Unconventional oil and gas development is frequently hindered by complex reservoir structures and unquantified rock mechanical properties. To optimize horizontal well trajectories and hydraulic fracturing designs—thereby expanding the stimulated reservoir volume, mitigating casing deformation risks, and ensuring efficient production—high-resolution rock mechanical models are essential. This study proposes an integrated workflow combining statistical regression and 3D prestack seismic inversion to accurately characterize the vertical and lateral variations of rock mechanical parameters in target zones.

Using core test results and well log data, quantitative empirical relationships were established between elastic attributes and rock mechanical properties. A 3D prestack seismic inversion was subsequently performed using integrated drilling and seismic datasets to extract precise elastic parameters, including compressive wave velocity, bulk density, Poisson's ratio, and Young's modulus.

The workflow was applied to tight glutenite reservoirs in the Bonan Sag, yielding three key findings: (1) Young's modulus demonstrated a strong correlation (e.g., R² > 0.75) with Uniaxial Compressive Strength (UCS) . In these low-porosity glutenites, rock mechanics are primarily governed by lithology and gravel content rather than burial depth, justifying a unified prediction model. (2) By integrating well logging and a time-depth velocity field, a depth-domain structural framework was established. Attribute extraction and property modeling generated a continuous 3D volume of rock mechanical parameters, bridging the spatial gaps inherent in discrete log and core measurements. (3) The resulting 3D mechanical volume provides critical input for hydraulic fracturing optimization. It enables precise profiling along horizontal wellbores, rational stage and cluster spacing, pumping parameter optimization, and improved fracture conductivity, thereby maximizing single-well productivity.

Integrating prestack seismic inversion with statistical regression offers a robust approach for 3D geomechanical characterization in heterogeneous tight glutenites. This methodology successfully bridges the gap between petrophysics and seismic geophysics, resolving key uncertainties in complex unconventional plays.

This study provides a practical and scalable framework for predicting 3D rock mechanical properties. The findings offer a reliable data foundation for defining mechanical boundaries and optimizing stimulation treatments in heterogeneous unconventional reservoirs.