Resumen de Multivariate statistical analysis for groundwater mixing ratios around underground storage caverns in Korea
Kyung-Seok Ko, Jeonghoon Lee, Kang-Kun Lee, Ho-Wan Chang
Understanding groundwater flow and chemistry is crucial to operate underground storage caverns because the groundwater flow and chemistry is mainly affected by the cavern operating conditions. Both classical (for example, graphical Durov diagram) and multivariate statistical methods [principal component analysis, (PCA) and cluster analysis, (CA)] were used to trace an effect of the injection of a sodium hypochlorite (NaOCl) disinfectant solution into the water curtain on the chemical compositions of groundwater and to calculate groundwater mixing ratio around liquefied petroleum gas underground storage caverns. Based on both hydrochemical analysis and statistical analysis, the groundwater samples in the study area were divided into four groups. Both CA and an end member mixing (EMM) calculation using the PCA results indicate that the chemical compositions of seepage water have been mostly influenced by waters from the water curtains. The propane seepage water is most influenced by the monitoring wells in the water curtain, with the largest mixing portion (34.2%), which indicates that a good performance of the water curtains in the propane area. The EMM model from the PCA results is used to calculate the mixing ratio and to identify the chemical evolution of groundwater evolution, providing clues to the groundwater flow and transport system around the underground storage caverns.
