Catastrophe models fail to account for “supershear” earthquake shockwaves, an oversight that could have consequences for high-risk areas such as California, researchers say.

An MS Amlin study found supershear events – causing shockwaves similar to jet aircraft sonic booms – have accounted for 66% of insured losses from quakes since 2016, equivalent to $US13.2 billion ($18.5 billion).

However, they are absent from seismic hazard models, building design codes and insurance catastrophe models.

A paper published in the Journal of Catastrophe Risk and Resilience warns the blind spot could have implications for capital and pricing decisions, particularly in major quake zones.

MS Amlin senior research analyst Luke Wedmore, who co-wrote the paper with research analyst William Sturgeon, says the 1906 San Francisco quake has been identified as a supershear event.

“Given the higher shaking intensities caused by supershear earthquakes, there is a significant chance that earthquake risk in California is markedly underestimated,” he said.

“With California potentially experiencing its longest major earthquake drought in 1000 years, now is a critical moment for the industry to address this blind spot.”

Supershear quakes occur when a rupture along a fault travels faster than usual, creating a shockwave.

The result can be far stronger ground shaking and a “double punch” effect caused by successive seismic waves.

According to the paper, supershear earthquakes can produce “unusual torsional forces” on buildings, particularly tall ones, and create stronger shaking that travels further from the fault.

Last year’s magnitude-7.7 Myanmar quake has also been identified as a supershear event.

“There are still lots of things we don’t know about supershear earthquakes, but the evidence now suggests they are more common – and potentially far more damaging – than previously understood,” Dr Wedmore said.

MS Amlin modelled the impacts of supershear events on representative insurance and reinsurance portfolios, finding losses at a 200-year return period increased by 5%-10%. At 500-year return periods, losses jumped 30%-60%.

Dr Wedmore says the insurer has updated its models and it is important the risks are captured more widely, particularly with major vendors preparing to update US quake models following revisions to the national seismic hazard framework.

Insurers and catastrophe modellers could begin addressing the issue now through stress tests and enhanced sensitivity scenarios to assess the risk in the short term, the paper says.