Anisotropic dynamic and static geomechanical property correlations in shale formations

Abstract

Many correlations have been developed for converting dynamic and static elastic moduli for various rock types. The ability to make quick and simple estimates of static elastic moduli from more easily obtained dynamic values provides several advantages to the oil and gas industry. This effectively makes geomechanical modeling a more flexible process which can be used to further optimize drilling and completions operations and lower the overall cost of resource development. In particular, this research presents several correlations that convert dynamic Young’s modulus to values of static Young’s modulus for shale. The novel approach of this study incorporates a large variety of shale formation data and considers the directional nature of core sample measurements. By making comparisons from the data and identifying trends, analysis using several statistical tools has also produced a correlation that uses rock property information as input to estimate static Young’s modulus for shale. These correlations are later validated by performing multiple statistical tests and by comparing their results to actual values of static Young’s modulus for accuracy.

To better capture the impact of shale anisotropy, the relationships developed in this study use Young’s modulus measurements from both vertical and horizontal core samples – with respect to the rock bedding plane. Also, multivariate analysis, in the form of multiple linear regression (MLR), was performed to consider variables such as mineralogy and natural fracture density to increase static Young’s modulus estimation accuracy. Comparison of P-values, Q-Q plots, residuals vs. fitted plots, and R2 values for fitting the data resulting from regression analysis are shown to support the findings of this study. Also, matching the correlation results to actual static Young’s modulus measurements demonstrated these correlations can be used reliably.

Evaluation of the correlations found in this study prove that they can be used to determine static Young’s modulus using measurements of dynamic Young’s modulus with confidence. It is highly encouraged to utilize the multivariate correlation when possible as it provides a more accurate estimation of static Young’s modulus and further considers shale anisotropy. This is due to it using known rock property information as input. While this research uses a specific set of data from the literature, the inclusion of more, similar data from a variety of shale formations should be considered in future analysis. The utilization of more data, combined with machine learning methods, would likely improve the accuracy of this study’s correlations if they were to be incorporated into the analysis.