other
Research is neither straightforward nor predictable, hence the
inclusion of this last section. Itmay become necessary to delve
deeper into a result whose meaning is still unclear. Moreover,
anadded study increasing the wavelength regime might also prove
useful in tying down observed cor-relations. Here are a few
“unforeseens” that I foresee; those I don’t will be left to your
imagination:
• Affiliated Research
– GALEX UV Archival Photometry Data - Constraints on the SED and
quasar contribu-tions in the UV. Constrain UV excesses from young
stellar population in the UV.
– Chandra X-Ray Data - X-Ray properties of PSQs, as related to
AGN accretion rate,lifetime and fuel availability.
• Follow-up
– Herschel Proposal and Data - Far-IR, ionic fine structure
lines probing highly ionizedgas excited by AGN and photoionized gas
from young stellar populations.
– Feedback Modeling - Closer involvement with theorists working
on merger induced feed-back spurring M -σ∗ correlations.
– Other proposed science as seen fit.
• Funding
– Through Summer 2008 - NSF K-12 Science Enrichment Teaching
– Winter 2008 - Propose for NASA based funding (NESSF, GSRP and
Wyoming SpaceGrant)
– Fall 2008 to Summer 2009 - NSF or NASA based funding, Summer
Teaching
– Winter 2009 - Propose for NASA based funding
– Fall 2009 to Summer 2010 - NASA based funding
– Fall 2010 to Summer 2011 - Advisor or NASA based funding (if
necessary), part-timeTA
-
– 18 –4Hopkinsetal.
FIG.1.—
Anschematicoutlineofthephasesofgrowthina“typical”galaxyundergoingagas-richmajormerger.ImageCredit:(a)NOAO/AURA/NSF;(b)REUprogram/NOAO/AURA/NSF;(c)NASA/STScI/ACS
ScienceTeam;(d)Optical(left):NASA/STScI/R.P.vanderMarel&J.Gerssen;X-ray(right):NASA/CXC/MPE/S.Komossaetal.;(e)Left:J.Bahcall/M.Disney/NASA;Right:GeminiObservatory/NSF/University
ofHawaiiInstituteforAstronomy;(f)J.Bahcall/M.Disney/NASA;(g)F.Schweizer(CIW/DTM);(h)NOAO/AURA/NSF.
Fig
.7.
—Sc
hem
atic
evol
utio
nof
gala
xies
usin
ghi
erar
chic
alm
echa
nism
for
build
ing
gala
xies
(Hop
kins
etal
.20
07).
-
– 19 –
–11–
Fig
.8.
—Sn
apsh
ots
ofth
esi
mul
ated
tim
eev
olut
ion
inm
erge
rsof
two
gala
xies
wit
h(t
op)
and
wit
hout
(bot
tom
)bl
ack
hole
sby
DiM
atte
oet
al.
(200
5).
The
full
tim
ese
quen
cefo
rth
issi
mul
atio
nca
nbe
view
edin
the
Supp
lem
enta
ryV
ideo
avai
labl
eon
line.
Inbo
thca
ses,
four
snap
shot
sat
diffe
rent
tim
esin
the
sim
ulat
ions
are
show
n.T
heim
ages
visu
aliz
eth
epr
ojec
ted
gas
dist
ribu
tion
inth
etw
oga
laxi
es,
colo
ur-
code
dby
tem
pera
ture
(blu
eto
red)
.T
heco
llidi
ngga
laxi
esha
veth
esa
me
init
ial
mas
sco
rres
pond
ing
toa
viri
alve
loci
tyofV
vir
=(M
tot
×10GH
0)1
/3
=16
0km
/s,
and
cons
ist
ofan
exte
nded
dark
mat
ter
halo
,a
stel
lar
bulg
e,an
da
disk
mad
eup
ofst
ars
and
20%
gas.
The
first
snap
shot
(t=
1.1
Gyr
)sh
ows
the
syst
ems
afte
rth
efir
stpa
ssag
eof
the
two
gala
xies
.T
hese
cond
snap
shot
(t=
1.4
Gyr
)de
pict
sth
ega
laxi
esdi
stor
ted
byth
eir
mut
ual
tida
lin
tera
ctio
n,ju
stbe
fore
they
mer
ge.
The
peak
inth
est
arfo
rmat
ion
and
blac
kho
leac
cret
ion
(see
also
Fig
ure
9)is
reac
hed
atth
eti
me
ofth
eth
ird
snap
shot
(t=
1.6
Gyr
),w
hen
the
gala
xies
final
lyco
ales
ce.
At
this
tim
e,a
stro
ngw
ind
driv
enby
feed
back
ener
gyfr
omth
eac
cret
ion
expe
lsm
uch
ofth
ega
sfr
omth
ein
ner
regi
ons
inth
esi
mul
atio
nw
ith
blac
kho
les.
Fin
ally
,th
ela
stsn
apsh
ots
show
the
syst
ems
afte
rth
ega
laxi
esha
vem
erge
d(t
=2.
5G
yr),
leav
ing
behi
ndqu
asi-
stat
icsp
hero
idal
gala
xies
.In
the
sim
ulat
ion
wit
hbl
ack
hole
s,th
ere
mna
ntis
very
gas
poor
and
has
littl
ega
sle
ftde
nse
enou
ghto
supp
ort
ongo
ing
star
form
atio
n.T
his
high
light
sth
atth
epr
esen
ceof
supe
rmas
sive
blac
kho
les,
whi
chac
cret
efr
omth
esu
rrou
ndin
gga
san
dhe
atit
wit
hth
eas
soci
ated
feed
back
ener
gy,
dram
atic
ally
alte
rsth
em
erge
rre
mna
nt.
-
– 20 –
– 13 –
Fig. 9.— Star formation rate (top), black hole activity (middle)
and black hole growth (bottom) plotted asa function of time during
the galaxy-galaxy merger of Figure 8 by Di Matteo et al. (2005).
The three linesin each panel correspond to models with galaxies of
virial velocity Vvir = 80, 160, and 320 km/s, (bottomto top lines
in each panel, also labeled in the bottom panel). For comparison
also shown is the evolution ofthe star formation rate for the model
without a black hole of Figure 8 (dashed line for the Vvir = 160
km/sgalaxy). We note, in particular, that owing to AGN feedback,
the peak amplitude of the starburst duringthe merger is lowered by
a significant factor. The black solid circles in the individual
panels identify thetimes of the corresponding snapshots shown in
Figure 8.
-
– 21 –
9. Timeline
Some research is currently underway (HST analysis of images, PSF
fitting/subtractions) andmuch of the data is already in hand. No
doubt, pursuing a thesis involves analyzing data, writingpapers and
proposals, making new discoveries, following up on those that are
interesting and/orperplexing and becoming an expert in the field by
keeping up to date on the literature and obser-vational techniques.
Nevertheless, it is possible to complete my thesis by the
Spring/Summer of2010. This allows roughly six months per
sub-project plus an extra six months for “Other” andwrapping up,
e.g.,
• Winter 2008 - Candidacy Exam, HST and IRTF Imaging and PSF
subtraction, Begin PaperI (Imaging of HST Sample), Begin
Dissertation (Chapters 1 and 2)
• Spring 2008 - Continue with HST data analysis (AGN to host
galaxy light fraction), Com-plete and Submit Paper I to
Astrophysical Journal, Continue with Dissertation (CompleteChapters
1 and 2)
• Summer 2008 - Reduce Kitt Peak 4-m data, Recalibrate Keck
data, EPS of optical dataconstrained by AGN to host galaxy light
fraction, Chandra proposal (Other), Begin Paper II(EPS of optical
data), Continue with Dissertation (Chapter 3)
• Fall 2008 - Travel?: Porto Alegre, Brazil (M. Brotherton,
Collaboration), Riverside, CA (G.Canalizo, EPS Tutorial), Finish
analysis (AGN and starburst demographics) for and writingPaper II
(Submit to ApJ), Prepare results for AAS meeting, Continue with
Dissertation(Complete Chapter 3)
• Winter 2009 - Present results at Long Beach, CA AAS meeting,
Reduce Spitzer data,IR diagnostics (Color and PAH features), Begin
Paper III (IR diagnostics), Continue withDissertation (Chapter
4)
• Spring 2009 - Finish IR diagnostics analysis (Color and PAH
features), Complete PaperIII (Submit to ApJ), Continue with
Dissertation (Complete Chapter 4), Travel? China (Z.Shang,
Collaboration)
• Summer 2009 - Collect data of other classes of objects and
modeling, Herschel proposal(Other), Begin writing Paper IV
(Comparisons of PSQs with objects and models), Teach asummer
Astronomy 1050 Course, Continue with Dissertation (Chapters 5 and
Other)
• Fall 2009 - Finish analysis for and writing Paper IV, Job
search and applications, Continuewith Dissertation (Complete
Chapter 5), Begin final analysis
• Winter 2010 - Wrap up “Other” if necessary, Job search and
applications, Continue withDissertation (Complete Final Chapters),
Prepare for Dissertation Defense and AAS Meeting
• Spring 2010 - Defend Dissertation and present at AAS Meeting,
Get a Job!
-
– 22 –
REFERENCES
Bennert, N., Canalizo, G., Jungwiert, B., Stockton, A.,
Schweizer, F., Peng, C. Y., & Lacy, M.2008, ArXiv e-prints,
801
Brotherton, M., Cales, S., Stoll, R., & Canalizo, G. 2008b,
ApJ
Brotherton, M., Grabelsky, M., Canalizo, G., vanBreugel, W.,
Filippenko, A., Croom, S., Boyle,B., & Shanks, T. 2002, PASP,
114, 593
Brotherton, M., Stoll, R., Paul, C., Diamond-Stanic, A., Shang,
Z., Cales, S., Ganguly, R., Canalizo,G., & Vanden Berk, D.
2008a, ApJ
Brotherton, M., vanBreugel, W., Stanford, S., Smith, R., Boyle,
B., Miller, L., Shanks, T., Croom,S., & Filippenko, A. 1999,
ApJ, 520, L87
Bruzual, G. & Charlot, S. 2003, MNRAS, 344, 1000
Burkert, A. & Silk, J. 2001, ApJ, 554, L151
Canalizo, G., Bennert, N., Jungwiert, B., Stockton, A.,
Schweizer, F., Lacy, M., & Peng, C. 2007,ApJ, 669, 801
Canalizo, G., Stockton, A., Brotherton, M. S., & van
Breugel, W. 2000, AJ, 119, 59
Charlot, S. & Bruzual, G. 2007, MNRAS
Cox, T. J., Dutta, S. N., Di Matteo, T., Hernquist, L., Hopkins,
P. F., Robertson, B., & Springel,V. 2006, The Astrophysical
Journal, 650, 791
Croom, S. M., Smith, R. J., Boyle, B. J., Shanks, T., Loaring,
N. S., Miller, L., & Lewis, I. J. 2001,MNRAS, 322, L29
Dale, D. A. et al. 2006, ApJ, 646, 161
Di Matteo, T., Springel, V., & Hernquist, L. 2005, Nature,
433, 604
Dietrich, M. & Hamann, F. 2004, ApJ, 611, 761
Ferrarese, L. & Merritt, D. 2000, ApJ, 539, L9
Gebhardt, K. et al. 2000a, ApJ, 539, L13
Gebhardt, K.et al. 2000b, ApJ, 543, L5
Genzel, R., Lutz, D., Sturm, E., Egami, E., Kunze, D., Moorwood,
A., Rigopoulou, D., Spoon, H.,Sternberg, A., TacconiGarman, L.,
Tacconi, L., & Thatte, N. 1998, ApJ, 498, 579
Granato, G. L., De Zotti, G., Silva, L., Bressan, A., &
Danese, L. 2004, ApJ, 600, 580
-
– 23 –
Hopkins, P. F., Hernquist, L., Cox, T. J., & Keres, D. 2007,
ArXiv e-prints, 706
Kaspi, S., Smith, P. S., Netzer, H., Maoz, D., Jannuzi, B. T.,
& Giveon, U. 2000, ApJ, 533, 631
Kauffmann, G. & Haehnelt, M. 2000, MNRAS, 311, 576
Kormendy, J. & Richstone, D. 1995, ARA&A, 33, 581
Krolik, J. 1999, in Active Galactic Nuclei, Vol. 119, 306–+
Magorrian, J., Tremaine, S., Richstone, D., Bender, R., Bower,
G., Dressler, A., Faber, S. M.,Gebhardt, K., Green, R., Grillmair,
C., Kormendy, J., & Lauer, T. 1998, AJ, 115, 2285
Maraston, C. 2005, MNRAS, 362, 799
Peterson, B. M. 1993, PASP, 105, 247
Sanders, D. B. & Mirabel, I. F. 1996, ARA&A, 34, 749
Sanders, D. B., Soifer, B. T., Elias, J. H., Neugebauer, G.,
& Matthews, K. 1988, ApJ, 328, L35
Schneider, D. P.et al. 2002, AJ, 123, 567
Springel, V., Di Matteo, T., & Hernquist, L. 2005, ApJ, 620,
L79
Toomre, A. 1977, in Evolution of Galaxies and Stellar
Populations, ed. B. M. Tinsley & R. B.Larson, 401–+
Tremaine, S. et al. 2002, ApJ, 574, 740
Tremonti, C., Moustakas, J., & DiamondStanic, A. 2007, ApJ,
663, L77
Veilleux, S. 2006, NewAR, 50, 701
Vestergaard, M. & Wilkes, B. J. 2001, ApJS, 134, 1
Zabludoff, A. I., Zaritsky, D., Lin, H., Tucker, D., Hashimoto,
Y., Shectman, S. A., Oemler, A., &Kirshner, R. P. 1996, ApJ,
466, 104
This preprint was prepared with the AAS LATEX macros v5.2.
-
– 24 –
Table 1. HST Sample Data Sets
Imaging Spectroscopy
Sample ID SDSS Target z HSTa IRTFb Keckc /KPNOd Spitzere
74 212843+002435 0.346 X X XKeck X43 124833+563507 0.266 X X
XKeck X10 040210−054630 0.270 X X XKeck · · ·38 115159+673604 0.274
X X XKPNO · · ·58 145640+524727 0.277 X X XKPNO X39 115355+582442
0.319 X X XKPNO X76 231055−090107 0.364 X X XKeck X07 023700−010130
0.344 X X XKeck X04 020258−002807 0.339 X X XKeck X14 075549+321704
0.420 X X XKeck X61 154534+573625 0.268 X X XKPNO X05 021447−003250
0.349 X X XKeck X13 075521+295039 0.334 X X XKeck X17 081018+250921
0.263 X X XKeck X12 075045+212546 0.408 X X XKeck X11 074621+335040
0.284 X X XKeck X32 105816+102414 0.275 X X XKPNO X69 170046+622056
0.276 X X XKPNO X65 164444+423304 0.317 X X XKPNO · · ·41
123043+614821 0.324 X X XKPNO · · ·71 210200+000501 0.329 X X XKeck
· · ·73 211838+005640 0.384 X X XKeck · · ·72 211343−075017 0.420 X
X XKeck · · ·75 230614−010024 0.267 X X XKeck · · ·80 234403+154214
0.288 X X XKeck · · ·02 005739+010044 0.253 X X XKeck · · ·59
145658+593202 0.326 X X · · · · · ·78 233430+140649 0.363 X X XKeck
· · ·01 003043−103517 0.296 X X XKeck · · ·
aHST/ACS WFC F606W broad V-band filter with spatial resolution
of ∼ 0.05”.bNASA’s IRTF/NSFCAM2 K-band (near-IR) imaging at
0.15”/pixel.
cKeck LRIS spectroscopy covering 3100 Å to 1 µm. The 300/5000
grating on the red
side and 400/3400 grism on the blue yield 2.54 Å/pixel and
1.071 Å/pixel, respectively.
dKPNO-4m RC-Spec using the KPC-10A 316/4000 grating spanning
3200 Å to 7200 Å
with a dispersion of 2.75 Å/pixel.
eSpitzer/IRS spectroscopy using the two low resolution (R ∼
60-127) modules covering5.2 to 38 µm.