Developing a novel permeability prediction method for tight carbonate reservoirs using borehole electrical image logging

Predicting permeability accurately is crucial for effective hydrocarbon extraction, but the intricate pore structures of tight carbonates, resulting from sedimentation, diagenesis, and tectonic activity, present significant challenges. Based on borehole electrical image logging and fractal theory, we developed a method to calculate the fractal dimension of the porosity spectrum to characterise the complexity of the pore structure of the reservoir. Fractal features of the porosity spectra were studied and fractal parameters were calculated, such as the left ( D f_left), middle ( D f_middle), and right fractal dimension ( D f_right). A permeability prediction model was proposed based on fractal parameters by investigating the linear relationship between fractal parameters and core permeability. The results indicate that D f_left and permeability have a coefficient of determination (R2) of 0.78, whereas R2 between porosity and permeability is only 0.03. D f_middle and D f_right have little correlation with core permeability. The prediction results of the D f_left -based permeability model are in good agreement with the experimental data with Pearson product-moment correlation coefficient of 0.93 in the field applications. Our findings suggest that large pores primarily contribute to the permeability of tight carbonates since D f_left corresponds to the macroporous part of the porosity spectrum. This study enhances our understanding of the factors that influence permeability and provides a useful tool for predicting permeability in tight carbonate reservoirs.