No abstract

No abstract

At 12.51 pm (NZST) on 22 February 2011 a shallow, magnitude MW 6.2 earthquake with an epicentre located just south of Christchurch, New Zealand, caused widespread devastation including building collapse, liquefaction and landslides. Throughout the Port Hills of Banks Peninsula on the southern fringes of Christchurch landslide and ground damage caused by the earthquake included rock-fall (both cliff collapse and boulder roll), incipient loess landslides, and retaining wall and fill failures. Four deaths from rock-fall occurred during the mainshock and one during an aftershock later in the afternoon of the 22nd. Hundreds of houses were damaged by rock-falls and landslide-induced ground cracking. Four distinct landslide or ground failure types have been recognised. Firstly, rocks fell from lava outcrops on the Port Hills and rolled and bounced over hundreds of metres damaging houses located on lower slopes and on valley floors. Secondly, over-steepened present-day and former sea-cliffs collapsed catastrophically. Houses were damaged by tension cracks on the slopes above the cliff faces and by debris inundation at the toe of the slopes. Thirdly, incipient movement of landslides in loess, ranging from a few millimetres up to 0.35 metres, occurred at several locations. Again houses were damaged by extension fissuring at the head of these features and compressional movement at the toe. The fourth mode of failure observed was retaining wall and fill failures, including shaking-induced settlement and fill displacement. These failures commonly affected both houses and roads. In the days and weeks immediately following the earthquake a major concern was how to manage the risks from another large aftershock or a long return period rainstorm, in the areas worst affected by landslides, should one occur. Each of the four identified landslide types required a different risk management strategy. The rock-fall and boulder roll hazard was managed by identifying buildings at risk and enforcing mandatory evacuation. In the days immediately following the earthquake this process was based on expert opinion. In the weeks after the earthquake this process was rapidly enhanced with empirical data to confirm the risk. The rock-falls associated with cliff collapse were managed by evacuating properties damaged by extensional ground cracking at the top of the cliffs, adjacent properties, and properties damaged by debris inundation at the toe of the cliffs. The incipient landslide hazard was managed by rapidly deploying movement monitoring technologies to determine if these features were still moving and to monitor their response to on-going aftershock activity. The fill and retaining wall failures were managed by encouraging public reporting of areas of concern for rapid assessment by a geotechnical professional. The success of the landslide risk management strategy was demonstrated by the magnitude MW 6.0 earthquake of 13 June when rock-falls and boulder roll damaged evacuated buildings and ground cracking and debris inundation further damaged evacuated areas. Some incipient landslides reactivated, producing similar movement patterns to the 22 February 2011 earthquake. Several retaining walls identified as dangerous and cordoned off also collapsed. No lives were lost and no serious injuries were reported from landslides in the 13 June 2011 earthquake.

No abstract

Landslides represent the most frequent geological hazard in mountainous environments. Most notably, landslides are a major source of fatalities and damage related to strong earthquakes. The main aim of this research is to show through three-dimensional engineer-friendly computer drawings, different mountain environments where coseismic landslides could be generated during shallow crustal and megathrust earthquakes in the Andes of central Chile. We have determined topographic, geomorphological, geological and seismic controlling factors in the occurrence of earthquake-triggered landslides from: (1) a comparison of local earthquake-induced landslide inventories in Chile (the Mw 6.2, shallow crustal Aysén earthquake in 2007 (45.3° S) and the Mw 8.8, megathrust Maule earthquake in 2010 (32.5°S–38.5°S)) with others from abroad; and (2) analysis of large, prehistoric landslide inventories proposed as likely induced by seismic activity. With these results, we have built four representative geomodels of coseismic landslide geomorphological environments in the Andes of central Chile. Each one represents the possible landslide types that could be generated by a shallow crustal earthquake v. those likely to be generated by a megathrust earthquake. Additionally, the associated hazards and suggested mitigation measures are expressed in each scenario. These geomodels are a powerful tool for earthquake-induced landslide hazard assessment.Thematic collection: This article is part of the Ground models in engineering geology and hydrogeology collection available at: https://www.lyellcollection.org/cc/Ground-models-in-engineering-geology-and-hydrogeology