Analysis of the Timing of Building Displacement during Intense Seismic Shaking for three Earthquakes, 2001 Mw 6.8 Nisqually (Washington State), 2017 Mw 7.1 Puebla (Mexico), 2018 Mw 7.1 2018 Anchorage (Alaska)

Abstract

Localized intense ground shaking from strong earthquakes pose significant hazards to humans including building damage that may lead to structural failure and possible collapse. Building failure can cause harm or loss of life and also result in economic loss. For example, in the 2017 Mw 7.1 Puebla Earthquake in Mexico, more than 300 people died with the majority of those deaths in Mexico City. More than 3000 buildings were damaged and 44 buildings collapsed (Grillo, 2017).

Structural engineers and designers study rigorously whether or not a building may experience damage and/or structural failure. However, little emphasis has been placed on the timing of displacement and/or failure during strong shaking. For example if a building were to fail late in a long-duration earthquake while the interstory drift rose to destructive levels, the advice to its occupants might be to attempt to evacuate.

Therefore, I undertook an analysis to explore factors that might impact and constrain the variability in timing to maximum building displacement. To provide a range of estimates of building response timing to shaking, I explored the buildings’ temporal responses to different ground motions from seismic inputs recorded at different seismic stations. I modeled the building responses as a simple single degree of freedom oscillator, programmed in the Python programming language. I used ground motion recordings from three deep “intraslab” subduction zone earthquakes characterized by normal faulting, the 2001 Nisqually (Washington), the 2017 Puebla Earthquake, and the 2018 Anchorage Earthquake ( Figures 1-3). Using computer simulations, I tested the responses of a range of concrete building heights from 1 to 30 stories tall.

Results show that the time to maximum building displacement changes with the height of a building and varies with distance to rupture as well as local site responses. While perhaps not an accurate indication of the absolute time when a building would fail, the time between the onset of an earthquake and the maximum displacement observed in the modeled building response can be interpreted as a proxy for time to failure and perhaps its relative sensitivity to seismic shaking.

Earthquake Early Warning (EEW) is a newly emerging technology that uses real-time seismic data, rapid telecommunications, and new real-time processing algorithms to provide several seconds to up to 3 minutes of advanced warning of strong shaking to an at risk population. The actual warning time depends on factors like the latency of the system and the distance the earthquake source is from the targeted at-risk population. Prior to EEW, the most efficient life saving action that disaster prevention officials could offer was, “when you feel shaking, drop, cover and hold on”. With EEW, there is now the possibility of an additional response. Namely, if the timing of maximum building displacement from strong shaking were known for different storied buildings, it might be possible that warnings could be issued with sufficient lead time to allow people to exit the structure rather than shelter in place in a collapsing building. An EEW-mediated evacuation strategy would depend on the range of timing between an earthquake and a building collapse.

The single degree of freedom oscillator shows it can be a first step in quantifying the timing of building displacement during strong shaking. Understanding the tectonic setting, the distance of the rupture to an at-risk population, earthquake recurrence intervals, aftershock sequencing, earthquake directivity and seismic amplification should also be put into context when modeling earthquakes.