Understanding Gamma-Ray Emission from the Galactic Center: Constraining the Millisecond Pulsar Population

We propose a study aimed at modeling the population of recycled millisecond pulsars (MSPs) in the inner ∼100 pc surrounding our galactic center (GC). Our study is motivated by Fermi-LAT observations indicating a diffuse γ-ray source with a 3D radial profile following ρ∼r−2.44±0.04 (Gordon & Macias 2013) in the inner Galaxy. Given its γ-ray spectrum, the origin of this source has been linked to either a population of MSPs or dark matter (Hooper & Linden 2011, Abazajian 2011). Given the γ-ray luminosity of MSPs detected by the Fermi- LAT, a population of ∼1000 MSPs would be required to explain the signal. Furthermore, these MSPs must have a radial morphology that is significantly steeper than the stellar profile. Our current understanding of MSP formation involves neutron-star recycling through long-term accretion in low-mass X-ray binaries (LMXBs; Bhattacharya & van den Heuvel 1991). Given the radial profile of the Fermi-LAT emission, the binary progenitors of MSPs would have to be produced via dynamical interactions following roughly the square of the stellar density (Abazajian & Kaplinghat 2012). There is an ongoing debate as to whether these dynamical processes can act over a large enough region to produce the necessary MSP distribution out to ∼100 pc from the GC. On the other hand, the GC is known to host massive clusters inside which frequent dynamical interactions might efficiently lead to the formation of pulsars (Figer et al. 1999). Thus, the migration and dissolution of clusters in the GC environment can represent an efficient source term of MSPs (Antonini 2014). We will assess the MSP hypothesis for the origin of the Fermi-LAT signal through two complementary avenues: (i) Monte-Carlo models of stellar clusters will be generated, taking into account stellar processes relevant to neutron stars and pulsars. N -body representations of the Monte-Carlo models will then be generated and evolved in a model of the GC, including the effects of dynamical friction and internal dynamical relaxation, until the clusters reach the center or dissolve in the process. The result will be a set of predictions of the distribution of MSP populations in the GC. (ii) We will constrain the absolute normalization of the putative MSP population in view of the magnetar recently discovered by NuSTAR and its implications for pulsar observability in the GC, by employing the recently developed code for modeling pulsar survey selection effects and applying it to GC MSPs (Bates et al. 2013).