File Formats: Source Table: Sources Extension
This document describes the column content of the source extension (HDU1) within the FITS format version of the UV catalogue. This extension records the measured quantities of each source detected by the XMM pipeline software. There are 127 individual properties (columns) within the table. Numerous columns report the same quality but for different instrument filters. For the sake of efficiency, these descriptions are not repeated six times i.e. once per filter. When the word FILTER is encountered within column descriptions you will find six columns within the table in which the word "FILTER" can be replaced by one of "UVW2", "UVM2", "UVW1", "U", "B" or "V", the broadband filter sequence of the OM.
Each column name listed below may also be followed with terms in round () or square [] brackets. The first brackets contain the form of the column data in FITS nomenclature. J represents a 4-byte integer, E is a 4-byte floating point number, D is a 8-byte double precision number and nA is a character string where n is the maximum length allowed for the string. The square brackets contain the units of the data, where appropriate.
NULL values within the filter-dependent table columns have two meanings: 1) no observations were obtained through the filter during the observation, and 2) No source was detected at this position through a filter which was used during an observation. The cases may be distinguished using the values FILTER_SIGNIF column. These contain the detection significances for each source. If FILTER_SIGNIF = NULL then no exposures were made through the filter. If FILTER_SIGNIF = -999 then an image through the filter was obtained, but no source was detected at the position (Known Issues). Any other negative significance indicates that the source count rate exceeds 1000 count s-1 and is saturated.
Beyond column descriptors, there are no additional non-standard keywords within the extension header structure.
- IAUNAME (22A)
The IAU designated source name of format XMMOM JHHMMSS.S+DDMMSS.- N_SUMMARY (J)
This is a reference number. HDU2 of the source catalogue contains a summary of the XMM pointings included within the catalogue. Each row of HDU2 refers to a specific pointing and also contains a N_SUMMARY column. So N_SUMMARY is cross-referencing system allowing users to couple a source to its parent observation.- OBSID (10A)
Each XMM pointing has an exclusive 10-digit observation identification number or OBSID. This column contains the OBSID of the observation from which the source detection originates.- SRCID (J)
Source detection is performed in the XMM pipeline when the SAS task omdetect passes across each image obtained during a pointing. Source lists created from observations with the same instrument filter are combined and parameters of an individual source detected within different exposures are averaged. Within each combined source list created by the pipeline a source is given a unique reference number. The combination of OBSID and SRCID allows the user to trace the origins of each source in the catalogue. Note well that SRCID is not a unique number within the UV catalogue.- SRCNUM (J)
This is the unique reference number for the source, internal only to the UV catalogue.- FILTERSRCDIST (E) [arcsec]
This parameter provides the distance of a source from its nearest detected celestial neighbour. It allows the user to assess the probability of source confusion, contamination or misidentification.- RA (D) [degrees]
The measured J2000 Right Ascension of the source. The onboard pointing 'knowledge' of the XMM-Newton spacecraft is good to a few arcsecs. All fields included within the UV catalogue have had small order astrometric corrections performed upon them by correlating detected sources with the positions of USNO-B catalogue of optical objects. Note that the error provided in the POSERR column of the table does not include the systematic uncertainty remaining after the astrometric correction (See ??? for an analysis of the systematic pointing error). No proper motion corrections have been attempted.- DEC (D) [degrees]
The measured J2000 Declination of the source. The onboard pointing 'knowledge' of the XMM-Newton spacecraft is good to a few arcsecs. All fields included within the UV catalogue have had small order astrometric corrections performed upon them by correlating detected sources with the positions of USNO-B catalogue of optical objects. Note that the error provided in the POSERR column of the table does not include the systematic uncertainty remaining after the astrometric correction (See ??? for an analysis of the systematic pointing error). No proper motion corrections have been attempted.- RA_HMS (13A)
Hours-minutes-seconds rendition of the RA column value for the source.- DEC_HMS (14A)
Degrees-arcminutes-arcseconds rendition of the DEC column value for the source.- POSERR (E) [arcsec]
The statistical 1σ error attached to the measured source position. The RA and Dec errors are averaged with no weights to produce POSERR.- LII (D) [degrees]
The Galactic longitude of the source at epoch 2000.0.- BII (D) [degrees]
The Galactic latitude of the source at epoch 2000.0.- N_OBSID (J)
Over the mission lifetime, XMM-Newton has revisited many fields or observed sources with overlapping fields. N_OBSID records the number times a source has been detected during separate observations. Note that this is distinct from the number of times XMM has revisited the field. There may have been occasions where an observations was not deep enough to re-detect a specific source, or the source itself is variable. The criteria for matching a source over different epochs is that the positions are separated by <1.5" and <3σ. In cases where multiple sources within the same observation meet the matching criteria for another observation, the nearest source is adopted as the match.- N_FILTER_EXP (I)
During an individual observation, the OM may obtain multiple images through the same filter. These may occur through repeat exposures or by mosaicing the field with small sub-windows (this was a common practice during the early phases of the mission in order to avoid telemetry saturation). This parameter provides the number of separate images of a specific filter, taken over a single observation that contain the source. The criteria for matching a source over different exposures is that the positions are separated by <???" and <???σ. In cases where multiple sources within the same exposure meet the matching criteria for another image, the nearest source is adopted as the match.- FILTER_SIGNIF (E) [sigma]
This column contains the significance of a source detection above the data noise. Data noise contains contributions from the background and source. Detector noise is negligible. Significance is calculated on raw count rates, uncorrected for photon coincidence losses.
Due to the nature of the OM detector, sources with count rates >5 count/sec suffer from increasing quantities of photon coincidence loss where two photons arriving at the same location of the detector within one readout cycle cannot be distinguished. The effect is correctable using a count rate-dependent function up to corrected count rates of 400 count/sec. This column contains the background-subtracted source count rate corrected for coincidence loss using the standard calibration polynomial contained within the XMM CAL.
The 1σ uncertainty determined for the coincidence corrected source count rate recorded in FILTER_RATE.
The conversion from coincidence-corrected source count rate to AB flux occurs through a simple filter-dependent, multiplicative factor: FILTER_AB_FLUX = CF * FILTER_CORR_RATE. These columns contain AB fluxes calculated using conversion factors (CF) contained within the XMM CAL.
The 1σ uncertainty determined for the AB flux of a source recorded in FILTER_AB_FLUX.
The conversion from coincidence-corrected source count rate to AB magnitudes occurs through the standard logarithmic expression with filter-dependent, zero points: FILTER_AB_MAG = AZP - 2.5log(FILTER_CORR_RATE) These columns contain AB magnitudes calculated using zero points (AZP) contained within the XMM CAL.
The 1σ uncertainty determined for the AB magnitude of a source recorded in FILTER_AB_MAG.
The conversion from coincidence-corrected source count rate to Vega (Johnson) magnitudes occurs through the standard logarithmic expression with filter-dependent, zero points: FILTER_VEGA_MAG = VZP - 2.5log(FILTER_CORR_RATE) These columns contain Vega magnitudes calculated using zero points (VZP) contained within the XMM CAL.
The 1σ uncertainty determined for the Vega magnitude of a source recorded in FILTER_VEGA_MAG.
Many single observations (OBSIDs) yield multiple detections of a source through the same filter. Multiple detections provide us with a means to detect variable sources on timescales of hours. The reduced-χ2 statistic (χ2 per degree of freedom) measures the short-term variability of a source: FILTER_CHI = Σi ((CCRi - FILTER_CORR_RATE)2 / FILTER_CORR_RATE) / (N - 1), where CCRi represents coincidence corrected count rates from the sample of N independent source detections per OBSID per filter. The larger the value of the reduced-χ2 statistic, the more variable the source. This statistic is most suitable for selecting sources which vary consistently over the duration of the observation.
Many single observations (OBSIDs) yield multiple detections of a source through the same filter. Multiple detections provide us with a means to detect variable sources on timescales of hours. The reduced-χ2 statistic FILTER_CHI can, in certain instances, be relatively insensitive to brief flare events sampled by a single measurement. Another statistic more suited for picking out flare events is FILTER_MAXDEV, which is simply a measure of the most deviant point from the median of CCRi in units of sigma.
Using 1st order moment analysis, the size on the sky and ellipticity of sources is determined. This column records the full-width half-maximum (FWHM) of the source along what is considered to be the long axis of the elliptical source. Note that not all sources in the sample are necessarily well-described by an ellipse.
Using 1st order moment analysis, the size on the sky and ellipticity of sources is determined. This column records the full-width half-maximum (FWHM) of the source along what is considered to be the short axis of the elliptical source. Note that not all sources in the sample are necessarily well-described by an ellipse.
The angle on the sky subtended by the major axis of the source and J2000 north. The angle increases as the axis rotates from north to east.
Each source passes through an automatic screening process in order isolate potential problems with either the location or brightness recorded within the catalogue. The results of these tests are stored in a Boolean sense (i.e passed or failed) within the integer number stored in this column. FILTER_QUALITY_FLAG is simply the sum of all individual flags ΣFi, where i = 0,1,2,..,8. The meaning of each flag is summarized below, whereas a full description of the algorithms used to set the flags is presented in the Data Processing : Quality Flags section. The word 'source' below includes both the source and background extraction regions. Caveats.
| i | Fi | Description |
|---|---|---|
| 0 | 1 | Source lies over a cosmetic detector defect (bad pixel) |
| 1 | 2 | Source lies potentially over a readout streak originating from a bright field object |
| 2 | 4 | Source lies potentially over a ghost ring originating from a bright field object |
| 3 | 8 | Source lies potentially over a diffraction spike originating from a bright field object |
| 4 | 16 | Source lies potentially within a region background containing a modulo-8 fixed pattern |
| 5 | 32 | Source lies within the central image, where the background is potentially contaminated by scattered light |
| 6 | 64 | Potential source confusion with an object detected nearby |
| 7 | 128 | Source aperture lies across the outer edge of a raw detector image |
| 8 | 256 | A point source which lies within, or partially-within, an extended source |
| 9 | 512 | Noise spike |
An alternative representation of FILTER_QUALITY_FLAG containing 10 characters, one for each quality flag. The example of a source situated over a bad pixel, on a diffraction spike and at the edge of an image will have a quality flag of 137 and FILTER_QUALITY_FLAG_STR = "TFFTFFFTFF"
Sources which have spatial extents consistent with the point spread function (PSF) of the detector are flagged with FILTER_EXTENDED_FLAG = 0. Sources with measured FWHM major-axes, FILTER_MAJOR_AXIS, greater than the calibrated PSF FWHM with > 3σ confidence are flagged with FILTER_EXTENDED_FLAG = 1. Calibrated PSF values for each filter are obtained from the XMM CAL???.