On the morphology of the electron-positron annihilation emission as seen by SPI/INTEGRAL

TL;DR

This study analyzes the morphology of the 511 keV positron annihilation emission in our Galaxy using SPI/INTEGRAL data, revealing detailed structure of the Galactic bulge and extended emission components.

Contribution

The paper introduces a new background modeling method and provides detailed morphological measurements of the Galactic positron annihilation emission.

Findings

Galactic bulge detected at ~70 sigma significance.

Bulge modeled with two Gaussian components of 3.2° and 11.8° FWHM.

Extended emission structure detected with flux 1.7 to 2.9 x10^-3 photons/(cm^2.s).

Abstract

The 511 keV positron annihilation emission remains a mysterious component of the high energy emission of our Galaxy. Its study was one of the key scientific objective of the SPI spectrometer on-board the INTEGRAL satellite. In fact, a lot of observing time has been dedicated to the Galactic disk with a particular emphasis on the central region. A crucial issue in such an analysis concerns the reduction technique used to treat this huge quantity of data, and more particularly the background modeling. Our method, after validation through a variety of tests, is based on detector pattern determination per ~6 month periods, together with a normalisation variable on a few hour timescale. The Galactic bulge is detected at a level of ~70 sigma allowing more detailed investigations. The main result is that the bulge morphology can be modelled with two axisymmetric Gaussians of 3.2 deg. and 11.8 deg. FWHM and respective fluxes of 2.5 and 5.4 x 10^-4 photons/(cm^2.s^1). We found a possible shift of the bulge centre towards negative longitude at l=-0.6 +/- 0.2 degrees. In addition to the bulge, a more extended structure is detected significantly with flux ranging from 1.7 to 2.9 x10^-3 photons/(cm^2.s^1) depending on its assumed geometry (pure disk or disk plus halo). The disk emission is also found to be symmetric within the limits of the statistical errors.