Probabilistic Addition of Reserves - A New Approach

Abstract

Abstract For multi-field projects, the total project ultimate recovery (UR) at a given confidence level is traditionally derived by the arithmetic addition of the UR's of its constituent fields. However, this overstates the uncertainty range at the total project level. The total project UR, at a given confidence level, is more correctly derived by probabilistic addition (PA) of the field UR's. As Probabilistic Addition requires the definition of the dependencies between the added elements, it is important that UR dependencies between fields are correctly estimated and incorporated into the calculation of the total project UR. This paper introduces a new statistically rigorous method for estimating UR dependencies between fields and for Probabilistic Addition of the field UR's to project level. The UR dependencies between fields are estimated based on the field UR sensitivity to field parameters, as commonly summarised in the UR Tornado Diagram. This data is routinely generated when the Multiple Realisation Approach is used to derive reserves estimates in a field study. The method of Probabilistic Addition is based on the assumption of Multivariate Normality. The method is very efficient in capturing field dependencies and eliminates the need for Monte Carlo simulation. Additionally, the method allows for a sensitivity analysis of the generated project UR, to the estimated field UR correlations. In this paper, the application of the methodology to an actual project case is presented. The Probabilistic Addition of the North West Shelf gas-condensate fields producing into a shared onshore domestic gas and LNG plant is discussed. The North West Shelf project reserves at P90 confidence level, calculated using Probabilistic Addition, are significantly larger than the arithmetic sum of the P90 reserves of the constituent fields.