“I don’t want to know how long it takes to get there – I want to know how deep it is!,” bellowed the grizzled driller to the naïve seismic interpreter. Prior to pre-stack depth migration becoming commonplace, this was all too easy a trap to fall into. Even these days, for non-salt projects, most seismic interpretations are done on time domain data. While pre-stack time migration has compensated for the lateral positioning error, we are still faced with making depth predictions from time data.
Post by Bruce Blake, Senior Geophysicist, Headwave
Time to depth conversion is business critical for drilling on time and on budget, for accurate reserve estimation and for understanding the uncertainty of various depth predictions. Standard time to depth conversion is fraught with errors and uncertainties typically of unspecified magnitude – and the process has been time consuming.
There are several accepted methods for making a depth conversion, usually dictated by available information as well as past practices. For example, an exploration manager may prefer the pseudo-average velocity method because of past bad experiences with an ill-constrained layer cake outcome. Or perhaps the prospect is in a relatively unexplored basin with few well penetrations to the zone of interest. In this case, using seismic velocities to predict target depths is required. These types of approach are tedious and typically result in a deterministic depth prediction.
Headwave has been developing a depth conversion module (T2D) to enable the interpreter to perform multiple conversions using various models and ranges of parameters to quantify the depth uncertainty. This uncertainty in depth is critical in two broad areas: prospective gross rock volume and depth to target. Modeled velocities can be determined from checkshot data, sonic logs, and seismic velocities (derived and calibrated within T2D). Multiple analytic relationships can be compared using Slotnick, Faust, and other relationships. Well marker mis-ties can be determined and applied. While these same features might be available through other software packages, only with Headwave can these multiple realizations be completed in real time.
Another advantage of Headwave is the ability to review, revise and even build from scratch seismic velocities using CDP gathers as input. Rarely does the interpreter have available the gathers once a processing project has been completed. With Headwave, one can add additional seismic velocity functions in an area of interest, revise velocity picks to be more geologic (perhaps the velocities were picked fast to bias against stacking multiples), and even perform horizon-based velocity picking at every 3D bin.
Multiple models plus probabilistic parameterization directly lead to quantified uncertainty. How big is the closure? Where is the culmination (or spill point)? How shallow might top of pressure come in? How deep might we have to plan the well to make certain we test the target? All these can be answered with T2D.
A supermajor oil company has commissioned Headwave to build modern, fast depth conversion software that can be used for better internal decision making, in equity determinations and in data rooms. It is intended to be fast and robust.