Pulsed Radio Emission from the Fermi-LAT Pulsar J1732-3131: Search and A Possible Detection at 34.5 MHz

We report our search for and a possible detection of periodic radio pulses at 34.5 MHz from the Fermi-LAT pulsar J1732-3131. The candidate detection has been possible in only one of the many sessions of observations made with the low-frequency array at Gauribidanur, India, when the otherwise radio weak pulsar may have apparently brightened many folds. The candidate dispersion measure along the sight-line, based on the broad periodic profiles from about 20 minutes of data, is estimated to be 15.44 +/-0.32 pc/cc. We present the details of our periodic & single-pulse search, and discuss the results and their implications relevant to both, the pulsar and the intervening medium.

💡 Research Summary

The authors present a search for low‑frequency radio emission from the Fermi‑LAT γ‑ray pulsar J1732‑3131 using the Gauribidanur low‑frequency array at 34.5 MHz. Ten observing sessions (≈20 min each) were recorded between 2002 and 2006; raw voltage streams were Fourier‑transformed into 256‑channel filter‑bank data covering a 1.05 MHz bandwidth with ~2 ms time resolution. Two complementary search strategies were employed: (1) a single‑pulse search that scans a range of trial dispersion measures (DM) and applies exponential match‑filtering to account for interstellar scattering, and (2) a periodicity search that folds the data at the known spin period (0.19652 s) after barycentric correction.

In the single‑pulse analysis, the authors identified ten events exceeding a low threshold of 4.5 σ at a trial DM of 15.55 pc cm⁻³ with a smoothing width of eight samples (≈15 ms). When these events were aligned and averaged, a dispersed pulse profile became visible, yielding a refined DM of 15.55 ± 0.04 pc cm⁻³. However, the authors caution that co‑adding a small number of marginal detections can artificially enhance a dispersive signature, and they performed control tests using pulses selected at other DMs, finding only modest differences in signal‑to‑noise ratio (S/N).

The periodicity search proved more compelling. After folding each frequency channel separately, the authors performed a three‑dimensional search over DM, smoothing width, and period offset. The peak S/N (using a conventional peak‑based merit) occurred at DM ≈ 15.44 pc cm⁻³ and a smoothing width corresponding to ~80–180° of phase, with a peak S/N of about 7 (conservative estimate) or >9 with a more optimistic noise estimate. To avoid bias toward narrow profiles, they also evaluated the sum‑of‑squares (SSQ) of the folded profile as a merit function; this yielded an isolated peak at DM = 15.44 ± 0.32 pc cm⁻³, consistent with the single‑pulse result. The inferred period, 0.19652(3) s, matches the ephemeris extrapolated to the epoch of observation (0.19652125 s).

Extensive sanity checks were performed. The data were split into the lower and upper halves of the band, each folded independently; both halves reproduced the same dispersed pulse at the same DM. The authors also examined the possibility of radio‑frequency interference (RFI) mimicking a ν⁻² dispersion law. By dedispersing the data with a linear delay law (Δt ∝ ν) and comparing the resulting S/N, they demonstrated that the ν⁻² law yields a significantly stronger detection, supporting an astronomical origin.

The measured DM of 15.44 pc cm⁻³, when interpreted with the NE2001 Galactic electron density model, corresponds to a distance of roughly 600 pc, consistent with estimates derived from γ‑ray energetics. This detection, albeit in only one of ten observing sessions, suggests that J1732‑3131 is not strictly radio‑quiet; rather, its radio beam may be detectable at low frequencies where the emission cone widens (radius‑to‑frequency mapping). The authors discuss that low‑frequency observations can probe emission geometry inaccessible at the typical ~1 GHz surveys that have yielded only four radio detections among the >30 Fermi‑LAT blind‑search pulsars.

Nevertheless, the authors acknowledge several limitations. The detection relies on a single epoch, raising the possibility of transient brightening (e.g., giant pulses or scintillation‑induced amplification). The modest S/N, even after conservative estimates, leaves a non‑zero false‑alarm probability given the low detection threshold (4.5 σ). While RFI mitigation steps were thorough, the authors cannot completely exclude exotic RFI that mimics a dispersive sweep. Consequently, they recommend follow‑up observations with higher sensitivity low‑frequency arrays (e.g., LOFAR, LWA, uGMRT) and multi‑epoch monitoring to confirm the periodic signal and to characterize its spectral and temporal stability.

In summary, this work reports a tentative but statistically significant detection of periodic radio emission from the γ‑ray pulsar J1732‑3131 at 34.5 MHz, with a dispersion measure of 15.44 ± 0.32 pc cm⁻³ and a period consistent with the γ‑ray ephemeris. The result highlights the value of low‑frequency searches for pulsars that appear radio‑quiet at higher frequencies and provides a potential avenue for probing pulsar emission geometry, interstellar scattering, and the Galactic electron distribution.