One of the most significant effects of the M9.0 Tohoku, Japan earthquake of March 11, 2011 is the now well-known long duration (>10 minutes) shaking of buildings in Japan – particularly those in Tokyo (~350-375 km from the epicenter) and in places as far as Osaka (~770 km from the epicenter). Although none collapsed, the strong shaking caused many tall buildings not to be functional for days and weeks.

The purpose of this paper is to discuss the behavior and performance of four tall buildings, considered to be representative of most tall buildings in Tokyo and other locations, and from which shaking data were retrieved. Of particular interest is a building in Osaka that almost reached an average of 0.5% drift ratio – this, with a ground level input motion of ~3% g is significant. What might have happened during an event with input level motions with similar low frequency content and in 10-20% g range is a legitimate question that must be pondered. The particular building had serious site effects and was in resonance.

The other three examples are from Tokyo. For example, based on records obtained from a 54-story building retrofitted with 288 oil dampers on 24 floors, computations show that average drift ratio may have reached ~0.3% and maximum drift ratio likely was > .3%. The maximum allowed by Japanese practice for buildings taller than 60 m and for collapse protection (level 2) motions is 1% (The Building Center of Japan, 2001).

Performances of tall buildings in many seismically active regions of the world (e.g. Chile, Turkey) or those tall buildings affected by long distance long period effects by sources at a distance (e.g. Abu Dhabi, Dubai) are of interest to the earthquake engineering community. Chile imposes 0.2 % drift limit that result in elastic design. USA and Turkey impose 2% drift limit. Such wide variations of drift limits in design practices deserve discussion in light of functionality and performance of tall buildings during the 2011 Tohoku event.