Determining representative elementary volume for hydraulic conductivity of fractured rock masses: Comparative analytical and numerical studies

In rock engineering, hydraulic properties are typically estimated by investigating and analyzing the spatial distribution and mechanical characteristics of fractures, which is supplemented by a surface geological survey and limited in situ hydrogeological tests. However, these approaches face challenges owing to considerable scale effects in fracture distribution and geometric parameters, as well as variability in hydraulic test results across different scales. To address these issues, herein, we develop a method to reliably evaluate the representative hydraulic conductivity of in situ fractured rock masses through quantifying the impacts of influencing factors. Using the Heshe hydrogeological test site as a case study, the research extends the crack tensor theory and compares the findings with numerical analyses based on a discrete fracture network. Results indicate that a consistent representative elementary volume with a size of 16 m was identified for the Heshe well test site, despite the approach used. Additionally, this study highlights the notable impacts of aperture distribution on hydraulic conductivity evaluation, underscoring the importance of site-specific investigations and detailed analysis. Investigation scale considerably influences the variability of hydraulic conductivity, with a positive relation between fracture aperture and trace length that must be carefully considered for accurate assessments. Finally, a comprehensive field assessment method for hydraulic characteristics of fractured rock masses is proposed. The analytical approach allows for a rapid preliminary assessment of representative elementary volume; however, the precise modeling of hydraulic conductivity and seepage direction requires further hydrogeological tests. Meanwhile, the numerical approach, though more time consuming, provides detailed and proper assessments. Overall, the assessment methodology developed in this study offers a feasible and robust approach for hydrogeological investigations in engineering applications.