Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://zebu.uoregon.edu/postscript/p12.ps Äàòà èçìåíåíèÿ: Fri Aug 5 05:36:13 1994 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 22:08:59 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: stars

Mon. Not. R. Astron. Soc. 000, 000--000 (0000) Printed 1 August 1994
Surface photometry of low surface brightness galaxies
W.J.G. de Blok, 1 J.M. van der Hulst, 1 G.D. Bothun, 2
1 Kapteyn Astronomical Institute, P.O. Box 800, 9700 AV Groningen, The Netherlands
2 Department of Physics, University of Oregon, Eugene, Oregon 97403, USA
Received date; accepted date
ABSTRACT
Low Surface Brightness (LSB) galaxies are galaxies dominated by an exponential disk
whose central disk surface brightness is much fainter than the canonical value of
¯B (0) = 21:65 \Sigma 0:30 mag=ut 00 found by Freeman (1970). In this paper we present
broadband photometry of a sample of 21 late type LSB galaxies. The median central
surface brightness of the sample is ¯B (0) = 23:2 mag=ut 00 and the median scale length is
3.2 kpc, showing that LSB galaxies are normal sized galaxies. We find colours that are
comparable to or bluer than those of the more widely studied ``normal'' High Surface
Brightness (HSB) galaxies. LSB galaxies are therefore not faded disks, that have no
current star formation. The colours can on the other hand not be ascribed entirely to
old metal­poor stars but we can explain the colours by assuming an occasional and
local surge in the star formation rate, superimposed on either a low, constant or absent
global star formation rate. LSB galaxies hence appear to be unevolved and quiescent
objects, which give us an insight in the processes of star formation in an unperturbed
environment.
Key words: galaxies: photometry -- galaxies: spiral -- galaxies: stellar content
1 INTRODUCTION
1.1 Freeman's law and selection effects
One of the most well­known early results derived from
surface photometry of spiral galaxies is Freeman's Law
(Freeman 1970). Freeman analyzed the photometry of disk
galaxies available at that time and found the remarkable
result that the central surface brightnesses of 28 of the
36 disks studied fell within the range of ¯B (0) = 21:65 \Sigma
0:30 mag=ut 00 . This result has been widely discussed. Some
showed the effect might be real (van der Kruit 1987; Bosma
& Freeman 1993), while others suggested that the ``law''
probably is a selection effect (Disney 1976; Allen & Shu
1979; Disney & Phillips 1983; Davies 1990), or results from
extinction (Valentijn 1990; Huizinga 1994) or a combination
of both. Galaxies that deviate strongly from Freeman's Law
are too faint or inconspicious and usually do not meet the
diameter criteria of the major galaxy catalogues, which is
why they are under­represented in these catalogues.
With the advance of CCD technology and multi­
wavelength studies it became apparent that many objects
that are almost invisible optically may in fact have a large
mass. The Hi clouds that were found at 21 cm are a good
example (Giovanelli & Haynes 1989), as are the damped Ly­
ff absorbers (Wolfe et al. 1986). Selection effects are thus
determining our view of the universe, but studies of Low
Surface Brightness (LSB) galaxies are beginning to show us
other, less visible parts of the galaxy population.
1.2 Low Surface Brightness galaxies
We have begun a systematic programme to use multicolour
and Hff surface photometry, spectrophotometry and Hi ob­
servations of a representative sample of ¸ 20 LSB galaxies,
to determine the distributions of light, colour, Hii­, Hi­, and
mass surface density, and elemental abundances with the
ultimate goal to unravel the star formation history of LSB
galaxies. In this paper we present structural parameters and
colours of our sample of LSB galaxies as derived from optical
multicolour broad­band data.
We define a LSB galaxy to be a galaxy which is domi­
nated by an exponential disk with a face­on central disk sur­
face brightness which is on average one or two magnitudes
fainter than the canonical ¯B (0) = 21:65 mag=ut 00 found by
Freeman (1970). This group of objects is relatively poorly
studied, and under­represented in many catalogues.
A typical LSB galaxy resembles a normal late­type spi­
ral, usually with a few ill­defined spiral arms. It contains
only a few Hii regions, which do not trace the spiral arms
very well and are usually found towards the edges of the
galaxy. It usually shows a central condensation, which is,
however, in most cases not very conspicious.
Many of these objects have been discovered on the
new, second Palomar Observatory Sky Survey plates that
are currently being obtained (Schombert & Bothun 1988;
Schombert et al. 1992; [also referred to as ``the Schombert

2 W.J.G. de Blok, J.M. van der Hulst and G.D. Bothun
lists'']). Some of the objects were already visible on the orig­
inal POSS plates, but were never catalogued.
LSB galaxies have Hi masses of a few times 10 9 M fi
(van der Hulst et al. 1993; McGaugh 1992; Schombert et al.
1992). Observations with the WSRT and VLA show low Hi
surface densities, usually close to or even below the critical
surface density for star formation as formulated by Ken­
nicutt (1989). The oxygen abundances in the few Hii re­
gions these galaxies possess are quite low: ¸ 0:1 to ¸ 0:5
solar metallicity (McGaugh & Bothun 1993; de Blok & van
der Hulst, in prep.). Previous studies have shown that LSB
galaxies are unusually blue compared to normal late type
galaxies (McGaugh 1992; van der Hulst et al. 1993; Knezek
1993). So far no CO­emission has been detected (Schombert
et al. 1990; van der Hulst & de Blok, in prep.). Thuan et
al. (1987) and Bothun et al. (1993) find that although LSB
galaxies follow the spatial distribution of HSB galaxies, they
tend to be more isolated from their nearest neighbours than
HSB galaxies.
If we assume that large amounts of star formation are
triggered by major tidal interactions or infall of companions
the above results and the lack of major star formation sug­
gest that LSB galaxies are relatively unperturbed, and have
not suffered from these events. These galaxies thus offer a
unique opportunity to extend the range of environments in
which we can study the properties of galaxies and star for­
mation.
Some LSB galaxies have remarkable characteristics: Ma­
lin 1 (Impey & Bothun 1989) has a huge (55 kpc, H0 =
100) scale length, very low disk central surface brightness
(¯B (0) ú 25:5 mag=ut 00 ), and an enormous neutral hydro­
gen mass (MHI ¸ 10 11 M fi ). Malin 1 is an extreme case,
but no exception. Studies by e.g. Knezek (1993) and Mc­
Gaugh (1992) show that the range in LSB galaxy sizes and
Hi masses is comparable with that of High Surface Bright­
ness (HSB) galaxies that define the Hubble sequence.
It should be stressed that the LSB galaxies we study
here are neither Malin­1­like giant galaxies, nor intrinsically
small galaxies such as the dwarf ellipticals and dwarf irreg­
ulars from our Local Group. As already stated, the LSB
galaxies we study resemble normal late type spirals in their
morphology.
Section 2 of this paper will discuss the sample and the
reduction techniques. In Section 3 we will show and discuss
the results. In Section 4 describes the colours in our sample.
Section 5 we will describe the properties of a few individual
galaxies. In Section 6 we will compare our data with photo­
metric evolution models from the literature and discuss age­
and metallicity­effects. In Section 7 possible star formation
histories will be discussed and in Section 8 we will summa­
rize the discussion, and give some concluding remarks.
2 SAMPLE AND REDUCTION
2.1 The sample
The sample was selected using the lists of LSB galaxies by
Schombert and Bothun (1988) and Schombert et al. (1992)
and the UGC (Nilson 1973). The selected galaxies are mor­
phologically similar to late Hubble type galaxies (Scd and
later) with ¯B (0) ? 23 mag=ut 00 and relatively face­on with
estimated inclinations (as derived from axis ratios on the
POSS­plates) i ! 60 ffi , for which WSRT or VLA Hi obser­
vations were available (van der Hulst et al. 1993; McGaugh
1992). The Hi observations of the UGC galaxies in our sam­
ple are discussed in van der Hulst et al. (1993). Those of
a few other galaxies are discussed in McGaugh (1992). Hi
observations of all galaxies in our sample will be discussed
in a future paper.
Figure 1 shows R­band images of our sample of LSB
galaxies. These images are presented using a linear intensity
scale adjusted to bring out the faint outer parts, so that the
brighter inner parts are saturated. In most cases this gives
the impression that there is a bulge present. This is however
an artefact of the intensity scale used. Table 1 lists the sam­
ple. The right ascencion for F561­1 as given in the Schombert
lists is off by 10 s . The distances were derived from the helio­
centric redshift (Schombert & Bothun 1988; Schombert et al.
1992). These redshifts were corrected for Galactic rotation
and Virgocentric flow, and converted to distances, assuming
a Hubble constant H0 = 100 km s \Gamma1 Mpc \Gamma1 .
In this table and the rest of this paper all Galactic ex­
tinction corrected magnitudes and colours have been cor­
rected assuming a standard stellar extinction curve, follow­
ing Allen (1973) and Burstein & Heiles (1982). Extinctions
in the B­band were furthermore determined using the NED
database. As the dust content of LSB galaxies is probably
low (McGaugh 1992) and as the optical inclinations have
a large uncertainty, we have not corrected for internal ex­
tinction in the LSB galaxies. Total magnitudes are given in
Table 2, and will be discussed later.
2.2 Observations
UBVRI images of the galaxies in the sample were taken dur­
ing 4 nights in the period from 8 until 14 January 1991, with
the 2.5 meter Isaac Newton Telescope at La Palma ? . The
nights were excellent, with good (¸ 1 00 ) seeing and mostly
photometric conditions. A GEC CCD was used, with a pixel
size of 0: 00 54.
Images were bias subtracted and flatfielded and co­
added with the MIDAS package using standard methods.
Flatfield images had been taken several times each night.
Deviations between flats made on different nights using the
same filter were ! ¸ %.
2.3 Flux calibration
The images were flux­calibrated using standard stars (Lan­
dolt 1983). Since many standard star exposures had been
taken through each filter we could compare these to deter­
mine the errors in the photometry over the 4 nights. We
found these to be 0.04 mag (1oe) for the U ­filter, based on
19 exposures. The deviation for the B­filter was the largest:
0.05 mag, based on 12 exposures. The deviation for the V ­
filter was 0.02 mag, based on 11 exposures. The R­filter
standard stars showed a deviation of 0.02 mag, based on
? The Isaac Newton Telescope is operated by the Royal Green­
wich Observatory at the Observatorio del Roque de los Muchachos
of the Instituto de Astrof'isica de Canarias with financial support
from the SERC (UK) and NWO (NL).