AN RAI OF DATA: GENERALIZING THE DATA-DRIVEN ROCKFALL ACTIVITY INDEX (RAI) BASED ON LONG-TERM OBSERVATIONS OF WELL-CHARACTERIZED SLOPES

Abstract

In previous PacTrans research, the research team developed the Rockfall Activity Index system (RAI), a point cloud-derived, high- resolution, morphology-based approach for identifying, assessing, and mapping rockfall hazards at a high resolution across the entire surface of a rock slope. Continued monitoring at sites in both Alaska and Oregon has shown that rockfall activity is variable as a function of geology and rock properties. In this research, we used geologic characterization, analysis of change from light detection and ranging (lidar) differencing, and collection of 4,800 Schmidt Hammer measurements to constrain relationships between geologic structure and rockfall activity at a variety of rock slopes. Generally, lower Schmidt hammer rebound measurements were observed in areas with higher activity rates, suggesting that rebound (especially when corrected per ASTM standards) may be an effective proxy for rockfall activity or susceptibility. However, the large variability in these measurements, particularly between sites, suggests that these measurements are best applied on a site-specific basis. Additionally, we computed rockfall volumes to analyze mobility impacts by using an empirical relationship derived from the Rockfall Impacts to Mobility (RIM) database, demonstrating the importance of linking rockfall activity and hazard. Future work could look at expanding databases of rockfall impacts to mobility, collect more Schmidt hammer measurements, collect more epochs of rock slope digital terrain models, and further connect rock slope weathering and structure to the RAI analysis.