Send comments to Pat Slane ( slane@cfa.harvard.edu )
Einstein data for SNR are available as images on magnetic tape or diskette and as tables of counting rates. These have been published in the Astrophysical Journal Supplements (Seward, 1990), and preprints are available on request. This Catalog contains 3 tables. For details please see the published article referenced above.
The sections of this document are given below. Click on Table # to go directly to that table.
Table 2. Count rates in Einstein Detectors.
Table 3. Unresolved X-ray Sources within or nearby the Cataloged Supernova Remnants
The rates (after background subtraction) given in table 2 (and 3) apply to the entire energy range of the IPC (pulse-height channels 1 to 15). Rates are also integrated over the entire IPC or HRI field of view. This is different from the way point-source count rates are sometimes presented in other Einstein publications. The rate is usually restricted to the "broad" energy band, (pulse-height channels 2 - 10) and sometimes to a circle of radius 3 minutes centered on the source. We have used all energy channels because soft, nearby SNR appear strongly in the lowest channels and because distant, strongly-absorbed SNR, have appreciable signal in the highest channels. Because many SNR are larger than the field of view, rates over the entire field of view (~ 60 minutes, diameter) are quoted for all SNR and for all point sources. This allows direct comparison of counting rates for different SNR. The relative contribution of point sources such as internal pulsars is also easily obtained. All rates have been corrected for vignetting, dead time, and scattering from the telescope mirror.
IPC rates for internal pulsars may be high because of a contribution from surrounding diffuse emission. For example, the HRI data for the Vela pulsar and for PSR 1509-58, show bright diffuse emission within 1 minute of the point-like sources, so the IPC rates for these objects include some diffuse emission.
For weak remnants, uncertainties listed for IPC rates are due to counting statistics and background subtraction. For the small, bright remnants, systematic uncertainties in scattering and vignetting corrections dominate. Rates for remnants made from merges of several fields were calculated by determining the total number of counts above background in the merged image and then using a bright spot to normalize to the IPC rate measured in one field. Here the largest uncertainties are in the determin- ation of the background and in the counting rate of the area used for normalization.
The conversion of counting rate to energy flux depends on the spectrum. Tables to aid in doing this are in the Einstein Observatory Revised User's Manual (Harris, 1984) which can be obtained by request.
The HRI was not as sensitive as the IPC to faint diffuse emission. Because the relative background was higher, diffuse emission detected with the IPC was often below threshold in HRI images. Thus only the brighter regions of 3C 58 and CTB 80 are visible in the HRI images. The HRI observ- ation of W 44 shows nothing when examined visually. A rate for W 44 was obtained by comparing the rate integrated over that part of the field covering the remnant with that from the part of the field outside the remnant.
HRI rates given in table 2 have been corrected for vignetting, scattering in the telescope, and deadtime. Uncertainties are both statistical and systematic. Counting statistics are the dominant uncertainty for weaker sources and short observations. Possible errors in scattering correction and HRI sensitivity are dominant for the bright sources. The scattering correction is uncertain because it depends on the sometimes unknown energy spectrum. The efficiency of the HRI is uncertain because it decreased during the observatory lifetime, dropping 20% in 28 months. HRI rates have been corrected to January 1979, the first use of HRI #3, the detector used for all SNR observations, and the closest time to the preflight calibration.
HRI coverage of the larger remnants was usually not complete (eg. Cyg Loop, IC 443). In seven cases, with coverage from 35% to 90%, we have calculated an HRI rate for the entire SNR by using the observed IPC data and assuming the ratio of HRI to IPC rates was the same for the observed and the unobserved parts of the remnant. Such rates are footnoted in table 2.
The dominant uncertainty in MPC counting rates is in background subtraction. Since the MPC is not an imaging detector, background must be measured in source free locations. The background rate, however, depended on both position of the satellite over the Earth's surface and time. Measured background must be corrected for these variations before being applied to those times when a source was in the field of view. Anti- coincidence rates are used to correct for positional variations. The normalization of the anti-coincidence rate to the X-ray rate is allowed to vary with time to fit the background data. Average background rate in the MPC is 10 counts/s in channels 1-6 and 7 counts/s in channels 7-8. Uncertainty in background subtraction is 1% or 0.1 counts/s in channels 1-6. We restrict the data to channels 1-6 (1 - 10 keV) since the highest energies, in channels 7 and 8, suffer the highest charged-particle-induced background rate.
The collimation was 45 minutes FWHM. Rates from remnants with diameters < 4 minutes, if centered in the field of view, needed no correction for transmission through the collimator. Rates for small remnants, off center, were corrected using the ground-measured collimator transmission. Remnants with sizes between 4 minutes and 60 minutes, were assumed to appear as uniform rings (SNR 1006) or disks (IC 443) of emission and correction factors were calculated.
The largest remnants, Vela XYZ and the Cygnus Loop, were observed using a raster of pointings with regular spacing of 60 minutes and 30 minutes respectively between field centers. Thus the Cygnus Loop was observed in entirety with efficiency (calculated by adding the collimator-transmission efficiency of overlapping observations) varying between 1.5 and 2.0, and an MPC rate for the entire remnant could be calculated. Similarly, the efficiency of the Vela SNR observations varied between 0 and 1.0. Regions with 0 efficiency were small and most of the SNR was observed with efficiency ~ 0.5 (50%).
A possible source of error in MPC rates could be caused by counts from sources at the edge of the MPC field of view but just outside the smaller field of view of the imaging detector. In this way, a bright source might produce a moderate rate in the MPC and give no other clue to its existence. Positions of the bright bulge sources and remnants in the galactic plane are known and the rates in table 2 are free of the effects of these. No MPC rate is given for Kes 69 because of nearby G21.5-0.9. Similarly, the rate for G327.1-1.1 is contaminated by 2S 1553-542 and other weak nearby sources. There are also transient sources, always a possibility, and an unusually high MPC rate should not be accepted without reservation.
Aside from the obvious interest in objects which might be neutron stars formed in the SN explosion, this table is a useful supplement to other catalogs which sometimes do not search "messy" fields for weak sources. IPC and HRI rates quoted are total counting rates in the detector. All energy channels of the IPC are included and corrections have been made for mirror-scattering and dead time. Identifications are given when known and in these cases the coordinates listed are those of the optical or radio counterpart. If the source is not identified, the coordinates are those derived from IPC or HRI data.
Sources are listed as being outside the boundary of the SNR(O), inside the boundary, (I), or at the approximate center of the SNR (C).