First-break QC for statics

Part 6 — QC during acquisition

Learning objectives

Explain why 5–20% of auto-picks are bad on noisy land data
Apply residual-based rejection: drop picks outside ± N ms of model fit
Recognise the iterative fit–reject–refit workflow
Link pick QC to statics-model accuracy

First-break picking is a high-volume operation: a 30,000-channel land survey produces hundreds of thousands of picks per day. Auto-pickers (Sercel FirstBreak, commercial packages) triggered on simple amplitude thresholds or on short-term-average/long-term-average (STA/LTA) ratios get 80–95% of picks right — the remaining 5–20% are mispicks that, if left unchecked, skew the near-surface statics model by tens of milliseconds per station.

Outlier mechanisms

Three typical failures. A noise spike triggers the picker before the real first break arrives (pick too early). A cycle-skip on a low-SNR trace makes the picker latch onto the second or third cycle of the wavelet (pick 10–20 ms too late). A deeper refraction crosses the first break at some offsets and steals the pick (pick corresponds to a different earth interface entirely). Each produces a visible outlier in the time-vs-offset scatter.

Robust QC flow

The standard flow is iterative: (1) fit a linear refraction model to all far-offset picks, compute predicted times; (2) reject any pick whose residual exceeds a threshold (typically 8–15 ms); (3) re-fit on the surviving picks; (4) repeat until the residual distribution is visually clean. Threshold too tight and a lot of noisy-but-correct picks are thrown away; too loose and outliers sneak into the fit.

From QC to statics

The QC’d first-break dataset is the input to the near-surface tomography (§5.7). Each rejected outlier is a station that will need an alternative source of static correction — typically the upstream residual statics step during processing, which solves for shifts that make the final stack coherent. Good QC reduces the work the residual step has to do.

References

Sheriff, R. E., Geldart, L. P. (1995). Exploration Seismology (2nd ed.). Cambridge University Press.
Yilmaz, Ö. (2001). Seismic Data Analysis: Processing, Inversion, and Interpretation of Seismic Data (2 vols.). SEG Investigations in Geophysics 10.
Cordsen, A., Galbraith, M., Peirce, J. (2000). Planning Land 3-D Seismic Surveys. SEG Geophysical Developments 9.