Morphology of 137 Fast Radio Bursts down to Microseconds Timescales from The First CHIME/FRB Baseband Catalog

TL;DR

This study analyzes 137 FRBs at microsecond resolution, revealing sub-structures, scattering properties, and diversity in morphology, providing new constraints on emission mechanisms and the nature of FRB classes.

Contribution

First detailed microsecond-scale analysis of a large FRB sample, uncovering sub-structures and morphological diversity with implications for emission models.

Findings

Sub-burst components as narrow as 23 μs (FWHM).

20% of bursts have sub-structures narrower than 100 μs.

No correlation between scattering time and dispersion or polarization.

Abstract

We present a spectro-temporal analysis of 137 fast radio bursts (FRBs) from the first CHIME/FRB baseband catalog, including 125 one-off bursts and 12 repeat bursts, down to microsecond resolution using the least-squares optimization fitting routine: fitburst. Our measured values are compared with those in the first CHIME/FRB intensity catalog, revealing that nearly one-third of our sample exhibits additional burst components at higher time resolutions. We measure sub-burst components within burst envelopes as narrow as $\sim$23 $\mu$s (FWHM), with 20% of the sample displaying sub-structures narrower than 100 $\mu$s, offering constraints on emission mechanisms. Scattering timescales in the sample range from 30 $\mu$s to 13 ms at 600 MHz. We observe no correlations between scattering time and dispersion measure, rotation measure, or linear polarization fraction, with the latter suggesting that depolarization due to multipath propagation is negligible in our sample. Bursts with narrower envelopes ($\leq$ 1 ms) in our sample exhibit higher flux densities, indicating the potential presence of sub-ms FRBs that are being missed by our real-time system below a brightness threshold. Most multicomponent bursts in our sample exhibit sub-burst separations of $\leq$ 1 ms, with no bursts showing separations $<$41 $\mu$s, even at a time resolution of 2.56 $\mu$s, but both scattering and low signal-to-noise ratio can hinder detection of additional components. Lastly, given the morphological diversity of our sample, we suggest that one-off and repeating FRBs can come from different classes but have overlapping property distributions.