Tuesday, April 10, 2018

Oklahoma earthquake scales - Part 2

We've seen the crude scales of 'effect', and now we wish to go into more precision.  That involves strong motion accelerometers which are expensive to deploy, although I have designed one using a raspi zero and 4 cheap accelerometer chips.  The expensive part is useful deployment and calibration.  Everybody has an accelerometer chip in their phone, but not useful.

The easiest thing to get from these, is peak ground acceleration, or PGA.  You just take the whole time history of acceleration and pick the tiny spot where it is the maximum.  Unfortunately, this saturates fairly quickly and is the same for all types of different intensities.  A totally useless measure, yet it used religiously by the big engineering groups.  They base this on 'shaketable physics' which is as good as nasa or usgs physics.  They get tons of money building bigger and bigger shake tables, and yet the buildings tilt at very low ground shaking.  I won't say any more on that, because you could kill yourself trying to change it.  We must simply wait for more damage.

Here I am on dangerous ground, but the one measure that aligns well with Newtonian Physics, and observable intensities is peak ground velocity, or PGV.  I use this exclusively, but it's only me.

PGV is easy to get from accelerometers, but it is incompatible with shaketable physics.  Thus, it is never used.  In general, there is a factor of two increase in pgv for every MM intensity unit.  I usually just confine the reporting to factors of 10, there's no use being more precise than the uncertainties, although 'institutional physics' loves doing that to add more gravitas. 

Knowing the intensities from damage is totally useless.  However, knowing damage from pgv is useful.  You can categorize the 'seismic capacity' of buildings or an entire city, and then you can relate that to 'seismic hazard' or the probability spread of pgv.

For example, a pgv of 10 cm/s, should just rattle off the Royal Dalton, unless you were smart enough to secure it with tack putty.  A crummy old brick building will collapse.  And here's where pgv is infinitely more useful than pga.  That's on soft soils, where pgv can amplify by a factor of 10 to 100, but pga stays the same.  At 40 cm/s houses start to crack and peel off unsecured brick, and all the modern high-rises start to tilt.  Once a building tilts just a little bit, it has to be destroyed.  That's lots of money for engineers!

In OK we can expect the old downtown of Pawnee or Perry to be destroyed at 20 cm/s.  We can expect all the oil tanks to rupture at 40 cm/s.  These levels can be achieved by an m6 thrust.  We've had many m6's in OK, but most were of the wrong type and too deep.  The pgv depends greatly on earthquake mechanism, distance, and depth.


End of this useless series.  :)

All of those young people who want jobs, you should align yourself with 'institutional physics'. 



No comments: