Depth conversion: the map you hand over is not the map you made
Learning objectives
- Fit a V0 plus kt velocity function to an imported checkshot and apply it to a time map
- Explain how a high-velocity overburden body creates apparent structure in time
- Predict the direction in which a single-well velocity model biases a prospect
- Say what a second well improves and what it leaves untouched
Everything so far has been in time, and time is not a place. Nobody drills to 550 milliseconds, and the volume you are about to compute is in cubic metres of rock. The conversion needs a velocity, and the only measured velocities you own are the checkshots delivered with the wells.
In the workstation
Open the Map viewport in DEPTH. The workstation fits @@z = V_0 t/2 + k t^2/4@@, the integral of a @@V_0 + kt@@ interval velocity, to whichever checkshots you tick, and quotes the fitted @@V_0@@ and @@k@@ beside the map with the crest, spill and closure it implies. Try Ogbon-3 alone, then add Ogbon-5, and watch which numbers move.
The error has a direction
Ogbon-3 is the discovery well, drilled where discoveries are drilled: on the crest. If anything anomalous sits in the overburden above a crest, a reef, a channel fill, a gas chimney, the crestal well sits in it, and a velocity function fitted there carries that anomaly to every bin in the survey. A high-velocity body above the crest makes reflectors beneath it arrive early, so part of the relief on your time map is pull-up rather than geology; convert with a function fitted inside the body and you do something worse than fail to remove the false relief. You apply the elevated velocity everywhere, and the prospect comes out larger than the time map suggested. The reader who does every step of the arithmetic correctly ends up further from the truth than when they started.
Adding the flank well roughly halves the depth error at the wells, because the fit now averages two different overburdens. Watch the closure, though: it barely moves. The error was never in the average velocity; it is that velocity varies LATERALLY and a single @@V_0 + kt@@ function cannot represent variation of any kind. A second well fixes your depths and leaves your structure. Fixing structure needs a velocity model that varies across the map, which means seismic velocities, or many wells, or an interpretation of what the anomalous body actually is. That is the work depth conversion really asks for, and it is why the step is rarely as quick as it looks on a schedule.