The stochastic modeling approach can result in a large number of realizations that must be ranked so that fewer numbers can be used in more detailed flow simulation to quantify the uncertainties in reservoir performance. The objective of this paper is to investigate the use of tracer production data for the purpose of ranking geostatistical realizations. Here we introduce ranking of geostatistical models based on the first moment of the tracer production data. In this study we have used single-phase tracer simulation to rank a large number of detailed geostatistical realizations. The results of the ranking based on single-phase tracer simulations are compared with rankings based on increasingly more accurate two-phase flow simulations of a waterflood recovery process. In addition to the analysis of the integrated tracer breakthrough curve for ranking, a new tracer test used for estimating reservoir heterogeneities, called Vertical Tracer Profiling (VTP), is used to rank the generated geostatistical models. This interwell tracer test is performed by injecting and/or collecting tracer samples at different locations along the wellbore Using the more detailed VTP ranking, we investigated the differences between the tracer ranking and waterflood simulation ranking.

The results indicate that there is no correlation between the ranking based on tracer breakthrough time or peak tracer concentration and ranking based on cumulative oil recovery. Ranking based on the tracer's mean swept pore volume, first moment of the tracer production data, shows the highest degree of correlation with ranking based on cumulative oil recovery from waterflood simulations.

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