Fatality rates of the M w ~8.2, 1934, Bihar–Nepal earthquake and comparison with the April 2015 Gorkha earthquake

Abstract

Large Himalayan earthquakes expose rapidly growing populations of millions of people to high levels of seismic hazards, in particular in northeast India and Nepal. Calibrating vulnerability models specific to this region of the world is therefore crucial to the development of reliable mitigation measures. Here, we reevaluate the >15,700 casualties (8500 in Nepal and 7200 in India) from the M w ~8.2, 1934, Bihar–Nepal earthquake and calculate the fatality rates for this earthquake using an estimation of the population derived from two census held in 1921 and 1942. Values reach 0.7–1 % in the epicentral region, located in eastern Nepal, and 2–5 % in the urban areas of the Kathmandu valley. Assuming a constant vulnerability, we obtain, if the same earthquake would have repeated in 2011, fatalities of 33,000 in Nepal and 50,000 in India. Fast-growing population in India indeed must unavoidably lead to increased levels of casualty compared with Nepal, where the population growth is smaller. Aside from that probably robust fact, extrapolations have to be taken with great caution. Among other effects, building and life vulnerability could depend on population concentration and evolution of construction methods. Indeed, fatalities of the April 25, 2015, M w 7.8 Gorkha earthquake indicated on average a reduction in building vulnerability in urban areas, while rural areas remained highly vulnerable. While effective scaling laws, function of the building stock, seem to describe these differences adequately, vulnerability in the case of an M w >8.2 earthquake remains largely unknown. Further research should be carried out urgently so that better prevention strategies can be implemented and building codes reevaluated on, adequately combining detailed ancient and modern data.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 42 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.