Suppression: First-Break Overload, Noise Attenuation, and Amplitude Scaling in Seismic Processing
In seismic acquisition and processing, suppression is the deliberate attenuation of selected amplitudes, applied either to reduce the effect of unwanted noise or to prevent overload of recording and display systems from the very high energy of first breaks. It is one of the routine conditioning steps that turns raw field records into an interpretable image, and it spans both the recording instrument and the data-processing flow. The need arises because a seismic trace carries an enormous dynamic range: the first arrivals, the direct and refracted energy that reaches a geophone or hydrophone closest to the source, can be hundreds or thousands of times stronger than the faint reflections from deep formations that the survey actually targets. If that early, high-amplitude energy is allowed to dominate, it can saturate a display, bias automatic gain functions, and mask the weak primary reflections beneath it. First-break suppression, often applied as a front-end mute or a tapered scaling zone, removes or tapers down this overpowering early energy so that the deeper reflection information can be balanced and seen. The same principle extends to coherent and random noise across the whole record. Ground roll, the strong low-velocity surface waves that swamp land data in the WCSB, air blast, powerline hum, swell noise on marine streamers, and multiple energy are all candidates for suppression through frequency filtering, velocity-based (f-k or tau-p) filtering, deconvolution, or amplitude editing of anomalously hot traces and samples. Suppression is closely related to but distinct from muting: a mute zeroes a defined region of the record entirely, while suppression more broadly describes any reduction in amplitude, which may be a hard zero, a smooth taper, or a frequency- or velocity-selective attenuation that leaves the underlying signal intact where it matters. Effective suppression is a balance, because over-aggressive attenuation can remove genuine reflection signal or introduce filtering artifacts, while too little leaves noise that degrades stacking, velocity analysis, and the final migrated image. In WCSB processing of Montney, Duvernay, and Cardium targets, careful suppression of ground roll and first-break overload directly improves the signal-to-noise ratio that controls how confidently an interpreter can pick a thin reservoir or a subtle stratigraphic edge. The step connects to the broader processing chain, feeding cleaner gathers into deconvolution, velocity analysis, stacking, and migration, each of which performs better when overwhelming noise and instrument overload have been controlled first.
Key Takeaways
- Two distinct goals: Suppression either attenuates noise (ground roll, air blast, multiples, random spikes) or prevents recording and display overload from the high energy of first breaks. Both protect the weak deep reflections that a survey targets from being masked or clipped by far stronger, unwanted energy on the same trace.
- Dynamic range problem: First arrivals can be hundreds to thousands of times stronger than deep primary reflections. Without front-end suppression that early energy saturates displays, biases gain functions, and hides the reservoir signal beneath it, so a tapered mute or scaling zone is applied to the early part of each trace.
- Many methods, one aim: Suppression is delivered through bandpass and notch frequency filters, f-k and tau-p velocity filters, deconvolution, and trace or sample amplitude editing. The choice depends on whether the unwanted energy is separable by frequency, by apparent velocity, by predictability, or simply by being anomalously hot.
- Suppression versus mute: A mute zeroes a region of the record outright; suppression is the broader idea of reducing amplitude, which can be a hard zero, a smooth taper, or a selective attenuation. Tapering matters because abrupt edits create high-frequency artifacts that can ripple into the stacked and migrated image.
- Balance is everything: Over-suppression strips real reflection signal and can leave filtering artifacts; under-suppression leaves noise that degrades velocity analysis, stacking, and migration. WCSB ground-roll-heavy land data demands tuned suppression so that thin Montney and Duvernay reflectors survive cleanly into the final volume.
First-Break Suppression and Instrument Overload
The first breaks are the earliest energy to arrive, the direct wave and shallow refractions that travel near the surface from source to receiver. On near-offset traces this energy is so strong it can drive recording electronics or a display toward saturation, clipping the waveform and corrupting any automatic gain applied across the trace. Front-end suppression applies a tapered mute or a steep scaling ramp over the first-arrival cone, removing or reducing that energy before gain balancing. The taper width matters: too narrow and it leaves a hard edge that rings, too wide and it eats into shallow reflection signal. In WCSB land surveys over shallow Cardium targets, careful first-break handling preserves the genuine shallow reflectors a deeper-focused mute would erase.
Ground Roll and Coherent Noise Attenuation
Ground roll, the dispersive low-velocity surface wave that dominates Western Canadian land data, is the classic coherent-noise suppression target. Because it travels far slower than reflected body waves and occupies low frequencies, it separates cleanly in the frequency-wavenumber (f-k) domain or by apparent velocity in tau-p, where a fan or low-cut filter attenuates it while leaving steep reflection energy. Air blast and powerline hum are removed by notch filtering. Suppressing these coherent noises before stacking sharpens velocity semblance and lets the processor build accurate velocity models for Montney and Duvernay imaging, where a few milliseconds of residual surface-wave energy can blur a thin, high-value reflector.
Fast Facts
The dynamic range that makes suppression necessary is staggering. A modern 24-bit seismic recording system captures amplitudes spanning more than 120 decibels, a ratio of over a million to one between the loudest first break and the quietest deep reflection it can resolve. Early analog systems had nowhere near this range, so first-break suppression was not a refinement but a survival requirement to keep the strong early energy from saturating the entire record and erasing the reservoir signal the crew had paid to acquire.
Related Terms
Suppression is one link in the seismic conditioning chain. Deconvolution both removes the source wavelet and suppresses short-period reverberations and multiples, overlapping directly with noise suppression goals. Stacking is itself a powerful suppressor of random noise, summing many traces so signal reinforces while noise cancels, which is why pre-stack suppression of coherent energy matters so much. Deterministic Deconvolution is a related inverse-filtering step that depends on a clean, suppressed input to estimate reflectivity accurately.
Real-World WCSB Scenario: Suppressing Ground Roll Over a Montney Survey
A processing contractor handling a 3D survey over a Montney play near Dawson Creek, British Columbia, finds the raw shot gathers dominated by ground roll that swamps the 8 to 20 Hz band and a strong first-break cone clipping the near offsets. The processor applies a tapered first-break mute to control overload, then an f-k fan filter and surface-consistent amplitude scaling to suppress the dispersive surface waves, taking care to preserve the steep primary reflections. The flow is iterated against velocity semblance panels so the suppression is tuned, not blanket.
The conditioned gathers lift the signal-to-noise ratio enough that the target Montney doublet, previously a smeared low-amplitude event, resolves into two pickable reflectors. The improved image lets the geoscience team high-grade landing zones across the 90 square kilometre survey, supporting a multi-well pad program worth well over 40 million CAD in drilling and completion capital.