GROUND-SHAKING IN THE STEEPEST TERRAIN ON EARTH: LANDSLIDES FROM THE 2015 GORKHA, NEPAL EARTHQUAKE SEQUENCE

When earthquakes occur in steep terrain, an inevitable consequence is landsliding. On April 25, 2015, a M7.8 earthquake shook central Nepal including a section of the Himalaya mountain range that has more than 6,000 m of relief. The earthquake and its aftershocks, including a M7.3 that occurred 17 days after the initial rupture, caused thousands of fatalities, destroyed entire villages, and triggered thousands of landslides.

In response to this disaster, members of the USGS Landslide and Earthquake Hazards Programs deployed to Nepal to assess landslide hazards and to collect perishable data related to the landslides’ effects. In collaboration with an international team of remote-sensing experts to identify priority areas for in-country reconnaissance, we conducted helicopter- and ground-based assessments in 12 priority areas and flew more than 3,200 km over an 8,000 km² region of central Nepal. We focused our work on landslides that directly affected villages and infrastructure – this included the ~2,000,000 m³ Langtang debris avalanche that killed more than 200 people, the ~250,000 m³ Kerauja rock slide that killed 1 person at the very edge of a small (2,700 person) village, and the ~300,000 m³ Baisari rock slide that dammed the Kali Gandaki River 29 days after the initial earthquake and impounded a lake that extended nearly 3 km upstream of the landslide dam. A report delivered upon returning from the field (http://dx.doi.org/10.3133/ofr20151142) outlines hazard assessments at 17 sites and 74 landslide dams.

In general, landslides were abundant near the major earthquake epicenters and in an east-west swath of terrain along the northern part of the Lesser Himalaya and throughout the Higher Himalayan zones. However, landslides were sparse in many steep areas where landslides would have been expected, which suggests that a complex combination of factors played a role in the landslide distribution. These include the complex geology of the Himalayan Main Central and Boundary Thrust zones and directivity effects of the rupture. Further, evidence of extreme ground-shaking (i.e., “shattered ground”) at many ridgetops indicates that shaking was likely asymmetrically amplified at topographic high points, which explains at least some of our observations related to the distribution of landslides.