2.1. -ray Line Emission
Galactic e+ / e- annihilation line radiation
Since observations in the late 1970s gave the first evidence of the 511 keV positron annihilation line (Leventhal et al. 1978), the Galactic center region has been observed by numerous experiments. These observations have not been able to determine the distribution of line emission due to limited spatial resolution (Tueller 1993, Skibo et al. 1992). The situation has also been complicated by apparently time-variable emission (Riegler et al. 1981, Leventhal et al. 1982), which was thought to be caused by positrons escaping from compact sources (Ramaty et al. 1992).
Recently, the data taken with OSSE have been combined with scanning observations by TGRS and SMM to produce maps of the narrow Galactic 511 keV line emission (Purcell et al. 1997). The resulting maps and modelling of the combined data give evidence for three distinct spatial features: the Galactic plane, a central bulge, and an extended emission region at positive latitudes above the Galactic center. Purcell et al. find this asymmetric distribution to be in good agreement with nearly all historic observations, without invoking time variability. Considering fluxes rather than the spatial distribution, supernovae seem capable of producing positrons at the rate required to account for the observed 511 keV emission (Purcell et al. 1997).
The positive latitude feature is suggestive of an outflow from the Galactic center. The extended nature of the emission together with the lack of a high-density target seem to argue against jet activity from one or more of the black-hole candidates residing near the Galactic center. As an alternative it has been proposed that the high-latitude feature is associated with a fountain of radioactive debris produced by enhanced supernova activity in the Galactic center region (Dermer and Skibo 1997).
Galactic Nuclear De-Excitation -ray Line Emission
Nuclear de-excitation -ray lines provide a unique tracer for low energy (~ 2-100 MeV/nuc) cosmic ray nucleons. These are well known from solar flares (Share et al. 1997), but no compelling evidence was seen from other sources prior to CGRO, although some claims were made.
An extensive evaluation of candidate -ray lines from nuclear interactions was presented by Ramaty et al. (1979). The main candidate lines are from 12C at 4.4 MeV and 16O at 6.1 MeV. The lines from energetic nuclei are broader than those of ambient nuclei, hence both can be distinguished. Gamma-ray spectroscopy of nuclear de-excitation lines thus provides a potentially powerful tool to study low energy cosmic ray nuclei and their relative acceleration. It was not predicted, however, that CGRO would be able to detect such lines.
Preliminary results of COMPTEL observations of the inner Galaxy show some, but not convincing evidence for line structure in the spectrum, completely in line with theoretical expectations (Bloemen and Bykov 1997). It came as a surprise when early COMPTEL observations of the Orion region revealed intense emission in the 3-7 MeV band which was soon attributed to Carbon and Oxygen de-excitation lines (Bloemen et al. 1994). Later COMPTEL observations seemed to confirm the detection, though a slightly different spatial distribution of the emission was obtained (Bloemen et al. 1997). OSSE has so far not detected this emission in Orion (Murphy et al. 1996), which can be reconciled with the COMPTEL results only if the source of emission is very extended.
The existing observational limits on diffuse X-ray emission from inverse bremsstrahlung, -ray continuum emission following 0-decay, and -ray line emission of heavier nuclei in the 1-3 MeV band, require substantial fine tuning in attempts to model the Orion source as Carbon and Oxygen de-excitation emission (Ramaty et al. 1997, and references therein). However, alternative models have not been able to explain both the observed spectrum and the apparent lack of time variability (see Bloemen and Bykov 1997).
Recently it has been found by the COMPTEL team that background subtraction techniques used so far may be insufficient. A re-analysis of the COMPTEL Orion data is on the way and it is unclear to what extent the results will change.