TWT: Two-Way Traveltime, Zero-Offset Normal Incidence, and Time-to-Depth Conversion

TWT, two-way traveltime, is the total elapsed time for a seismic wave to travel down from its source at the surface to a subsurface reflector and back up to a receiver, and it is the fundamental measurement axis of all reflection seismology. Because a seismic survey records arrivals as a function of time rather than depth, every reflection event on a stacked section, every interpreted horizon, and every velocity pick is initially expressed in TWT, conventionally in seconds or milliseconds. The reference case is minimum two-way traveltime, the time of a normal-incidence wave with zero offset, meaning the idealized situation where source and receiver coincide and the ray travels straight down to the reflector and straight back along the same path with no horizontal separation. Real acquisition never places source and receiver at the same point, so field traces are recorded at finite offset and show longer traveltimes that increase with offset along a hyperbolic moveout curve; the processing step of normal-moveout correction removes that offset-dependent stretch to recover the equivalent zero-offset TWT before stacking. The link between TWT and the depth an interpreter actually wants is velocity, through the basic relationship that depth equals one half of the two-way time multiplied by the average velocity to the reflector, the factor of one half appearing precisely because the recorded time is for the round trip rather than a one-way path. Across the Western Canadian Sedimentary Basin this conversion is the daily bread of exploration: a Montney reflector might sit at 1.55 seconds TWT, a Leduc reef at 2.1 seconds, and turning those times into the 2,500 m or 3,200 m (8,200 ft or 10,500 ft) depths that a drilling engineer can use requires a carefully built velocity model from well checkshots, sonic logs, and stacking velocities. Errors in that velocity field translate directly into depth errors at the bit, so the gap between a clean TWT pick and a reliable depth prediction is where much interpretation risk lives. TWT also governs vertical resolution and tuning, controls the time-migration domain in which most WCSB volumes are interpreted, and underlies synthetic seismograms that tie a well's measured depths to the seismic time section. In short, two-way traveltime is the native currency of seismic data, and converting it faithfully to depth is one of the central tasks of subsurface mapping.

Key Takeaways

  • Round-Trip Time Measurement: TWT is the time for energy to travel from surface source down to a reflector and back to a surface receiver. Because the path is there and back, any depth conversion divides the time by two before multiplying by velocity. A Leduc reef at 2.1 seconds TWT under central Alberta corresponds to a one-way time of 1.05 seconds, the value that maps to true vertical depth.
  • Zero-Offset Normal Incidence Reference: Minimum two-way traveltime assumes a coincident source and receiver with the ray going straight down and back at normal incidence. Field data acquired at real offsets must be normal-moveout corrected to this zero-offset datum before stacking, so that a stacked WCSB section approximates what a single coincident source-receiver pair would have recorded.
  • Depth Equals Half-Time Times Velocity: The working formula is depth = (TWT / 2) x average velocity. A Montney target at 1.55 seconds TWT with an average velocity near 3,250 m/s converts to roughly 2,520 m (about 8,270 ft). The accuracy of that number rests entirely on the velocity model, which is why checkshots and sonic calibration dominate depth-conversion effort.
  • Velocity Model Controls The Risk: Time picks are usually repeatable, but the velocity field that converts them to depth carries the real uncertainty. Lateral velocity changes across Cretaceous channels or Devonian carbonate buildups can shift a predicted Duvernay landing point by tens of metres, enough to miss a thin target, so multi-well velocity control is essential before committing a CAD 9 million horizontal pad.
  • Foundation Of Interpretation And Ties: TWT is the axis on which horizons are mapped, isochrons measured, tuning and vertical resolution assessed, and synthetic seismograms built. The well-to-seismic tie matches log-derived synthetic reflections in time to the recorded section, anchoring every depth-converted map to hard well data and exposing any miscalibration in the velocity model.

Normal Moveout And The Path To Zero Offset

Field traces record reflections at a range of source-receiver offsets, and a single reflector therefore appears not as a flat event but as a hyperbola whose curvature is set by velocity and offset. Normal-moveout correction stretches each trace to remove this offset-dependent delay, collapsing the hyperbola toward the zero-offset TWT so that traces can be summed coherently in the stack. The semblance velocity analysis that drives this correction outputs coherency as a function of trial velocity against zero-offset time, and the picked stacking velocities become a first-pass velocity field. On WCSB land data, getting the moveout right at long offsets is what sharpens deep Montney and Duvernay events without the residual stretch that would otherwise smear the wavelet.

Time-To-Depth Conversion And Its Pitfalls

Converting a TWT map to a depth map is rarely the simple multiplication the formula suggests. Stacking velocities are root-mean-square quantities that must be converted to interval and then average velocities, and they are biased by anisotropy, dip, and lateral heterogeneity. Best practice in the basin ties the time-domain interpretation to checkshot and vertical seismic profile velocities at control wells, builds a layer-cake or gridded velocity model honouring those ties, and applies it to the picked horizons. A common pitfall is using uncalibrated stacking velocities directly, which can place a Nisku or Leduc horizon tens of metres off true depth and lead to a mis-landed well or a dry penetration above or below the target.

Fast Facts

The factor of one half in the depth equation is so fundamental that forgetting it is one of the classic blunders in seismic interpretation, instantly doubling every predicted depth. Equally telling is how little absolute time is involved: a deep 3,500 m (11,480 ft) WCSB Devonian target sits at only about 2.2 seconds of two-way time, and the entire sedimentary column from surface to Precambrian basement across much of Alberta returns in under three seconds. Seismic interpreters thus spend their careers reasoning in milliseconds, where a 10-millisecond pick error can equate to 15 to 20 m of depth uncertainty at reservoir level.

TWT is woven through the core vocabulary of reflection seismology. It measures the round trip of a downgoing spherical wave launched by the source, and the zero-offset assumption invokes the plane wave idealization used in stacking. The normal moveout correction is what reduces finite-offset arrivals to the zero-offset TWT, while a checkshot survey supplies the direct time-depth pairs that calibrate the velocity model used to convert two-way time into the depths that drive WCSB drilling decisions.

Real-World WCSB Scenario: Time-To-Depth On A Duvernay Lease Near Kaybob

An operator mapping a Duvernay landing zone near Kaybob, Alberta, picks the target reflector at 2.34 seconds TWT on a reprocessed 3D volume. Using uncalibrated stacking velocities the first depth estimate places the landing point at 3,290 m (about 10,790 ft), but a CAD 120,000 checkshot and sonic-log calibration program across three offset wells reveals the stacking velocities ran fast, and the true average velocity puts the reflector closer to 3,340 m (10,960 ft), a 50 m correction. For a 35 m thick Duvernay window steered along a single bench, that difference is the gap between staying in zone and drilling out the top.

With the recalibrated velocity model the geosteering plan lands the lateral inside the organic-rich interval, and the CAD 11 million two-well pad delivers liquids-rich gas rates in line with type-curve expectations. The episode shows that the value of a TWT map is only as good as the velocity field that turns time into depth.