On Sunday I caught a bus from Taksim to the airport hotel for an early flight. We drove under an aqueduct that a Google search says was built about 368 AD. It says it is still in use. It evidently did not collapse in earthquakes, or if parts of it did, it was repaired (this is likely known). Then the bus ran along the edge of Marmara Sea and by the Atakoy area, where I had met the ship. That area is low, and presumably has bad site condistions for quakes (although I don't know the exact areas where there were 2000 fatalities in 1999). It is densely packed with apartment buildings. One could question whether these would collapse in certian quake scenarios. Then, close to the airport there was a block of new buildings: were these designed and built for quakes? At the airport hotel I asked the check in person when the hotel was built (6 years ago) and he said it was over-built/safe; is owned by the airport (also is only 2 or 3 stories-high). He said he had heard that 2 million could be killed in an Istanbul quake. I probably said that I doubted that because site conditions are good in large parts of Istanbul (unlike Los Angeles). I had once heard a seismologist say that 1 million could be killed in an Istanbul quake, and Roger Bilham (not same person as early in sentence) was on TV saying that there would be 1 million fatality quakes; he may have gestured at the general middle east larger region.

You could probably get a reasonable idea of what was possible from existing work; scenario earthquakes, site conditions, how many people live in badly constructed high-rise (10 stories?) apartments in areas of bad site conditions. I don't think it will be as bad as the numbers above, but I'm guessing (Istanbul has about 12 or 15 million people, if 10% of area has bad site conditions and 20% fatalities in these areas, you get a number).

So, why aren't the poorly constructed buildings in the bad areas being torn down?

I suppose you could ask that of certain buildings in certain places in California.

Chris