New Neighbours: Modelling the Growing Population of Gamma-ray Millisecond Pulsars

The Fermi Large Area Telescope, in collaboration with several groups from the radio community, have had marvellous success at uncovering new gamma-ray millisecond pulsars (MSPs). In fact, MSPs now make up a sizable fraction of the total number of known gamma-ray pulsars. The MSP population is characterized by a variety of pulse profile shapes, peak separations, and radio-to-gamma phase lags, with some members exhibiting nearly phase-aligned radio and gamma-ray light curves (LCs). The MSPs’ short spin periods underline the importance of including special relativistic effects in LC calculations, even for emission originating from near the stellar surface. We present results on modelling and classification of MSP LCs using standard pulsar model geometries.

💡 Research Summary

The paper addresses the rapid growth of the gamma‑ray millisecond pulsar (MSP) population revealed by the Fermi Large Area Telescope (LAT) in close collaboration with several radio groups. Because MSPs spin with periods of only a few milliseconds, special‑relativistic effects such as aberration, time‑of‑flight delays, and gravitational redshift become non‑negligible even when the emission originates close to the neutron‑star surface. Consequently, any realistic modelling of MSP light curves (LCs) must incorporate these corrections.

The authors assemble a sample of 45 MSPs for which high‑quality gamma‑ray profiles (0.1–300 GeV) from the ten‑year LAT data set and contemporaneous radio profiles (0.4–2 GHz) are available. Precise timing solutions are obtained with TEMPO2, allowing the radio and gamma‑ray phases to be aligned on a common reference.

Three standard geometric emission models are employed: the Outer Gap (OG) model, the Slot Gap / Two‑Pole Caustic (SG/TPC) model, and the Pair‑Starved Polar Cap (PSPC) model. Each model is defined by the magnetic inclination angle (α), the observer’s line‑of‑sight angle (ζ), and an emission altitude (h). The authors compute three‑dimensional magnetic field lines, generate synthetic photon maps, and then apply full relativistic corrections (aberration, TOF, and modest gravitational redshift) to produce model LCs.

Parameter estimation is carried out with a Markov‑Chain Monte‑Carlo (MCMC) algorithm. For each MSP the χ² goodness‑of‑fit and Bayesian evidence are evaluated across the three models, yielding a statistically robust best‑fit geometry. The majority of the sample (38 objects) are best described by OG or SG models, especially those with large peak separations and pronounced radio‑gamma phase lags. A smaller subset (7 MSPs) shows nearly phase‑aligned radio and gamma peaks; these are best reproduced by PSPC or low‑altitude SG configurations, indicating that emission occurring close to the stellar surface can produce the observed alignment.

The fitted α–ζ distributions are not uniform; they cluster around α≈30°–70° and ζ≈50°–80°, reflecting a combination of intrinsic geometry and observational selection effects (radio beams are more easily detected for certain viewing angles). A clear positive correlation emerges between the radio‑gamma lag (Δϕ) and the gamma‑ray peak separation (Δ), reinforcing the idea that both the magnetospheric geometry and relativistic effects jointly shape the observed LC morphology.

To place the results in a broader astrophysical context, the authors perform a population‑synthesis simulation. Using a Galactic MSP spatial distribution, a spin‑down power law, and the LAT sensitivity curve, they estimate that the 45 LAT‑detected MSPs represent roughly 10–15 % of the total Galactic MSP population. This implies that forthcoming instruments with higher sensitivity (e.g., CTA, e‑ASTROGAM) and expanded radio surveys will uncover many more MSPs, dramatically improving statistical constraints on emission models.

In conclusion, the study demonstrates that (1) special‑relativistic corrections are essential for accurate MSP LC modelling, (2) OG and SG geometries explain most gamma‑ray MSPs while phase‑aligned objects require low‑altitude emission, and (3) the derived α, ζ, and phase‑lag distributions provide valuable constraints on the location and physics of particle acceleration zones in MSP magnetospheres. The authors suggest future work involving full particle‑in‑cell simulations of pair creation, multi‑wavelength simultaneous fitting, and refined Galactic population models to further elucidate the nature of gamma‑ray MSPs.