Galactic bulge millisecond pulsars shining in x rays: A $\gamma$-ray perspective

If the mysterious Fermi-LAT GeV $\gamma$-ray excess is due to an unresolved population of millisecond pulsars (MSP) in the Galactic bulge, one expects this very same population to shine in x rays. For the first time, we address the question of what is the sensitivity of current x-ray telescopes to an MSP population in the Galactic bulge. To this end, we create a synthetic population of Galactic MSPs, building on an empirical connection between $\gamma$- and x-ray MSP emission based on observed source properties. We compare our model with compact sources in the latest Chandra source catalog, applying selections based on spectral observables and optical astrometry with Gaia. We find a significant number of Chandra sources in the region of interest to be consistent with being bulge MSPs that are as yet unidentified. This motivates dedicated multiwavelength searches for bulge MSPs: Some promising directions are briefly discussed.

Figure 1

From left to right, source density of the MSP Galactic population for the disk, BB, and NB components, and their sum.

Figure 2

KDE 2D joint PDF (colored background) of ${\log }_{10}(F_{\gamma }/F_X)$ and x-ray spectral index $\mathrm{\Gamma }$ from the 40 x-ray observed MSPs having a $\gamma$-ray counterpart. Original data from [27] are shown by the white dots.

Figure 3

ACIS limiting sensitivity map for the TRUE and MARGINAL detection likelihood classes. White regions are not covered by the Chandra observations used to build the CSC catalog.

Figure 4

X-ray energy flux distribution (0.5–7 keV) of the synthetic MSP population, averaged over 100 Monte Carlo simulations: Total MSPs in the ROI (orange filled), total detectable MSPs (green solid) including MSPs from BB (red dot-dashed), NB (blue dashed) and disk (not shown). The vertical dotted lines illustrate the validity range of our model extrapolation (see text for details). Errors from Monte Carlo dispersion are not shown here for clarity, see text for details.

Figure 5

Spectral index density histograms for all MSPs in the ROI (oranged filled) and all detectable MSPs (green solid).

Figure 6

As in Fig. 4, showing Monte Carlo predictions for the total number of detectable MSPs (green solid), together with the conservative (orange dashed) and the aggressive (blue dotted) CSC selections.

Figure 7

Left panel: 2D ($l$, $b$) histogram of detectable MSPs, averaged over 100 Monte Carlo simulations. Central panel: Same as left panel for the 3153 objects of the conservative selection. Right panel: Detectable-to-candidate MSPs ratio.