🔭 astrophysics

Pulsar searches of Fermi-LAT gamma-ray sources with the MWA

This paper presents the largest low-frequency (154 MHz) radio survey of unassociated Fermi-LAT gamma-ray sources to date, utilizing a new semi-coherent search pipeline on MWA data which, despite yielding no new pulsar discoveries due to insufficient sensitivity, establishes flux limits and demonstrates the potential for a ~30% increase in detectable gamma-ray pulsars with the upcoming MWA Phase III upgrade.

Original authors: C. P. Lee, N. D. R. Bhat, B. W. Meyers, W. van Straten, D. A. Smith

Published 2026-06-08

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: C. P. Lee, N. D. R. Bhat, B. W. Meyers, W. van Straten, D. A. Smith

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Hunting for Cosmic Clocks

Imagine the universe is filled with millisecond pulsars. These are dead stars (neutron stars) that spin incredibly fast—hundreds of times a second—and act like lighthouses, beaming radio waves out into space. They are so precise they make atomic clocks look sloppy.

Astronomers have found many of these by looking at the sky with radio telescopes. However, there is a "missing link" group: about 1,300 bright gamma-ray sources spotted by the Fermi space telescope that haven't been matched to a radio pulsar yet. The scientists in this paper wanted to see if they could find these missing pulsars by looking at those specific gamma-ray spots with a radio telescope.

The Challenge: The "Cosmic Fog"

The team used the Murchison Widefield Array (MWA), a massive radio telescope in the Australian outback. They chose to look at a low radio frequency (154 MHz).

Think of the space between us and these stars as a giant, invisible ocean of plasma (charged gas). When radio waves travel through this ocean, they get slowed down. The lower the frequency (like a deep bass note), the slower it gets. This causes the sharp "tick" of the pulsar to smear out into a long, blurry "whoosh."

The Problem: If you try to listen to a fast ticking clock through a thick fog, the ticks blur together until you can't hear them at all.
The Old Way: Previous surveys mostly listened at higher frequencies (where the fog is thinner) or used a method that only partially cleared the fog.
The New Tool: The team developed a new computer pipeline (a set of instructions) that acts like a smart noise-canceling headphone. It uses a "semi-coherent" method. Imagine trying to clean a muddy window: instead of wiping the whole thing at once (which takes too much effort) or just wiping a tiny spot (which leaves the rest dirty), they wipe it in smart, overlapping sections. This allows them to clear the "fog" much better than before, specifically for these low-frequency signals.

The Hunt: A Shallow Dive

The researchers pointed their telescope at 308 of these unassociated gamma-ray sources.

The Strategy: They didn't stare at each spot for hours. Instead, they looked at each one for 20 minutes. Think of this as a "shallow dive" into a pool to see if you can spot a fish near the surface.
The Test: Before hunting the unknowns, they tested their new "noise-canceling" software on five known pulsars. It worked! It successfully found four of them, proving the tool was sharp enough to see the signal through the fog.

The Result: No New Fish Found

Despite having the best tools and a solid plan, the team did not find any new pulsars in this survey.

Why?

Not Deep Enough: The "shallow dive" (20 minutes) wasn't long enough to see the faint signals. The pulsars they were looking for were likely too dim for this specific telescope setup at this frequency.
The Limits: They calculated that their telescope could only see pulsars if they were relatively bright (like a lighthouse in a storm). If the pulsars were dimmer (like a candle in the dark), they remained invisible.

They did, however, accidentally spot two known pulsars that were shining brightly in the "side beams" of their telescope (like seeing a streetlight through a gap in a fence), which proved the telescope was working correctly.

What's Next? The Upgrade

The paper concludes that while this specific hunt came up empty, the technology is ready for a bigger hunt.

The Upgrade: The MWA telescope is getting a major upgrade (Phase III) that will double its size and sensitivity.
The Promise: With this bigger telescope, they could find about 30% more pulsars in the same amount of time. If they spend longer looking (1 hour instead of 20 minutes), they could find 60% more.
The Future: They plan to use this new, super-sensitive setup to revisit these gamma-ray spots, looking deeper into the "fog" to finally catch the missing pulsars. They also plan to use this method to look at other cosmic objects like supernova remnants and star clusters.

In summary: The team built a better radio receiver to listen through cosmic fog, tested it successfully, and went hunting for missing cosmic clocks. They didn't find any new ones this time because the clocks were too quiet for their current setup, but they are now ready to build a bigger, louder microphone to find them in the future.

Technical Summary: Pulsar searches of Fermi-LAT gamma-ray sources with the MWA

Problem
Millisecond pulsars (MSPs) are critical tools for astrophysics, including gravitational wave detection via pulsar timing arrays. Approximately one-fifth of known Galactic MSPs have been discovered through targeted radio searches of unassociated gamma-ray sources identified by the Fermi Large Area Telescope (Fermi-LAT). However, these searches have historically been conducted at radio frequencies above 300 MHz to mitigate interstellar dispersion and scattering. This leaves a gap in the parameter space below 300 MHz, where MSPs with steep radio spectra might be detectable. Low-frequency surveys face significant challenges due to chromatic effects in the cold ionised interstellar medium (IISM), specifically the computational cost of correcting large dispersive delays and intrachannel smearing, which limits sensitivity to short-period pulsars. While a Low Frequency Array (LOFAR) survey successfully utilized a semi-coherent dedispersion strategy to discover three MSPs at 135 MHz, a comparable survey using the Murchison Widefield Array (MWA) had not been performed.

Methodology
The authors developed a new search pipeline tailored for the MWA to target unassociated Fermi-LAT sources using archival data from the Southern-sky MWA Rapid Two-metre (SMART) survey. The methodology involves:

Data Source: The survey utilized 20-minute beamformed observations from the SMART survey at 154 MHz, targeting 308 unassociated Fermi-LAT sources (selected from the 4FGL-DR4 catalogue with declinations < +25°).
Pipeline Architecture: The pipeline employs a semi-coherent dedispersion scheme to balance computational cost and dispersive smearing.
- Phase-Coherent Dedispersion: Using the GPU-accelerated CDMT software, complex voltages are coherently dedispersed to 27 trial Dispersion Measures (DMs) ranging from 1.5 to 79.5 cm⁻³ pc in steps of 3 cm⁻³ pc.
- Incoherent Dedispersion: The output is then incoherently dedispersed using the DEDISP library to fill the gaps between coherent steps, resulting in a total of 27,000 DM trials.
- Acceleration Search: The ACCELSEARCH routine from PRESTO performs a Fourier-domain acceleration search (FDAS) to detect binary pulsars, searching for orbital accelerations corresponding to $z$ values between -200 and +200.
- Candidate Sifting: Candidates are filtered based on signal-to-noise thresholds, harmonic consistency, and DM validity. Final candidates are folded and optimized using PREPFOLD.
Validation: The pipeline was validated by blindly detecting five known MSPs (four in short-orbit binaries) within the SMART data, confirming its ability to detect spider systems despite the low frequency.

Key Contributions

Largest Low-Frequency Survey: This work represents the largest radio survey of unassociated Fermi-LAT sources to date and the second such survey conducted below 300 MHz.
Pipeline Implementation: The authors successfully adapted and implemented a semi-coherent dedispersion pipeline for MWA Voltage Capture System (VCS) data, enabling deeper searches for MSPs than previously possible with fully incoherent methods at these frequencies.
Sensitivity Analysis: The study provides detailed sensitivity estimates for the MWA at 154 MHz, calculating minimum detectable flux densities ( $S_{min}$ ) ranging from approximately 30 to 220 mJy (equivalent to 0.7–5.2 mJy at 1.4 GHz) for a 2 ms period pulsar with a 28% duty cycle.

Results

Discovery Yield: No new pulsars were identified in the survey of 308 sources. The authors attribute this null result to insufficient sensitivity.
Blind Detections: The pipeline successfully detected two known non-recycled pulsars (PSR J1752−2806 and PSR J2048−1616) in the sidelobes of the tied-array beam, demonstrating the pipeline's functionality.
Sensitivity Limits: The survey was sensitive to between 8% and 44% of the known population of radio-loud gamma-ray pulsars (below the DM limit imposed by scattering). In contrast, the LOFAR gamma-ray survey was sensitive to 86–95% of the population and discovered three MSPs. The authors note that even if the MWA survey had matched the LOFAR yield, the expected number of discoveries would have been less than one, consistent with the null result.
Blind Test: Of the five test MSPs used to validate the pipeline, four were blindly detected. One (PSR J0952−0607) was not detected, likely due to low signal-to-noise and significant fractional smearing.

Significance and Future Outlook
The paper concludes that while the pilot survey did not yield new discoveries, it establishes a validated workflow for targeted MSP searches with the MWA. The authors argue that the null result is consistent with the survey's sensitivity limits rather than a failure of the method.

The significance of this work lies in its preparation for future upgrades:

MWA Phase III: The upcoming Phase III upgrade, which doubles the number of tiles to 256, is projected to improve instantaneous sensitivity by a factor of two. This would increase the expected yield of detectable gamma-ray pulsars by approximately 30% for the same integration time (20 min).
Real-Time Beamformer (RTB): The development of an RTB will allow for targeted observations with the primary beam centered closer to the source, reducing sensitivity losses from beam offsets and enabling longer integration times.
Extended Integration: Increasing integration time to 60 minutes could increase the expected yield by ~60%, though this reduces sensitivity to very short binary orbits.
Broader Applications: The developed semi-coherent pipeline is applicable to searches for supernova remnants, globular clusters, and steep-spectrum radio sources, providing a foundation for future SKA-Low surveys.

The authors emphasize that observing at lower frequencies remains advantageous for pulsars with steep spectra, provided that scattering limits and integration times are managed effectively through future hardware and software upgrades.

The Big Picture: Hunting for Cosmic Clocks

The Challenge: The "Cosmic Fog"

The Hunt: A Shallow Dive

The Result: No New Fish Found

What's Next? The Upgrade

More like this