It is a common experience among geoscientists that structural uncertainties are crucial, especially when development decisions have to be taken on the basis of a small number of wells and a 3D seismic interpretation. Changes in the approach used for time-to- depth conversion and, to a lesser degree for horizon picking, may significantly impact the value of the Gross-Rock Volume (GRV). The methods presented here assume that maps of uncertainty on the time pick and on the average velocity have been produced, and they solve the problem of evaluating the impact of these uncertainty maps on GRV uncertainties.

The methods proposed consist of generating "possible" error maps that are all within the range provided by uncertainty maps. A "possible" depth map is obtained by adding an error map to the reference case. A simulated GRV can then be computed between the simulated top of the reservoir, the base of the reservoir (deduced by adding a thickness map to the simulated top, a thickness uncertainty may have been accounted for) and the Hydrocarbon-Water Contact (HWC) (its depth may have been simulated). By iterating many times (several hundreds) histograms and expectation curves of the GRV can be derived.

Introduction

The industry shows a growing interest for managing seismic uncertainties in a reservoir as they can dramatically impact the Gross-Rock Volume (GRV) estimates, especially when dealing with gently dipping structures at field appraisal stage.

Methods to simulate errors are presented as well as the use of resulting error maps (depth or time and velocity error maps) to generate "possible" depth maps. These "possible" depth maps can then be used to compute the simulated GRV. These methods are fast, and can be used to generate a great number of realisations (several hundreds) in a few hours on a workstation. Such an approach allows statistical analysis using histograms and expectation curves to derive key values such as 15%, 50% (median) and 85% quantiles.

This content is only available via PDF.
You can access this article if you purchase or spend a download.