ar X iv : a st ro - p h / 99 12 31 1 v 1 1 5 D ec 1 99 9 COMPTEL Time - Averaged All - Sky Point Source Analysis

We use all COMPTEL data from the beginning of the CGROmission (April ’91) up to the end of CGRO Cycle 6 (November ’97) to carry out all-sky point source analyses in the four standard COMPTEL energy bands for different time periods. We apply our standard maximum-likelihood method to generate all-sky significance and flux maps for point sources by subtracting off the diffuse emission components via model fitting. In addition, fluxes of known sources have been determined for individual CGRO Phases/Cycles to generate lightcurves with a time resolution of the order of one year. The goal of the analysis is to derive quantitative results – significances, fluxes, light curves – of our brightest and most significant sources such as 3C 273, and to search for additional new COMPTEL sources, showing up in time-averaged maps only. INTRODUCTION The imaging COMPTEL experiment aboard CGRO is the pioneering satellite experiment of the MeV-sky (∼1 30 MeV). For a detailed description of COMPTEL see [1]. One of COMPTEL’s prime goals is the generation of all-sky maps, which provide a summary on the MeV-sky in total. This goal has been achieved by e.g. [2], [3] who generated maximum-entropy all-sky images and by [4], who generated the first COMPTEL all-sky maximum-likelihood maps, which – compared to maximum-entropy ones – have the advantage of providing quantitative results like significances and fluxes of source features. Here we present all-sky maximumlikelihood maps from which models of the diffuse emission have been removed. Our emphasis is on AGN. For a discussion on the method see [5] in these proceedings. The main analysis goals are 1) to derive a summary of known COMPTEL point sources, 2) to search for further point sources, 3) to derive time-averaged quantitative parameters (’first order’) of our brightest point sources, i.e., significances, fluxes, MeV-spectra, and possible time variability, and 4) to further investigate our data and analysis methods. DATA AND ANALYSIS METHOD Using all data from the beginning of the CGRO mission (April ’91) to the end of CGRO Cycle VI (Nov. ’97), we generated a consistent database of relevant COMPTEL data sets (events, exposure, geometry) for individual CGRO viewing periods (VPs) in the 4 standard energy bands (0.75-1, 1-3, 3-10, 10-30 MeV) in galactic coordinates by applying consistent data selections. This database was supplemented by relevant data sets containing models describing the galactic diffuse γ-ray emission (HI, CO, and inverse-Compton components) and the isotropic extragalactic γ-ray background emission. To check for time variability of γ-ray sources these data sets were combined for different time periods: the six individual CGRO Phases/Cycles, the sum of all data (CGRO Phases I-VI; April ’91 Nov. ’97) as well as the first (CGRO Phases I-III; April ’91 Oct. ’94) and the second half (CGRO Phases IV-VI; Oct. ’94 Nov. ’97). Each set of all-sky data is analysed by our standard maximum-likelihood method which simultaneously ’handles’ individual VPs, generates, iteratively, a background model (see [6]), and finally generates significance and flux maps and/or significances and fluxes for individual sources. Because we are interested in point sources, the diffuse emission is always removed in the fitting procedure (e.g. Figure 1). For the derivation of the source fluxes (see Figure 2 as an example), the point sources of interest (e.g. 3C 273, Cyg X-1) have additionally been included in the fitting procedure. We like to mention however, that the results derived by such all-sky fits should be considered correct to first order only. To derive final/optimal results for a particular source, a dedicated analysis has to be carried out, which e.g. makes several cross checks by applying different background models and would take into account the presence of other source features in the region of interest. Also, along the galactic plane the results depend on the ’goodness’ of the applied diffuse emission models for the MeV-band. RESULTS The significance maps in Figure 1, which contain all data of the first 6.5 years of the COMPTEL mission, are the first COMPTEL all-sky point source maps in the continuum bands. They provide a summary of the on-average brightest and most significant MeV-sources. Similar maps focussing on the Galactic plane only are given elsewhere in these proceedings ([5]). The Crab – for display reasons removed in all maps of Figure 1 – is by far the most significant COMPTEL point source. In the 1-3 MeV band for example it reaches a significance of ∼110σ (i.e. a likelihood ratio of ∼12000) for the CGRO Phase I-VI period. With significances of ∼11σ, ∼10σ, and ∼6σ in the 1-3, 3-10, and 10-30 MeV bands is the quasar 3C 273 found to be on average the second most significant point source. Its fluxes in these bands are between 10% and 15% of the Crab flux. Several other extragalactic (e.g. 3C 279, PKS 0528+134, Cen A) and galactic (e.g. Cyg X-1, PSR 1509-58, a known COMPTEL 1−3 MeV All−Sky Significance Map; CGRO Phase I−VI ( W. Collmar August 12, 1999 file_1−3.fits Layer 0 0 30 60 90 120 150 180 210 240 270 300 330