Differential Morphology Between Rest-frame Optical and UV Emission from 1.5 < z < 3 Star-forming Galaxies

Abstract

We present the results of a comparative study of the rest-frame optical and
rest-frame ultraviolet morphological properties of 117 star-forming galaxies
(SFGs), including BX, BzK, and Lyman break galaxies with B<24.5, and 15 passive
galaxies in the region covered by the Wide Field Camera 3 Early Release Science
program. Using the internal color dispersion (ICD) diagnostic, we find that the
morphological differences between the rest-frame optical and rest-frame UV
light distributions in 1.4<z<2.9 SFGs are typically small (ICD~0.02). However,
the majority are non-zero (56% at >3 sigma) and larger than we find in passive
galaxies at 1.4<z<2, for which the weighted mean ICD is 0.013. The lack of
morphological variation between individual rest-frame ultraviolet bandpasses in
z~3.2 galaxies argues against large ICDs being caused by non-uniform dust
distributions. Furthermore, the absence of a correlation between ICD and galaxy
UV-optical color suggests that the non-zero ICDs in SFGs are produced by
spatially distinct stellar populations with different ages. The SFGs with the
largest ICDs (>~0.05) generally have complex morphologies that are both
extended and asymmetric, suggesting that they are mergers-in-progress or very
large galaxies in the act of formation. We also find a correlation between
half-light radius and internal color dispersion, a fact that is not reflected
by the difference in half-light radii between bandpasses. In general, we find
that it is better to use diagnostics like the ICD to measure the morphological
properties of the difference image than it is to measure the difference in
morphological properties between bandpasses.

Full text

Page 1

arXiv:1010.1525v1 [astro-ph.CO] 7 Oct 2010
Draft version October 8, 2010
Preprint typeset using LATEX style emulateapj v. 11/10/09
DIFFERENTIAL MORPHOLOGY BETWEEN REST-FRAME OPTICAL AND UV EMISSION FROM
1.5 < z < 3 STAR-FORMING GALAXIES
Nicholas A. Bond, Eric Gawiser
Physics and Astronomy Department, Rutgers University, Piscataway, NJ 08854-8019, U.S.A.
and
Anton M. Koekemoer
Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218, U.S.A.
Draft version October 8, 2010
ABSTRACT
We present the results of a comparative study of the rest-frame optical and rest-frame ultraviolet
morphological properties of 117 star-forming galaxies (SFGs), including BX, BzK, and Lyman break
galaxies with B < 24.5, and 15 passive galaxies in the region covered by the Wide Field Camera 3
Early Release Science program. Using the internal color dispersion (ICD) diagnostic, we find that the
morphological differences between the rest-frame optical and rest-frame UV light distributions in 1.4 <
z < 2.9 SFGs are typically small (ICD ∼ 0.02). However, the majority are non-zero (56% at > 3σ)
and larger than we find in passive galaxies at 1.4 < z < 2, for which the weighted mean ICD is 0.013.
The lack of morphological variation between individual rest-frame ultraviolet bandpasses in z ∼ 3.2
galaxies argues against large ICDs being caused by non-uniform dust distributions. Furthermore, the
absence of a correlation between ICD and galaxy UV-optical color suggests that the non-zero ICDs in
SFGs are produced by spatially distinct stellar populations with different ages. The SFGs with the
largest ICDs (? 0.05) generally have complex morphologies that are both extended and asymmetric,
suggesting that they are mergers-in-progress or very large galaxies in the act of formation. We also
find a correlation between half-light radius and internal color dispersion, a fact that is not reflected
by the difference in half-light radii between bandpasses. In general, we find that it is better to use
diagnostics like the ICD to measure the morphological properties of the difference image than it is to
measure the difference in morphological properties between bandpasses.
Subject headings: cosmology: observations — galaxies: formation – galaxies: high-redshift – galaxies:
structure
1. INTRODUCTION
The Hubble Sequence of galaxies (Hubble 1936)
is seen out to z
∼
1.5 (Glazebrook et al. 1995;
van den Bergh et al.1996;
Brinchmann et al. 1998; Lilly et al. 1998; Simard et al.
1999; van Dokkum et al. 2000; Stanford et al. 2004;
Ravindranath et al. 2004) beyond which high-redshift
galaxies typically appear compact and irregular (e.g.
Giavalisco et al. 1996; Lowenthal et al. 1997; Dickinson
2000; van den Bergh 2001).
were initially identified using the Lyman-break tech-
nique (Steidel et al. 1996), wherein z > 2.5 galaxies
are selected using a flux discontinuity in the continuum
(e.g. a U-band dropout) caused by absorption from
intervening neutral hydrogen (e.g. Rafelski et al. 2009;
Yan et al. 2009; Oesch et al. 2010; Bunker et al. 2009;
Hathi et al. 2010). Subsequent techniques, such as BzK
(Daddi et al. 2004) and ‘BX’ (Adelberger et al. 2004)
color selection, as well as narrow-band selection of Lyα
emitters (Hu & McMahon 1996), have allowed for the
photometric identification of star-forming galaxies over
a wider redshift range.
The morphological properties of high-redshift galaxies
are quantified using a wide range of techniques, including
the concentration and asymettry indices (Conselice et al.
Griffiths et al.1996;
Galaxies at high redshift
nbond@physics.rutgers.edu
gawiser@physics.rutgers.edu
koekemoer@stsci.edu
2005), Gini coefficients (Lotz et al. 2006), and S´ ersic pro-
file fitting (e.g. Cassata et al. 2005), among others. At
2 ? z ? 3.5, star-forming galaxy (SFG) sizes range from
< 1 kpc to ∼ 5 kpc, with the largest often exhibiting mul-
tiple photometric components (e.g. Bouwens et al. 2004;
Ravindranath et al. 2006; Oesch et al. 2009). Morpho-
logical analyses have revealed that most of these sys-
tems are disturbed and disk-like (i.e., with exponential
light profiles), with only ∼ 30% having light profiles
consistent with galactic spheroids (e.g., Ferguson et al.
2004; Lotz et al. 2006; Ravindranath et al. 2006). Be-
cause high-redshift galaxies are typically faint and
clumpy/irregular, it can be difficult to quantify their
morphology in a meaningful way. Using a sample of 97
Lyman-α emitters, Bond et al. (2009) found that even
the half-light radius could not be accurately measured
without S/N ? 30 within a 0.′′6 aperture.
In 2003, Papovich et al. (2003, P03) introduced a dif-
ferential morphological diagnostic known as the inter-
nal color dispersion (ICD) that could be used to quan-
tify the morphological differences between bandpasses
for astronomical objects. When applied to galaxies, this
proved to be particularly useful for comparing the rest-
frame ultraviolet light distribution, which often appears
clumpy even in low-redshift galaxies, to the rest-frame
optical light distribution. If the young stars in a galaxy
are distributed differently from the old stars (as is usu-
ally the case at low redshift), then that galaxy will

Page 2

2 Bond et al.
have a large ICD. Similarly, inhomogeneous dust dis-
tributions can lead to spatial variations in the level of
obscuration of rest-frame UV light, leading to a large
ICD. Their intial study focused on low-redshift galax-
ies and found that mid-type spirals exhibited the largest
ICDs (ξ ∼ 0.2), irregular galaxies were intermediate
(ξ ∼ 0.1), and early-type galaxies had negligible mor-
phological variation between bands (ξ ∼ 0). The authors
also concluded that ICDs could be reliably measured for
high-redshift galaxies so long as the angular resolution
(? 0.5 beam kpc−1≃ 0.′′1 at z = 2) and signal-to-noise
(S/N ? 80) were sufficient.
Following up on this exploratory study, Papovich et al.
(2005)(P05) computed the ICDs of separate flux-limited
galaxy samples at z ∼ 1 and z ∼ 2.3 using a combi-
nation of Hubble Space Telescope (HST) WFPC2 and
NICMOS imaging. The authors found that the ICDs of
z ∼ 1 galaxies were generally larger and more varied than
those of their higher-redshift counterparts, with the for-
mer sample including many spiral galaxies and the latter
being dominated by irregulars and interacting systems.
They interpret these changes as being due to increas-
ing stellar population heterogeneity at low redshift, with
gaseous disks forming around older spheroids. In addi-
tion, they speculated that the high-redshift galaxies with
large ICDs were actually mergers-in-progress based upon
their large apparent asymmetries.
With the installation of the Wide Field Camera 3
(WFC3) on HST, we now have a new, improved win-
dow into the rest-frame optical light of galaxies at
1 ? z ? 3. The deep NIR images being taken as
part of the WFC3 Hubble Ultra Deep Field 2009 (GO
11563: PI Illingworth) and WFC3 Early Release Science
(ERS, Windhorst et al. 2010) programs allow for a de-
tailed study of high-redshift galaxy morphologies down
to H160∼ 26 mag. In this paper, we present the results
of a comparative study of the rest-frame optical and rest-
frame ultraviolet morphological properties of 132 SFGs
in the WFC3 ERS region of GOODS-S, using the ICD
as our primary diagnostic. Throughout we will use AB
magnitudes and assume a concordance cosmology with
H0 = 71 km s−1Mpc−1, Ωm = 0.27, and ΩΛ = 0.73
(Spergel et al. 2007). With these values, 1′′= 8.2 physi-
cal kpc at z = 2.5.
2. DATA AND SAMPLES
2.1. Advanced Camera for Surveys Imaging
In the Chandra Deep Field-South, the southern half
of the GOODS survey (Giavalisco et al. 2004) covers ∼
160 arcmin2of sky and has HST/ACS observations in
the B435, V606, I775, and z850filters. For this study, we
use only the I775 image, which was multidrizzled to a
pixel scale of 60 mas. The effective exposure time of this
survey is variable across the GOODS area, but for 1.′′2
fixed aperture, a typical I775-band, 5σ detection limit is
mAB= 27.4.
2.2. Wide Field Camera 3 Imaging
In order to probe the rest-frame optical light of z ∼ 2.3
galaxies, we use HST/WFC3 F160W images taken as
part of the WFC3 ERS. The F160W imaging consisted
of 20 orbits over ten visits to the GOODS-S field (Pro-
gram 11563), leading to a total of 60 exposures. The
TABLE 1
Galaxy Sample Properties
SampleNumber Redshift RangeReference
SFGmain
SFGhighz
PassGal
100
17
15
1.4 < z < 2.9
2.9 < z < 3.5
1.4 < z < 2.0
1
1
2
References.
Cameron et al. 2010
— (1) Balestra et al.2010; (2)
exposures were then reduced and calibrated (Koekemoer
et al.2010, in prep.), incorporating SPARS100 dark
frames and correcting for residual gain differences be-
tween the mosaic quadrants. This also included the re-
moval of large-scale scattered light residuals and satellite
trails. The exposures were then combined using Mul-
tiDrizzle (Koekemoer et al. 2002) using a pixfrac of 0.8
and a square kernel, to produce final drizzled images with
a pixel scale of 60 mas. For a 1.′′2 fixed aperture, a typ-
ical H160-band, 5σ detection limit is mAB = 27.2. In
order to ensure accurate alignment of the IR imaging
with the GOODS ACS imaging, the WFC3 exposures
were individually aligned to the GOODS-S v2.0 z-band
catalog and to each other. Even for single exposures, the
number of matching sources was ∼ 500 and the final as-
trometric solution appears to be robust to ∼ 10 mas or
better. Further details on this reduction are presented in
a forthcoming paper (Koekemoer et al. 2010, in prep.)
2.3. Galaxy Samples
SFG sample Ourprimaryisdrawn fromthe
VLT/VIMOS spectroscopic observations described in
Popesso et al. (2009) and Balestra et al. (2010). Their
pointings targeted high-redshift galaxies in GOODS-S
with B < 24.5, including U-band dropouts, BzK color-
selected objects (Daddi et al. 2004), sub-U-dropouts
(similar to ’BX’ selection, Adelberger et al. 2004),
and X-ray sources in the Giacconi et al. (2002) and
Lehmer et al. (2005) catalogs. For this analysis, we ex-
clude objects with redshift quality flag C (poor quality),
objects targeted as X-ray sources, and objects satisfying
the BzK color selection for passively-evolving galaxies.
In a further attempt to exclude unobscured AGN, we re-
move from our sample six objects that are within 2′′of
any X-ray source in the expanded catalogs of Virani et al.
(2006) and Luo et al. (2008).
P03 showed that the internal color dispersion is most
effective at identifying the signatures of heterogeneous
stellar populations when the two images cover opposite
sides of the Balmer and 4000˚ A breaks. Since we are us-
ing images taken through the F160W and F775W filters
on HST, we restrict the main SFG sample (SFGmain) to
galaxies in the redshift range, 1.4 < z < 2.9, of which
there are 100 BX and BzK galaxies in the WFC3 ERS
region. We also analyze a control sample (SFGhighz)
of 17 Lyman break galaxies with 2.9 < z < 3.5, for
which the I775and H160filters both contain light blue-
ward of the 4000˚ A break in the galaxies’ rest frame. For
these galaxies, both filters are tracing light from young
stars in the galaxies rest frame and we only expect a
large ICD if there is a large column of inhomogeneously
distributed dust. Finally, we include a passive galaxy

Page 3

Differential Morphologies of 1.5 < z < 3 SFGs3
TABLE 2
Galaxy Samples
Number Sampleαδz
447
455
462
479
482
SFGmain
SFGmain
SFGmain
SFGmain
SFGhighz
3:32:01.117
3:32:01.302
3:32:01.509
3:32:02.067
3:32:02.167
−27:44:04.830
−27:42:43.973
−27:44:22.574
−27:45:13.878
−27:42:30.428
2.6355
2.2998
2.5119
2.3131
3.2630
*This table is only a stub. A manuscript with complete tables
is available at http://www.nicholasbond.com/Bond1004.pdf
sample (PassGal), containing 15 objects found in the
WFC3 ERS region using a YHVz color selection tech-
nique (Cameron et al. 2010). All of these passive galax-
ies have 1.4 < z < 2.
The sample properties are summarized in Table 1 and
the objects are listed in Table 2, including a total of 132
galaxies in the WFC3 ERS. For detailed information on
the physical properties of the objects in our sample, see
the references given in the table. In this paper, we will
refer to SFGs according to their line number in version
2.0 of the VIMOS master catalog (Balestra et al. 2010),
excepting the passive galaxies, which will be denoted P1-
P15. sAlthough it is listed in Table 2, we exclude object
1379 from the majority of our analysis because it was not
detected in either the H160image or the I775image.
3. METHODOLOGY
3.1. Fixed-aperture magnitudes and half-light radii
Because star-forming galaxies are often clumpy, con-
sist of multiple components, and may be interacting with
nearby galaxies, there is some ambiguity in choosing the
appropriate aperture for morphological analysis. We will
use a simple approach in this paper, measuring all quan-
tities (with the exception of the internal color dispersion,
see Section 3.2) within a fixed 1.′′2 aperture.
Our methodology for measuring object magnitudes,
centroids, and fixed-aperture half-light radii is described
in detail in Bond et al. (2009). All of these measurements
are performed on the original drizzled images, prior to
the point spread function (PSF) convolution described
below.
Using SourceExtraction
Bertin & Arnouts 1996), we estimate the maximum
contamination by interloping sources within the 1.′′2
apertures to be ∼ 17% based on the background sky
density of sources with H160< 26. However, because the
SFGs in our sample were selected using ground-based
photometry, the contamination rate is probably much
lower because the presence of a bright interloper within
∼ 1′′would generally have caused galaxies to fall out of
their respective color selection regions.
software(SExtractor,
3.2. Internal color dispersion
The internal color dispersion attempts to quantify the
differences in the distribution of an object’s light be-
tween two observed bandpasses without making model-
dependent assumptions. It was first presented in P03
Fig. 1.— Internal color dispersion as a function of centroid offset
for a series of Monte Carlo simulations of an SFG in the WFC3
ERS.
and, for a noise-free cutout, is given by,
ξ(I1,I2) =
N ?
i=0(I2,i− αI1,i− β)2
N ?
i=0(I2,i− β)2
, (1)
where the sum is performed over N pixels in a chosen
aperture, I1,i and I2,i are the fluxes in pixel i, α is the
ratio of the total fluxes between cutout 2 and cutout 1,
and β is the difference in the background between the two
cutouts. Since we perform SExtractor sky subtraction
on all cutouts prior to analysis (see Bond et al. 2009),
we will assume β = 0 for all of our measurements.
If we wish to accurately compute a differential mor-
phological diagnostic between two observed bandpasses,
we must ensure that (1) the effective PSF is the same
in the two images and (2) there is no astrometric offset.
If the PSF is well approximated by a Gaussian in both
frames, the former requirement can be satisfied by con-
volving the lower-resolution image with a Gaussian that
has σ =
?σ2
H160-band imaging is not well approximated by a Gaus-
sian (Bond et al. 2007), so we must convolve each image
with the PSF of the other. We estimated the PSF of each
image using a stack of stars from Altmann et al. (2006)
that fell within the WFC3 ERS region and had temper-
atures consistent with K and M-type stars, a regime in
which photometric confusion with galaxies is minimal.
We further restricted our stack to 29 stars that were iso-
lated (i.e., the only photometric component within 0.′′6),
unambiguously stellar (SExtractor stellarity > 0.9), and
unsaturated. Experiments with stars of a variety of spec-
tral types reveal that the internal color dispersion is in-
2− σ2
1. Unfortunately, the PSF of the WFC3

Page 4

4Bond et al.
sensitive to the spectral shape of the PSF across the H160
filter, with ICDs varying by < 10% of the typical ICD
uncertainty in SFGmain.
In real images of high-redshift galaxies, the terms in
Equation 1 will have non-negligible contributions from
noise in the sky background. In order to estimate the
amplitude of this contribution, we extracted 400 cutouts
from random positions on the F160W and F775W im-
ages. Cutouts were discarded if they contained any de-
tected sources and we computed a noise term,
ξn(α,N) = Median(
N
?
i=0
I2
2,i) − 2αMedian(
N
?
i=0
I1,iI2,i)
+α2Median(
N
?
i=0
I2
1,i),
(2)
where the medians were computed over the sample of
blank cutouts. This was then subtracted from the nu-
merator and denominator of Equation 1 when computing
the ICDs of the objects in our sample.
We extracted 5′′×5′′cutouts from the HST/WFC3
and HST/ACS images, each centered at the flux-
weighted centroid of all SExtractor detections within 1.′′2
of the object positions reported in Balestra et al. (2010)
and Cameron et al. (2010).
uncertainties on the ICD measurements, we computed
them within circular apertures, ranging from 0.′′6 to 1.′′5
in radius, that were selected on an object-by-object basis
based upon a visual inspection of the cutouts.
In order to minimize the
3.3. Monte Carlo Simulations
For this study, it was important to obtain accurate es-
timates of the true uncertainties on the internal color
dispersion measurements for individual galaxies, so we
performed object-by-object Monte Carlo simulations. In
order to do this, we first extracted the galaxies from their
H160cutout using SExtractor (DETECT MINAREA =
5 and DETECT THRESH = 1.65). Mock H160cutouts
were then created by adding realizations of the noise from
random positions on the F160W image (see Section 3.2).
We used the same H160-extracted galaxy light distribu-
tions to generate mock I775cutouts, normalizing the light
distribution such that the detected flux was equal to that
in the real I775image (that is, accounting for the object’s
I775−H160color). We then added noise realizations from
the F775W image.
The other significant source of uncertainty in our mea-
surements is the unknown astrometric shift between the
WFC3 and ACS imaging. In order to simulate this, we
applied random two-dimensional shifts to the mock H160
cutouts, with a mean amplitude of 10 mas (our astro-
metric uncertainties, see Section 2.2).
Our simulations are designed to approximate the scat-
ter induced by the sky background on ICD measure-
ments of an object with an intrinsic ξ = 0 and will allow
us to distinguish objects with intrinsic non-zero ICDs.
Systematic sources of error, such as astrometric offsets
and non-uniform sky backgrounds, only produce positive
scatter, so the error distribution is non-Gaussian and we
estimated the bias and positive scatter using the median
Fig. 2.— Ratio between the uncertainty in internal color disper-
sion as determined by our Monte Carlo simulations and the un-
certainty determined by the analytical approximation of P03. All
points are drawn from our z ∼ 2.5 SFG sample. All points have
σm
ξ,sims/σξ,P05> 1 as a result of correlated noise in the drizzled
HST images.
and third quartile, where we define,
σm
ξ= 1.496[Q3(ξ) − Median(ξ)].(3)
We performed a total of 400 simulations for each object.
Figure 1 shows the dependence of the ICD on the rel-
ative astrometric shift between I775and H160for one of
our simulations. The dashed line indicates an iterative
quadratic fit to the data (discarding outliers). Each mea-
surement includes random realizations of the sky noise,
leading to the scatter seen about the best-fit curve. For
the majority of objects in our sample, astrometric shifts
become important at ∼ 0.5 pixels, although the precise
behavior depends upon the shape of the light distribu-
tion. Point sources, for example, will exhibit more sensi-
tivity to astrometric errors than well-resolved galaxies.
We also compare the simulated uncertainties for the
SFGs in our sample to those that would be derived from
the analytical approximation of P03, in Figure 2. As
the authors of that paper note, their expression will un-
derestimate the uncertainties in the presence of corre-
lated noise. Since HST images are typically drizzled to
correct for oversampling of the PSF, our simulations do
in fact yield larger uncertainties than predicted by P03.
P05 accounted for this discrepancy in drizzled images
by correcting the uncertainty by a single empirically-
determined multiplicative factor,
δ(ξ) = C(ξ)
?2/Npix
N ?
i=0(B2
N ?
2+ α2B2
1)
N ?
i=0(I2− β)2−
i=0(B2− αB1)2
.(4)

Page 5

Differential Morphologies of 1.5 < z < 3 SFGs5
Fig. 4.— Internal color dispersion (top) and ICD signal-to-noise
(bottom) as a function of redshift. We indicate 1.4 < z < 2.9
SFGs with triangles, 2.9 < z < 3.5 SFGs with open squares, and
1.4 < z < 2 passive galaxies with stars. Galaxies with σm
are not plotted. In the top panel, the dotted line indicates ξ = 0.01
and in the bottom panel, the two dotted lines indicate ξ > 0 by 1σ
and 3σ.
ξ
> 0.05
Although C(ξ) = 6 would be a reasonable approximation
at σξ ? 0.01, it would underestimate the uncertainty
for some of the brighter objects in our sample (σξ,P05?
0.01) by factors up to ∼ 3. Hereafter, we will only use the
uncertainties derived from our Monte Carlo simulations.
4. RESULTS
The photometric properties, including centroids, mag-
nitudes, half-light radii, and ICDs, of the galaxies in our
samples are compiled in Table 3.
4.1. Internal color dispersion
4.1.1. Star-forming galaxies at 1.4 < z < 2.9
Some example I775and H160cutouts for 15 SFGs with
ξ > 0.05 are shown in Figure 3. Of the galaxies shown, all
are in SFGmain except for object 863 (z = 3.1105), which
was identified as a Lyα emitter (LAE) in Gronwall et al.
(2007). Given its large ICD, distinct disk-like morphol-
ogy, and large size compared to the majority of LAEs
at that redshift (Bond et al. 2009), it may be a low-
redshift galaxy superimposed on a high-redshift LAE.
The remaining galaxies exhibit complex morphologies,
many with multiple components in both the rest-frame
optical and rest-frame ultraviolet. Some of the objects,
such as 1204 and 1598, resemble the “chain galaxies”
of Cowie et al. (1995), which are believed to be mas-
sive galaxies in the act of formation, while others (447
and 560) have two-component morphologies suggestive
of mergers.
The distribution of internal color dispersions is shown
as a function of redshift in Figure 4. The 100 galaxies
in SFGmain exhibit a wide range of ICDs, 56 of which
are non-zero at > 3σ (see bottom panel). Of the galax-
ies with small uncertainties in the ICD (σm
only 22% have ξ < 0.01, suggesting that the majority
of color-selected SFGs in this redshift range have under-
lying populations of old stars that are spatially distinct
from the young stars. The median ICD in this sample is
0.23.
There is no apparent correlation between ξ and z in
SFGmain, except for a possible downturn at z ? 2.6. It
may be that the ICD is becoming less sensitive to stellar
population heterogeneity as the F160W filter approaches
the rest-frame 4000˚ A break. Alternatively, it may be
due to evolution - galaxies are expected to evolve towards
less stellar population heterogeneity and lower dust mass
with increasing redshift. There is a small subset of 21
SFGs with a very large ICD (ξ > 0.05), which appear
to make up a much larger fraction of the SFG sample at
z < 2. This is consistent with the findings of P05, who
found that galaxies with very large ICDs went from 7%
of their H-band-selected sample at z ∼ 2.3 to 25% at
z ∼ 1.
Figure 5 plots ξ as a function of I775− H160 color.
We find no evidence for a correlation in SFGmain. Al-
though this is in agreement with the bulk properties of
the P05 sample, they found that the galaxies with very
large ICDs had very red I775−H160colors. We see no ev-
idence for this effect in our sample, but considering that
they only had 7 galaxies with ξ > 0.05, the apparent
differences may be due solely to small-number statistics.
The lack of a correlation between ξ and I775−H160sup-
ports the hypothesis that the morphological differences
between the rest-frame ultraviolet and rest-frame opti-
cal are due primarily to stellar population heterogeneity
rather than inhomogeneous dust distributions, as greater
dust columns would both redden the galaxies and pro-
duce larger ICDs.
ξ
< 0.01),
4.1.2. Passive galaxies
The I775and H160cutouts for the 15 passive galaxies in
our PassGal sample are shown in Figure 6. Objects P1,