J/ApJ/796/105 Catalog of X-ray sources in the NARCS (Fornasini+, 2014)
The Norma arm region Chandra survey catalog: X-ray populations in the spiral
arms.
Fornasini F.M., Tomsick J.A., Bodaghee A., Krivonos R.A., An H., Rahoui F.,
Gotthelf E.V., Bauer F.E., Stern D.
<Astrophys. J., 796, 105 (2014)>
=2014ApJ...796..105F 2014ApJ...796..105F (SIMBAD/NED BibCode)
ADC_Keywords: Milky Way ; X-ray sources ; Binaries, X-ray ; Photometry, CCD
Keywords: binaries: general - catalogs - Galaxy: disk -
novae, cataclysmic variables - X-rays: binaries - X-rays: stars
Abstract:
We present a catalog of 1415 X-ray sources identified in the Norma Arm
Region Chandra Survey (NARCS), which covers a 2°x0.8° region
in the direction of the Norma spiral arm to a depth of ∼20 ks. Of
these sources, 1130 are point-like sources detected with ≥3σ
confidence in at least one of three energy bands (0.5-10, 0.5-2, and
2-10 keV), five have extended emission, and the remainder are detected
at low significance. Since most sources have too few counts to permit
individual classification, they are divided into five spectral groups
defined by their quantile properties. We analyze stacked spectra of
X-ray sources within each group, in conjunction with their fluxes,
variability, and infrared counterparts, to identify the dominant
populations in our survey. We find that ∼50% of our sources are
foreground sources located within 1-2 kpc, which is consistent with
expectations from previous surveys. Approximately 20% of sources are
likely located in the proximity of the Scutum-Crux and near Norma arm,
while 30% are more distant, in the proximity of the far Norma arm or
beyond. We argue that a mixture of magnetic and nonmagnetic
cataclysmic variables dominates the Scutum-Crux and near Norma arms,
while intermediate polars and high-mass stars (isolated or in
binaries) dominate the far Norma arm. We also present the cumulative
number count distribution for sources in our survey that are detected
in the hard energy band. A population of very hard sources in the
vicinity of the far Norma arm and active galactic nuclei dominate the
hard X-ray emission down to fX∼10-14 erg/cm2/s, but the
distribution curve flattens at fainter fluxes. We find good agreement
between the observed distribution and predictions based on other surveys.
Description:
We performed Chandra ACIS-I observations in faint mode of a 2°x0.8°
region of the Norma spiral arm in 2011 June. The ACIS-I consists of
four 1024x1024 pixel CCDs, covering a 17'x17' field of view (Garmire et al.
2003SPIE.4851...28G 2003SPIE.4851...28G). The on-axis spatial resolution of the ACIS-I is
fully sampled by the 0.492"x0.492" CCD pixel but it increases greatly
off-axis. The PSF increases in size and becomes more elliptical at
large off-axis angles, such that at an off-axis angle of 10', the PSF
has ellipticity ∼0.3 and semi-major axis ∼15' for an enclosed-count
fraction (ECF) of 90% for 4.5 keV photons (Allen et al.
2004SPIE.5165..423A 2004SPIE.5165..423A). The CCDs are sensitive to incident photons with
energies in the 0.3-10.0 keV range, and have a resolution of about
50-300 eV. The time resolution of the CCDs, which is determined by
the read-out time, is 3.2 s.
Objects:
----------------------------------------------------------
RA (ICRS) DE Designation(s)
----------------------------------------------------------
16 00 -50.0 Norma arm = NAME Norma Arm
----------------------------------------------------------
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This fileil
table3.dat 147 1415 Catalog of Point and Extended Sources:
Detection and Localization
table4.dat 142 1415 Catalog of Point and Extended Sources:
Photometry
table5.dat 60 1415 Catalog of Point and Extended Sources:
Infrared Counterparts
--------------------------------------------------------------------------------
See also:
IX/45 : The Chandra Source Catalog, Release 1.1 (Evans+ 2012)
II/348 : VISTA Variable in the Via Lactea Survey DR2 (Minniti+, 2017)
J/ApJS/134/77 : ASCA Galactic Plane Survey faint X-ray sources (Sugizak+, 2001)
J/ApJ/635/214 : Chandra X-ray sources and NIR identifications (Ebisawa+, 2005)
Byte-by-byte Description of file: table3.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 4 I4 --- NARCS [1/1415] NARCS catalog number
6- 20 A15 --- ID Chandra source identifier
(CXOU JHHMMSS.s+DDMMSS in Simbad)
22- 29 A8 --- ObsID Observation identifier (1)
31- 40 F10.6 deg RAdeg Right Ascension in decimal degrees (J2000) (2)
42- 51 F10.6 deg DEdeg Declination in decimal degrees (J2000) (2)
53- 57 F5.2 arcsec PosUnc Positional uncertainty (3)
59- 71 A13 arcmin Off Offset angular separation of source and aim point
73- 78 F6.1 --- SigFB Full 0.5-10 keV band significance (4)
80- 84 F5.1 --- SigSB Soft 0.5-2 keV band significance (4)
86- 91 F6.1 --- SigHB Hard 2-10 keV band significance (4)
93-106 A14 arcsec Radius Radius of aperture source region (5)
108-120 A13 arcsec PSF Point spread function (6)
122-147 A26 --- Flag Source flag(s) (7)
--------------------------------------------------------------------------------
Note (1): Observation(s) in which wavdetect detects the source. The format of
ObsID numbers is 125XX, where the last two digits are those provided
in the catalog. See Section 2.1 for details about wavdetect usage.
Note (2): If the source is detected in multiple observations, the position
reported is the weighted average of its positions in different
observations.
Note (3): For a source detected in a given observation, this uncertainty is
equal to the quadrature sum of the 95% statistical uncertainty based
on Equation (5) of Hong et al. (2005ApJ...635..907H 2005ApJ...635..907H) and the average
systematic uncertainty of positions in that observation after
astrometric refinement (see Column 5 in Table 2). For sources detected
in multiple observations, the uncertainties associated with the source
position in different observations were combined to provide the
uncertainty of the weighted average of the source positions.
Note (4): It is calculated by finding the probability that the source is a noise
fluctuation using Equation (5) and using the Gaussian cumulative
distribution function to determine the corresponding source
significance. If the source is detected in multiple observations, the
reported significance is the sum in quadrature of the source
significance in individual observations.
Note (5): For most sources, the aperture source region is defined as a circle
with radius equal to the 90% ECF for 4.5 keV photons (see Column 12).
For potentially extended sources, flagged with "e" (see Column 13),
the radius is instead equal to the semi-major axis of the aperture
region defined by wavdetect. In cases where two or more sources have
overlapping circular regions, the regions are redefined as a circular
core plus an annular pie sector following the guidelines in Table 6;
in such cases, the radius provided in the catalog represents the outer
radius of the pie sector. For sources detected in multiple
observations, a semicolon-separated list of the aperture region radius
used in different observations is provided; the order of radii matches
the order of ObsIDs reported in Column 3.
Note (6): PSF radius for 90% ECF for 4.5 keV photons at the detector location
of the source. The PSF radius varies with detector position, generally
increasing with increasing offset angle from the observation aim
point. For sources detected in multiple observations, a
semicolon-separated list of the PSF radius at the source detector
position in different observations is provided; the order of PSF radii
matches the order of ObsIDs reported in Column 3.
Note (7): Flag as follows:
b = "blended": blended source that is unblended in another observation.
c = "created": source noticed by eye but not detected by wavdetect. The
source aperture region was created manually based on the visible
position and extent of the source. The positional uncertainties
calculated for such sources underestimate the true uncertainties,
since the source is found by eye and not by wavdetect.
e = "extended": possibly extended source. The semi-major axis of the
smallest aperture region defined by wavdetect for such sources is
larger than twice the PSF radius reported in Column 12. These sources
are typically detected in images that have been binned by
4 x 4 or 8 x 8 pixels.
id = "inspected duplicate": possible duplicate source flagged for manual
inspection. A "duplicate" source refers to a single source detected
in multiple overlapping observations; sources were considered to be
duplicates of one another if the distance between them was smaller
than the quadrature sum of their positional uncertainties. Sources
were flagged for manual inspection if (1) they were separated by a
distance greater than the quadrature sum of their positional
uncertainties but smaller than the simple sum of their positional
uncertainties, or (2) they were separated by a distance smaller than
the quadrature sum of their positional uncertainties but differed in
a substantial way (e.g., one is flagged as possibly extended while
another is not, one is found to have two duplicates by the distance
criterion but these two duplicates of the first source are not found
to be duplicates of one another by the distance criterion). Generally,
if sources flagged with "id" showed consistent photon fluxes and
quantile parameters, they were determined to be true duplicates.
m1, m2 or m3 = "modified"; in cases where the circular source aperture region
overlaps with the aperture region of another source, the source
region is modified to reduce overlapping. See Table 6 for
details;
nb = "near bright"; source near a very bright source which may be a spurious
detection;
nd = "not detected"; source is located where at least two observations overlap
but it is only detected in one observation;
s = "surrounding"; a possibly extended source that completely surrounds one
or more point sources. The aperture regions of the surrounded sources are
excluded from the aperture region of the source flagged with "s";
vl = "variable long"; source determined to be variable on long (hours-days)
timescales. The photon flux in at least one energy band (full, soft, or
hard) varies by ≥ 3σ between different observations;
vp = "variable probable"; source is probably variable on short (second-hour)
timescales. The K-S test finds the source light curve within a single
observation to be inconsistent with a constant light curve at ≥95%
confidence;
vs = "variable short"; source is variable on short (second-hour) timescales.
The K-S test finds the source light curve within a single observation to
be inconsistent with a constant light curve at ≥3σ confidence.
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table4.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 4 I4 --- NARCS [1/1415] NARCS catalog number
6- 10 I5 ct FBNet Net full 0.5-10 keV band source counts (1)
12- 14 I3 ct E_FBNet ? Upper 1σ error in FBNet
16- 18 I3 ct e_FBNet ? Lower 1σ error in FBNet
20- 22 I3 ct SBNet Net soft 0.5-2 keV band source counts (1)
24- 25 I2 ct E_SBNet ? Upper 1σ error in SBNet
27- 28 I2 ct e_SBNet ? Lower 1σ error in SBNet
30- 34 I5 ct HBNet Net hard 2-10 keV band source counts (1)
36- 38 I3 ct E_HBNet ? Upper 1σ error in HBNet
40- 42 I3 ct e_HBNet ? Lower 1σ error in HBNet
44- 50 F7.2 10-6/cm2/s fphFB Full 0.5-10 keV band photon flux (2)
52- 55 F4.2 10-6/cm2/s E_fphFB ? Upper 1σ error in fphFB
57- 60 F4.2 10-6/cm2/s e_fphFB ? Lower 1σ error in fphFB
62- 66 F5.2 10-6/cm2/s fphSB Soft 0.5-2 keV band photon flux (2)
68- 71 F4.2 10-6/cm2/s E_fphSB ? Upper 1σ error in fphSB
73- 76 F4.2 10-6/cm2/s e_fphSB ? Lower 1σ error in fphSB
78- 84 F7.2 10-6/cm2/s fphHB Hard 2-10 keV band photon flux (2)
86- 89 F4.2 10-6/cm2/s E_fphHB ? Upper 1σ error in fphHB
91- 94 F4.2 10-6/cm2/s e_fphHB ? Lower 1σ error in fphHB
96- 98 F3.1 keV E50 Median source energy (3)
100-102 F3.1 keV e_E50 The 1σ error in E50
104-106 F3.1 keV E25 Energy below which 25% of total source
counts reside (3)
108-110 F3.1 keV e_E25 The 1σ error in E25
112-115 F4.1 keV E75 Energy below which 75% of total source
counts reside (3)
117-119 F3.1 keV e_E75 The 1σ error in E75
121-126 F6.2 10-17W/m2 fXFB Full 0.5-10 keV band flux (4)
128-131 F4.2 10-17W/m2 E_fXFB ? Upper 1σ error in fXFB
133-136 F4.2 10-17W/m2 e_fXFB ? Lower 1σ error in fXFB
138-140 A3 --- PFlag Photometric flag (5)
142 A1 --- Q [ABCDE] Quantile group (6)
--------------------------------------------------------------------------------
Note (1): Calculated as described in Section 2.2. For cases in which the
estimated background counts in a source aperture region were
determined to be greater than or equal to the total number of counts
in the source region, then the catalog presents the 90% upper
confidence limit to the net source counts based on the method
described in Kraft, Burrows & Nousek (1991ApJ...374..344K 1991ApJ...374..344K); in such
cases, the error columns are left blank. For sources detected in
multiple observations, net counts from different observations were
added together and errors combined in quadrature.
Note (2): The photon flux was calculated by dividing the net counts by the
exposure time and the mean effective area within the source region.
For sources with zero or negative net counts, the catalog provides
the 90% upper limit on the photon flux and leaves the error columns
blank. For sources detected in multiple observations, the average
photon fluxes are reported; if a source was found to be variable
between observations (flagged as "vl") then its photon fluxes from
individual observations were simply averaged, but otherwise its
photon fluxes were weight-averaged.
Note (3): It is determined from the total counts (not background corrected) in
the source region. For sources detected in multiple observations, the
simple average of the energies from individual observations is
reported if a source is found to be variable between observations or
the weighted-average is reported otherwise.
Note (4): This estimate of the energy flux is calculated by multiplying the full
band photon flux and the median energy of the source provided in the
catalog. In cases where only an upper limit to the photon flux is
available, the 90% upper limit to the energy flux is reported and
the error columns are left blank.
Note (5): If the photometric values provided for a source are 90% upper limits
in the full, soft, or hard energy bands, this column displays an F,
S, or H, respectively.
Note (6): The spectral group defined using quantile diagrams to which the
source belongs. Group as follows:
A = The group A stacked spectrum is best fit by a two-temperature thermal
plasma model. The column density associated with each temperature
component is low (~<1021 cm-2), suggesting most sources in this group
are foreground sources, located at a distance ~<1 kpc.
B = Similar to group A, the group B stacked spectrum is also best-fit by a
two-temperature thermal model with low hydrogen column densities.
However, the temperature of the hotter component is significantly higher
(kT∼7 keV) for the B sources than for the A sources. The origin of the
low-temperature component may be low-mass X-ray active stars, ABs, and
symbiotic binaries, the dominant sources in group A, while the
high-temperature component is more typical of CVs.
C = The group C stacked spectrum is best-fit by an absorbed power-law with
Γ∼1.1 and NH=1.4x1022 cm-2, which suggests that these sources
are located at a distance of 3-5 kpc, in the Scutum-Crux and near Norma
spiral arms. The luminosities spanned by group C sources are
LX=1031-1032.7 erg/s, assuming a distance of 4 kpc. Possible classes
of X-ray sources present in this group are magnetic and nonmagnetic CVs,
hard-spectrum symbiotic binaries, low-mass X-ray binaries (LMXBs), and
HMXBs.
D = The group D stacked spectrum has a very hard photon index (Γ∼0.7),
a prominent Fe line, and a high NH, indicating that these sources
typically lie on the far side of the Galaxy, near, in, or beyond the far
Norma arm. The presence of this strong, non-redshifted Fe line suggests
that many of the sources in this group must be Galactic; otherwise, if
this group were dominated by AGNs, their spread in redshift would result
in a smearing out of the Fe line.
E = The group E stacked spectrum is best-fit by a two-temperature thermal
model, making it significantly different from the group D power-law
spectrum, even though some of the same classes of X-ray sources must be
present in both groups D and E since they are not sharply separated in the
quantile diagram. The photometric and spectral properties of group E are
also consistent with symbiotic binaries.
See Section 3 for details about quantile analysis and the spectral groups
defined in this work.
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table5.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 4 I4 --- NARCS [1/1415] NARCS catalog number
6- 17 A12 --- VVV VVV source identifier; if available
(Minniti et al. 2010NewA...15..433M 2010NewA...15..433M)
19- 28 F10.6 deg RAdeg ? Right Ascension in decimal degrees (J2000)
30- 39 F10.6 deg DEdeg ? Declination in decimal degrees (J2000)
41- 44 F4.2 arcsec Delta ? Angular separation between Chandra & VVV source
46- 53 E8.2 --- pnoise ? Probability VVV source is a noise fluctuation
55- 60 F6.4 --- Rel ? Reliability (1)
--------------------------------------------------------------------------------
Note (1): Of the VVV counterpart calculated according to the method of
Sutherland & Saunders (1992MNRAS.259..413S 1992MNRAS.259..413S). The reliability depends
on the distance between the X-ray and IR sources, the positional
uncertainties of the X-ray and IR sources, and the spatial density of
IR sources. The reliability is expressed as a fraction between zero
and one; VVV sources with a higher reliability are more likely to be
true IR counterparts to the Chandra sources.
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Prepared by [AAS], Tiphaine Pouvreau [CDS] 07-Aug-2017