Comparison of different approaches and development of improved formulas for estimating GSI

The Geological Strength Index (GSI) is an important parameter for estimating rock mass properties and can be estimated either through direct observations or by using a quantitative GSI chart or equations. This study aims to discuss the applicability of the RMR₈₉ method and the latest version of GSI|₂₀₁₃ by applying them to a case study and to the data published in the earlier papers. The relevant rock mass properties, including blockiness and discontinuity conditions, of the walls of 15 exploratory adits with a total length longer than 1100 m in a hydropower project in China have been assessed to provide data for the analysis. In addition, a repeat analysis of the earlier published data has been conducted to validate the results derived from the case study. The results show that the GSI values calculated with the RMR₈₉ are generally larger than those obtained from the qualitative GSI chart, with an average magnitude of about five. The GSI values predicted from the values of JCond89 and RQD are consistent with those derived through the qualitative observational method for a medium-quality rock mass but are relatively smaller or larger for a poor-quality (i.e., GSI<45) or a high-quality (i.e. GSI>65) rock mass, respectively. By comparing the results of the RMR₈₉ method and the qualitative GSI chart method and analyzing the classification factors considered by the two methods, two new equations based on the RMR₈₉ system have been developed for estimating the GSI. Their application to the case study and the published data prove that these two equations have higher predictive accuracy than the original RMR₈₉ method and the latest version of GSI|₂₀₁₃. It is expected that the acquisition of more engineering examples will enable the two formulas to be further verified and improved.