Descriptive statistics and probability distributions of volumetric parameters of a Nigerian heavy oil and bitumen deposit

The absence of geostatistical modeling of volumetric parameters of the long-discovered Nigerian heavy oil and bitumen deposits is responsible for the inconsistencies surrounding estimates of hydrocarbon-in-place contained therein. An exploratory data analysis (EDA) is a pre-cursor to such modeling. As part of EDA, this work presents the descriptive statistics and probability distributions of the volumetric parameters of a Nigerian heavy oil and bitumen deposit. Raw data from the existing works have been assembled into a database. Using basic principles, porosity have been computed, from the raw data, for several core samples retrieved from the two bituminous horizons in the deposit. The computed database has been partitioned into the two horizons, using depth-to-top and thickness data. Furthermore, this work has conducted detailed analyses and offers robust discussions on the descriptive statistics and probability distributions of the porosity, depth-to-top, and thickness databases. The statistics and distribution curves obtained are observed to exhibit good correlations with existing geologic, stratigraphic, and textural data. An hypothesis suggesting the two horizons belong to same geological population has been formulated and tested at field and well levels; with results affirming the hypothesis. The descriptive statistics and probability distributions obtained offer a significant understanding of the characteristics and features of the available data. In addition, the distributions now become prior information to which reservoir descriptions would be constrained, in the future conditional simulation stage of this work. The correlation of core data obtained here with the existing geologic, stratigraphic, and textural data would promote data integration in the characterization of this deposit.