Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://hea-www.harvard.edu/CHAMP/PUBLICATIONS/ChaMP_NEWFA.ps Äàòà èçìåíåíèÿ: Wed Nov 15 20:34:52 2000 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 02:33:00 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: m 31

**TITLE**
ASP Conference Series, Vol. **VOLUME**, **PUBLICATION YEAR**
**EDITORS**
The Chandra Multi­wavelength Project (ChaMP): a
serendipitous survey with Chandra archival data.
B. J. Wilkes, P. Green, R. Brissenden, R. Cameron, A. Dobrzycki, J.
Drake, N. Evans, A. Fruscione, T. Gaetz, M. Garcia, H. Ghosh, J.
Grimes, J. Grindlay, E. Hooper, M. Karovska, V. Kashyap, D.­W. Kim,
K. Kowal, H. Marshall, A. Mossman, D. Morris, J. Nichols, A.
Szentgyorgyi, H. Tananbaum, L. van Speybroeck, A. Vikhlinin, S.
Virani, P. Zhao
Smithsonian Astrophysical Observatory (SAO), 60 Garden St.,
Cambridge, MA 02138, USA
J. Baldwin, A. Kindt (MSU), F. Chaffee (Keck), A. Dey, B. Jannuzi
(NOAO), C. Foltz (MMTO), S. Mathur (OSU), B. McNamara (Ohio),
H. Newberg (RPI)
Abstract. The launch of the Chandra X­ray Observatory in July 1999
opened a new era in X­ray astronomy. Its unprecedented, ! 0:5'' spatial
resolution and low background are providing views of the X­ray sky 10­100
times fainter than previously possible. We have initiated a serendipitous
survey (ChaMP) using Chandra archival data to flux limits covering the
range between those reached by current satellites and those of the small
area Chandra deep surveys. We estimate the survey will cover ¸ 5 sq.deg.
per year to X­ray fluxes (2\Gamma10 keV) in the range 10 \Gamma13 \Gamma 6 \Theta 10 \Gamma16 erg
cm \Gamma2 s \Gamma1 discovering ¸2000 new X­ray sources, ¸ 80% of which are ex­
pected to be active galactic nuclei (AGN). The ChaMP has two parts,
the extragalactic survey (ChaMP) and the galactic plane survey (ChaM­
Plane). Over five years of Chandra operations, the ChaMP will provide
both a major resource for Chandra observers and a key research tool for
the study of the cosmic X­ray background (CXRB) and the individual
source populations which comprise it. ChaMP promises profoundly new
science return on a number of key questions at the current frontier of
many areas of astronomy including (1) locating and studying high red­
shift clusters and so constraining cosmological parameters, (2) defining
the true population of AGN, including those that are absorbed, and so
constraining the accretion history of the universe, (3) filling in the gap in
the luminosity/redshift plane between Chandra deep and previous sur­
veys in studying the CXRB, (4) studying coronal emission from late­type
stars as their cores become fully convective and (5) search for cataclysmic
variables (CVs) and quiescent Low­Mass X­ray Binaries (qLXMBs) to
measure their luminosity functions.
In this paper we summarize the status, predictions and initial results
from the X­ray analysis and optical imaging.
1

2 Wilkes et al.
1. Introduction
The large fields­of­view (FoVs) typical of X­ray imaging instruments have long
been exploited for serendipitous surveys, resulting in far­reaching and fundamen­
tal advances in our knowledge of the X­ray universe and indeed the universe as a
whole (e.g., the Einstein Medium Sensitivity Survey (EMSS, Stocke et al. 1991,
Maccacaro et al. 1982); the Cambridge­Cambridge ROSAT Serendipitous Sur­
vey (CCRSS, Boyle, Wilkes & Elvis 1997); ROSAT International X­ray/Optical
Survey (RIXOS, Page et al. 1996)). The next decade promises to be rich for
X­ray astronomy as two major observatories, Chandra and XMM, open new
vistas with their deeper flux limits and superior spatial and spectral resolution.
Figure 1. Predicted source
counts per flux bin for
ChaMP Cycle 1.
To take full advantage of the rich serendipitous dataset afforded by Chan­
dra's large FoVs (ACIS­I: 16 0 \Theta 16 0 ) and to maximize the science per ksec of
observing time, we have organized an extragalactic serendipitous X­ray survey,
the ChaMP and a similar Galactic plane survey (ChaMPlane). We estimate a
ChaMP X­ray sample of ¸ 2000 sources in the ¸ 5 sq. degs. of ChaMP fields
in Cycle­1. ¸ 80% of these are likely to be AGN of various types, including
those with low luminosity and/or substantial low energy X­ray absorption ex­
pected based on the spectrum of the CXRB and already being found in early
results from Chandra surveys (Hornschemeier et al. 2000, Mushotzky et al.
2000, Barger et al. 2000, Giacconi et al. 2000). The remaining sources will in­
clude clusters of galaxies, stars, X­ray binaries, and supernova remnants. Unlike
targeted, small area, deep Chandra surveys, which reach fainter flux levels, the
ChaMP provides the wide area essential for finding rare, bright and/or unex­
pected sources, generating statistically meaningful samples of rare source types
such as BL Lac objects, quiescent X­ray binaries and high­redshift clusters. As
such the ChaMP complements existing deep Chandra surveys (Figure 1).
2. ChaMP:Data Analysis
2.1. Field List: ChaMP, extragalactic survey
Of prime importance in carrying out any survey is an accurate understanding
of the selection effects present. We have defined a subset of Chandra fields

The Chandra Multiwavelength Project 3
to minimize bias due to the effects of known extended sources and to include
a variety of target types. Each field will be analyzed as it becomes available
in the archive using uniform X­ray source detection and analysis techniques
(Section 2.2) resulting in a well­defined X­ray sample over the full survey area.
The 172 fields in the ChaMP extragalactic Cycle­1 and Cycle­2 lists have
jbj ? 20 ffi . We exclude targeted survey fields, e.g. Hubble Deep Field, Lockman
Hole (being studied optically by their Chandra PIs); special instrument modes
(e.g., grating observations, ACIS sub­arrays); unstable detector periods; ! 4
ACIS chips and fields containing bright optical or X­ray sources covering ? 10%
of the Chandra FoV on the DSS optical or ROSAT/other X­ray images.
2.2. X­ray Data Analysis
We are currently refining our source detection procedures using two methods:
sliding cell search and wavelet transformation; the former is optimized for point
source detection while the latter provides efficient detection of diffuse/extended
sources and those in crowded fields. The Chandra point­spread­function (PSF) is
a strong function of off­axis angle and energy so we are carrying out simulations
of source detection efficiency as a function of size, off­axis angle, background
characteristics, source spectrum etc. to understand survey limits and minimize
false sources while optimizing source detection. X­ray source properties will be
derived for sources in the master (merged) list including: accurate positions,
count rates and fluxes, spatial extent, variability and spectrum/hardness ratio
when possible given the source signal. ACIS sources with 100 counts will con­
strain simple power law fits to within 25%, 30 counts provide hardness ratios
adequate to distinguish between the high energy spectra of AGN and typically
softer stellar sources and to provide an estimate of the soft X­ray absorption.
Our log N \Gamma log S estimates suggest ¸ 400 AGN per year will yield useful hard­
ness ratios, and ¸ 130 will furnish enough counts for a spectrum.
2.3. Optical Imaging
The second essential ingredient for an X­ray survey, which opens up a much
wider range of scientific questions, is optical identification of the X­ray sources.
We are obtaining optical images of the ChaMP fields using the same filters with
matched flux limits: F x =F opt ! 0:5 designed to identify ?90% of the normal AGN
population and automatically includes larger fractions of other known source
types. This is very efficient, observing all the X­ray sources in one Chandra
field at one time, minimizing the large, ground­based telescope time required to
reach the faint optical limiting magnitudes (21 !
¸ V !
¸ 25, with mode of 24). The
choice of filters, Sloan Digital Sky Survey (SDSS) g 0 ,r 0 ,i 0 , provides direct benefit
from the SDSS photometric/spectroscopic database and simulations at brighter
limits, allowing excellent photometric classification and redshift determination
(Vikhlinin et al. 1998), invaluable at the faintest optical fluxes. To date we have
observed 41 ChaMP fields in 31 observing nights (9 nights:4­m KPNO (Kitt
Peak National Observatory), 6:0.9m KPNO, 3:4m CTIO (Cerro Tololo Inter­
American Observatory), 17:1.2m SAO). The uniformity of the dataset and our
analysis methods will provide a well­understood and highly uniform sample.

4 Wilkes et al.
Figure 2.
ACIS data for the
MS1137.5+6625 clus­
ter showing the X­ray
sources (circles, size
/ log(significance)).
Crosses indicate 48
possible optical coun­
terparts !2 00 .
2.4. Optical Identification
Chandra's small PSF and unprecendented, for X­ray telescopes, astrometric cal­
ibration minimize the number of possible optical counterparts per X­ray source.
We are developing procedures to reduce and analyze the optical images, derive
source lists and cross­correlate (solving for aspect differences), to identify the
X­ray sources. In Figure 2 we show preliminary results for a 110 ksec ACIS­I ob­
servation based on a preliminary 900 sec i 0 ­band image taken with the 4­shooter
CCD imager on the SAO 1.2m telescope, 12 Feb 2000. The flux limit, i 0 ¸ 18, is
¸ \Theta15 brighter than the ChaMP required limit. The 165 detected X­ray sources
found using the CIAO (Chandra Interactive Analysis of Observations) wavdetect
tool and those with optical identifications ! 2 00 of the X­ray centroid.
We will determine upper limits in optical/X­ray as appropriate for sources
not detected in both bands. Sources will be classified using X­ray size, optical
and X­ray colors employing color­color plane methods (Newberg et al. 1999).
2.5. ChaMP Archive
We will release a list of X­ray sources and their properties along with optical
images and source lists in to a web­based public archive (the X­ray data will be
available via the Chandra public archive). The ChaMP database will provide
an invaluable, multi­wavelength database to the community, helping to ensure
that the science return from Chandra reaches its full potential.
3. ChaMPlane: galactic survey
ChaMPlane selected 15 fields in Cycle­1 such that jbj ! 10 ffi and exposure time
? 30 ksecs. The X­ray data analysis is being carried out in close collaboration

The Chandra Multiwavelength Project 5
with the ChaMP (Section 2.2). However optical imaging in R and Hff filters
is aimed at optimizing the identification of CVs and qLMXBs, primarily black
hole X­ray novae in quiescence, in order to constrain and ultimately measure
their luminosity functions. Secondary objectives are to determine the Be X­
ray binary content and stellar coronal source distributions in the Galaxy. The
deep Chandra galactic fields will detect 2­6 keV fluxes (allowing for low energy
absorption) of F x (2­6 keV) = 2 \Theta10 \Gamma15 erg s\Gamma1 and thus CVs or qLMXBs with
L x = 10 31 erg s \Gamma1 out to ¸7 kpc. Thus most CVs and qLMXBs in the Galaxy can
be reached, and the ChaMPlane survey offers the best chance for constraining
their formation/evolution and the stellar BH content of the Galaxy. CVs and
qLMXBs will be identified by their ubiquitous Hff excess as ``blue'' objects in
the R vs. (Hff ­ R) plane down to R ¸ 24. We have demonstrated this technique
for crowded fields with our HST discovery of the first CVs in globular cluster
cores and have now conducted a successful pilot ChaMPlane survey at CTIO.
ChaMPlane has been awarded 31 4­m nights of observing time at KPNO and
CTIO as part of the NOAO Survey Program.
4. Preliminary Results
4.1. X­ray log N \Gamma log S
We have generated a preliminary log N \Gamma log S from the CIAO wavdetect list of
165 background X­ray sources in the MS1137.5+6625 field and the 127 sources
in the field of the cluster: CL 0848.6+4453. Our results, which assume a power
law with ff E =1.0, (Figure 3) agree well with those from ROSAT (Hasinger et
al. 1998) and extend to fainter fluxes.
Figure 3. Prelimi­
nary ChaMP log N \Gamma
log S (crosses: MS
1137.5+6625; squares:
CL 0848.6+4453) with
1oe error bars, com­
pared with the ROSAT
deep survey (circles,
Hasinger et al. 1989).
4.2. Optical Spectroscopy
We plan to take advantage of the new generation of large area, multi­object
spectrographs to efficiently obtain spectra and provide redshifts for a subset of

6 Wilkes et al.
¸ 1500 sources, sufficient to constrain to ¸ \Sigma20% the points on AGN luminosity
functions (LF) for 0!z!4. Currently available instruments (e.g. the 6.5m MMT,
on which we have 10 nights per year) limit to V !
¸ 22 but we plan to expand our
sample beyond this once the new generation of large telescopes come on­line.
CIV CIII] MgII
114024.8+660216
SiIV+OIV]
CIV CIII]
MgII
114028.1+660320
5000 7500
1
2
CIII]
MgII
FeII
114022.0+660429
114015.7+660143
[OIII]
CaII H&K
5000 7500
1
Figure 4. Spectra
of 3 ChaMP QSOs
and 1 galaxy from
Figure 2 taken April
29, 2000 on the Keck
with LRIS, courtesy
of the Keck Director.
Initial spectra were obtained the Low­Resolution Imaging Spectrograph
(LRIS) on the Keck telescope in director's discretionary time, 29 April 2000.
Slitlets were placed at the positions of the closest optical couterparts to 12 X­
ray soruces in a region of the MS1137.7+6625 field several of which had a low
probability due to the large separation. The resulting 3600sec spectra revealed
3 stars, 5 QSOs, 2 early­type galaxies, a narrow emission line galaxy (NELG),
and one mystery object with very smooth continuum (star/BL Lac?). Four of
the spectra: 3 QSOs and 1 galaxy are presented in Figure 4.
This work was supported by NASA contract: NAS8­39073 (CXC)
References
Boyle, B.J., Wilkes, B.J. & Elvis, M.S. 1997, MNRAS, 285, 511
Barger, A.L, et al., 2000, AJsubmitted (astro­ph/0007175)
Giacconi, et al., 2000, ApJ, submitted (astro­ph/007240)
Hasinger, G. et al., 1998, A&A, 329, 482
Hornschemeier, A.E., et al. 2000, ApJ, submitted , astroph/0004260
Maccacaro, T., et al. 1982, ApJ, 253, 504
Mushotzky, R.F., et al., 2000, Nature, 404, 459
Newberg, H., J., Richards, G.T., Richmond, M. & Fan, X. 1999, ApJS, 123, 377
Page, M.J., et al. 1996, MNRAS, 281, 579
Stocke, J. T. et al. 1991, ApJS, 76, 813
Vikhlinin, A. et al. 1998, ApJ, 502, 558