Testing rebound hardness for estimating rock properties from core and wireline logs in mudrocks

Abstract

Rebound hardness (RHN) has become a widely applied rock mechanical parameter in the petroleum industry due to economic and convenient testing procedures. However, the RHN data can be under-utilized when lacking detailed integration with other rock properties. Targeting the unconventional “Mississippian Limestone”/STACK play in north-central Oklahoma, USA, and outcrops of the Vaca Muerta Formation in Argentina, this study aims to test the value of RHN in predicting rock properties. RHN data from the “Mississippian Limestone”/STACK cores show correlative trends with mineralogy and porosity. All the correlations show clusters by facies groups with overlaps being present among different clusters. Within these correlations, mineralogy and porosity show variable significance levels in affecting RHN among different facies groups. Leverage analysis suggests that bulk clay content and porosity exhibits the most significant control on RHN for the MISS/STACK data, with variabilities being present in different facies groups. These partitioning patterns of data by facies groups imply that facies variability affects the statistical pattern and that RHN can assist in rock typing, and hence, sample selection for detailed laboratory analyses. Forward regression analysis reveals that the confidence level of predicting porosity and sonic velocity can be enhanced using RHN. In addition to the correlative trends between RHN and rock properties, results from forward regression analysis indicate that RHN can help estimate these properties in a faster, cheaper, and non-destructive way relative to conventional laboratory analyses. Correlative trends are also observed in Vaca Muerta data, suggesting the value of RHN in characterizing similar types of mixed carbonate-siliciclastic reservoirs. • Rebound hardness (RHN) can help estimate bulk clay content, porosity, sonic velocity, and dynamic elastic parameters. • RHN may help estimate static rock mechanical properties, but more data are needed for further investigation. • RHN cluster by facies and thus can help design an integrated rock typing and sample selection scheme. • Multivariate methods indicate RHN is affected by multiple factors and can help apply RHN to predict rock properties. • A facies-based, integrated approach with a well-designed testing protocol is critical for using RHN in reservoir analysis.