Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.stsci.edu/ftp/science/singg/halpha_hi.ps.gz Äàòà èçìåíåíèÿ: Wed Nov 22 06:40:02 2000 Äàòà èíäåêñèðîâàíèÿ: Sat Dec 22 10:58:18 2007 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: http www.astronomy.ru forum index.php topic 4644.0.html

NOAO Observing Proposal Survey proposal Panel: For office use.
Date: March 15, 2000 Category: Resolved Galaxies
Star Formation in Hi Selected Galaxies
PI: Gerhardt R. Meurer Status: P Affil.: The Johns Hopkins University
Department of Physics and Astronomy, 34th and Charles Streets, Baltimore, MD 21218 USA
Email: meurer@pha.jhu.edu Phone: 410 516 5154 FAX: 410 516 8260
CoI: Henry Ferguson Status: P Affil.: STScI
CoI: Rachel Webster Status: P Affil.: University of Melbourne
CoI: Rob Kennicutt Status: P Affil.: Steward Observatory
CoI: Patricia Knezek Status: P Affil.: STScI
CoI: Sally Oey Status: P Affil.: STScI
CoI: Chris Smith Status: P Affil.: NOAO/CTIO
M. Drinkwater
K. Freeman
V. Kilborn
M. Putman
L. Staveley­Smith
Abstract of Scientific Justification (will be made publicly available for accepted proposals):
Cosmological evolution is mapped using tracers of star formation. Usually the selection of tracers is
biased by stellar luminosity. We propose to determine star formation properties of a sample selected
free of the stellar bias: by gas content from the HiPASS survey. Since an interstellar medium is
a prerequisite for star formation, our survey will uniformly sample all galaxies that could form
stars, and hence provide a fair view of the local star formation demographics. The sample will
consist of the 500 HiPASS galaxies, each of which will be imaged in Hff and the stellar continuum.
Integrated, this yields the local star formation rate density, which will be used as benchmark to
estimate the bias in other Hff samples. This survey will also provide high resolution images of
the star formation morphology over a wide range of galaxies, particularly towards irregular dwarfs
which are often missed in other studies. The Hi content of galaxies combined with the SFR will
be used to predict the evolution of the Hi mass function and the epoch when the cosmological gas
tank runs dry.
Scheduling constraints and non­usable dates (up to four lines).
The moon must be within 10 days of new to reach our surface brightness limit. Two observing runs
per semester of four to five nights each are requested.

NOAO Proposal Page 2 This box blank.
Scientific Justification Be sure to include overall significance to astronomy. For standard proposals
limit text to one page with figures, captions and references on no more than two additional pages.
Introduction
Statistical studies of galaxies in the local universe are hampered by the existence of severe selection
biases, which make it difficult to tell whether the quantities being measured pertain to all galaxies or
simply those favored by the survey's selection criteria. The large uncertainties in the local density of
low­surface­brightness galaxies (Impey & Bothun 1997), the failure to verify or disprove Stochastic­
Self­Propagating Star Formation (SSPSF) models for dwarf galaxies (Gerola et al. 1980), and the
missing evolutionary links between exotic objects such as luminous infrared galaxies or blue compact
dwarf galaxies and the rest of the Hubble sequence are all examples of the difficulties brought on
by selection effects. The selection biases are particularly severe for existing Hff surveys of galaxies
often used to study both the physics of star formation and to estimate the integrated star formation
density for cosmological studies. Previous Hff imaging surveys of local galaxies either covered a
broad range of morphologies (e.g. Hodge & Kennicutt 1983; Young et al. 1996) or concentrated
on a particular interesting type of galaxy (e.g. Seyfert galaxies: Mulchaey et al. 1996; edge­on
starbursts: Lehnert & Heckman 1996; Early type spirals: Hameed & Devereux 1999). But in all
cases the selection of targets was based on optical or infrared properties. Wide­field emission­line
surveys have been used to detect star forming galaxies and estimate the local star formation rate
density (e.g. Gallego et al. 1995) but are intrinsically biased toward galaxies with the highest star
formation intensities and a proper treatment of these biases integrated over the Hff luminosity
function could introduce substantial corrections (Schade & Ferguson 1994).
Selecting galaxies in a new way (e.g. via infrared radiation in the case of IRAS) can lead into new
insights in the physics of galaxies, especially if those surveys are accompanied by detailed follow­up
studies. The recently completed Hi Parkes All­Sky Survey (HiPASS) in the southern hemisphere
is the first large survey to select galaxies entirely by their Hi 21­cm emission. Because hydrogen is
the essential fuel for star formation, this is an ideal sample to use in star formation surveys. The
Hi survey is completely unbiased with respect to optical luminosity, surface brightness, or Hubble
type and covers the entire sky south of ffi = +2 ffi over a velocity range ­1200 to 12,700 km s \Gamma1 with a
detection limit ¸ 10 6 D 2 M fi where D is measured in Mpc. Hence it can offer a unique opportunity
to study not only the intrinsic trends of Star Formation Rates (SFR) with galaxy properties (e.g.
mass, surface brightness) but also the distribution about the mean trends.
We propose an Hff imaging survey of a large, unbiased subset of HiPASS selected to evenly sample
the Hi mass function (Fig. 1). The survey will address evolution issues (detailed below) ranging
from how star formation is regulated within galaxies, to the star formation history of the universe
as a whole. The size, completeness, and quality of the survey insures that it will be the reference
Hff dataset of nearby galaxies for at least the next decade.
Science Drivers
Star Formation Density of the Local Universe. The redshift evolution of the SFR density,
ae SFR (z), is fundamental to understanding cosmological evolution, and the local SFR, ae SFR (0), is
a key datum in discriminating between various models. The ae SFR (z) plot of Madau et al. (1996)
shows a factor of ú 10 decline in ae SFR from z ú 1 to z = 0 which is difficult to reconcile with
models of cosmological evolution (e.g. Baugh et al. 1998). Recent estimates of ae SFR (0) range
over a factor of ¸ 3 (0.03 to 0.08 M fi Mpc \Gamma3 yr \Gamma1 for H 0 = 75 km s \Gamma1 Mpc \Gamma1 , q 0 = 0:5) with
individual measurements not always agreeing within their quoted errors. This is in part due to the
various sample biases. For example, Hff samples (e.g. Gallego et al. 1995) are biased towards high
equivalent width (starburst) systems; the UV sample of Treyer et al. (1998) is biased against very

NOAO Proposal Page 3 This box blank.
dusty systems; conversely, FIR samples (e.g. Sanders & Mirabel 1996) are biased against low dust
systems. No survey is completely free from selection biases (ours may be slightly biased towards
high SFR galaxies if photodissocation of H 2 significantly increases Hi content). However, our survey
will be free from most of the selection biases that plague the optical and IR studies, providing an
independent estimate of ae SFR (0). Strictly speaking, we will measure a lower limit to ae SFR (0), since
our observations do not address extinction. Additional observations in the literature (e.g. ae SFR (0)
from IRAS) can be used with our results to constrain the total SFR.
The Distribution of Star Formation Rates. A key aspect of our proposal is that it will allow
a measurement of the distribution of star formation (rates and surface densities) as a function of
other galaxy properties. The largest uncertainties in cosmological models of galaxy evolution are the
prescriptions for star formation (quiescent and merger­induced) and feedback. These prescriptions
are poorly tested by measurements of mean quantities such as the Tully­Fisher relation or the color--
luminosity relation. Much more stringent tests are possible when the full distribution functions can
be measured. A powerful aspect of the present generation of semi­analytic galaxy­formation models
is that they predict such distribution functions (e.g. Sommerville & Primack 1999). The HiPASS
Hff survey will provide a sample for testing the cosmological predictions for the distribution of
SFR at fixed Hi mass. The sample will also test more schematic models of self­propagating star
formation in dwarf galaxies (Gerola et al. 1980; Tyson & Scalo 1988) and provide evidence for or
against intermittent star formation in Milky­Way­like galaxies (Rocha­Pinto et al. 2000).
Hii region luminosity functions. HiiLFs are important diagnostics of current star formation.
At the upper end, there may be a universal power­law in the number of stars per cluster, as
suggested by the HiiLF shape in most galaxies (Oey & Clarke 1998), the LF of open and globular
clusters (Elmegreen & Efremov 1997), and super star clusters (Meurer et al. 1995). The HiiLFs
in our sample will be a crucial test to see whether this universal law is upheld. A turnover at the
faint end of the HiiLF sensitively probes the high mass end of the stellar IMF.
Census of star formation modes. Our observations will provide new demographics on how and
where stars form in galaxies. We will catalog different star forming structures (i.e. spiral arms,
bars, rings, and nuclei) and determine how frequently they occur. Of particular interest, central
starbursts will be recognized by their high Hff effective surface brightness (¯ e
(Hff)). Thus we will
make an accurate estimate of the starburst incidence rate. When combined with estimates of
starburst duration (Meurer 2000), the average number of bursts per galaxy can be determined.
Gas consumption timescales. While star formation occurs in molecular gas, to astronomi­
cal precision, the neutral mass, 1:3M Hi , makes a good proxy for the total gas mass (the mean
log(M Hi =MH 2 ) ú 0:0 in the sample of Young et al. 1996) and is a lot easier to measure than
MH2 . Hence to first order the the gas consumption time t gas ¸ 1:3M Hi =SFR. On a galaxy by
galaxy basis t gas is an alternative for distinguishing between starburst (t gas Ü 1=H 0 ) and quiescent
systems (t gas ¸ ? 1=H 0 ), which we will compare with the ¯ e
(Hff) method. On a sample wide scale
we will measure the cosmological gas consumption timescale t gas;cosmo ú 1:3ae Hi =ae SFR , where ae Hi is
the local Hi density in M fi Mpc \Gamma3 . This tells how long present day evolution can be maintained
before the cosmic fuel tank starts to run dry. We will also estimate the gas consumption rate over
each 0.5 dex bin of the HiMF which will indicate where evolution is occurring most quickly.
The Reference Hff sample for the local universe. An essential aspect of the survey will be
the release to the community of high quality calibrated Hff and continuum images. This will enable
comparative studies with other measurements of gas and star formation in galaxies. For example,
UV observations with HST or GALEX (Martin et al. 1997) can be used to calibrate the extinction
correction for UV star formation rates and constrain the high mass end of the IMF; The Hii regions
can be used to derive the true [O/H] distribution of galaxies, avoiding the current bias towards

NOAO Proposal Page 4 This box blank.
Figure 1: HiPASS Hi mass function of Kilborn et al. (1999)
towards high excitation systems with [O=H] ú 0:1 (Kunth & ¨
Ostlin 1999); and Hi and 12 CO imaging
will allow a detailed reexamination of the star formation law of Kennicutt (1989; 1998) without a
bias toward optically prominent galaxies.
Sample Selection
HiPASS (Staveley­Smith et al. 1996) is a blind survey for 21 cm Hi line emission of the entire
sky with ffi ! +2 ffi using the Parkes 64m radio telescope and the multibeam receiver. It covers the
velocity range V r
= \Gamma1200 to 12,700 km s \Gamma1 with a peak flux limit of ¸ 40 mJy; survey observations
have just been completed. Preliminary analysis of ú10% of the survey reveals a relatively steep
faint end slope (ff ú 1:3) to the HiMF as shown in Fig. 1 (Kilborn et al. 1999). HiPASS is
detecting many previously uncatalogued galaxies (¸ 20% of detections). However, so far only one
has not been detected in optical broadband follow­up observations.
We will select 500 HiPASS sources from the whole high galactic latitude (jbj ? 30 ffi ) southern
sky. The survey will include galaxies in the Hi mass range 10 7 ! M Hi ! 10 11 , with 150 galaxies
chosen per decade of Hi mass in the range 10 8 ! M Hi ! 10 10 , and the remaining 50 from the
extreme high­ and low­mass ends of the sample (this will include virtually all HiPASS galaxies
at the extremes). Except for the high­mass extreme, M Hi ¸ ? 10 9:75 , the sample will have radial
velocities V r ¸ ! 1800 km/s, with selection of specific galaxies guided by their velocity relative the
the center of the bandpasses of the Hff filters.
A sample of this size is the minimum necessary to accomplish the essential goal of this proposal: to
measure not only the mean but also the distribution about the mean of star formation rates among
galaxies of different Hi masses, Hubble types, surface brightnesses, and in different environments.
A sample of 150 galaxies per decade of Hi mass in the core sample allows the mean SFR per
decade to be measured to better than 10% accuracy, the mean SFR per magnitude central surface
brightness 21:6 ! ¯ c
! 24:6 within each decade of Hi­mass to be measured to 15% accuracy, and
the mean SFR per decade local density 10 \Gamma2 ! ae ! 10 2 Galaxies/Mpc 3 to be measured to better
than 20% accuracy. Within each decade of M Hi , the rms width of the Hff luminosity distribution
can be determined to ¸ 10% (SSPSF models predict that the width should increase with decreasing
galaxy mass; Gerola et al. 1980), and sensitive tests for non­Gaussianity can be performed. This
sample will allow a search for extremely rare objects: galaxies with significant Hi masses but no

NOAO Proposal Page 5 This box blank.
star formation (such as transition dI/dSph systems), and galaxies with low Hi masses but very
high star formation rates.
Summary: Advancement of Astronomy
The HiPASS Hff survey will be the first to measure the star formation properties of galaxies in
a sample free from optical or IR selection biases. The data set will provide an independent and
more precise estimate of the local star formation density ae SFR (0), and a measurement of which
galaxies contribute the most to that star formation. By measuring the distribution of L(Hff)=M Hi
the survey will provide a definitive test of stochastic star formation models for low mass galaxies,
and a sensitive test for sporadic variations of the star formation rate in more massive galaxies
along the traditional Hubble sequence. The sample will provide a new set of demographics on the
morphology of star formation and vast improvement in the constraints on Hii region luminosity
functions. We intend this sample to provide the reference Hff data set for nearby galaxies for the
next decade. The public release of uniform high­quality data from a large sample with well­defined
selection criteria will provide an ideal basis for complementary studies at other wavelengths and
with other techniques, advancing our general understanding of star formation in galaxies.
References
ffl Baugh, Cole, Frenk, & Lacey 1998, ApJ, 498, 504
ffl Elmegreen, & Efremov 1997, ApJ, 480, 235
ffl Ferguson, Gallagher, & Wyse 1998, AJ, 116, 673
ffl Gallego, Zamorano, Arag'on­Salamanca, & Rego 1995, ApJ, 455, L1
ffl Gerola, Seiden, & Schulman 1980, ApJ, 242, 517
ffl Hameed, & Devereux 1999, AJ, 118, 730
ffl Hodge, & Kennicutt 1983, AJ, 88, 296
ffl Impey, & Bothun 1997, ARA&A, 35, 2671
ffl Kennicutt 1989 ApJ, 344, 685
ffl Kennicutt 1998 ApJ, 498, 541
ffl Kilborn, Webster, & Staveley­Smith 1999, PASAu, 16, 8
ffl Kunth, & Ostlin 1999, ARA&A, in press (astro­ph/9911094)
ffl Lehnert, & Heckman 1996, ApJ, 462, 651
ffl Madau, Ferguson, Dickinson, Giavalisco, Steidel, & Fruchter 1996, MNRAS, 283, 1388
ffl Martin 1997, BAAS, 191, 63.04
ffl Meurer, Heckman, Leitherer, Kinney, Robert, & Garnett 1995, AJ, 110, 2665
ffl Meurer 2000, in Massive Stellar Clusters, eds. Lan¸con & Boily (astro­ph/0003161)
ffl Mulchaey, Wilson, Tsvetanov 1996, ApJS, 102, 309
ffl Oey & Clarke, 1998, ApJ, 115, 1543
ffl Rocha­Pinto, Scalo, Maciel, & Flynn 2000, ApJ, 531, L115
ffl Ryder, & Dopita 1993, ApJS, 88, 415
ffl Sanders, & Mirabel 1996, ARA&A, 34, 749
ffl Schade, & Ferguson 1994, MNRAS, 267, 889
ffl Sommerville & Primack 1999, MNRAS, 310, 1087
ffl Staveley­Smith, et al. 1996, PASAu, 13, 243
ffl Treyer, Ellis, Milliard, Donas, & Bridges 1998, MNRAS, 300, 303
ffl Tyson & Scalo 1988 ApJ 329, 618
ffl Young, Allen, Kenney, Lesser, & Rownd, 1996, AJ, 112, 1903

NOAO Proposal Page 6 This box blank.
Management Plan Describe the overall plan for conducting the proposed survey, including the ex­
perimental design, survey deliverables, staffing requirements, and a list of observing runs. See the Survey
Program instructions for details.
The observations. For each source we will obtain images with a narrowband filter whose passband
encompasses redshifted Hff and the broad band R ``continuum''. We will use FWHM ú 35 š A Hff
filters for this study. This width provides a good compromise between low widths which minimize
sky contribution on the one hand, and broader widths which allow the full velocity width of galaxies
to be image, and keep costs reasonable. Our team has one such filter available (Smith's 6573/30).
We will purchase additional 3 00 filters to cover the V r
range of our sample. The data will be obtained
in three frames per band with small offsets between frames for cosmic­ray and bad­pixel removal.
Southern spectro­photometric standards will be observed to flux­calibrate the data.
As detailed in the technical section, a total of fifty nights on the 1.5m with CFIM8 + T2K are
required to obtain high­quality observations of all 500 galaxies in our sample. The observations
will be spread over six semesters (2000B to 2003A) to insure enough time for the HiPASS team to
extract sources to its full depth. Two runs per semester are requested to evenly sample the sky.
Accuracy. Previous studies (e.g. Young et al. 1996) demonstrate that continuum subtraction with
R band images will be adequate for our purposes, producing Hff fluxes accurate to ¸ 20%. The
satellite [Nii]6548,6584 š A lines also limit the accuracy of Hff flux determinations. From typical Hii
region spectra and the pass band of the filter we estimate that the Hff fluxes will typically be
boosted ¸ ! 20% by [Nii] contamination. We will correct for this effect statistically, using published
Hii region spectra. The correction factor will be tuned to the Hi velocity profile of each galaxy.
Products. For each galaxy, the products that we will make available to the community are:
(1) Flux calibrated continuum subtracted Hff image (with error array).
(2) Flux calibrated continuum image (with error array).
(3) Hff and continuum radial surface brightness profiles (with error arrays).
(4) Catalog of Hii region positions and their Hff fluxes.
The final manufacture of these products will be done in a pipeline consisting of a series of IRAF
scripts. Error arrays are important for determining which sources are real. The error arrays will
include Poissonian and read noise terms, and the Hff error array will include the properly propagated
continuum errors and a term for the uncertainty in the Hff/continuum ratio. The radial profiles will
be extracted in elliptical annuli with constant center and axial ratio. These orientation parameters
will be determined from the continuum image using a moment analysis technique. Code to perform
the surface photometry has been developed by Meurer and will be encorporated into the pipeline.
Hii region detection and photometry will be done within the DAOPHOT package, or a specialized
Hii region photometry package (we are currently assessing the options).
Labor. The initial breakdown of responsibilities is listed in Table 1. We expect to also attract one
or two graduate students. They will assist in all aspects of the project. These include observing,
pipeline development, pipeline runs, scientific analysis and paper writing.
Project support. The initial observations will be made with funds available to the individual
observers and team members. Full financial support for the USA based investigators will be re­
quested from the NSF. These funds will support graduate students, buy computer resources, and
archive resources, and support observing runs. We expect enough funds from JHU's Center for As­
trophysical Sciences general research funds, STScI's Director's Discretionary Research Fund, and
Hubble Fellowship grants (Putman) to allow at least five observing trips from the USA per year
until our own grant is obtained. Note, one of our team members (Smith) is on the CTIO staff and
can cover some of the runs with little cost.

NOAO Proposal Page 7 This box blank.
Table 1: Responsibilities
Member Responsibilities
Gerhardt Meurer PI, observer, pipeline construction and overview
Harry Ferguson co­PI, observer, science overview
Rachel Webster Australian­PI, coordination with full HiPASS team
Michael Drinkwater Pipeline development
Ken Freeman Science advisor
Rob Kennicutt Science and technical advisor
Virginia Kilborn Sample selection, pipeline development
Patricia Knezek Observer, source positions
Sally Oey Observer, Hii region photometry specialist
Mary Putman Observer
Chris Smith Observer, filter specialist, local CTIO contact
Lister Staveley­Smith Source positions
Release of Data Describe the timeline and mechanism for the release of data subsets, the complete
dataset, and catalogs to the astronomical community.
Our intention is to make the pipeline as fully automated as possible. We expect that much of
the effort expended in the first year will be spent in constructing scripts of sufficient robustness
to accomplish this goal. We will also construct tools to efficiently examine the output products
so as to insure that the pipeline is behaving. Appropriate flags and comments will be placed in
the image/product headers if human intervention is required. The final check on quality will be
comparison of the derived fluxes with previously published Hff imaging (e.g. Ryder & Dopita 1993;
Young et al. 1996)
The raw data will be archived on CDROM in Baltimore and Melbourne and will be available upon
request. Project web pages will be established at JHU. The pipeline software will be available on
these pages, as will ASCII observing logs, and logs of the pipeline runs. The final data products
listed above will be archived in STScI's Multi­Mission archive, and hence will be available to the
community via the web (http://archive.stsci.edu/mast.html). The images will be stored in fits
format, while the radial profiles and Hii region catalogs will be ASCII tables. The first data
delivery will occur a year after the observations commence, with semesterly data ingestions there
after.
Previous Use of NOAO Facilities List allocations of telescope time on facilities available through
NOAO to the Principal Investigator during the past 2 years, together with the current status of the data
(cite publications where appropriate). Mark with an asterisk those allocations of time related to the current
proposal.
Why CTIO? (For CTIO proposals only.) Explain why access to the southern hemisphere is needed to
achieve your scientific goals.
HiPASS is primarily a southern sample: it results from observations made with the 64m Parkes
Radio Telescope in Australia. It can only be efficiently accessed from a southern site.

NOAO Proposal Page 8 This box blank.
Observing Run Details for Run 1: CT­1.5m/CFIM8 + T2K
Technical Description Describe the observations to be made during this observing run. Justify the
specific telescope, the number of nights, the instrument, and the lunar phase. List objects, coordinates, and
magnitudes (or surface brightness, if appropriate) in the Target Tables section below (required for WIYN­2hr,
WIYN­SYN, YALO, and Gemini runs).
Our scientific goals require fairly deep Hff images, especially to probe the HiiLFs, but also to
measure the integrated Hff fluxes for the galaxies, including the contribution from diffuse emission.
While useful Hff images could be obtained on the 0.9m, the gain in efficiency of using the 1.5m
(3\Theta the collecting area) and the associated improvement in image quality (3\Theta fewer cosmic rays,
less susceptibility to guiding errors) make it the preferred telescope. The equal sensitivity at given
surface brightness often associated with small telescopes is not relevant to the 0.9m because it has
the same focal length as the 1.5m (the pixel scale of the Schmidt is too large for distinguishing Hii
regions). With the exposure times detailed below and about 11 minutes of overhead per galaxy
(filter changes, telescope offsets, and readouts) we require 47 minutes per galaxy, and can observe
about 10 galaxies a night. The entire survey can be completed in 50 nights. In contrast, with the
0.9m two hours per galaxy would be required, only four galaxies a night can be observed, and we
would require 125 nights to complete the survey.
The primary goal of the HiiLF work is to test the models of Oey & Clarke (1998). This requires
that we reach the turn­over in the LF, which typically occurs at LHff ú 10 37 erg s \Gamma1 . Tracing the
HiiLF this deep also insures that the total Hff flux from Hii regions is recovered. This turn­over
corresponds to FHff ú 2 \Theta 10 \Gamma16 erg cm \Gamma2 s \Gamma1 at D = 20 Mpc. Beyond this distance crowding
makes it difficult to probe the peak of the HiiLF. For the 1.5m, this flux corresponds to a count
rate of ú 0:2 e s \Gamma1 , compared to the sky count rate of ú 1:6 e s \Gamma1 through a 35 š A wide filter and a 2 00
aperture at 10 days lunation. This flux can be measured with 20% random errors in 30 minutes of
exposure with the 1.5m. As noted earlier, the flux error from continuum subtraction is also ¸ 20%
so there is no need to go deeper.
For diffuse emission detection, a limiting surface brightness ¸ ? 1% of sky is usually set by flatfielding
uncertainties. The observations are easily sky limited even in dark time. One percent of sky
corresponds to SHff ú 10 \Gamma17 erg cm \Gamma2 s \Gamma1 arcsec \Gamma2 for 10 days lunation and a 35 š A wide filter. This
is about the state of the art in simple narrow­band Hff surface photometry (e.g. Ferguson et al.
1998). Hence 10 days lunation is the most we can tolerate for these observations. The R band
continuum image will be obtained in 3 \Theta 2 minutes, and will easily have high enough S=N that its
subtraction will not add significant Poissonian or readnoise to the Hff image.
A complete list of targets is not yet available. However, ú 3000 HiPASS detections spread over
the whole southern sky will be available for selection, so we will have sufficient targets for any time
allocations we are assigned. Ideally, each observing run will last 4­5 nights and will occur every
three months or so.
Instrument Configuration
Filters: R band and our own Hff filters. Slit: Fiber cable:
Grating/grism: Multislit: Corrector:
Order: – start : Collimator:
Cross disperser: –end : Atmos. disp. corr.:

NOAO Proposal Page 9 This box blank.
Special Instrument Requirements Describe briefly any special or non­standard usage of instru­
mentation.
NOAO observing proposal L A T E X macros v2.2.