Earthquakes trigger landslides in mountainous regions. Recent research suggests that the stability of hillslopes during and after a large earthquake is influenced by legacy effects of previous seismic activity. However, the shear strength and strain response of ductile hillslope materials to sequences of earthquake ground shaking of varying character is poorly constrained, inhibiting our ability to fully explain the nature of earthquake‐triggered landslides. We used geotechnical laboratory testing to simulate earthquake loading of hillslopes and to assess how different sequences of ground shaking influence hillslope stability prior to, during, and following an earthquake mainshock. Ground‐shaking events prior to a mainshock that do not result in high landslide strain accumulation can increase bulk density and interparticle friction. This strengthens a hillslope, reducing landslide displacement during subsequent seismicity. By implication, landscapes in different tectonic settings will likely demonstrate different short‐ and long‐term responses to single earthquakes due to differences in the magnitude, frequency, and sequencing of earthquakes.

Predictive models used to assess the magnitude of coseismic landslide strain accumulation in response to earthquake ground shaking typically consider slope-parallel ground accelerations only and ignore both the influence of coseismic slope-normal ground accelerations and the phase relationship between dynamic slope-normal and slope-parallel accelerations. We present results of a laboratory study designed to assess the significance of the phase offset between slope-normal and slope-parallel cyclic stresses on the generation of coseismic landslide displacements. Using a dynamic back-pressured shearbox that is capable of simulating variably phased slope-normal and slope-parallel dynamic loads, we subjected sediment samples to a range of dynamic loading scenarios indicative of earthquake-induced ground shaking. We detail the variations in strain accumulation observed when slope-normal and slope-parallel stresses occur independently and simultaneously, both in and out of phase, using a range of dynamic stress amplitudes. Our results show that the instantaneous phasing of dynamic stresses is critical in determining the amount of coseismic landslide displacement, which may vary by up to an order of magnitude based solely on wave-phasing effects. Instantaneous strain rate is an exponential function of the distance normal to the Mohr Coulomb failure envelope in plots of shear stress against normal effective stress. This distance is strongly controlled by the phase offset between dynamic normal and shear stresses. Our results demonstrate that conditions considered by conventional coseismic slope stability models can either overestimate or underestimate earthquake-induced landslide displacement by up to an order of magnitude. This has important implications for accurate assessment of coseismic landslide hazard.

No abstract

Citation for published item:ep¡ ulvedD FeF nd etleyD hFxF nd frinD wFtF nd unstllD xF @PHITA 9he e'et of dynmi loding on the sher strength of pyrolsti sh deposits nd implitions for lndslide hzrd X the se of udhuel sgnimriteD ghileF9D ingineering geologyFD PHS F ppF SRETIF Further information on publisher's website: Use policyThe full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes provided that:• a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.Please consult the full DRO policy for further details.