3 The first COMPTEL source catalogue

This section consists of four different parts. The first part (Sect. 3.1) lists all observations on which the catalogue is based. Section 3.2 contains COMPTEL all-sky maps in continuum and line emission. Section 3.3 is the catalogue of detected sources, which is subdivided into detections of spin-down pulsars, galactic sources ($\mid$b$\mid$ $< 10^{\circ}$), active galactic nuclei, unidentified high-latitude sources, gamma-ray line sources, gamma-ray burst sources within the COMPTEL field-of-view, and solar flare detections. Section 3.4 lists COMPTEL upper limits on source candidates, namely galactic objects, active galactic nuclei, and possible gamma-ray line sources.

Figure 1: Effective exposure of COMPTEL from the sum of all observations. Top: Sum of all observations up to Phase IV/Cycle-5. Bottom: Sum of all observations up to Phase IV/Cycle-7

 

Table 1: COMPTEL 3$\sigma$ point source sensitivity limits

	3$\sigma$ Flux Limits [10-5 photons cm-2 s-1]
$E_{\gamma}[{\rm MeV}]$	2 weeks in Phase 1	Phase 1+2+3	Phase 1+2+3+4 (Cycle-5)
0.75 - 1	20.1	7.4	3.7
1 - 3	16.8	5.5	3.8
3 - 10	7.3	2.8	1.7
10 - 30	2.8	1.0	0.8
1.157	6.2	2.0	1.6
1.809	6.6	2.2	1.6
Note. From this table rough upper limits can be derived for those objects, which are not contained
in the later Tables 10 to 12 by deriving the effective exposure from Fig. 1.

3.1 Observations and exposure maps

This first COMPTEL source catalogue contains mainly results from Phase-I to IV/Cycle-5 of the Compton mission. The relationship of the viewing periods (VP) to the actual dates of the observations is given in Table 2 (for completeness, the data of Phase-IV/Cycle-6 and 7 have been added in the table). The table also lists the pointing direction of the z-axis (COMPTEL telescope axis) in celestial coordinates, the duration of the pointing and the effective COMPTEL observation time.

The effective COMPTEL exposure of the entire sky from the sum of all observations from the beginning of the mission to Phase-IV/Cycle-5 and Phase IV/Cycle-7 are illustrated in Fig. 1. The deepest exposures were obtained in the Galactic center and anticenter region, where effective observation times up to $6 \ 10^{6}$ s have been obtained (see also Table 1).

3.2 COMPTEL all-sky maps

COMPTEL all-sky maps exist for continuum emission in the three standard energy ranges 1-3 MeV, 3-10 MeV, and 10-30 MeV, and for the 1.8 MeV line from radioactive ²⁶Al. These maps are shown in Figs. 2 and 3.

Figure 2 is a maximum-entropy map using all data from Phase I to Phase IV/Cycle-6 (Strong et al. 1999). The background method used in this map is based on averaging high-latitude observations (see Sect. 2).

Well known sources appear in the map:

Crab ( $l=184.5^{\circ}$, $b=5.9^{\circ}$), Vela (263.6$^{\circ}$, $-2.5^{\circ}$) above 3 MeV, Cyg X-1 (71.1$^{\circ}$, $+3.3^{\circ}$), as well as striking excesses at (18$^{\circ}$, 0$^{\circ}$) and near the Galactic center. At higher latitudes the sky is dominated by extragalactic sources: Cen A (309$^{\circ}$, +19$^{\circ}$) below 10 MeV ([Steinle et al. 1995]), 3C 273 (290$^{\circ}$, +64$^{\circ}$) and 3C 279 (305$^{\circ}$, +57$^{\circ}$) ([Williams et al. 1995b]). Various "MeV blazars'': 3C 454 (86$^{\circ}$, $-38^{\circ}$), PKS 0208-512 (276$^{\circ}$, $-62^{\circ}$), GRO J 0516-609 ( $270^{\circ}$, $-35^{\circ}$) appear in one or more of the energy ranges. Next to the Crab the quasar PKS 0528+134 is clearly visible ([Collmar et al. 1994]; [Collmar et al. 1997a]). Details on these sources can be found in Blom et al. (1996) and references therein.

Note that since these sources are variable, their appearance in these time-averaged maps may not reflect their published fluxes or spectra precisely. Another interesting feature is the apparent presence of significant areas of diffuse emission away from the plane; in particular, the regions around (170$^{\circ}$, +50$^{\circ}$) and (85$^{\circ}$, 35$^{\circ}$), which have been presented as candidates for assocations with high-velocity cloud complexes ([Blom et al. 1997b]). Details in the structure of this emission should, however, be viewed with caution and are under further study.

An alternative approach to derive all-sky continuum maps is described in Bloemen et al. (1999a). This is a new approach combining model fitting, iterative background modelling and maximum entropy imaging, using the first five years of COMPTEL observations. On a coarse scale the maps derived by both methods are similar. However, on a fine scale, there are differences which are not yet fully understood. The uncertainties especially effect the identification of sources in the Galactic plane (see above).

Figure 3 is a COMPTEL maximum-entropy map at 1.809 MeV from all observations up to VP 522.5 ([Oberlack 1997]). The brightest regions are in the inner Galaxy ( $-35^{\circ}< l < +35^{\circ}$), near Carina ( $l \sim 285^{\circ}$), and Vela ( $l \sim 267^{\circ}$). Other regions of enhanced emission are Cygnus ( $l\sim 80^{\circ}$), and Aquila ( $l \sim 45^{\circ}$).

More recently, new 1.809 MeV COMPTEL all-sky maps have been produced using different imaging and background modelling methods. In one case ([Knödlseder et al. 1999]) the analysis uses a multi-resolution version of the Richardson-Lucy algorithm, based on wavelets. In the other case ([Bloemen et al. 1999b]), the maximum entropy method is combined with model fitting and iterative background modelling. All three maps are consistent with each other within their statistical and systematic uncertainties, although the multi-resolution map shows substantially less structure along the Galactic plane.

So far, no all-sky map in the light of the 1.157 MeV ⁴⁴Ti gamma-ray line exists. But imaging analysis along the Galactic plane have been performed by Dupraz et al. (1997) for data from Phases I to III, and by Iyudin et al. (1999) from Phases I to IV/Cycle-6. Two ⁴⁴Ti gamma-ray line sources have been discovered so far, these are Cas A ([Iyudin et al. 1997a]) and GRO J 0852-4642 ([Iyudin et al. 1998]), a supernova remnant near the Vela region ([Aschenbach 1998]).

A maximum-entropy map in the light of the 2.2 MeV neutron-capture line based on data from the first five years of the mission (VP 1.0 through VP 523.0) has been produced by Mcconnell et al. (1997b). In general, the sky at 2.2. MeV is relatively featureless, e.g. the galactic plane is not visible. There is, however, evidence for significant ( $\sim 3.7\sigma$) emission from a point-like feature near $(l, b) = (300.5^{\circ}$, $-29.6^{\circ}$), the origin of which remains unknown at this time ([McConnell et al. 1997b]). Flux limits for any candidate source are typically in the range (1 to 2.0) 10^-5 cm^-2 s^-1 (at the $3\sigma$ significance level).

Figure 4 is an all-sky map of the statistical location contours of 31 gamma-ray bursters, which happened to be in the COMPTEL field-of-view from the beginning of the mission up to viewing period 419.5 ([Kippen et al. 1998a]).

3.3 COMPTEL source detections

The COMPTEL source detections are summarized in Tables 3 to 9:

• Table 3: Detected Spin-Down Pulsars.
• Table 4: Galactic Sources $\mid$b$\mid$ $< 10^{\circ}$.
• Table 5: Active Galactic Nuclei.
• Table 6: Unidentified High-Latitude Sources.
• Table 7: Gamma-Ray Line Sources.
• Table 8: Gamma-Ray Burst Locations.
• Table 9: Solar-Flare Detections.

Out of the 7 spin-down pulsars listed in Table 4 COMPTEL has made firm detections from PSR B0531+21 (Crab), PSR B0833-45 (Vela), and from PSR B1509-58. The analysis of the Vela pulsar, however, is not yet finally settled; the results presented are presently based on Phases 0 and I, only. Because of the good statistics, the Crab pulsar fluxes are listed for smaller energy intervals than the standard ones (0.75-1, 1-3, 3-10 and 10-30 MeV). Only indications for emission in the COMPTEL energy range were found from the four pulsars PSR B1951+32, PSR J0633+1746 (Geminga), PSR 0656+14, and PSR B1055-52.

Figure 2: Full-sky maximum entropy intensity maps from all data between Phases I to IV/Cycle-6. Energy ranges are (from top to bottom): 1-3 MeV, 3-10 MeV, 10-30 MeV (from [Strong et al. 1999])

Figure 3: COMPTEL maximum entropy map from VP 1 to VP 522.5 at 1.809 MeV (from [Oberlack 1997])

The Galactic Sources with $\mid$b$\mid$ $< 10^{\circ}$ listed in Table 4 are all objects, which were seen by at least one other experiment of the Compton Observatory. These are Cyg X-1, the two EGRET sources 2EG 2227+61 and 2EG J0241+6119 (which both are also COS-B sources), Nova Persei 1992 (GRO J0422+32), the Crab nebula, and an unidentified bright source at $l = 18^{\circ}$ within the plane (coincident with the EGRET source 2ES J1825-1307). The COMPTEL fluxes for these sources are listed in the standard energy ranges. For some of the sources the fluxes are also given for other energy intervals. The possible contribution of the diffuse galactic emission to the listed source fluxes constitutes a basic uncertainty (see Col. 11 of the table).

Nine of the Active Galactic Nuclei listed in Table 5 are of the $\gamma$-ray Blazar type, discovered by EGRET (with the exception of 3C 273, earlier discovered by COS-B, Swanenburg et al. (1978). The only non-blazar type object in the table is the radio galaxy Cen A. All $\gamma$-ray blazars are highly variable in intensity.

Three of the five unidentified high-latitude sources in Table 6 are not point-like, but cover an extended region. Their extent may actually be due to a larger number of - so far - unresolved point sources (GRO J1823-12 and the two High-Velocity Cloud complexes).

Figure 4: Statistical (1-, 2-, and 3-sigma) location contours, in Galactic coordinates of 29 gamma-ray bursts observed through viewing period 419.5. The extent of the contours depends on the strength of the burst (from [Kippen et al. 1998a])

The gamma-ray line sources listed in Table 7 are ordered with increasing line-energy. Apart from the four point-like sources SN 1991T, Cas A, Carina, and the SNR GRO J0852-4642, also the three extended emission regions of the inner Galaxy and of the Cygnus and Vela regions are included in the table.

The 31 gamma-ray bursts listed in Table 8 were all recorded in the "telescope mode". The error radius of the burst location (Col. 4) is defined as the angular radius, having the same area as the irregularily shaped COMPTEL 1$\sigma$ confidence region (see also Fig. 5). Columns 5, 6, and 7 provide informations on the COMPTEL accumulation time, the measured 0.75 to 30 MeV fluence and the COMPTEL detection significance of each burst. The parameters for a power-law fit to the spectrum of each burst are listed in Cols. 8 and 9. Column 10 states, whether variability of the burst spectrum during accumulation was observed in the more sensitive "burst mode".

Solar Flare gamma-ray measurements taken in the burst mode are summarized in Table 9. Column 1 contains the COMPTEL flare identification, including the year/month/day and UT of the flare start (taken from the GSFC Solar Data Analysis Center (SDAC)). Truncated Julian Day, in Col. 2 and Col. 3, contains the GOES X-ray start time (taken from the Solar Geophysical Reports). The X-ray classification is in Col. 4 and the BATSE flare number, as assigned at the SDAC at Goddard Space Flight Center, is in Col. 5. The corresponding BATSE trigger number is in Col. 2. The COMPTEL data measured in this burst mode (see Sect. 6) have been inspected and the flare (integration time) duration as judged by the visible signal in the spectrometer is listed in Col. 7 with the corresponding integrated counts in the full energy range in Col. 8. The peak counts (with an integration time of x s) are listed in Col. 9. The peak count rate as measured by BATSE is in Col. 10 as obtained from the SDAC at GSFC. Finally, in Col. 11 is the integrated amount of spacecraft and instrument material between the spectrometer and the Sun. This material attenuates the gamma-ray flux and degrades the spectrum. A small number is better. Large amounts of intervening material can, in principle, be modeled away when producing a photon spectrum, but the results have large errors and are often not unique.

3.4 COMPTEL upper limits to source candidates

The limits are given at the 2$\sigma$ confidence level for the following objects:

• Table 10: Galactic Objects $\mid$b$\mid$ $< 10^{\circ}$.
• Table 11: Active Galactic Nuclei.
• Table 12: Possible Sources of Gamma-Ray Line Emission.

Table 10 contains upper limits to Galactic Sources with $\mid$b$\mid$< 10$^{\circ}$. This source list is restricted to black-hole candidates. Due to the above mentioned, still existing, uncertainty in modelling the diffuse Galactic emission, the source limits are at present rather conservative.

Presented in Tables 11a and 11b are the cumulative two-sigma upper limits to the MeV-emission measured with COMPTEL from active galactic nuclei (AGN) and other unidentified gamma-ray sources detected at high Galactic latitudes. These limits were derived using composite COMPTEL all-sky maximum-likelihood maps for the 4.5 year period covering Phases I through IV/Cycle-4 of the CGRO mission (1991-1995). A description of the data-processing procedure used to obtain the composite all-sky maps can be found in Stacy et al. (1997).

In the choice of candidate objects, emphasis was placed on known or suspected gamma-ray sources, particularly those detected in neighbouring energy bands to COMPTEL by the CGRO/EGRET and OSSE instruments (e.g., [von Montigny et al. 1995]; [McNaron-Brown et al. 1995]). The flux-extraction routine applied to the composite all-sky maps computes average output values within one-pixel radius of a specified source location. To minimize the number of spurious false detections, only those objects for which the summed log-likelihood ratio exceeds the equivalent of a three-sigma source detection (adopting $\chi^{2}_{1}$statistics, appropriate for a previously known source at a specified location) are considered to exhibit potentially significant MeV emission.

Table 11a lists the COMPTEL cumulative upper limits for MeV-emission from AGN through Phase IV/Cycle-4 of the CGRO mission. Column 1 gives the object name in coordinate format; Col. 11 gives another common name for the source; Col. 10 lists the object "type", which is either the object class (SY for Seyfert galaxy, from the target list of Maisack et al. 1995), or a reference to a previously reported gamma-ray detection of this source (1EG for the First EGRET Catalog of [Fichtel et al. 1994], 2EG for the Second EGRET Catalog of [Thompson et al. 1995], 2EGS for the Supplement to the Second EGRET Catalog of [Thompson et al. 1996]).

Table 11b lists the COMPTEL cumulative upper limits for MeV-emission from unidentified high-latitude gamma-ray sources detected by the CGRO/EGRET instrument. In both Tables 11a and 11b the recommended COMPTEL team-standard corrections (for time-of-flight effects, livetime, etc.) are applied to obtain final fluxes and upper limits (see [Diehl 1996]).

Inspection of Tables 11a and 11b shows only in a few isolated cases the cumulative detection with COMPTEL of significant emission from high-latitude sources. This is no contradiction to the detections listed in Table 5. Note that all these sources are time-variable, and their detection in individual viewing periods does not mean that they are visible in cumulative maps.

In general, the flux limits presented in Table 11a show that COMPTEL does not detect cumulative MeV-emission from a majority of the extragalactic blazar sources detected by EGRET. This result is similar to that obtained by Blom (1997), for the case of individual CGRO viewing periods. Ultimately, these cumulative results will be further compared with other source studies for individual CGRO viewing periods, and used in statistical investigations of source properties by object class (e.g. [Blom 1997]; [Williams et al. 1997]).

The upper limits to possible sources of gamma-ray line emission in Table 12 are again ordered with increasing line energies. The line sources considered are SN 1993J (⁵⁶Ni $\rightarrow$ ⁵⁶CO $\rightarrow$ ⁵⁶Fe decay), four supernovae as possible ⁴⁴Ti line emitters at 1.157 MeV, eleven recent novae as possible sources of ²²Na line emission at 1.275 MeV, five nebula or cloud complexes as 1.809 MeV ²⁶Al sources, and six possible 2.223 MeV neutron capture sources.