Fastest or Significant: A Systematic Framework for Validating Global Minimum Variability Timescale Measurements of Gamma-ray Bursts

The minimum variability timescale (MVT) is a key observable used to probe the central engines of Gamma-Ray Bursts (GRBs) by constraining the emission region size and the outflow Lorentz factor. However, its interpretation is often ambiguous: statistical noise and analysis choices can bias measurements, making it difficult to distinguish genuine source variability from artifacts. Here we perform a comprehensive suite of simulations to establish a quantitative framework for validating Haar-based MVT measurements. We show that in multi–component light curves, the MVT returns the most statistically significant structure in the interval, which is not necessarily the fastest intrinsic timescale, and can therefore converge to intermediate values. Reliability is found to depend jointly on the MVT value and its signal-to-noise ratio ($\mathrm{SNR}{\mathrm{MVT}}$), with shorter intrinsic timescales requiring proportionally higher $\mathrm{SNR}{\mathrm{MVT}}$ to be resolved. We use this relation to define an empirical MVT Validation Curve, and provide a practical workflow to classify measurements as robust detections or upper limits. Applying this procedure to a sample of Fermi-GBM bursts shows that several published MVT values are better interpreted as upper limits. These results provide a path toward standardizing MVT analyses and highlight the caution required when inferring physical constraints from a single MVT measurement in complex events.

💡 Research Summary

Gamma‑Ray Bursts (GRBs) are among the most luminous transients in the universe, and the minimum variability timescale (MVT) is widely used to constrain the size of the emitting region and the bulk Lorentz factor of the outflow. However, the MVT is notoriously sensitive to statistical noise and to the details of the analysis method, leading to ambiguous physical interpretations. In this paper the authors present a comprehensive simulation‑based study that quantifies the reliability of the Haar‑wavelet based MVT estimator originally introduced by Golkhou & Butler (2014).

The simulation framework systematically explores a multi‑dimensional parameter space: intrinsic pulse shapes (Gaussian, triangular, Norris), pulse widths ranging from milliseconds to seconds, peak amplitudes (i.e., signal‑to‑background ratios), and complex multi‑component light curves built from up to eleven Norris and Gaussian sub‑pulses. For each combination 300 independent time‑tagged event (TTE) realizations are generated with a fixed background of 1000 counts s⁻¹, and the MVT is measured on binned light curves with a range of bin widths (BW). The authors adopt a Monte‑Carlo approach: the median of the MVT distribution is taken as the reported value, and the 16th/84th percentiles provide asymmetric 1σ uncertainties.

Two key behaviours emerge from the simulations. First, the choice of BW determines whether the measurement is “bin‑limited” (BW comparable to or larger than the intrinsic timescale) or “source‑dominated” (BW ≪ intrinsic timescale). In the former regime the MVT is systematically over‑estimated and depends on the binning; in the latter it converges to a stable plateau that accurately reflects the true variability timescale. Second, the reliability of a given MVT depends jointly on the measured MVT value and its signal‑to‑noise ratio (SNR_MVT). Short intrinsic timescales (≲ few ms) can only be recovered if SNR_MVT is sufficiently high; otherwise the algorithm either returns an over‑estimated value or fails altogether. By mapping the success rate across the (MVT, SNR_MVT) plane the authors derive an empirical “MVT Validation Curve”. This curve provides a quantitative criterion: measurements lying above the curve have a high probability (> 90 %) of being genuine detections, while those below should be treated as upper limits.

Armed with this validation curve, the authors propose a practical workflow for real data: (1) compute the Haar‑based MVT and its uncertainty; (2) evaluate SNR_MVT; (3) locate the point on the validation curve to assess the success probability; (4) classify the result as a robust detection or an upper limit. Applying this procedure to a sample of Fermi‑GBM bursts, they find that several previously published MVT values fall below the validation curve. In those cases the reported MVTs are better interpreted as upper limits rather than precise measurements of the fastest variability. Consequently, physical constraints derived from such values—e.g., emission radius or Lorentz factor—must be revisited.

The paper concludes that the Haar‑based MVT estimator reliably identifies the most statistically significant structure in a light curve, but its physical meaning is contingent on the measurement’s SNR. The introduced validation framework standardizes the assessment of MVT measurements, mitigates the risk of over‑interpreting noise‑driven features, and paves the way for more robust inferences about GRB central engines and emission mechanisms.