Reservoir Characterization of The Cretaceous Sandstones of Sembar Formation, Lower Indus Basin, Pakistan

Abstract

The Lower Cretaceous sandstones of the Sembar Formation in the Lower Indus Basin, Pakistan, exhibit good hydrocarbon reservoir potential. However, a detailed understanding of the formation’s characteristics has been limited mainly due to insufficient critical data analysis. While studies have identified the shale gas potential of the Sembar Formation, no extensive research has been conducted to evaluate the reservoir potential of thick packages of sandstones encountered within it. This study was intended to fill this research gap for the first time by a comprehensive analysis of these sandstones. This study was important because it comprehensively integrated primary and secondary data analysis using classical and cutting-edge techniques, including machine learning and fluid inclusion studies, to reveal a range of reservoir properties from microscopic to seismic scale. Data available for this study included well cuttings, well logs, and seismic reflection data. The X-ray diffraction (XRD) analysis, petrography, scanning electron microscopy (SEM) analysis, X-ray fluorescence (XRF) analysis, inductively coupled plasma mass spectrometry (ICP-MS), inductively coupled plasma optical emission spectroscopy (ICP-OES) analyses, petrophysical modeling, rock physics analysis, seismic structural analysis, seismic inversion, multi-attribute transformation, artificial neural network (ANN) modeling and fluid inclusion studies were integrated for the assessment of reservoir quality. Five volume-based seismic attributes were sequentially calculated through forward stepwise regression. These seismic attributes were cross-validated when a Probabilistic Neural Network (PNN) was trained in a non-linear mode integrated with multi-attribute transformation, where correlation (r2) was improved from 72% to 88% between seismic attributes and porosity log. Viable prospective pay zones (up to a maximum of ~200m continuous thick sandstone package with good reservoir properties) were identified through petrophysics, rock physics, and seismic data coupled with ANN analyses. Petrographic analysis revealed that the sandstones were primarily quartz arenite, with fine to medium, subrounded to subangular grains and moderate sorting. Key diagenetic processes included quartz overgrowths and the presence of clay minerals (constituting less than 4 weight% of each sample), primarily illite-smectite and illite-mica with minor chlorite and kaolinite. Intergranular porosity was common due to the dissolution of feldspar, chert, and rock fragments, but intragranular porosity was sparse. Since sandstones of the Sembar Formation were encountered at a depth of around ~4000m, compaction and overburden pressure at such a great depth reduced its permeability. The PNN-derived porosity distribution was geologically more accurate and realistic compared with linear regression approaches and in excellent match with the porosity log, confirming the validity of the porosity model. Fluid inclusion studies have shown gas and condensate spectra in the studied wells and highlighted the temperature conditions and fluid sources associated with the development of quartz overgrowth. Fluid inclusion petrography confirmed the presence of yellow-fluorescent, moderate to upper-moderate gravity oil inclusions and non-fluorescent methane-rich gas inclusions in detrital quartz and along quartz dust rims, respectively. These findings challenged previous views of the formation as solely unconventional and offer valuable insights for geoscientists to explore both its conventional (newly recognized sandstone reservoir) and unconventional (shale gas) potential.