In this study, an innovative physical–chemical approach was integrated with site investigation techniques and real-time monitoring data for the past 3 years to understand the mechanisms of large-scale landslides. To investigate and understand mechanism of slope movements, various real-time monitoring data—including rainfall,electrical conductivity, borehole biaxial displacements, groundwater levels, and multitemporal satellite and UAV images—were collected from the site of the Babaoliao landslide. This research site is located at Dongxing Village,Zhongpu Township, Chiayi County, Taiwan. According to a geological investigation, downhole geophysical logging,and in situ monitoring data analysis, the key precursors to the landslide were identified by the occurrence of slope movement events and changes in groundwater chemistry. After assessing the aforementioned approach and actual sliding events, we could find that electrical conductivity to be the most sensitive factor as the landslide precursors. Furthermore,the groundwater composition of this research site were high correlation with electrical conductivity such as
the concentration of bicarbonate ions, sulfate ions, calcium ions, magnesium ions, and sodium ions.
Key Words: landslide precursor, electrical conductivity, cation and anion concentration