| Annu. Rev. Astron. Astrophys. 1994. 32:
115-52
Copyright © 1994 by Annual Reviews. All rights
reserved
|
2. OVERVIEW: THE MORPHOLOGICAL DEPENDENCE OF FUNDAMENTAL
PROPERTIES
In an early study of the integral properties of galaxies,
Roberts (1969)
analyzed 98 spiral and irregular galaxies for which total mass,
neutral hydrogen content, luminosity, color and radius were available.
Over the last several decades, the database of observed quantities for
extragalactic objects has expanded at an enormous rate. The
availability today of large catalogs of galaxies and compilations of
data in digital form makes statistical and graphical analysis possible
as it has not been before. In preparing this review, we have drawn
upon such catalogs to explore several of the morphological dependence
issues. In this section, we present the results of our own analysis
which we discuss in comparison with the findings of others in later
sections.
Construction of Samples for Analysis
For the current purpose, we make use of two primary compilations:
first, the RC3 and second, a private catalog maintained
by R. Giovanelli and M. Haynes that we refer to by its familiar name,
the Arecibo General Catalog (AGC). The latter catalog primarily adds a
significant body of HI line data including upper limits for
non-detected objects, a variety of measurements of the 21 cm line
width, and qualitative indicators of profile shape.
Currently, redshift surveys extend relatively deeper in the north
than in the south (as visible in Figure 2 of
Giovanelli & Haynes
1991).
Because of the northern hemisphere bias in redshift survey depth,
we use as the prime deep sample the compilation of objects that
are included both in the RC3 and in the Uppsala General Catalogue
(Nilson 1973:
UGC). We refer to the sample of objects common
to both catalogs as the ``RC3-UGC sample'' It should be noted that,
because the RC3 is intended to be complete only for objects of high
surface brightness, the lowest surface brightness objects
are found only in the UGC, and are underrepresented in the current analysis.
Likewise, the UGC, being angular-diameter limited, is biased
against high surface brightness, compact objects and becomes
incomplete for early-type galaxies especially at the larger distances.
When one selects galaxies of fixed flux, the volume element containing
the more distant, intrinsically brighter objects is larger than that
occupied by the nearer, intrinsically fainter population. This
``Malmquist bias'' affects all galaxy catalogs that are flux-limited.
In order to examine (and counteract) the effects of Malmquist bias, we
have also constructed a nearby volume-limited one that should be
complete but has relatively fewer galaxies. Since the volume occupied
by the Local Supercluster has been well-studied by most available
multiwavelength techniques, we have identified 4972 RC3 objects with
redshifts implying membership in the Local Supercluster VLG <
3000 km s-1. This subset is referred to as the ``RC3-LSc sample.''
Note that it contains galaxies that are not in the RC3-UGC sample.
Since the backbone of our compilation is the RC3, the reader is
referred to its first volume for an explanation of its contents. Our
general philosophy has been to use all of the corrected parameters
directly from the RC3 when available since its authors have gone to
considerable length to reduce parameters obtained from different
sources to a standard system. For the present comparative purposes,
the consistency of approach is perhaps more critical than absolute
prescription. Most parameters as detailed below have been taken
directly from the RC3. Additional radial velocities, 21 cm parameters
and far infrared data from IRAS come from the AGC.
DISTANCES
In order to convert velocities to distances and
to further calculate intrinsic parameters, it is necessary to adopt a
value of the Hubble constant and a model of the local velocity field.
Heliocentric velocities V are taken from the AGC
preferentially if a good quality 21 cm spectrum is available;
otherwise, the available optical velocity is used. The velocity with
respect to the Local Group VLG was calculated by applying the
standard correction to V given in the RC3: 300 sinl
cosb. For objects in the Local Supercluster, a non-linear
infall model was used to calculate the distance to an object with the
observed VLG. The model adopted follows the outline of
Schechter (1980)
with the assumptions of a distance of 20.0 Mpc and an
overdensity of 2 for Virgo and an infall velocity at the Local Group
of 300 km s-1. The Local Supercluster boundary is taken to be at
VLG = 3000 km s-1. For more distant
objects, distance is computed merely from the Hubble ratio from
VLG and is not referenced to
any other frame. The assumptions that we have made are not
intended to be an endorsement of any particular solution but are
chosen for convenience. Most important is our emphasis on consistency.
Throughout this paper, we adopt a Hubble constant H0 of 50
km s-1 Mpc-1.
OPTICAL SIZE, LUMINOSITY AND SURFACE MAGNITUDE
The linear size follows from the RC3 and the calculated distance. Likewise,
the prescriptions outlined in the RC3 for correcting magnitudes are
adopted along with a value of the solar absolute magnitude MB
of +5.48. The surface magnitude B used here is defined simply as
B =
BT0 + 2.5 log ab,
with a equal to D25 and b the corresponding
minor axis obtained from the RC3 axial ratio.
NEUTRAL HYDROGEN MASS AND SURFACE DENSITY
The total neutral hydrogen mass MHI in solar units is
calculated
from the integrated 21 cm line emission MHI = 2.36 x
105 D2 SdV, where D is the distance in
Mpc and SdV
is the HI line flux in Jy kms-1. For objects for which only a
value of the rms noise per velocity interval in the emission spectum
is available, the upper limit to MHI is calculated
assuming the emission is rectangular, of amplitude 1.5 times the rms
noise and width equal to that expected for an Sa-Sb galaxy of similar
luminosity, properly corrected for inclination. The latter
relationship was derived from the detected objects. Objects showing
emission confused with other sources or HI in absorption cannot be
used properly in the analysis and have been ignored. The HI surface
density, HI, has
been calculated as MHI / R2 where R is the optical linear
radius. Although the use of the optical area makes HI a hybrid quantity,
Hewitt et al. (1983)
have shown that on average, the HI and optical
sizes scale linearly. Most authors use the quantity HI
or some variant thereof as the indicator of HI content in comparative
studies.
FAR INFRARED LUMINOSITY AND SURFACE DENSITY
The far infrared luminosity is derived from the fluxes measured by IRAS
in the 60 and 100 micron bands FFIR (Jy) = 2.58
F60µ +
F100µ, as LFIR
(L) = 3.86
x 105 D2 FFIR. Similar to
HI, a hybrid far
infrared luminosity surface density FIR is
calculated as LFIR /
R2.
TOTAL MASS AND SURFACE DENSITY FOR SPIRALS
For the
galaxies for which 21 cm line emission is detected, profiles widths
are available to provide an estimate of the circular rotation
velocity. Since widths are often measured using different algorithms,
we have selected a subset of the available data that meet the
following criteria: 1. the level at which the width was measured must
be either at 20% of one or more peaks or 50% of the mean
intensity; 2. the detection must be a good one, that is, not poor or
confused; 3. the inclination must be greater than 40° for
the width to be corrected to edge-on. While these restrictions cut
down the number of galaxies for which corrected 21 cm line widths are
available, they insure greater certainty of the resultant
correlations. The total mass MT is available only for non
E-type systems, and is calculated according to MT(< R)
(M) = 2.325 x
105 R V2rot. As a
practical application,
we use the corrected 21 cm line width as the measure of
2Vrot and D25 as the indicator of 2
R. Note that we have not applied a
correction for turbulent velocity. The total mass surface density
T is likewise
calculated as MT /
R2.