The Thousand-Pulsar-Array programme on MeerKAT: -- VI. Pulse widths of a large and diverse sample of radio pulsars

TL;DR

This study measures pulse widths of a large sample of radio pulsars using MeerKAT, analyzing their frequency dependence and population properties, revealing new insights into pulsar emission characteristics and magnetic obliquity.

Contribution

It provides the largest pulse width dataset from MeerKAT, refines the period-width relationship with a steeper power law, and investigates frequency-dependent width changes with a novel model.

Findings

Pulse width follows a power law with period, with a steeper index when using orthogonal distance regression.

Identifies pulsars with width broadening at higher frequencies.

Monotonic width changes across the population are observed, excluding bias and scattering as causes.

Abstract

We present pulse width measurements for a sample of radio pulsars observed with the MeerKAT telescope as part of the Thousand-Pulsar-Array (TPA) programme in the MeerTime project. For a centre frequency of 1284 MHz, we obtain 762 $W_{10}$ measurements across the total bandwidth of 775 MHz, where $W_{10}$ is the width at the 10% level of the pulse peak. We also measure about 400 $W_{10}$ values in each of the four or eight frequency sub-bands. Assuming, the width is a function of the rotation period P, this relationship can be described with a power law with power law index $\mu=-0.29\pm 0.03$. However, using orthogonal distance regression, we determine a steeper power law with $\mu=-0.63\pm 0.06$. A density plot of the period-width data reveals such a fit to align well with the contours of highest density. Building on a previous population synthesis model, we obtain population-based estimates of the obliquity of the magnetic axis with respect to the rotation axis for our pulsars. Investigating the width changes over frequency, we unambiguously identify a group of pulsars that have width broadening at higher frequencies. The measured width changes show a monotonic behaviour with frequency for the whole TPA pulsar population, whether the pulses are becoming narrower or broader with increasing frequency. We exclude a sensitivity bias, scattering and noticeable differences in the pulse component numbers as explanations for these width changes, and attempt an explanation using a qualitative model of five contributing Gaussian pulse components with flux density spectra that depend on their rotational phase.