The 1st Revision to WGACAT

Nick White, NASA/Goddard Space Flight Center-HEASARC
Giommi , ESA/ESRIN-ESIS
Lorella Angelini , NASA/Goddard Space Flight Center-HEASARC

Last modified on April 28, 1995.

WGACAT Rev 1

The 1st revision, rev 1, to the WGACAT differs from the original release as follows:

An offset in the positions given in the original, rev 0, catalog has been corrected.

It includes about 800 more fields to give a total of 3007 unique RORs (compared to 2229 in the current version of ROSATSRC). Rev 1 has a total of ~68,000 detections and ~62000 unique sources within 20 arc seconds radius of each other (compared to 49721 entries and 44542 unique sources in rev 0).

A lightcurve and spectrum has been accumulated for each source. These are provided as products available through the HEASARC browse interface for download or display. The lightcurve and spectrum are given in gif format for instant display, and also as an event file and a pha spectrum in FITS format for further analysis. The background is provided accumulated over the whole field with the detected sources removed. These are intended as quick look products, and should be used with care. Please note: these products take some time to generate. This is on going and they will not be complete until the end of May 1995.

The removal of the overlap region between the inner and outer regions, by removing sources in the outer region at a radius less then 21 arc min. In this region edge effects combined with the presence of the inner rib caused a spurious peak in the number of detected sources.

The source designations have been revised to be prefixed by 1WGA to distinguish them from sources designated in rev 0.

The lower channel used for the soft band has been changed from zero to 12. Two additional ratios (HR SOFT and HR HARD) are provided in the range covered by the original hardness ratio, to give more sensitivity to spectral changes.

Errors in the calculation of the AOX and AOR parameters have been corrected.

Additional catalogs have been added to the cross-correlation and the order of these correlations was changed to give the best results.

A cross correlation against ROSATSRC has been made and the results written to the WGACAT database.

Radial Distribution of Source Detections
The rev1 source detection distribution as a function of offaxis angle for WGACAT is given in Figure 1.

Figure 1: The radial distribution of sources detected in WGACAT. The gap at an offset of about 20 arc min is caused by the inner rib, and the removal of this area of the detector from the catalog.
The large peak at about 2 arcminutes includes most of the targets. Almost all other sources are serendipitous. For a detector with constant sensitivity the number of sources detected at a given offset increases linearly. This trend is seen in figure 1 only up to about 10 arcminutes where the detector response is uniform. Between 10 and 25 arcminutes the number of sources is lower than expected because the window support structure reduces the exposure of this area as a result of the wobble of the satellite. At higher offset values the number of sources sharply decreases because of the strongly reduced sensitivity in the outer parts of the detector.
Further Comparison with ROSATSRC

We have made a further comparison of the Rev1 WGACAT with ROSATSRC with the following key points:

The rev 1 WGACAT was cross-correlated against the MPE generated ROSATSRC catalog to see how many WGACAT sources are seen in ROSATSRC. Radii of 15 and 60 arc sec were used for the inner and outer regions respectively. Only those fields that appear in both catalogs were used. The results are given in Table 1.

Table 1: WGACAT sources that appear in ROSATSRC | Inner | Outer ------------------------------------------------------------------------- | 78% (15568/19898) | 35% (12846/36686)
In the inner region a large fraction of the WGACAT sources also appear in ROSATSRC. In the outer region the situation is quite different. Only 29% of the WGACAT sources appear in ROSATSRC. An explanation for this large difference is that the threshold used in ROSATSRC is higher than in WGACAT in the outer region. If we make the reverse correlation, i.e. how many ROSATSRC sources appear in WGACAT we find the opposite situation (Table 2).

Table 2: ROSATSRC sources that appear in WGACAT | Inner | Outer ------------------------------------------------------------------------- | 46% (16160/35023) | 71% (10935/15391)
In the inner region only half of the ROSATSRC detections appear in WGACAT. This is because in this part of the detector the detection threshold in ROSATSRC is set below that used in WGACAT. In the outer region 71% of the ROSATSRC detections appear in WGACAT. It is worth emphasising that WGACAT uses a constant detection threshold across the entire field, whereas in ROSATSRC the threshold is raised for the outer region.

The radial distribution of sources detected in ROSATSRC is given in Figure 2 and can be compared with the WGACAT distribution given in Figure 1.

Figure 2: The radial distribution of sources detected in ROSATSRC.
Note that in the inner region many more sources are detected in ROSATSRC, whereas in the outer region far fewer are detected. This is caused by the MPE processing using a lower threshold in the inner region and a higher threshold in the outer region. The offaxis angle defined in the ROSATSRC catalog is from the target position. In WGACAT it is defined as from the center of the detector.

Positional Accuracy
To estimate the positional accuracy of the WGA sources we have made an accurate comparision of the WGA coordinates with those of the Hipparcos Input Catalog. The results are shown in figures 3 and 4.

Figure 3: The offset in declination from the positions given in the Hipparcos input catalog for the sources in WGACAT.

Figure 4: The offset in RA from the positions given in the Hipparcos input catalog for the sources in WGACAT.
The large majority of the stars in the Hipparcos catalog that have a counterpart in WGACAT are within 10 arcseconds in alpha and delta in the inner 20 arcminutes of the detector. The error significantly increases at larger off-axis angles.
Limitations of Rev 1
The known limitations of WGACAT are as follows:

Count rates of sources located near the window support structure may be underestimated. This is because the processing assumed a constant exposure time over the entire image, neglecting the obscuration effects due to the support structure moving in sky coordinates while the satellite wobbles.

The catalog cannot be used for statistical purposes unless only the central 10 arcminutes are considered.

A number of spurious sources due to low brightness extended emission may be present.

Future Revisions
Apart from regular updates when new data is added to the archive, the following more detailed 2nd revision is planned before the end of this year (1995):

Make use of a background map. This will better account for gradients in the background across the detector, and result in more accurate count rates and a more sensitive detection threshold
Make use of a vignetting map. This will give better estimates of the source counting rates close to the detector ribs.

Acknowledgements
We thank: Steve Fantasia for doing the quality checking; Pat Tyler for help with the database creation; Mike Corcoran for maintaining the ROSAT archive; Frank Haberl and Wolfgang Zimmerman at MPE for useful discussions.
Additional documentation available in the WGACAT home page