The primary objective of this work is to leverage downhole temperature measurements to enhance inverse modeling of the subsurface, particularly to enable more accurate estimations of fluid volumes in place.
We examine nonisothermal closed reservoirs with impermeable boundaries, under the assumption of sufficiently high Péclet numbers where thermal diffusion is negligible. Considering an axisymmetric cylindrical domain, a dimensionless formulation is developed to simplify the analysis and to elucidate key parameters that govern the system’s overall behavior. An analytical expression for temperature under a pseudosteady pressure regime is derived using the method of characteristics. The results are verified by comparing them with a commercial numerical thermal simulator under varying degrees of thermal forcing at sandface conditions and further validated against wellbore temperature data. The application of the solution to reservoir parameter estimation is tested on two synthetic examples, one of which accounts for reservoir heterogeneity and the influence of an aquifer.
The key question we address is what additional insights does temperature analysis offer that pressure analysis alone cannot? In conventional pressure analysis, estimating stock-tank original oil in place (STOOIP) requires an assumed value for water saturation as input. This study leverages the full potential of temperature data—readily available and typically recorded alongside pressure at minimal cost—to directly estimate fluid saturations and reduce uncertainty in volumetric assessments.
From an even more compelling perspective, our solution significantly enhances the accuracy of reservoir volume estimates, in scenarios involving hydraulic support systems such as aquifers. We present a real field case study featuring the presence of an aquifer to demonstrate the effectiveness of our approach.
This abstract is taken from paper SPE 228263 by C. S. Guimaraes, Stanford University and Petrobras; M. S. Galvao, Petrobras; and R. N. Horne and D. M. Tartakovsky, Stanford University. The paper has been peer reviewed and is available as Open Access in SPE Journal on OnePetro.