The displacement of a given fluid by another in a porous media can occur either in miscible or immiscible conditions. An immiscible displacement happens, for instance, when water displaces oil in a reservoir. It can also occur when the displacing fluid is a gas, if the interfacial tension with the oil is high enough. These processes are usually described by the fractional flow theory, which assumes isothermal flow of two immiscible and incompressible fluids in an one-dimensional, homogeneous porous media; dissipative effects, such as capillary pressure, compressibility and thermal conductivity, are neglected. Miscible floods, on the other hand, can take place in gas-oil displacements when the interfacial tension approaches zero. Common dissipative effects are fingering and dispersion, the former related to mobility ratio magnitudes and the later related to diffusion and velocity contrast commonly caused by the presence of heterogeneities. In miscible displacements through porous media, dispersion is described as the result of diffusion, local velocity gradients, streamlines dimensions in heterogeneous regions, and mechanical mixing into pores. The present work discuss these concepts and relate them to data from an actual deep offshore field. Estimates of field dispersivity by history matching tracer production profiles are based on the analytical solution for the convection-diffusion equation. Gas tracers injected and produced from wells located in the field area were evaluated assuming miscible displacement. The estimated dispersivity allowed the evaluation of mixing/spreading zones. The solution for intermittent tracer injection was used to history match tracer production data. Results shows the scale dependence of the dispersivity, which depends upon the distance and degree of heterogeneity between wells. The field scale dispersivity obtained by history match was coherent with numerical dispersion observed in a field scale numerical model. They were also compared against literature data (laboratory and field scale) in a log-log plot denoting a good agreement of present data with literature.

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