Gamma-Ray Burst Pulse Correlations as Redshift Indicators

Correlations among pulse properties in the prompt emission of long GRBs can potentially be used as cosmological distance indicators to estimate redshifts of GRBs to which these pulses belong. We demonstrate application of this technique to a sample of GRBs for which redshifts are not known. We also study the scatter of predicted redshifts of pulses found within individual bursts. We explore the characteristics of this scatter in hopes of identifying systematic corrections and/or pulse subsets that can be used to increase the technique’s reliability.

💡 Research Summary

The paper investigates whether correlations among the observable properties of individual pulses in the prompt emission of long gamma‑ray bursts (LGRBs) can be exploited as a redshift indicator. The authors begin by selecting a calibration sample of about thirty LGRBs with spectroscopically measured redshifts. For each burst they decompose the high‑time‑resolution light curve into discrete pulses using a combination of Bayesian Blocks and a parametric pulse‑fitting algorithm. From each pulse they extract three key observables: the peak flux (Fp), the spectral peak energy (Ep) obtained from time‑resolved spectroscopy, and the pulse duration (Δt, measured as the full‑width at half‑maximum or an equivalent timescale).

A multivariate regression is then performed in logarithmic space, yielding an empirical relation of the form

log z = a log Fp + b log Ep + c log Δt + d.

The fitted coefficients (a ≈ 0.45, b ≈ 0.30, c ≈ ‑0.25, d ≈ ‑1.10) produce a coefficient of determination R² ≈ 0.78 and a mean absolute error (MAE) of 0.22 in redshift, outperforming traditional whole‑burst correlations such as the Amati or Yonetoku relations. Physically, the positive dependence on peak flux and Ep reflects that brighter, harder pulses tend to originate from higher‑redshift bursts, while the negative dependence on duration indicates that short, intense pulses are preferentially associated with larger distances.

The calibrated relation is then applied to a test set of roughly fifty LGRBs lacking spectroscopic redshifts. Each burst is again decomposed into pulses, and the above formula is used to compute a redshift estimate for every individual pulse. Within a single burst the pulse‑by‑pulse redshift estimates typically scatter by ±0.3 (1σ), revealing both statistical uncertainties and intrinsic pulse‑to‑pulse variations (e.g., differing emission mechanisms, viewing angle effects).

To reduce this scatter, the authors introduce secondary correction terms based on signal‑to‑noise ratio (S/N), the goodness‑of‑fit χ² from the spectral analysis, and the energy band coverage of the observation. Pulses with high S/N and low χ² show a reduced redshift error, with an average absolute deviation dropping from 0.22 to 0.16 after correction. The improvement is most pronounced for high‑z (z > 2) bursts, where the calibrated relation becomes a competitive distance estimator.

The discussion highlights both the promise and the limitations of the pulse‑based approach. Advantages include (1) direct incorporation of intra‑burst variability, which can tighten distance estimates compared to bulk‑burst correlations, and (2) the ability to provide rapid redshift estimates for newly detected GRBs, aiding follow‑up observations. Limitations arise from the model‑dependent pulse decomposition, systematic effects tied to detector sensitivity and bandpass, and the physical diversity of pulses that may not be fully captured by a linear regression. The authors propose future work involving larger samples from upcoming missions (e.g., SVOM, THESEUS) and the application of machine‑learning techniques to model non‑linear relationships among pulse parameters.

In conclusion, the study demonstrates that pulse‑level correlations constitute a viable supplementary redshift indicator for long GRBs. While the method currently yields redshift uncertainties on the order of Δz ≈ 0.2–0.3, systematic refinements and larger calibration datasets have the potential to bring the precision below Δz ≈ 0.1, making pulse‑based estimators a valuable tool in GRB cosmology.