A Core Logging, Machine Learning and Geostatistical Modeling Interactive Approach for Subsurface Imaging of Lenticular Geobodies in a Clastic Depositional System, SE Pakistan

Abstract

Facies models are essential tools for imaging subsurface geobodies and for reducing exploration and development risks efficiently. The Lower Goru Formation is one of the principal formations in the Lower Indus Basin, Pakistan. Its substantial hydrocarbon potential is unexplored, as most of the wells within the Sawan gas field are facing relatively low production yields. This study aimed to delineate subsurface geobodies by developing a facies model to study the depositional processes and facies distributions that have been neglected previously. The interactive approaches used in this research consisted of petrophysical, mineral composition, well-log facies, and horizon attribute analyses, as well as an unsupervised vector quantizer artificial neural network (UVQ–ANN) and sequential indicator simulation (SIS) modeling. A series of E–W-oriented lenticular geobodies were delineated. These geobodies had variable thicknesses, and they pinch out to the NW and prograde to the NE. The results of the SIS, UVQ–ANN, petrographic analysis, and attribute analysis show a fluvial fan-delta sedimentary system. The reservoir sands were deposited in distributary mouth bars and deltaic channels in proximal delta front settings. The coarse- to very fine-grained reservoir sands prograde toward the NE. Thinly laminated beds of fine-grained black shales and lime muddy siltstones were deposited under low-energy conditions in mid-shelf marine settings. The adopted methodology for the generated facies model can be extended to different basins within Pakistan with the same geological settings, and it can be used for prospect evaluation, future drilling, and development plans within the Sawan gas field in the Lower Indus Basin.