A Lithology-lndependent Porosity Model Using Three Neutron Measurements: Part II—Applications and Error Analysis

Abstract

Abstract A lithology independent porosity model, N3LIP, using three neutron measurements, thermal and epithermal neutron porosity and thermal absorption cross section, was developed and presented at the 33rd SPWLA annual logging symposium (reference 1). This model considers any formation, regardless of its mineralogical composition, as two major regions. The first region is the pore space which is filled with fluid, while the second, the matrix, is the rest of the formation. N3LIP deals with the second region, the matrix, as a black box. The model calculates matrix diffusion length, which is the most lithology dependent parameter of a formation, without a need to know its lithological composition. This matrix diffusion length is calculated using two different techniques while iterating on porosity. The final solution, porosity, is deterrnined by analyzing the difference of the diffusion lengths. In Part I (reference 1), the methodology of N3LIP was described in detail and applied to some artificial logs generated using SNUPAR (reference 2) as a forward model. Here, the applications of this model to actual reservoirs are presented, and an error analysis of the calculated porosity is discussed. N3LIP has been used to calculate porosity in two different types of reservoirs, shaly sandstone and carbonates. The shaly sand reservoirs are Kuparuk in the North Slope and Long Beach in California, while the carbonate reservoir is Sycamore in the Permian Basin. Due to concerns of the environmental effects on the calculated neutron measurements, especially the borehole rugosity on the epithermal tool, the uncertainty in the calculated porosity using N3LIP in those reservoirs has been studied by applying uncertainties to the logs that have been used. The effect of these uncertainties are presented and discussed in detail.