X.B.1 The Machine Readable Version of the Point Source Catalog

IRAS Explanatory Supplement
X. The Formats of the IRAS Catalogs and Atlases
B. Point Sources
B.1 The Machine Readable Version of the Point Source Catalog

Chapter Contents | Authors | References
Table of Contents | Index | Previous Section | Next Section

The point source catalog tape is divided into six individual files, covering the range 0 to 24 hr in right ascension in blocks of four hours each. Each file contains from 20,000 to 90,000 sources arranged in order of increasing right ascension. Each of the six catalog files is preceded by a file containing a single 80-character ASCII record which lists the date and version number of the catalog. Thus to read the entire catalog one must read sequentially through 12 files, six containing the data and six containing dates and version numbers.

Table X.B.1 describes each entry in the catalog tape. Those columns that are also included in the printed version are marked. Each catalog entry requires 160 + NID × 40 bytes of ASCII data where NID is the number of positional associations for each source. In the table the column "Format" refers to the length and type of the (FORTRAN) character field used to read or write each entry. Figure X.B.1 describes the format of the printed version of the catalog.

The tape is written with 80-character (ASCII) logical records and blocked with 256 logical records per physical record so that one can regard the tape as a sequence of card images. The entries are arranged so that the source data fits into two records. Association information requires an additional 40 characters per association and appears in subsequent records, two associations per record. If one assumes an average of two associations per source, then a file for one of the right ascension blocks will require about 10 Mbytes.

Format of Point Source Catalog Tape
Table X.B.1
Start Byte Name Description Units Format
00 NAME1 Source Name --- 11A1
11 HOURS Right Ascension 1950. Hours I2
13 MINUTE Right Ascension 1950. Minutes I2
15 SECOND1 Right Ascension 1950. deci-Seconds I3
18 DSIGN1 Declination Sign ± A1
19 DECDEG Declination 1950. Arc Deg I2
21 DECMIN Declination 1950. Arc Min I2
23 DECSEC1 Declination 1950. Arc Sec I2
25 MAJOR1 Uncertainty ellipse major axis Arc Sec I3
28 MINOR1 Uncertainty ellipse minor axis Arc Sec I3
31 POSANG1 Uncertainty ellipse position angle Degree (East of North) I3
34 NHCON1 (<25) Number of times observed --- I2
36 FLUX1 Averaged non-color corrected flux densities (1 value per band) Jansky (10-26W m-2 Hz-1) 4E9.3
72 FQUAL1 flux density quality. (1 value per band) --- 4I1
76 NLRS Number of significant LRS spectra --- I2
78 LRSCHAR1 Characterization of averaged LRS spectrum. --- 2A1
(---------------new record--------------)
80 RELUNC1 percent relative flux density uncertainties (1 value per band). --- 4I3
92 TSNR ten times the minimum signal-to-noise ratio in each band --- 4I5
112 CC1 point source correlation coefficient (1 value per band). --- 4A1
116 VAR1 percent Likelihood of Variability   I2
118 DISC Discrepant Fluxes flag (one per band, hex-encoded) --- A1
119 CONFUSE1 Confusion flag (1 flag per band, hex-encoded) --- A1
120 PNEARH1 Number of nearby hours- confirmed point sources --- I1
121 PNEARW1 Number of nearby weeks- confirmed point sources --- I1
122 SES11 Number of seconds-confirmed nearby small extended sources. (1 value per band) --- 4I1
126 SES21 Number of nearby weeks-confirmed small extended sources. (1 value per band) --- 4I1
130 HSDFLAG1 Source is located in high source density bin. (1 value per band, hex-encoded) --- A1
131 CIRR11 Number of nearby 100 µm only WSDB sources --- I1
132 CIRR21 Spatially filtered 100 µm sky brightness ratio to flux density of point source (see text) --- I1
133 CIRR3 Total 100 µm sky surface brightness MJy sr-1 I3
136 NID1 (<25) Number of positional associations. --- I2
138 IDTYPE Type of Object --- I1
139 SPARE 21 spare bytes --- 21A1
(---------------new record--------------)
160 CATN01 Catalog number --- I2
162 SOURCE1 Source ID --- 15A1
177 TYPE1 Source Type/Spectral Class --- 5A1
182 RADIUS1 Radius Vector from IRAS Source to Association Arc Sec I3
185 POS Position Angle from IRAS Source to Association Degree E of N. I3
188 FIELD12 object field #1 (magnitude/other) catalog dependent I4
192 FIELD22 object field #2 (magnitude/other) catalog dependent I4
196 FIELD3 object field #3 (size/other) catalog dependent I4
200   continuation of associations3    
-240   in blocks of 40 bytes    

1 Quantities listed in printed version of catalog.

2 FIELD1 is listed in printed version of catalog, except for catalogs 2 and 19, where FIELD2 is listed.

3 CATNO, SOURCE, TYPE, RADIUS, POS, FIELD 1-3 are repeated in blocks of 40 bytes, 2 per logical record, as necessary. The definition and formats of FIELD1-3 depend on the individual catalog in which the association is found. See Table X.B.4.

Figure X.B.1 Explanation of format of printed version of point source catalog.
larger largest

In general, for quantities that have a value in each wavelength band, subscripts or array indices range from 1 to 4 and refer, respectively, to 12, 25, 60 and 100 µm. A number of the flags discussed below have values in each of the four wavelength bands. For compactness these are encoded into a single base-16 (Hex) digit (values 0-F) in the following manner (Table X.B.2). The four bits of that hex digit correspond to the four wavelength bands with bit 0 (Least Significant Bit) for 12 µm, bit 1 for 25 µm, bit 2 for 60 µm and bit 3 for 100 µm. The presence of a flag in a band is denoted by setting its bit to 1. Thus a source with a particular flag, e.g. CONFUSE, set at 12 and 25 µm would have CONFUSE = 0011 = 3(Hex) while one confused in 25,60 and 100  µm would have CONFUSE = 1110 = E(Hex). A flag encoded in this manner will be referred to as "hex-encoded by band".

The remainder of this section discusses individual entries in the catalog. At the end of the section a few remarks relevant only to the printed version of the catalog are given.

Meaning of Hex Encoded Flags
Table X.B.2
Flag Set in A Particular Band Resultant Value
100 µm (Bit 3) 60 µm (Bit 2) 25 µm (Bit 1) 12 µm (Bit 0) of Encoded Flag xxxx = HEX=Decimal
0 0 0 0 0000=0=0
0 0 0 1 0001=1=1
0 0 1 0 0010=2=2
0 0 1 1 0011=3=3
0 1 0 0 0100=4=4
0 1 0 1 0101=5=5
0 1 1 0 0110=6=6
0 1 1 1 0111=7=7
1 0 0 0 1000=8=8
1 0 0 1 1001=9=9
1 0 1 0 1010=A=10
1 0 1 1 1011=B=11
1 1 0 0 1100=C=12
1 1 0 1 1101=D=13
1 1 1 0 1110=E=14
1 1 1 1 1111=F=15

Source Name: NAME

The IRAS source name is derived from its position by combining the hours, minutes and tenths of minutes of right ascension and the sign,° and minutes of the declination. In obtaining the minutes of right ascension and declination for the name, the positions were truncated. The letters 'A','B','C', etc. are appended to names of sources so close together that they would otherwise have had identical names. Names were uniquely assigned to both catalog and reject file sources, with catalog sources receiving letters first.


Positions are given for the equinox 1950.0, and epoch 1983.5. Hours (HOURS) and minutes (MINUTE) of right ascension are given as integers while seconds (SECOND) are rounded to integer deciseconds. The declination is given as a character sign (DSIGN) followed by integer values of° (DECDEG), minutes (DECMIN) and seconds (DECSEC). Values of 60 seconds are given when the truncated minutes (given here) disagree with the minutes in the rounded position.

Position Uncertainty: MAJOR, MINOR, POSANG

As discussed in Section VII.B.2, the uncertainty in the position for a source depends on its brightness in the various wavelength bands, its path across the focal plane and the number of sightings. The final uncertainty after position refinement is expressed as a 95% confidence uncertainty ellipse (see Section V.D.9) whose semi-major (MAJOR) and semi-minor (MINOR) axes are given in seconds of arc. The orientation (POSANG) of the ellipse on the sky is expressed in terms of the angle between the major axis of the ellipse and the local equatorial meridian. It is expressed in° east of north

Number of Sightings: NHCON

The number of hour-confirmed sightings is given. This number of flux entries will be found in the WSDB.

Flux Density: FLUX(4)

Each of the four wavelengths has a non-color-corrected flux density in units of Janskys, (1 Jy = 10-26 W m-2 Hz-1). The quoted value is an average of all the hours-confirmed sightings as obtained by the prescription described in Section V.H.5. The quality of each flux density is designated by FQUAL (see below).

The flux densities have been calculated assuming an intrinsic source energy distribution such that the flux density fv is proportional to v-1. Corrections to other spectral shapes can be made by consulting Section VI.C.

The flux densities for sources so bright that they saturated the analog-to-digital converter on every sighting are lower limits based on the brightest value recorded. The uncertainties are given as ten times the quoted flux density and a flag is set indicating saturation has occurred (Table X.B.7a).

Signal-to-Noise Ratio: TSNR(4)

The signal-to-noise ratio given for an individual hours-confirmed sighting is the highest value of the detections comprising that sighting (Section V.C.2). The values quoted in the catalog are ten times the minimum of the signal-to-noise ratios for the various sightings (HCONs) of the source. A value is given for each wavelength band with a high or moderate quality measurement and for those upper limits coming from a non-seconds-confirmed detection. Values of TSNR greater than 30,000 are given as 30,000.

Source Variability: VAR

VAR is the percent probability (0-99) that a source is variable based on an analysis of the 12 and 25 µm flux densities and their uncertainties (see Section V.H.5). The value "-1 " indicates that the source was not examined for variability.

Discrepant Fluxes: DISC(4)

The DISC flag indicates whether any one of the fluxes in a given band disagrees with others in that band, hex-encoded by band (Section V.H.5).

Flux Density Quality: FQUAL(4)

As described in Section V.H.5, a flux density measurement can be either high quality (FQUAL=3), moderate quality (FQUAL=2) or an upper limit (FQUAL= 1).

Low-Resolution Spectra: NLRS,LRSCHAR

The Low-Resolution Spectrometer obtained 8-25 µm spectra of bright 12 and 25 µm sources (Chapter IX). NLRS gives the number of statistically meaningful spectra available for the source. LRSCHAR gives a short characterization of the nature of the spectrum (Table IX.D.1). All of the LRS spectra are available in tape and printed forms (Astronomy and Astrophysics Supplement, 1985) as described below (Section X.E) and in Chapter IX.

Flux Density Uncertainties: RELUNC (4)

Each flux density measurement other than an upper limit has an associated uncertainty expressed as a 1 value in units of l00 × fv/fv. Uncertainties are discussed in Sections V.H.5 and VII.D.2.

Point Source Correlation: CC(4)

As described in Section V.C.4, the point source correlation coefficient can have values between 87-100%. These are encoded as alphabetic characters with A=100, B=99...N=87, one value per band. The value quoted is for the highest correlation coefficient seen for that source on any sighting.


As described in Section V.D.2, a great deal of care went into trying to untangle instances of confusion between neighboring sources. In parts of the sky where the source density is low, confusion processing was sometimes able to separate sources that are quite close together. The CONFUSE flag is set if two or more sightings of the source in a given band had confusion status bits set indicating confusion in the seconds-confirmation or band-merging processes. This flag is hex-encoded by band.

Other indicators of possible confusion are given by PNEARN and PNEARW which are, respectively, the numbers of hours-confirmed and weeks-confirmed point sources located within a 4.5 cross-scan and 6' in-scan (half-widths) window centered on the source. Values larger than 9 are given as 9.

Regions of high source density received special processing to improve the reliability of the quoted sources (see Section V.H.6). The regions are band-dependent. If a particular band of a given source went through high source density processing, then the appropriate bit in HSDFLAG (Table X.B.7) is set. HSDFLAG is hex encoded by band.

Small Extended Sources: SES1 (4), SES2(4)

SES1 is the number of seconds-confirmed, small extended source detections in a given band found within a window centered on the source. The size of the window is 6' in-scan × 4.5' cross-scan (half-widths). As described in Sections V.H.3-4 and VII.H.1, values of SES1 greater than 1 should caution the reader that significant extended structure may exist in the region and that the source in question may be a point-source like piece of a complex field.

SES2 is the number of weeks-confirmed small extended sources in a given band, located within a 6' in-scan × 4.5' cross-scan window (half-width) centered on the source. SES2 greater than 0 means that the point source flux measurement should be treated with caution as the source in question may, in fact, be extended. The flux quoted in the catalog of small extended sources may provide a better value for the source.

Cirrus Indicators: CIRR1, CIRR2, CIRR3

Over a large range of Galactic latitudes the infrared sky at 100 µm is characterized by emission from interstellar dust on a wide range of angular scales. The so-called "infrared cirrus" can seriously hamper efforts to extract point source detections from the data. To aid the user in interpreting the quoted 100 µm measurements three quantities have been established (Section V.H.4 and VII.H).

CIRR1 gives the number of 100 µm-only WSDB sources located within a ± ½° box in ecliptic coordinates centered on the source. The sources included in this count are the weeks-confirmed sources prior to high source density region processing (if applicable) plus those sources hours-confirmed but not weeks-confirmed. Values of CIRR1 greater than 3 may indicate contamination by cirrus with structure on the point source size scale.

CIRR2 gives a cirrus indication on a larger scale than CIRR1 and compares a "cirrus flux" with the source flux at 100 µm (see (Eq. V.H.2)). Values larger than 4~5 indicate the presence of considerable structure in the 100 µm emission on a ½° scale. A value of 0 indicates that no ½° data were available for the source in question.

CIRR3 is the total surface brightness of the sky surrounding the source in a ½° beam at 100 µm, clipped to exclude values greater than 254 MJy sr-1. Values of CIRR3 greater than 30 MJy sr-1 indicate emission from dust with an appreciable column density. A value of CIRR3 = -1 means that no data were available.

Positional Associations: NID, IDTYPE, CATNO, SOURCE, TYPE, RADIUS, POS, FIELD1-3

Much of the utility of the IRAS catalog comes from the association of infrared objects with sources known to exist from other astronomical catalogs. As described in Section V.H.9, a large number of catalogs have been searched for positional matches. The total number of matches found is given by NID. Each match results in a forty-character description (2 per record).

IDTYPE ranges from 1 to 4 and states whether an association was found in an extragalactic catalog (1), a stellar catalog (2), other catalogs (3), or matches in multiple types of catalogs (4). CATNO is the number of the catalog in which the match was found (Tables V.H.1, X.B.4).

SOURCE is the name of the object in that catalog and TYPE its character or spectral type, if available. A vector is drawn from the IRAS position to the associated object. RADIUS is the length of that vector in arc seconds. POS is the angle between the vector and the local equatorial meridian expressed in° east of north. Three fields (FIELD1-3) have values depending on the catalog in question (Table X.B.4). Typically FlELD1,2 are magnitudes (in decimag) and FIELD3 a size.

Chapter Contents | Authors | References
Table of Contents | Index | Previous Section | Next Section