Time for another series!
My career has spanned the era where earthquake engineering has separated from earthquake science. Now it is so far apart, I fear there will be trouble in the future, especially with the new plants. For example, in Canada, we have no mechanism to 'certify' a given site, for a given nuclear plant design. I was hoping to work out something before I got kicked out of the process.
However, there is one area where engineering has tracked science, and this is in the ancient art of 'experience data'. In the 80's, this was pushed by EQE International. Unfortunately, it was totally rejected by the mainstream seismic engineers, since it would put them all out of job. EQE was swallowed by a Giant Borg Engineering Conglomerate, and has never been heard of since.
We start with the famous 1985 Mexico earthquake. This was an M8.1 subduction earthquake way out on the sunny coast. As far as earthquakes go, it was only about a 30% g ground motion, with probably 50 cm/s on firmer ground. As such, it never damaged well-built structures on firm ground.
Digression: One reason that experience data may be relegated to the dustbin of history, is their bloody-minded adherence to 'peak acceleration'. The sinusoid engineers can stick to this with their shake tables, since they have lost contact with reality long ago, but field studies need something correlated with physics, and the cause of damage. The best parameter is PGV, and if I were KingoftheWorld, then I would convert all the old experience data to this value.
Ok, when EQE studied this earthquake, they went beyond the pretty pictures of destroyed buildings and looked at what was undamaged. This became a keystone of the technique. At the time, I was mostly interested in the bizarre soil resonances that caused the damage, but EQE found that most of those buildings were junk anyway. Although there was a lot of amplification, the better buildings still survived.
--to be continued, with enough pennies in the jar.