Heterogeneities and Rock types Characteristics in a tide-dominated Estuarine Sandstone Reservoir Deposit, Coastal Swamp Depobelt, Niger Delta basin, Nigeria

Abstract

An oil-down-to and oil-up-to sandstone reservoir at the crest of a rollover anticline was characterised to document the petrophysical properties of the different sedimentary facies and to achieve optimum petroleum production. Characterisation methods include logs stacking pattern analysis, lithofacies analysis of cores, special core analysis data interpretations, evaluation/quantification of heterogeneity, and Winland r35 and hydraulic unit rock-typing. The depth zone of interest was interpreted as transgressive systems tract (TST) due to the backstepping logs' pattern. Twelve lithofacies described the cores, while the lithofacies characteristics/associations resulted in the identification of depositional environments that include fluvial channels and inner, outer and middle estuaries and transgressive and regressive offshore, transgressive offshore-transition and shoreface environments. Porosity and permeability values of all the sandstone facies indicate high reservoir quality due to moderated to well-sorted grains. Winland's R35 rock-typing categorised 75% of the rock to be mega-porous, but the twelve lithofacies, the Dykstra-Parson’s coefficient of 0.84 of core permeability distribution and the normal probability plot of irreducible water saturation indicate a high level of heterogeneities, with potential for compartmentalization and consequent poor sweep efficiency. Hydraulic unit rock-typing categorised the reservoir into eight hydraulic units (HUs) based on flow zone indicators (FZIs). HU rock-types 1 to 6 are reservoir hydraulic flow units (HFU) with FZI that range from good (0.6 µm) to best (1.52 µm). The plot of hydraulic units against depth indicates fluctuating tidal current speed and energy as the dominant controls on reservoir quality. The HU 7 and 8 with FZI 0.24 and 0.03 µm, respectively, serve as baffles and compartmentalized the reservoir into three compartments as non-reservoirs. Perforating the depth interval with the best HFUs in each compartment is a beneficial option for optimum petroleum production. The multiple linear regression machine learning algorithm predicted the flow zone indicators of the same reservoir in a distant non-cored oil well.