Waveform: Amplitude Versus Time, Seismic Wavelets, and Sampling in WCSB Reflection Surveys
A waveform is the shape of a wave, most often displayed as a graph of amplitude, or some other quantity of interest, plotted against time. In oil and gas exploration the term is central to reflection seismology, where the energy sent into the ground by a source and the energy that returns to a receiver are both recorded as time-varying voltages and drawn as waveforms called seismic traces. The vertical axis of a seismic trace is amplitude, proportional to the strength of ground motion or pressure the sensor detects, and the horizontal axis is two-way travel time, the time for energy to travel down to a reflecting boundary and back up to the surface. The particular waveform of a single reflection event is the seismic wavelet, and its shape, dominant frequency, and phase govern how sharply a geophysicist can resolve thin beds and pick the exact time of a reflector. Waveforms are not confined to seismic traces: the source signature of a vibroseis sweep or an air-gun pulse is a waveform, the impulse response of a formation is a waveform, and the outputs of sonic and other borehole tools are recorded as waveforms too. Because a waveform is a continuous physical signal but computers store numbers, every recorded seismic waveform is sampled at a fixed interval, commonly 1, 2, or 4 milliseconds in land acquisition, which sets the Nyquist frequency and therefore the highest frequency the data can faithfully represent without aliasing. The frequency content of a waveform is the flip side of its time shape: a short, sharp waveform is broadband and resolves thin layers, while a long, ringing waveform is narrowband and blurs them, a relationship made explicit by the Fourier transform that converts a waveform between the time and frequency domains. In the Western Canadian Sedimentary Basin, waveform quality is what determines whether a survey can distinguish the tight, laminated pay of the Montney or the Duvernay from surrounding shale, so processors spend enormous effort shaping waveforms through deconvolution, which compresses the wavelet toward a spike, and through careful attention to sampling, filtering, and phase so that a picked event corresponds to a real geological boundary rather than a processing artifact. Waveform analysis also underpins amplitude-versus-offset work, seismic inversion, and 4D monitoring, where subtle changes in the shape or amplitude of returning waveforms are interpreted as changes in rock and fluid properties at depth.
Key Takeaways
- Amplitude Against Time: A waveform plots amplitude, or another quantity of interest, versus time. A seismic trace is the canonical example, with amplitude on the vertical axis and two-way travel time on the horizontal. The shape of that curve carries the geological information: where the peaks and troughs fall in time indicates the depth of reflectors, and how large they are indicates the contrast in rock properties across each boundary.
- The Wavelet Sets Resolution: The waveform of an individual reflection is the seismic wavelet. Its dominant frequency and phase control vertical resolution, roughly a quarter of the dominant wavelength, so a broadband, compact wavelet resolves thin WCSB pay while a low-frequency, ringing wavelet smears adjacent beds together. Much of seismic processing exists to sharpen this waveform toward an ideal spike.
- Sampling and Aliasing: A continuous waveform must be digitized at a fixed sample interval, typically 1, 2, or 4 milliseconds on land. That interval fixes the Nyquist frequency, the highest frequency recordable without aliasing. Sampling too coarsely folds high frequencies back into the data as false low-frequency energy, corrupting the waveform, which is why acquisition parameters are chosen against the target's expected bandwidth.
- Time and Frequency Are Two Views: Every waveform has an equivalent frequency spectrum, and the Fourier transform moves between them. A short waveform is broadband; a long, oscillatory waveform is narrowband. Interpreters use both views, since filtering, deconvolution, and spectral decomposition are all operations that reshape the waveform to enhance the signal geologists need.
- Foundation for Advanced Analysis: Waveform shape and amplitude drive amplitude-versus-offset analysis, seismic inversion, and 4D time-lapse monitoring. In each case, subtle differences in returning waveforms are read as changes in lithology, porosity, or fluid, so the fidelity of the recorded waveform directly limits how much reservoir insight a survey can deliver.
Wavelet Shape and Vertical Resolution
The practical value of a seismic waveform comes down to how compact the wavelet is. Vertical resolution is often approximated as a quarter of the dominant wavelength, so raising the usable frequency of the wavelet directly thins the beds a survey can separate. A Montney or Duvernay operator wanting to map a laminated interval only a few metres thick needs a waveform rich in high frequencies and stable in phase. Processors pursue this through deconvolution, which attempts to collapse the recorded wavelet toward a spike, and through zero-phasing, which centers the waveform's energy on the reflector so a picked peak lands on the true boundary rather than beside it.
Source Signatures as Controlled Waveforms
The energy put into the ground is itself a designed waveform. A vibroseis truck emits a swept-frequency signal, a chirp whose waveform rises through a chosen band over several seconds, and correlation later compresses that long sweep into a short wavelet. An explosive or air-gun source instead emits a sharp impulse whose waveform approximates a spike. In the WCSB, vibroseis dominates land acquisition because its waveform is repeatable and its frequency band can be tuned to the depth and target, giving processors a clean, known input waveform to work back from.
Fast Facts
The mathematics that lets a waveform be split into its component frequencies dates to Joseph Fourier's 1822 work on heat conduction, more than a century before reflection seismology existed. His insight that any waveform can be represented as a sum of sinusoids is the reason a single seismic trace can be decomposed, filtered, deconvolved, and reassembled. Every frequency filter, every spectral-decomposition color blend, and every deconvolution operator applied to modern WCSB seismic data rests on that two-hundred-year-old theorem about the shape of a wave.
Related Terms
Waveform links to the core vocabulary of seismic work. The Wavelet is the waveform of a single reflection and the object most processing tries to sharpen. Amplitude is the quantity the waveform's vertical axis measures and the carrier of rock-property information. Deconvolution is the operation that reshapes a recorded waveform toward a spike to recover resolution. And Two-Way Time is the horizontal axis of a seismic waveform, converting the wave's shape into a depth-ordered image of the subsurface.
Real-World WCSB Scenario: Resolving Duvernay Pay With Waveform Processing
A geophysics team evaluating a Duvernay shale block west of Edmonton acquired a 3D vibroseis survey with a 2 ms sample rate and a sweep engineered to push usable energy toward the high end of the band. Raw stacks showed the Duvernay as a smeared low-frequency waveform that could not be separated from the overlying Ireton shale. The processing shop applied spiking deconvolution and zero-phasing to compress and center the wavelet, then ran spectral decomposition to isolate the frequencies that best illuminated the target. Processing costs for the shaping and imaging ran into the low hundreds of thousands CAD across the survey.
The reshaped waveform separated the Duvernay from adjacent shale by several milliseconds of two-way time and revealed thickness variation the raw data had hidden. That improved image let the operator place horizontal landing points more precisely, reducing the risk of drilling out of zone across multi-well pads worth tens of millions CAD in development capital.