Joint European and National Astronomical Meeting (2004)
THE BULGES OF INTERMEDIATE REDSHIFT GALAXIES
L. Domı́nguez-Palmero,1 M. Balcells,1 M. Prieto,1,2 D. Cristobal-Hornillos,1 P. Erwin,1 C. Eliche-Moral,1
ABSTRACT
We have studied the colors of the bulge component of 99 galaxies from the HST Groth Strip Survey (Groth
et al. 1994), covering redshifts 0.3 < z < 1.0. We selected all objects with apparent radii R > 1.4, and with
inclination above 50o in order to avoid reddening from dust in the disk on one side of the bulges. We find that,
as in the Local Universe, the minor axis color profiles are negative (bluer outside), and fairly gentle, indicating
that bulge colors are not distinctly different from disk colors. In most cases, dust bands are the most important
morphological structure in the color maps.
In a subsample of galaxies with spectroscopic redshifts, we analyze central rest-frame colors using Kcorrections. Bulge colors do not globally become bluer at higher redshifts. This suggests that there were “old”
bulges at z = 0.8. The color-magnitude distribution of intermediate-z bulges is steeper than that of bulges in
the Local Universe. The most massive bulges are as red as local bulges, while the remainder are significantly
bluer, a possible sign of late bulge formation.
Completeness of the sample is ensured by computing V /Vmax , which allows to convert the sample to
volume-limited. Comparison to local galaxies is done by artificially redshifting a similar local sample using K
and evolutionary corrections, and determining what local galaxies would be visible at each redshift.
Key Words: BULGE FORMATION — COLOR PROFILES — SELECTION EFFECTS — VOLUME
CORRECTION
1. COLOR PROFILES AND COLOR MAPS
We measured color profiles on both semi-minor
axes of each galaxy. Different colors on both sides
most likely result from dust reddening on the side of
the bulge seen through the disk. Color profiles on the
“clean” side are generally smooth, with mild negative gradients (bluer outward). Bulge-disk decompositions indicate that bulge colors are not significantly
different from disk colors. This result is identical to
that for bulges in the Local Universe, where bulgedisk color differences are smaller than the color differences between galaxies (Peletier & Balcells 1996),
and where dust bands are the most important morphological structures observed in the color maps.
All of our objects have been imaged in six bands
as part of the GOYA Photometric Survey (U, B with
INT/WFC, F606W and F814W with HST/WFPC2,
J and Ks with WHT/INGRID).
2. PHYSICAL CORRELATIONS WITH
CENTRAL COLORS
We estimated bulge colors from the colors measured on the “clean” side of the galaxies at 1.3 times
the PSF of the HST images. This corresponds to
1 Instituto de Astrofı́sica de Canarias, La Laguna, Tenerife,
Spain.
2 Departamento de Astrofı́sica, Universidad de La Laguna,
Tenerife, Spain
0.87 kpc at z = 0.3 to 1.56 kpc at z = 1.0. Kcorrections for colors and magnitudes were computed
using COSMOPACK (Balcells et al. 2003).
In figure 1 we show rest-frame central colors vs
redshift. The upper envelope of the distribution does
not become bluer with redshift, which indicates that
there were old bulges already at z = 0.8.
In figure 2 we show rest-frame central (B − R)
colors vs R-band galaxy absolute magnitude (solid
circles). More luminous galaxies have redder central
colors, as in the Local Universe. However, the relation at z = 0 is much shallower (solid line in figure 2,
from Schweizer & Seitzer 1992). Bulges at intermediate z are bluer, hence they are younger and/or
less metal-rich than at z = 0. The crosses and diamonds are Local Universe bulge colors with and
without dust, respectively, from Peletier & Balcells
(1996). In the luminous range of local bulges, high-z
bulges show a spread of colors. Some are as red as
local bulges, indicating that their populations have
comparatively similar ages and metallicities: ages of
these bulges must approach 10 Gyr. Bluer bulges
probably harvest young populations, a likely indication of late bulge formation. Finally, all the fainter
galaxies have significantly bluer central colors than
those at z = 0. Their colors indicate an active star
formation activity, and it is unclear whether these
galaxies harbor “bona-fide” bulges.
1
2
DOMı́NGUEZ-PALMERO ET AL.
3. SELECTION EFFECTS
In order to study selection biases in our
intermediate-z sample and to understand how local
and distant samples compare to each other, comparisons of local and distant samples have been carried
out by artificially shifting images of local galaxies to
redshifts from z = 0.1 to z = 1.0, using the COSMOPACK package. Cosmological dimming, K and
evolutionary corrections, sky brigthness, pixelation,
etc., are taken into account to mimic the F814W
HST/WFPC2 images of the Groth Strip, with Kcor
and Ecor computed using GISELXX (Bruzual &
Charlot 2003).
We used the inclined local sample of Balcells &
Peletier (1994), excluding those galaxies whose colors were affected by dust. In total, we have 14 galaxies, all of them with major axis diameter larger than
2 arcmin and inclinations higher than 50o , (see figure 3).
By plotting different local galaxies parameters
as superficial central brightness, scale length of the
disks or bulge absolute magnitudes vs. the redshift
at which each galaxy is lost, we can find interesting
correlations and uncorrelations. The main conclusions of this analysis are: 1.) selected galaxies will
be intrinsically larger and brigther as their redshifts
are higher (fig. 4a), 2.) selected galaxies will have
larger scale length than those ones that are not selected (fig. 4b), 3.) the central surface brightness
of disks decreases slightly as galaxy redshift become
higher (fig. 4c), 4.) there is no correlation between
the absolute magnitude of the bulges and the redshift
of the galaxy (fig. 4d).
4. VOLUME CORRECTION
In order to perform volume corrections in our distant sample, we have also shifted images of a subsample with known spectroscopic redshift. In total, we
have 40 objects with z > 0.1. The COSMOPACK
package is used in the same way as before, but in
this case evolutionary corrections are not taken into
account, since we want to obtain the accessible volume of each galaxy as it is in the moment we are
observing it, and not that one occupied by its predecessors.
We calculated the range of redshift at which each
galaxy would satisfy the size criteria we imposed to
the original sample: 1.4′′ < sma < 6.0′′ , this gives us
the minimal and maximal distance of the galaxy to
be included in the sample. In this way, each galaxy
is weighted by the inverse of its maximum accessible
volume. To check that the sample is statistically
complete, we have applied a V /Vmax test (Schmidt
Fig. 1. Rest-frame central B - R colors of the subsample with spectroscopic redshifts vs redshift. K corrections were computed using COSMOPACK (Balcells et
al. 2003).
1968; Thuan & Seitzer 1979). The average value we
have obtained is
V /Vmax = 0.54 ± 0.05
(1)
which is a very good result since we are testing
only a sub-sample. The test will be applied to the
whole sample when we have photometric redshift for
all the objects.
5. DISCUSSION
The evolutionary corrections are uncertain given
the uncertainties in the formation redshift and the
cosmological model.
Are the bluest bulges really “bulges” We defined
bulges through a bulge-disk decomposition of the 1D
surface brightness profiles. Such decompositions are
difficult due to the small size of the images and the
high galaxy inclinations.
REFERENCES
Balcells, M., Cristobal-Hornillos, D., & Eliche-Moral, C.
2003, RMxAC, 16, 259
Balcells, M & Peletier, R. 1994, AJ, 107, 135
Barbaro & Poggianti, 1997, A&Ap, 324, 490
Bruzual & Charlot, 2003, MNRAS, 000, 1-35
de Jong, R. S., van der Kruit, P. C., 1994, A&AS, 106,
451
Domı́nguez-Palmero, L., DEA, 2003
Groth, E. J. at al., 1994, BAAS, 185, 5309
Lilly, S. et al. 1998, ApJ, 500, 75
Peletier & Balcells, 1996, AJ, 111, 2238
JENAM 2004
3
Fig. 2. Rest-frame central B - R colors of the subsample
with spectroscopic redshifts vs R-band galaxy absolute
magnitude (solid circles). The crosses and diamonds are
Local Universe bulge colors with and without dust, respectively, from Peletier & Balcells (1996). The solid
line represents the color distribution of eliptical galaxies
in the Local Universe.
Fig. 3. Apparent sma (arcsec) of shifted galaxies measured by SExtractor vs redshift. Each galaxy is represented by a different color, the horizontal black line is the
value of the limit angular size of our intermediate redshift sample. We observe that as the redshift increases
the apparent sizes become smaller and that we are losing
galaxies in the whole range of redshift.
Peletier & Balcells, 1997, New Astronomy, 1, 349
Poggianti, B., 1997, A&AS, 122, 399
Schmidt, M., 1968, ApJ, 151, 393
Schweizer & Seitzer, 1992, AJ, 104, 1039
Thuan & Seitzer, 1979, ApJ, 231, 680
Fig. 4. (a) Physical sma (kpc) vs cut z, (b) disk scale
length vs cut z, (c) central surface brigthness in H vs cut
z, (d) bulge K absolute magnitude vs cut z.
4
DOMı́NGUEZ-PALMERO ET AL.
Fig. 5. V /Vmax of each galaxy vs its redshift. We can
see that V /Vmax for galaxies at z < 0.8 spreads in a
wide range of values, which means that objects selected
are distributed randomly in their accessible volume. For
0.8 < z < 1.0 do we find a selection bias that reflects
both the 4000 amgstrom break entering the I band, and
the selection criteria for the spectroscopic redshift survey.
Author and collaborators address: Instituto de Astrofı́sica de Canarias, c Vı́a Láctea s/n, 38200 La Laguna,
Tenerife. Spain.
Author e-mail: ([email protected])