Einstein@Home Searches for Gamma-ray Pulsars in the Inner Galaxy

TL;DR

This work uses Einstein@Home to search Fermi-LAT data for gamma-ray pulsations from sources in the inner Galaxy, aiming to identify bright members of a putative bulge MSP population that could explain the GC GeV excess. By combining updated source selection, photon weighting, and a semi-coherent pulsation search with a coherence time of about 48 days, the authors discover four new gamma-ray pulsars, including an isolated MSP, and perform phase-resolved imaging to separate pulsar flux from GC emission. Gamma-ray timing with MCMC templates reveals timing noise in two young pulsars and yields precise ephemerides, while radio searches with MeerKAT and GBT yield no detections; gravitational-wave analyses place stringent upper limits on continuous GW emission. Distance estimates based on gamma-ray energetics favor foreground disk pulsars, though one MSP could be bulge-associated if located at large distance; overall, the findings do not rule out an MSP origin for the GC excess but suggest only a small fraction of bulge MSPs would be detectable with current gamma-ray pulsation searches. The work demonstrates the viability of deep, volunteer-computing powered gamma-ray pulsation surveys in the inner Galaxy and motivates future searches with longer data sets and improved GC region modeling.

Abstract

The Fermi Large Area Telescope (LAT) has revealed a mysterious extended excess of GeV gamma-ray emission around the Galactic Center, which can potentially be explained by unresolved emission from a population of pulsars, particularly millisecond pulsars (MSPs), in the Galactic bulge. We used the distributed volunteer computing system Einstein@Home to search the Fermi-LAT data for gamma-ray pulsations from sources in the inner Galaxy, to try to identify the brightest members of this putative population. We discovered four new pulsars, including one new MSP and one young pulsar whose angular separation to the Galactic Center of 0.93° is the smallest of any known gamma-ray pulsar. We demonstrate a phase-resolved difference imaging technique that allows the flux from this pulsar to be disentangled from the diffuse Galactic Center emission. No radio pulsations were detected from the four new pulsars in archival radio observations or during the MPIfR-MeerKAT Galactic Plane Survey. While the distances to these pulsars remain uncertain, we find that it is more likely that they are all foreground sources from the Galactic disk, rather than pulsars originating from the predicted bulge population. Nevertheless, our results are not incompatible with an MSP explanation for the GC excess, as only one or two members of this population would have been detectable in our searches.

Figures (5)

• Figure 1: Photon phases and pulse profiles for the new gamma-ray pulsars discovered in this work. Lower panels show the phases for individual photons according to our best-fitting gamma-ray timing models, with the photon probability weights indicated by the grayscale. Upper panels show the integrated pulse profiles. The dashed blue lines show the estimated background level, calculated from the photon weights as $\sum_j w_j (1 - w_j) / N_{\rm bins}$2PC. Flux below these lines is attributed to the diffuse background or other nearby point sources. The period of enhanced exposure towards the Galactic Center is visible here as darker bands containing more photons between MJDs 56600 and 57000.
• Figure 2: MeerKAT radio images at 1.28 GHz covering the locations of two new gamma-ray pulsars. No plausible radio counterparts are detected. Left panel: cutout from the SARAO MeerKAT Galactic Plane Survey SMGPS covering the position of PSR J1736$-$3422. Right panel: cutout from the MeerKAT Galactic Center Mosaic Heywood2022+MKGC covering the position of PSR J1748$-$2815. Inset panels contain the $1\arcmin \times 1\arcmin$ region around the pulsars from which we estimated the RMS noise level to obtain flux-density upper limits, with the gamma-ray timing positions of the pulsars marked by red crosses. The positional uncertainties are smaller than these markers. The grayscale has the mean intensity subtracted, and a square-root normalisation applied, as indicated on the color bar.
• Figure 3: Distance dependence of inferred properties of PSR J1649$-$3012. The top panel shows the intrinsic spin-down luminosity ($\dot{E}_{\rm int}$), after correcting for the Shklovskii effect according to our best-fitting timing model (dashed black line) and also the acceleration in the Galactic potential (solid black line). The grey shaded region shows 1-$\sigma$ uncertainties on the intrinsic spin-down luminosity, propagated from the uncertainty on the total proper motion, which we assume to be much larger than the uncertainty in the Galactic acceleration. The gamma-ray luminosity ($L_{\gamma}$) and 1-$\sigma$ uncertainties according to the observed energy flux ($G_{\gamma}$) from 4FGL-DR4 are shown by the red line and shaded region. The middle panel shows the corresponding gamma-ray efficiency, with the ratio of the red and black lines from the top panel shown by the dashed red and black line. The relative contributions to the efficiency uncertainty from the proper motion and observed flux uncertainties are shown by grey and red shaded regions, respectively. The dotted black line shows a loose upper limit corresponding to 100% efficiency. The bottom panel shows the ratio between the predicted density of MSPs in the putative Galactic bulge population from the model of Calore2016 and in the Galactic disk according to the model of Levin2013+GalMSPs.
• Figure 4: Illustration of timing noise in PSR J1736$-$3422. The left panel shows the estimated variations in the rotational phase, after subtracting a cubic spin-down model, with the 2-$\sigma$ confidence interval indicated by the blue shaded region. The right panel shows the resulting power spectrum of these phase variations. The black error bars show the $1\sigma$ posterior uncertainties on the Fourier powers, marginalised over the underlying timing model and template pulse profile shape. The blue dot-dashed lines show the best fitting models for the two noise components, while the red line and shaded region shows the best-fitting model and uncertainty regions for the full noise model prior. The estimated white-noise level is shown by the black dashed line. The posterior distributions for high frequencies are all below this level, and closely follow the priors as a result.
• Figure 5: Gamma-ray photon counts maps of the Galactic Center region. Each row shows counts maps for logarithmically-spaced energy bands, convolved with the Fermi-LAT point-spread function for the highest energy in that band. The gamma-ray timing position of PSR J1748$-$2815 is highlighted by black markers. In each row panels show, from left to right: the total photon counts, on-pulse photon counts (divided by the on-pulse phase fraction to maintain the same color scale as the total count maps), off-pulse photon counts (divided by the off-pulse phase fraction), and the difference between the rescaled on- and off-pulse maps in signal-to-noise units. Counts maps have a square-root scaling, indicated on the color bars. Contour lines on the difference images are at the $2\sigma$ and $3\sigma$ level. The point source visible at $(l,b) = (359.3^{\circ},-0.8^{\circ})$, highlighted by white markers, is another gamma-ray pulsar, PSR J1747$-$2958.