An artificial neural network application for predicting the aspect ratio and permeability images of a heterogeneous carbonate aquifer in South Florida

Abstract Characterizing carbonate aquifers presents significant challenges, especially when it involves determining groundwater flow zones at the field scale. The combination of seismic data and well logs, along with a suitable processing approach, supports this effort. High-resolution reflection seismic data from the Port Mayaca aquifer were used, located in Martin County, South Florida. Standard well logs and Formation MicroImager (FMI) were applied to define the carbonate structure in the interwell region associated with dense, tight limestone layers. These FMI anomalies, combined with microresistivity logs, led to the existence of fractures. To enhance the interpretation, an artificial neural network (ANN) was applied, which predicted small-scale heterogeneities by feeding the seismic attributes into the ANN. In this approach, seismic traces are sourced from a seismic survey. The initial image produced by the ANN algorithm illustrated the pore space based on the aspect ratio log, which served as the target data for the neural network model. The Python library Keras provided the framework needed to develop these network models. The application of the Keras allowed us to create images of pore aspect ratio, Stoneley (ST) wave permeability, and intrinsic permeability. The interpretation and analysis of these images revealed the connectivity of the pore space between narrow and wide channels, as well as the flow paths linked to the permeable zones. The ST wave permeability image highlighted the major flow zones present in the inter-well region. The results of the data analysis demonstrated that a flow zone structure, defined by the aspect ratio, is part of a groundwater channel. Permeable zones were identified in the ST image representing the total flow created by the interconnected vugs and fractures. Intrinsic permeability images were generated to examine the aquifer’s pore structure, and the areas with low intrinsic permeability correlated with dense, fractured limestone layers.