Fluid Prediction in Near-Field Prospects Using PVT Modeling

Caillet, Gerard, Lucy Burns, Francois Montel, Total, Pau, France

This paper presents a methodology for a better prediction of the fluids in near-fields prospects, to reduce the uncertainties linked to the nature of the fluids. A well developed understanding of the distribution of the fluids in a basin can be approached with a reason­able confidence by PVT modelling.

The methodology consists first in a review of the different fluid types to get a synthetic description of the fluid distribution, including composition and property gradients (Saturation pressure, Rs, density). The mechanisms responsible for compositional grading give valuable information about the conditions of trapping.

Then, a reference fluid is selected, corresponding to a typical “source” fluid. A unique equation of state, calibrated to this reference fluid, is used to represent and compare all the fluids of a same area.

When the fluid types are not directly explained by the pressure and temperature condi­tions in the reservoirs, they can be derived of mixtures, either fluids from different source­rocks, or fluids from same source-rocks at different maturity levels.

In a case study the fluids were classified into three groups: 1) under-saturated light oils, the “source” type fluid; 2) gas condensate, a mixture of the “source” fluid and dry gas; 3) gas and oil at equilibrium, a mixture of the “source” fluid and wet gas. The dry gas comes from source-rocks at a high level of maturity. The wet gas is thermodynamically interpreted as the result of stripping.

With integration of the PVT and geological approaches, it becomes possible to accurate­ly predict the fluid types in the prospects of the studied area.