J/ApJ/740/87 Chandra observations of radio transients (Croft+, 2011)
X-ray observations of radio transients without optical hosts.
Croft S., Tomsick J.A., Bower G.C.
<Astrophys. J., 740, 87 (2011)>
=2011ApJ...740...87C 2011ApJ...740...87C
ADC_Keywords: X-ray sources ; Radio continuum
Keywords: radio continuum: general - stars: activity - stars: flare -
stars: neutron - supernovae: general - X-rays: general
Abstract:
We present a 50ks Chandra/ACIS-I X-ray observation of the Bower et al.
(2007ApJ...666..346B 2007ApJ...666..346B) Very Large Array archival field. The
observations reach a limiting sensitivity of ∼10-4counts/s,
corresponding to a flux of a few times 10-15erg/s/cm2 for the
models we explore. The Chandra observations were undertaken to search
for X-ray counterparts to the eight transient sources without optical
counterparts, and the two transient sources with optical counterparts
seen by Bower et al. Neither of the sources with optical counterparts
was detected in X-rays. One of the eight optical non-detections is
associated with a marginal (2.4σ) X-ray detection in our Chandra
image. A second optically undetected Bower et al. transient may be
associated with a z=1.29 X-ray-detected quasar or its host galaxy, or
alternatively is undetected in X-rays and is a chance association with
the nearby X-ray source. The X-ray flux upper limits, and the one
marginal detection, are consistent with the interpretation of Ofek et
al. (2010ApJ...711..517O 2010ApJ...711..517O) that the optically undetected radio
transients are flares from isolated old Galactic neutron stars. The
marginal X-ray detection has a hardness ratio that implies a
temperature too high for a simple one-temperature neutron star model,
but plausible multi-component fits are not excluded, and in any case
the marginal X-ray detection may be due to cosmic rays or particle
background. The X-ray flux upper limits are also consistent with flare
star progenitors at ≳1kpc (which would require the radio luminosity
of the transient to be unusually high for such an object) or less
extreme flares from brown dwarfs at distances of around 100pc.
Description:
Bower et al. (B07; 2007ApJ...666..346B 2007ApJ...666..346B) analyzed archival data at 5GHz
and 8.4GHz of a Very Large Array (VLA) calibration field centered at
RA=15:02:20.53, DEC=78:16:14.905; J2000. The data spanned ∼20yr with
944 epochs. The images had an FWHM of 8.6' and 5.1' at 5 and 8.4GHz,
respectively.
Optical and infrared (OIR) observations of the field were also
reported by B07, both with the Low Resolution Imaging Spectrometer
(LRIS) on Keck I.
X-ray observations were obtained with Chandra during Cycle 11, using
the Advanced CCD Imaging Spectrometer (ACIS)-I with no grating or
filter. The exposure time was 49441s and the observation start date
was UT 2010 July 3 01:14:32. The observation was centered at
RA=15:01:50.58, DEC=78:16:50.90 (J2000) so as to encompass the
majority of the VLA field.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table1.dat 121 18 VLA sources and their Chandra counterparts
table2.dat 66 128 Chandra sources
table4.dat 49 10 X-ray fluxes for radio transients
--------------------------------------------------------------------------------
See also:
B/chandra : The Chandra Archive Log (CXC, 1999-2012)
VIII/87 : The Allen Telescope Array 20cm Survey (ATATS). I. (Croft+ 2010)
J/ApJ/725/1792 : Pi Ghz Sky Survey (PiGSS). I. (Bower+, 2010)
J/ApJ/698/1398 : H1743-322 2003 outburst (McClintock+, 2009)
J/ApJ/663/81 : SED of hard X-ray selected AGN in XMDS (Polletta+, 2007)
J/ApJ/578/405 : Neutron stars in NGC 5139 (Rutledge+, 2002)
J/MNRAS/298/692 : Radio Survey of X-ray sources (Fender+ 1998)
Byte-by-byte Description of file: table1.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 7 A7 --- VLA VLA identification (NAN/NAN) (1)
9- 10 A2 --- --- [RT]
12- 19 I8 --- RT [19840502/20010331]? Transient from B07 (G1)
21- 22 I2 h RAh [14/15] VLA Hour of right ascension (J2000)
24- 25 I2 min RAm VLA Minute of right ascension (J2000)
27- 31 F5.2 s RAs VLA Second of right ascension (J2000)
33- 36 F4.2 s e_RAs RAs uncertainty (2)
38 A1 --- DE- [+] VLA Sign of declination (J2000)
39- 40 I2 deg DEd [78] VLA Degree of declination (J2000)
42- 43 I2 arcmin DEm VLA Arcminute of declination (J2000)
45- 48 F4.1 arcsec DEs VLA Arcsecond of declination (J2000)
50- 52 F3.1 arcsec e_DEs [0.2/4] DEs uncertainty (2)
54- 57 A4 --- CID Chandra identification (as in table 2)
58- 59 A2 --- f_CID [NC/i ] Remarks (3)
61 A1 --- l_Cts Limit flag on Cts (4)
62- 66 F5.1 ct Cts [3.2/167]? ACIS counts (0.3-10keV) (5)
68- 71 F4.1 ct e_Cts ? Cts uncertainty
73 A1 --- l_Rate Limit flag on Rate
74- 80 E7.4 ct/s Rate ? 0.3-10keV count rate
82- 86 F5.2 --- HR [-0.62/0.52]? Hardness constraint (G2)
88- 91 F4.2 --- e_HR ? HR uncertainty
93- 97 F5.2 arcsec Offset [,11]? Offset between X-ray and radio positions
99-102 F4.2 arcsec e_Offset ? Offset uncertainty (6)
103 A1 --- f_Offset [h] h: X-ray source is extended (7)
105-108 F4.2 --- z ? Redshift from Bower et al. 2007ApJ...666..346B 2007ApJ...666..346B
110-121 A12 --- OIR Optical/IR identification (8)
--------------------------------------------------------------------------------
Note (1): 5 or 8 stand for the 5GHz or 8.4GHz observations, respectively.
Letters stand for (see section 2.1):
T = "transient",
S = "steady": Bower et al. (B07; 2007ApJ...666..346B 2007ApJ...666..346B) also constructed a deep
image of the field using all data at each frequency. The resulting
images at 5 and 8.4GHz had an rms of 2.6 and 2.8uJy, respectively.
Sources which were seen in many of the individual epochs and also in the
deep field image were termed "steady" by B07. They saw eight steady
sources at 5GHz and four of these correspond to steady sources in the
deep image at 8.4GHz.
L = "long-timescale" transient: B07 also compared images made by making
images from two months' worth of data at a time, and found two additional
"long-timescale" transients, one at each frequency.
Note (2): 90% position uncertainties computed by multiplying the uncertainties
from B07 by 1.65.
Note (3): Flag as follows:
NC = outside the coverage of the Chandra image or fall in the gaps between
the chips;
i = Although B07 (2007ApJ...666..346B 2007ApJ...666..346B) and O10 (2010ApJ...711..517O 2010ApJ...711..517O)
consider the radio source to be unidentified, the nearby X-ray source
is associated with a z=1.29 quasar (Section 3.5.5). The radio source
may or may not be physically associated with the X-ray and
optical/IR source (Section 4.7).
Note (4): In cases where upper limits are shown, these are 90% confidence
Poisson limits computed following Gehrels (1986ApJ...303..336G 1986ApJ...303..336G).
Note (5): ACIS counts are reported for 0.3-10keV for the four sources with
cross-matches in the Chandra catalog (Table 2) noted in the column
headed "CID". For these sources, we tabulate the corresponding count
rate from Table 2, and the offset between the X-ray and radio
positions. For sources with no match in the Chandra catalog, we
tabulate counts in apertures centered at the radio positions, and
corresponding 90% confidence upper limits to the 0.3-10keV rate.
Aperture radii were 5", except for 5S7 and 5T6, where a 7.5"-radius
aperture was used, and 5L1, where a 10"-radius aperture was used
(these three sources are rather far off-axis, where the Chandra PSF
is somewhat elongated).
Note (6): The uncertainty in the offset is computed by adding the 90% positional
uncertainties of the X-ray and radio positions in quadrature.
Note (7):
h = The X-ray source is extended, and the offset is computed from the
centroid of the X-ray position to the centroid of the radio position.
Note (8): Classes are Galaxy, Galaxy group, or None.
The two transients with IDs with question marks were considered by
Bower et al. (2007ApJ...666..346B 2007ApJ...666..346B) as possible counterparts, but we
argue here that they are in fact too far away from the OIR sources to
be likely to be associated.
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table2.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 4 A4 --- CID Chandra identification (XNNN)
6- 7 I2 h RAh Hour of right ascension (J2000)
9- 10 I2 min RAm Minute of right ascension (J2000)
12- 16 F5.2 s RAs Second of right ascension (J2000)
18 A1 --- DE- Sign of declination (J2000)
19- 20 I2 deg DEd Degree of declination (J2000)
22- 23 I2 arcmin DEm Arcminute of declination (J2000)
25- 28 F4.1 arcsec DEs Arcsecond of declination (J2000)
30- 33 F4.2 arcsec ePos [0.64/2.28] Chandra position uncertainty (1)
35 I1 --- Bin [1/4] Binning (2)
37- 41 F5.1 ct Cts [3.7/332.7] ACIS counts (0.3-10keV); background
subtracted
43- 46 F4.1 ct e_Cts Cts uncertainty
48 A1 ---- l_HR Limit flag on HR (G1)
49- 53 F5.2 --- HR [-0.62/0.89]? Hardness constraint (G1)
55- 58 F4.2 --- e_HR ? HR uncertainty
60- 66 A7 --- VLA VLA identification (as in table 1)
--------------------------------------------------------------------------------
Note (1): 0% positional uncertainty radius, including the 0.64" systematic
Chandra pointing uncertainty added in quadrature.
Note (2): Binning in which the source was best detected. We used four different
pixel binnings (1, 2, 4, and 8) in order to detect sources that are
either truly extended or sources that appear extended because the
point-spread function (PSF) broadens off-axis. We combined the
catalogs from these four runs. See section 2.2 for further details.
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table4.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 3 A3 --- VLA VLA identification (NAN)
5- 6 A2 --- --- [RT]
8- 15 I8 --- RT [19840502/20010331] Transient from B07 (G1)
17- 20 A4 --- CID Chandra identification
22 A1 --- l_Rate Limit flag on Rate
23- 29 E7.3 ct/s Rate [0.000109/0.00082] 0.3-10keV count rate
31 A1 --- l_Sns Limit flag on Sns
32- 36 F5.2 aW/m2 Sns [3.36/25.3] Unobscured 0.3-10keV flux
(neutron star model) in 10-15erg/s/cm2 (1)
37 A1 --- f_Sns [d] Remark on source (2)
39 A1 --- l_Sfs Limit flag on Sfs
40- 43 F4.2 aW/m2 Sfs Unobscured 0.3-10keV flux (flare star model)
in 10-15erg/s/cm2 (3)
44 A1 --- f_Sfs [d] Remark on source (2)
46- 49 F4.2 mJy S5GHz [0.37/7.04] 5GHz flux density (4)
--------------------------------------------------------------------------------
Note (1): Unobscured flux for a basic neutron star model (blackbody with
kT=0.1keV).
Note (2):
d = The X-ray source appears to be an AGN, and values from more sophisticated
spectral fits for this source from Table 3 (see also Section 3.5.1)
should be used. The X-ray source may not be physically associated with
5T7 (Section 3.5.5).
Note (3): Unobscured flux for a basic flare star/brown dwarf model (thermal
bremsstrahlung with kT=1.0keV).
Note (4): From Column 4 of Tables 2 and 3 of Bower et al. (2007ApJ...666..346B 2007ApJ...666..346B).
For the 8GHz transients, we assume a flat spectral index between 5GHz
and 8GHz, consistent with the discussion in Ofek et al.
(2010ApJ...711..517O 2010ApJ...711..517O).
--------------------------------------------------------------------------------
Global notes:
Note (G1): Bower et al. (B07; 2007ApJ...666..346B 2007ApJ...666..346B) compared the images from
epoch to epoch, and detected seven single-epoch transient sources at
5GHz and one at 8.4GHz. B07 denoted these sources with identifiers
beginning "RT" and the epoch in which they were observed.
Note (G2): Based on the fact that the hardness must lie in the range -1 to
1. Hardness ratios were computed as HR=((C2-C1)/(C1+C2)), where C1 is
the counts (after background subtraction) in the 0.3-2keV band and C2
is the counts in 2-10keV. In some cases the formal errors on HR
resulted in values outside the range -1<HR<1; here we truncate the
values at these limits and report upper or lower limits on HR as
appropriate. See section 2.2.
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Emmanuelle Perret [CDS] 05-Feb-2013