Detection of Partial Coherence due to Multipath Propagation for FRB 20220413B with CHIME/FRB

Fast radio bursts (FRBs) are a $\sim$ millisecond-long transient phenomenon that propagate across extragalactic distances and are effectively a point source. Radio wave propagation through inhomogeneous distributions of plasma can act as a lens, generating multiple images of the emitted electric field. A lens can produce images of a point source where the phase of the electric field between images remains coherent when observed by a radio telescope. FRB 20220413B shows a complicated pulse structure with time separated components that may be image copies of the main components due to plasma lensing. We perform several analyses to determine if FRB 20220413B is consistent with expectations of a plasma lensed FRB. We analyze and fit the morphology of the burst to a plasma lens model and find consistency in the spectro-temporal profile but not the observed flux. Using the complex-valued channelized voltage data from the CHIME telescope, we perform a time-lag correlation analysis and report correlation signatures present in the electric field of FRB 20220413B. We find that there exists an excess correlation signature only in absolute power and not in phase. We perform a frequency-lag correlation analysis on the spectra of all subcomponents of the burst and find a consistent scintillation bandwidth across all components. We find the scintillation bandwidth is consistent with expectations of scattering due to the Milky Way. We interpret this as all burst components propagating through the same scintillation screen located in the Milky Way, which would generate the excess variance signature observed, even in the absence of phase coherence between burst components. We find that while the burst morphology can be modeled by a plasma lens, the coherent signature present in the time-lag correlation is consistent with the expectations of a common scattering screen, but not coherent plasma lensing.

💡 Research Summary

This paper presents a comprehensive investigation of FRB 20220413B using CHIME/FRB’s baseband voltage data to test whether the burst’s complex morphology arises from plasma lensing and whether any phase coherence between multiple images is present. The authors first describe the CHIME instrument, its 400–800 MHz band, 1024 frequency channels, and 2.56 µs time resolution, and they outline the baseband processing pipeline that coherently dedisperses the data and forms a single‑beam product. The intensity profile of the burst shows four main components (A–D), with a striking bifurcation around 550 MHz that resembles a near‑caustic plasma lens signature.

To assess the lensing hypothesis, the authors adopt a one‑dimensional Gaussian plasma lens model. They write the Fermat potential as T = µ_g|x − y|² + µ_l Φ(x) with µ_g containing geometric distances and µ_l ∝ DM_l f⁻². Using the Kirchhoff–Fresnel diffraction integral in the semi‑classical limit, they compute stationary points (image locations) for each frequency channel, obtaining image delays τ_l and magnifications ε_l. The model then synthesizes an intrinsic Gaussian burst, applying the frequency‑dependent delays and magnifications to generate a synthetic intensity map M(f,t). Parameters µ_g, µ_l, and source offset y₁ are varied to fit the observed spectro‑temporal structure.

Despite reproducing the overall “branching” morphology, the fitting process struggles: the parameter space is highly non‑linear and discontinuous because the number of images (1–3) can change abruptly with small parameter shifts. Consequently, the residuals between data and model do not converge, and the flux ratios predicted by the lens model are significantly higher than observed. This suggests that plasma lensing alone cannot explain the burst’s flux distribution.

The authors then turn to a direct search for phase coherence using the complex voltage data. They compute two time‑lag correlation functions: (i) a complex cross‑correlation preserving phase, C_phase(Δt)=⟨V(t)V*(t+Δt)⟩, and (ii) an absolute‑power correlation, C_amp(Δt)=⟨|V(t)||V(t+Δt)⟩. Only C_amp shows a statistically significant excess at lags of a few microseconds, while C_phase remains consistent with noise. This indicates that while the burst components share excess power, their electric‑field phases are uncorrelated, contradicting the expectation for coherent plasma‑lensed images.

A frequency‑lag analysis of each component’s spectrum yields a consistent scintillation bandwidth of order kilohertz across all components. This bandwidth matches predictions for scattering by a single turbulent screen in the Milky Way, implying that all components have traversed the same Galactic scattering screen. The authors argue that such a common screen can produce the observed excess variance in absolute power without requiring any inter‑image phase coherence.

In summary, the morphological features of FRB 20220413B can be qualitatively reproduced by a near‑caustic plasma lens model, but quantitative discrepancies in flux and the lack of phase coherence in the voltage data favor an alternative explanation. The consistent scintillation bandwidth across components points to a common Milky Way scattering screen as the origin of the observed “partial coherence” signature. The study demonstrates that high‑time‑resolution voltage data, combined with both morphological modeling and phase‑sensitive correlation analyses, provide a powerful toolkit for disentangling intrinsic burst structure from propagation effects such as plasma lensing and interstellar scattering.