Transcript The Luminosity-Metallicity Relations for Star

The Dwarf Galaxy Duty Cycle:
Measurements from a Complete
Sample of the Local Volume
1
Lee ,
1
Kennicutt ,
Janice C.
Robert C.
1
1,2
3
Sanae Akiyama , Jose G. Funes , Shoko Sakai
1
Steward Observatory, University of Arizona, Tucson, AZ, USA
2 Vatican Observatory Research Group, University of Arizona, Tucson, AZ, USA
3 Division of Astronomy and Astrophysics, UCLA, Los Angeles, CA USA
Abstract
We have carried out an Ha and R-band imaging survey of a
volume-limited sample of 340 spiral and irregular galaxies
within a distance of 11 Mpc. Our goal is to fully characterize the star formation properties of complete samples of
galaxies in the local universe. In particular, we are using
this dataset to understand the importance of starbursts in
dwarf galaxies. A preliminary analysis suggests that about
1 out of 10 dwarf galaxies is currently forming stars at a rate
that is at least two times higher than its average past rate.
These bursty systems are responsible for 20-30% of the
total current star formation in dwarfs.
The 11 Mpc Ha Imaging Survey
Construction of the Sample
To build a sample of the local volume, we began by
compiling spiral and irregular galaxies from:
● the Tully Nearby Galaxies Catalog (NBG) with
d < 11 Mpc, where the tabulated NBG distances
are based on the flow model of Tully & Shaya (1984)
● NED, with the constraints that the object lie outside
of the galactic plane (|b| > 20°), with B < 15.0
and vLG < 825 km/s.
After excluding probable Virgo Cluster and Coma
Group members and collecting all available direct
distance estimates, our resulting catalog contains 340
galaxies within 11 Mpc, where Ho = 75 km/s/Mpc has
been adopted to convert flow-corrected velocities to
distances for galaxies without direct distance estimates.
Observations
Over a four year period (2000-03), galaxies in our
11 Mpc catalog have been observed through KronCousins R-band and narrow-band (FWHM~70Å) Ha
filters at the Bok 2.3m (KPNO), the VATT 1.8m (MGIO)
and the CTIO 0.9m.
Completeness??
Ultimately our dataset is a composite of numerous catalogs with diverse selection criteria. Thus, there is no a
priori selection function that can be adopted. Therefore,
we will seek to establish the completeness of our
sample in a variety of ways, as illustrated to the left.
The Role of Bursty Systems: A Preview
Figure 3 – The relation
between b and EW(Ha)
as predicted by the
KTC94 models (red), and
from a coarse empirical
calibration between the
continuum flux and the
R-band apparent mag for
two choices of constant
M/LR that bracket the
typical range (black).
This type of analysis can
be used to set upper
limits on the # of bursty
systems, as well as their
fractional contribution to
the SFR.
Figure 4 – Distributions
of b by MB (solid curves)
and t-type (dotted
curves). The median b
for the entire sample is
0.5, and increases from
0.3 when only the most
luminous galaxies are
considered, to 0.5 for the
low-luminosity samples.
The median b also
increases with later
Hubble type.
What is a burst?
For the purposes of this first
analysis, we will use the Scalo “b”
(birthrate) parameter as a measure
of the burstiness of a galaxy where
b= current SFR/ past average SFR
To derive b, we use the relationship
between EW(Ha) and b predicted
by the synthesis models of
Kennicutt, Tamblyn & Congdon
(1994) (KTC94).
What is a dwarf?
Here we use the blue luminosity of
a galaxy as an observable
indicator of mass. While B-band
light is greatly affected by dust and
recent star formation, preliminary
analyses using stellar masses
computed from the models of
KTC94 indicate that these results
based on samples defined by MB
are consistent with dwarf samples
defined by mass with M<2e9Msun.
U8091 (GR8), a bursting dwarf galaxy
with b> 2 is shown in the background.
Figure 6 – The
contribution of bursty
systems to the total
current SF. For the
entire sample,
systems with b>2 are
responsible for ~6%
of the total current
SF, while they
account for 20-30%
of the SF in dwarfs.
Figure 5 – The fraction
of bursty systems in
our sample. For both
the entire sample and
the low-luminosity
samples about 7%
have b>2. If we take
this as a threshold
over which a galaxy is
considered bursting,
and assume that all
dwarfs have an equal
probability of bursting,
then the duty cycle for
dwarf galaxies (i.e. the
fraction of time each
system spends in the
bursting state) is also
about 7%.
Figure 1 - By construction,
our sample is limited at
B~15 (the limit of the parent
catalogs from which we
extract the majority of our
sources). Thus, our dataset should be complete to
MB~-15 for d<11 Mpc. This
is confirmed by comparison
to independently determined B-band luminosity
functions. Comparison also
shows that our 7 Mpc subsample is complete to MB~13.5, and that the 11 Mpc
sample can be reasonably
corrected down to that
magnitude.
Figure 2 - Our 11 Mpc
catalog fully samples the
HI mass function (based
on HI blind surveys) down
to 7.5.e8 Msun. By
comparing our HI mass
densities to those from
HIPASS we can estimate
the completeness of our
sample to be 90% at 2.4e8
and 83% at 1.3e8.
Coming Soon…
Clearly, much work lies ahead. We will:
● use updated synthesis models, and also include the
effects of dust
● incorporate incompleteness corrections, and assess errors
● use the entire sample to quantify the incompleteness in the
SFR density based on UV or emission-line selected surveys
● take advantage of the wealth of 2-D information
available in our images to understand how
asymmetries, concentrations, and radial extents
vary with burstiness and overall star formation.
…a GALEX Legacy
● During the current cycle, we will
obtain FUV and NUV images for a
complete subset (N=110) of our
catalog to probe star-formation on
a longer timescale. This will
provide powerful constraints on
systematic errors in Ha inferred
star formation related quantities.
Figure 7 – The top line shows the distribution of MB for the complete
11 Mpc sample. The filled histogram shows the current coverage of
this volume by the GALEX PI NGS and MIS surveys and illustrates the
roughly 5-fold under-representation of galaxies fainter than MB=-17.
The middle histogram shows the total coverage of the local population
when our GALEX Legacy program targets are combined with the NGS
and MIS surveys.