OilAuthority.com

Synthetic Seismograms

A synthetic seismogram is a computed seismic trace constructed from well log data — primarily sonic (acoustic velocity) and density logs — that predicts what the seismic reflection response at a specific well location would look like if the entire seismic wavelet were being reflected off each formation boundary recorded by the logs; it is created by converting the measured acoustic impedance (the product of velocity and density at each depth) into a reflection coefficient series (which quantifies the amplitude and polarity of seismic reflections at each formation interface), and then convolving this reflection coefficient series with an estimated seismic wavelet (characterizing the shape and frequency content of the seismic signal) to produce a synthetic trace that resembles a real seismic trace recorded at that location; the primary purpose of the synthetic seismogram is to tie the seismic data to the well — establishing a direct correspondence between specific reflections visible on the seismic section (identified by their two-way travel time) and specific formation tops identified by the log analysis (identified by their measured depth); without this well-to-seismic tie, interpretations of seismic data are geological guesses; with a well-calibrated synthetic seismogram, every reflection event on the seismic section can be associated with a geological formation, allowing the interpreter to map formation tops, identify unconformities, correlate between wells, and understand the geological significance of the amplitude and polarity of each reflection; the quality of the synthetic seismogram — and therefore the reliability of the well-to-seismic tie — depends on the quality of the input sonic and density logs (which must be corrected for borehole effects, cycle-skipping in the sonic measurement, and mud invasion), the accuracy of the wavelet estimate, and the accuracy of the depth-to-time conversion that aligns the log-derived synthetic with the seismic data in the two-way travel time domain.

Key Takeaways

  • The wavelet used in synthetic seismogram construction is the single most influential factor in determining how well the synthetic matches the real seismic data — the seismic wavelet (the shape of the seismic pulse recorded by geophones and embedded in the seismic data) determines the bandwidth, the phase, and the amplitude envelope of every reflection in the dataset; if the wavelet used to construct the synthetic has the wrong phase (zero-phase versus minimum-phase wavelet assumptions), the peaks and troughs of the synthetic will be systematically shifted relative to the seismic data, creating a poor tie that appears to show the well is in the wrong location when the issue is actually wavelet phase error; extracting the correct wavelet from the seismic data (using statistical wavelet extraction from the seismic traces near the well, or deterministic wavelet extraction using the known reflectivity from the sonic and density logs combined with the seismic trace) is a specialist processing step that requires both technical skill and geological understanding of what a good well-to-seismic tie should look like; a well-calibrated wavelet that produces a synthetic seismogram matching the real seismic data at the well gives the interpreter confidence that the same wavelet, applied in amplitude inversion or AVO analysis, will produce reliable lithology and fluid predictions across the dataset.
  • Synthetic seismograms reveal the "tuning thickness" at which thin beds become individually resolvable in seismic data — when a thin formation (a sand layer, a coal seam, a carbonate reef) is thinner than approximately one-quarter of the dominant seismic wavelength, the reflections from its top and base constructively or destructively interfere, creating a composite reflection whose amplitude and shape are controlled by the thickness rather than the impedance contrast; the synthetic seismogram for a well with a thin sand shows this tuning behavior directly — the computed reflection coefficient series has two reflections (top and base of sand), but the synthetic trace shows only a single, broadened reflection event if the sand is below tuning thickness; comparing the synthetic to the actual seismic trace at the well confirms whether the thin bed is resolved or is in the tuning regime; this comparison is essential for correctly interpreting amplitude anomalies on seismic data — a "bright spot" at the location of a thin gas sand may be a tuning reflection rather than a direct impedance effect, leading to incorrect quantitative interpretation of porosity or fluid content if the tuning behavior is not recognized and corrected in the analysis.
  • Check-shot surveys and vertical seismic profiles (VSPs) provide the depth-to-time conversion needed to align synthetic seismograms correctly with surface seismic data — sonic logs measure the interval transit time of acoustic waves in the formation, and integrating the transit time from surface to each depth provides the one-way travel time to each formation (the "time-depth curve" or "T-D curve") that converts the depth domain of the log data to the two-way travel time domain of the seismic data; however, sonic logs measure the formation velocity at very high frequency (tens of kilohertz) and at very small spatial scale (the wavelength of the sonic pulse in the mud is centimeters), while seismic waves travel at much lower frequency (10-100 Hz) and much larger scale; the velocity seen by the sonic may be slightly different from the velocity seen by the seismic wave (because of frequency dispersion and because the seismic wave averages over larger volumes than the sonic log borehole), causing the integrated sonic travel time to diverge from the actual seismic travel time at depth; check-shot surveys (firing a seismic source at surface and recording the direct arrival at downhole receivers at known depths) provide direct measurement of the seismic travel time to specific depth points, allowing calibration of the T-D curve derived from sonic log integration to the actual seismic velocity and ensuring that the synthetic seismogram is correctly positioned in time relative to the real seismic data.
  • Polarity conventions in synthetic seismograms must match the seismic data or the tie will be incorrect in a systematic way that can cause formation misidentification — different seismic acquisition, processing, and display conventions can result in seismic data that is displayed in normal polarity (where a peak corresponds to a positive reflection coefficient — an increase in impedance going downward) or reverse polarity (where a trough corresponds to a positive reflection coefficient); additionally, the definition of positive reflection coefficient (hard kick or soft kick) varies between companies, creating further potential for confusion; if the polarity assumed in the synthetic seismogram construction (or in the wavelet estimation) does not match the actual polarity of the seismic data, every peak in the synthetic will correspond to a trough in the real seismic and vice versa — a systematic polarity reversal that causes the interpreter to correlate the wrong seismic events to the formation tops, potentially mapping reflections that are 10-30 milliseconds shallower or deeper than the true position of the formation top; verifying polarity convention by checking whether known hard reflectors (limestone over shale, for example) appear as peaks or troughs on the seismic is a mandatory quality control step before relying on any synthetic seismogram tie for exploration or development decisions.
  • Synthetic seismograms are the bridge that connects reservoir models to seismic forward modeling for time-lapse (4D) feasibility studies — before investing in a repeat seismic survey to monitor reservoir fluid changes during production (4D seismic), operators use synthetic seismograms to evaluate whether the expected fluid substitution effects (replacing oil with water or gas with water as the reservoir is produced) will produce seismic amplitude changes large enough to be detectable above the seismic noise level; this 4D feasibility study uses the Gassmann fluid substitution equations to calculate the change in acoustic impedance caused by the expected saturation change, converts the impedance change to a change in reflection coefficient (and therefore seismic amplitude), and uses the seismic signal-to-noise ratio of the baseline survey to assess whether the predicted amplitude change exceeds the noise level by a sufficient margin to be reliably detected; the synthetic seismogram thus becomes a prediction tool for the future seismic response, not just a historical calibration tool for the existing well tie, allowing the 4D survey design (repeat acquisition timing, source and receiver geometry, processing requirements) to be optimized to detect the specific fluid changes expected in the specific reservoir interval being monitored.

Fast Facts

The first synthetic seismogram was computed in the 1950s by Wuenschel (1960) using a reflection coefficient series derived from well log impedance and a simple boxcar (rectangular) wavelet. It looked remarkably like the real seismic data at the well, confirming for the first time that seismic reflections were indeed generated by impedance contrasts at formation boundaries — something that had been assumed but not directly demonstrated before. Within a decade, synthetic seismograms had become standard practice in seismic interpretation, enabling the systematic correlation of seismic data to well data that transformed reflection seismology from an art of ambiguous pattern recognition to a quantitative geological mapping science.

What Is a Synthetic Seismogram?

A synthetic seismogram is the translation between two completely different measurement languages: well logs (which measure formation properties at each depth) and seismic data (which measures acoustic reflections in two-way travel time). Without this translation, interpreting seismic data is like reading a book in a language you don't speak — you can see the patterns, but you can't connect them to specific geological meaning. The synthetic seismogram is the Rosetta Stone: it shows which seismic reflection event corresponds to which formation top in the well, giving every reflection in the dataset a geological identity. With that identity established at the well, the interpreter can trace the reflection across the seismic section and map the formation everywhere the seismic data covers — turning subsurface geology into a mappable, three-dimensional picture rather than a collection of colored wiggles.

Synthetic seismograms are also called synthetics, seismic synthetics, or computed seismograms. Related terms include acoustic impedance (the log-derived product that generates the reflection coefficient series), reflection coefficient (the amplitude of each reflection in the synthetic), wavelet (the seismic pulse convolved with the reflection series to make the synthetic), well-to-seismic tie (the calibration exercise that the synthetic enables), check shot (the direct seismic travel time measurement that calibrates the sonic log), VSP (vertical seismic profile, a high-resolution alternative for well-seismic tie), polarity (the display convention that must match between synthetic and real data), and seismic inversion (the advanced analysis technique that synthetic seismograms calibrate).

Why the Quality of the Synthetic Seismogram Determines the Quality of Everything Built on Top of It

Every seismic interpretation that relies on correlating seismic reflections to geological formation tops is built on a synthetic seismogram tie at one or more wells. Get the tie right — correct wavelet, correct polarity, properly calibrated T-D curve, well-matched synthetic to real seismic — and the geological interpretation is anchored to ground truth. Get it wrong and the interpreter is mapping a formation that's actually the reflection from 20 milliseconds (and hundreds of meters) deeper, correlating amplitude anomalies to the wrong lithology, and building a reservoir model on a foundation that's systematically offset from reality. The synthetic seismogram is one of the most consequential computations in exploration and development geology, and yet it's routine enough that it's sometimes rushed, assumed to be "good enough," or not quality-controlled rigorously enough. The wells drilled on structural leads that turned out to be in the wrong structural position, the amplitude anomalies mapped as gas sands that turned out to be tuning effects — many of them trace back to a well-to-seismic tie that was accepted rather than validated. The synthetic seismogram deserves the scrutiny of any measurement that all subsequent work depends on.

Browse the Oil and Gas Glossary