Forensic rockfall scar analysis : development of a mechanically correct model of rockfall failure.

The mechanical controls on small (< 10 m3), individual rockfall in jointed rock masses are not well constrained. We use forensic analysis of rockfall detachment surfaces (scars) which display fractured surfaces broken through intact rock, termed rock bridges as well as pre-existing discontinuities, to understand failure mechanisms. The relative significance of intact rock fracture versus release along pre-existing surfaces in stability has not been thoroughly investigated using field data. The relative role of each of these components determines where weakening, is important in controlling the nature and timing of rockfall. This is vital for defining mechanically accurate models of failure. An initial inventory of rockfall scars from coastal rock cliffs was captured using high-resolution gigapixel imaging and terrestrial laser scanning to determine these relationships. Fracture mapping, planar surface identification, and weathering classification were undertaken to identify similarities in the mechanical controls on failure. Preliminary analysis reveals that even small rockfall display a multi-stage failure history, whereby final failure occurs through fracture of a single unweathered rock-bridge. Intact rock breakage accounts for 22 ±12% of the full scar surface. The rock bridges are commonly clustered at the scar crest or base, while planar pre-existing joint surfaces dominate the scar center. This suggests that although cantilevered, most rockfalls in this inventory are more likely to fail through tension. We consider volumetric and lithologic controls on failure mode, and consider the wider potential of this approach.