J/MNRAS/446/911 X-ray sources in the AKARI NEP deep field (Krumpe+, 2015)
Chandra survey in the AKARI North Ecliptic Pole Deep Field.
I. X-ray data, point-like source catalogue, sensitivity maps, and number counts.
Krumpe M., Miyaji T., Brunner H., Hanami H., Ishigaki T., Takagi T.,
Markowitz A.G., Goto T., Malkan M.A., Matsuhara H., Pearson C., Ueda Y.,
Wada T.
<Mon. Not. R. Astron. Soc., 446, 911-931 (2015)>
=2015MNRAS.446..911K 2015MNRAS.446..911K
ADC_Keywords: Active gal. nuclei ; Galactic pole, north ; Galaxies, Seyfert ;
X-ray sources
Keywords: methods: data analysis; surveys; galaxies: active; X-rays: galaxies
Abstract:
We present data products from the 300ks Chandra survey in the AKARI
North Ecliptic Pole (NEP) deep field. This field has a unique set of
9-band infrared photometry covering 2-24 micron from the AKARI
Infrared Camera, including mid-infrared (MIR) bands not covered by
Spitzer. The survey is one of the deepest ever achieved at ∼15 micron,
and is by far the widest among those with similar depths in the MIR.
This makes this field unique for the MIR-selection of AGN at z∼1.
We design a source detection procedure, which performs joint Maximum
Likelihood PSF fits on all of our 15 mosaicked Chandra pointings
covering an area of 0.34 square degree.
The procedure has been highly optimized and tested by simulations.
We provide a point source catalog with photometry and Bayesian-based
90 per cent confidence upper limits in the 0.5-7, 0.5-2, 2-7, 2-4, and
4-7keV bands. The catalog contains 457 X-ray sources and the spurious
fraction is estimated to be ∼1.7 per cent. Sensitivity and 90 per cent
confidence upper flux limits maps in all bands are provided as well.
We search for optical MIR counterparts in the central 0.25 square
degree, where deep Subaru Suprime-Cam multiband images exist. Among
the 377 X-ray sources detected there, ∼80 per cent have optical
counterparts and ∼60 per cent also have AKARI mid-IR counterparts. We
cross-match our X-ray sources with MIR-selected AGN from Hanami et al.
Around 30 per cent of all AGN that have MID-IR SEDs purely explainable
by AGN activity are strong Compton-thick AGN candidates.
Description:
The fits images labelled SeMap* are the sensitivity maps in which we
give the minimum flux that would have caused a detection at each
position. This flux depends on the maximum likelihood threshold chosen
in the source detection run, the point spread function, and the
background level at the chosen position.
We create sensitivity maps in different energy bands (0.5-2, 0.5-7,
2-4, 2-7, and 4-7keV) by searching for the flux to reject the
null-hypothesis that the flux at a given position is only caused by a
background fluctuation. In a chosen energy band, we determine for each
position in the survey the flux required to obtain a certain Poisson
probability above the background counts. Since ML=-ln(P), we know from
our ML=12 threshold the probability we are aiming for. In practice, we
search for a value of -ln P_total that falls within Delta ML=±0.2 of
our targeted ML threshold. This tolerance range corresponds to having
one spurious source more or less in the whole survey. Note, that
outside the deep Subaru/Suprime-Cam imaging the sensitivity maps
should be used with caution since we assume for their generation a
ML=12 over the whole area covered by Chandra.
More details on the procedure of producing the sensitivity maps,
including the PSF-summed background map and PSF-weighted averaged
exposure maps are given in the paper, section 5.3.
The fits images labelled u90* are the upper limit maps, where the
upper 90 per cent confidence flux limit is given at each position.
We take a Bayesian approach following Kraft, Burrows & Nousek,
1991ApJ...374..344K 1991ApJ...374..344K. Consequently, we obtain the upper 90∼per cent
confidence flux limit by searching for the flux such that given the
observed counts the Bayesian probability of having this flux or larger
is 10∼per cent.
More details on the procedure of producing the upper 90 per cent flux
limit maps are given in the paper, section 5.4.
Main source catalog:
The source catalog uses an internal threshold of ML=9.5 which
corresponds to MLempir∼12 (see Sect. 4.3.3 of the paper for
details). In total, 457 sources are detected, of which 377 objects
fall in the deep Subaru imaging region. This catalog is designed to
identify X-ray emitting objects in the Chandra/AKARI NEP deep field.
Together with the optimized cross- identification procedure, the clear
advantage of the catalog is the very high reliability, while the
catalog sacrifices completeness for objects with low counts (see
Figure 9 in the paper). Only ∼1.7 per cent of the objects listed in
the source catalog are expected to be spurious source detections.
Considering the uncertainty in the astrometric calibration, all
sources should be considered as possible X-ray counterparts that are
within a radius of rmatch=sqrt(σtotal2+σastro2)
with σtotal=5*sqrt(σsys2+σstat2) and
σsys=0.1arcsec and σastro=0.2arcsec (astrometric
uncertainty).
Low probability source catalog:
WARNING: Due to the significant number of spurious sources in the
catalog, it should not be used select X-ray sources or to increase
the sample size of X-ray-selected objects. This catalog can be of
interest if the scientific goal requires one to exclude potential
X-ray emitting objects from a sample with a high completeness. Using
this strategy, one accepts those objects that are excluded are not
associated with an X-ray-emitting object.
The source catalog has a lower maximum likelihood threshold (internal
threshold of ML=5; corresponding to MLempir∼9.5). The catalog
contains the same columns as main source catalog. This catalog
contains 626 detected sources, of which 506 are located within the
deep Subaru imaging region. Based on our simulated data, we conclude
that 19 per cent of all catalog entries are false detections.
Considering only the deep Subaru imaging area the spurious source
fraction drops to 15 per cent. We significantly increase the
completeness to 61 per cent for objects with at least four counts, ∼85
per cent for objects with seven counts or more, and >95 per cent for
sources with at least 12 counts.
Considering the uncertainty in the astrometric calibration, all
sources should be considered that are within a radius of
rmatch=sqrt(σtotal2+σastro2) with
σtotal=5*sqrt(σsys2+σstat2) and
σsys=0.1arcsec and σastro=0.2arcsec (astrometric
uncertainty).
When using information of this catalog please cite the paper (Krumpe
et al., 2015, MNRAS, 446, 911).
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
mainscat.dat 554 457 Main source catalog
mainscat.fits 2880 39 Main source catalog in FITS format
lowpscat.dat 554 626 Low probability source catalog
lowpscat.fits 2880 51 Low probability source catalog in FITS format
list.dat 108 10 List of FITS images
img/* . 10 Fits images
--------------------------------------------------------------------------------
See also:
B/chandra : The Chandra Archive Log (CXC, 1999-2014)
J/PASJ/59/S529 : NEP deep survey at 11um (Lee+, 2007)
J/A+A/517/A54 : 20cm survey of the AKARI NEP (White+, 2010)
J/ApJS/190/166 : Optical catalog of AKARI NEP-wide survey (Jeon+, 2010)
J/A+A/537/A24 : AKARI NEP-Deep field mid-IR source catalogue (Takagi+, 2012)
J/A+A/548/A29 : IR source catalog of AKARI NEP-Wide field (Kim+, 2012)
J/A+A/559/A132 : AKARI NEP Deep Survey revised catalog (Murata+, 2013)
J/ApJS/207/37 : IR sources spectroscopy in the AKARI NEP (Shim+, 2013)
J/A+A/566/A60 : Optical-NIR catalog of AKARI NEP Deep Field (Oi+, 2014)
J/ApJS/214/20 : AKARI NEP field J- and H- band source catalog (Jeon+, 2014)
Byte-by-byte Description of file: lowpscat.dat mainscat.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 3 I3 --- CID [1/626] Chandra source identification number
within the catalog (CID)
5- 17 A13 --- Name Name of object (4)
19- 27 F9.5 deg RAdeg Right ascension (J2000.0) (RA)
29- 37 F9.5 deg DEdeg Declination (J2000.0) (DEC)
39- 46 F8.5 arcsec ePos [0.01/2.3] 1-σ statistical error of
source position (RADEC_ERR) (1)
48- 59 E12.6 ct C0.5-7 [1.7/7302] 0.5-7keV source counts (CTS057)
61- 68 F8.5 ct e_C0.5-7 1-σ uncertainty of C0.5-7 (CTSERR05_7)
70- 81 E12.6 ct/s R0.5-7 0.5-7keV vignetting-corrected source
count rate (RATE057)
83- 94 E12.6 ct/s e_R0.5-7 1-σ uncertainty of R05.-7 (RATEERR05_7)
96-107 E12.6 mW/m2 F0.5-7 ?=0 0.5-7keV source flux corrected for Galactic
absorption); if null, upper limit in e_F0.5-7
(FLUX057)
109-120 E12.6 mW/m2 e_F0.5-7 [] 1-σ uncertainty of F0.5-7 or upper 90%
limit if negative (FLUXERR05_7) (2)
122-133 E12.6 --- L0.5-7 0.5-7keV internal maximum likelihood of the
source detection (ML057)
135-146 E12.6 ct C0.5-2 [0/5490] 0.5-2keV source counts (CTS052)
148-156 F9.6 ct e_C0.5-2 1-σ uncertainty of C0.5-2 (CTSERR05_2)
158-169 E12.6 ct/s R0.5-2 0.5-2keV vignetting-corrected source count rate
(RATE052)
171-182 E12.6 ct/s e_R0.5-2 1-σ uncertainty of R0.5-2 (RATEERR05_2)
184-195 E12.6 mW/m2 F0.5-2 ?=0 0.5-2keV source flux corrected for Galactic
absorption; if null, upper limit in e_F0.5-2
(FLUX052)
197-208 E12.6 mW/m2 e_F0.5-2 [] 1-σ uncertainty of source flux or 90%
upper limit if negative (FLUXERR05_2) (2)
210-221 E12.6 --- L0.5-2 0.5-2keV internal maximum likelihood of the
source detection (ML052)
223-234 E12.6 ct C2-7 2-7keV source counts (CTS27)
236-244 F9.6 ct e_C2-7 1-σ uncertainty of source counts
(CTSERR2_7)
246-257 E12.6 ct/s R2-7 2-7keV vignetting-corrected source count rate
(RATE27)
259-270 E12.6 ct/s e_R2-7 1-σ uncertainty of R2-7 (RATEERR2_7)
272-283 E12.6 mW/m2 F2-7 2-7keV source flux (corrected for Galactic
absorption; if null, upper limit in e_F2-7
(FLUX27)
285-296 E12.6 mW/m2 e_F2-7 [] 1-σ uncertainty of F2-7 or upper 90%
limit if negative (FLUXERR2_7) (2)
298-307 F10.6 --- L2-7 2-7keV internal maximum likelihood of the
source detection (ML27)
309-320 E12.6 ct C2-4 2-4keV source counts (CTS24)
322-330 F9.6 ct e_C2-4 1-σ uncertainty of C2-4 (CTSERR2_4)
332-343 E12.6 ct/s R2-4 2-4keV vignetting-corrected source count rate
(RATE24)
345-356 E12.6 ct/s e_R2-4 1-σ uncertainty of R2-4 (RATEERR2_4)
358-369 E12.6 mW/m2 F2-4 ?=0 2-4keV source flux corrected for Galactic
absorption; if null, upper limit in e_F2-4
(FLUX24)
371-382 E12.6 mW/m2 e_F2-4 [] 1-σ uncertainty of F2-4 or upper 90%
limit if negative (FLUXERR2_4) (2)
384-395 E12.6 --- L2-4 2-4keV internal maximum likelihood of the
source detection (ML24)
397-406 F10.6 ct C4-7 4-7keV source counts (CTS47)
408-416 F9.6 ct e_C4-7 1-σ uncertainty of C4-7 (CTSERR4_7)
418-429 E12.6 ct/s R4-7 4-7keV vignetting-corrected source count rate
(RATE47)
431-442 E12.6 ct/s e_R4-7 1-σ uncertainty of R4-7 (RATEERR4_7)
444-455 E12.6 mW/m2 F4-7 ?=0 4-7keV source flux corrected for Galactic
absorption); if null, upper limit in e_F4-7
(FLUX47)
457-468 E12.6 mW/m2 e_F4-7 [] 1-σ uncertainty of F4-7 or upper 90%
limit if negative (FLUXERR4_7) (2)
470-481 E12.6 --- L4-7 4-7keV internal maximum likelihood of the
source detection (ML47)
483-491 F9.6 --- HRsh [-1/1]?=9999 Hardness ratio in the 0.5-2keV and
16- 21 energy bands (HRSOFTHARD)
493-500 F8.6 --- e_HRsh ?=9999 1-σ uncertainty fo HRsh
(HRERRSOFT_HARD)
502-513 E12.6 --- HRsm [-1/1]?=9999 Hardness ratio in the 0.5-2keV and
16- 18 energy bands (HRSOFTMED)
515-526 E12.6 --- e_HRsm ?=9999 1-σ uncertainty of HRsm
(HRERRSOFT_MED)
528-539 E12.6 --- HRmu [-1/1]?=9999 Hardness ratio in the 2-4keV and
18- 21 energy bands (HRMEDULT)
541-552 E12.6 --- e_HRmu ?=9999 1-σ uncertainty of HRmu
(HRERRMED_ULT)
554 I1 --- In [0/1] Note (IN_AREA) (3)
--------------------------------------------------------------------------------
Note (1): The values in this column do not include the additional systematic
uncertainty to optimize the source detection (see Sect. 4.3.1 of the paper),
nor do they contain the uncertainty of the astrometric calibration of the
Chandra data to the Subaru/Suprime-Cam reference frame in the AKARI NEP deep
field (Sect. 5.1 of the paper). To determine the counterparts of our X-ray
sources in other wavelength regimes, we recommend using the equation given
in Sect. 5.2 of the paper.
Note (2): If a source is not detected in an energy band (internal ML-threshold
in this energy band below 9.5), we list the upper 90 per cent limit (see
Sect. 5.4 in the paper) as the flux error.
We make such cases visible by giving it a negative value in the error flux
column and and setting the flux to zero. The two sources that have a
ML-threshold in the 0.5-7keV band below 9.5 originate from a 0.5-7keV single
band source detection run. To quote similar ML values for all objects we list
the total 0.5-7keV ML values from the joint 3 energy band source detection
run. The listed counts, count rates, fluxes, and the corresponding
uncertainties in the 0.5-7keV band are taken from the single band detection
run.
Note (3): If the source falls within the deep Subaru/Suprime-Cam imaging
region, this value = 1, otherwise 0.
Note (4): AKARI North Ecliptic Pole Deep Field Chandra X-ray Observatory
(ANEPD-CXO) followed by the Chandra source identification number (CID),
ANEPD-CXONNN.
In case of the low probability source catalog the name is ANEPD-CXOs
plus Chandra source identification number (CID) of the low prob.
source catalog. The additional 's' stands for supplement, ANEPD-CXOsNNN.
--------------------------------------------------------------------------------
Byte-by-byte Description of file: list.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 9 F9.5 deg RAdeg Right ascension of center (J2000)
10- 18 F9.5 deg DEdeg Declination of center (J2000)
20- 24 F5.3 arcsec/pix Scale Scale of the image
26- 29 I4 --- Nx [4361] Number of pixels along X-axis
31- 34 I4 --- Ny [5121] Number of pixels along Y-axis
36- 54 A19 "datime" Obs.Date Observation date (YYYY-MM-DDThh:mm:ss)
56- 60 I5 --- Kibyte size [87255] Size of FITS file
62- 73 A12 --- FileName Name of FITS file in subdirectory img
75-108 A34 --- Title Title of the fits image
--------------------------------------------------------------------------------
Acknowledgements:
Mirko Krumpe, mkrumpe(at)ucsd.edu
History:
* 01-Apr-2015: First on-line version
* 21-Jun-2016: names added in lowpscat.dat and mainscat.dat, from author
(End) Mirko Krumpe [ESO, UCSD], Patricia Vannier [CDS] 21-Jun-2015