Rock-slope failure following Late Pleistocene deglaciation on tectonically stable mountainous terrain

Abstract

The ages of 31 postglacial rock-slope failures (RSFs) in Scotland and NW Ireland, derived from 89 cosmogenic isotope exposure ages, are employed to analyse the temporal pattern of failure and its relationship to the timing of deglaciation, rates of glacio-isostatic crustal uplift and periods of rapid climate change. RSF ages span almost the whole period since ice-sheet retreat, from 18.2 ± 1.2 ka to 1.7 ± 0.2 ka, or from 17.1 ± 1.0 ka to 1.5 ± 0.1 ka, depending on the production rate used in 10Be age calculation, but catastrophic failure of rock slopes was ∼4.6 times more frequent prior to ∼11.7 ka than during the Holocene. 95% of dated RSFs at sites deglaciated during retreat of the last ice sheet occurred within ∼5400 years after deglaciation, with peak RSF activity 1600–1700 years after deglaciation. This time lag is inferred to represent (1) stress release initiated by deglacial unloading, leading to (2) time-dependent rock mass strength degradation through progressive failure plane development, and ultimately (3) to either spontaneous kinematic release or failure triggered by some extrinsic mechanism. By contrast, 11 dated RSFs at sites reoccupied by glacier ice during the Younger Dryas Stade (YDS) of ∼12.9–11.7 ka exhibit no clear temporal pattern, suggesting that glacial reoccupance during the YDS was ineffective in preconditioning a renewed cycle of enhanced RSF activity. Comparison of timing of individual RSFs with that of deglaciation and rapid warming events at ∼14.7 ka and ∼11.7 ka suggests that glacial debuttressing, enhanced joint water pressures during deglaciation and thaw of permafrost ice in rock joints could have triggered failure in only a small number of cases. Conversely, the timing of maximum RSF activity following ice-sheet deglaciation corresponds broadly with maximum rates of glacio-isostatic crustal uplift, suggesting that the two are linked by enhanced seismic activity. A seismic failure trigger is consistent with full-slope failure at all sites where failure planes are clearly defined. Our results indicate that numerous RSFs must have occurred in areas that were reoccupied by glacier ice during the YDS, but have not been identified because runout debris was removed by YDS glaciers. More generally, they provide the first reliably-dated body of evidence to support the view that retreat of the last ice sheets in tectonically-stable mountainous terrain initiated a period of enhanced rock slope failure due to deglacial unloading and probably Lateglacial seismic activity, implying that most undated RSFs in such areas are probably of Lateglacial or very early Holocene age. They also demonstrate, however, that a low frequency of RSF activity extended throughout the Holocene.