****************** THIS IS A RUSH, UNEDITED TRANSCRIPT *******************

BROOKE: After a calamitous earthquake, we crave what is least obtainable, certainty. When will the the next one be? Where? How bad? Hells bells! Why can’t you tell us?

[CLIP]

FOX: God forbid, Dr. Graves, that if that big one that everybody has been predicting, that would make its way to California, Los Angeles, San Francisco, maybe there would be not as much damage as we’re seeing in the Nepal situation, because there’s better infrastructure. Or am I wrong?

Earthquakes always bring out the armchair expert, says Susan Hough, a seismologist at the US geological survey, but predicting earthquakes, or even estimating their potential harm, is a thorny exercise that hinges, she says, on countless crucial imponderables…

HOUGH: For example, if an earthquake happens in the dead of night it can make a big difference. If a population center experiences weak shaking, which might inspire people to run outside, in a place like Nepal where buildings can and do collapse, that may end up saving lives. Little things can matter - there was a 2001 earthquake in Western India. It happened on Republic Day, a holiday in India. More people were inside watching television, watching parades than normally would have been the case, and that made a difference for the death toll.

BROOKE: So, if you do get those factors, the vulnerability of the building stock, the time of day, the day of the week, the length of the shaking - can you plug those things into a computer then and get a really good estimate?

HOUGH: We really can't come all that close. The USGS has a system called Pager that does these calculations. The numbers I saw, there was a one third probability in Nepal that the fatalities would be between 1,000 and 10,000, and there was almost as a high a probability that there'd be between 10,000 and 100,000. And there was actually a 17% chance that they would be greater than 100,000. So that's a huge range, but that really does reflect the extent of uncertainty dealing with human life losses as a math problem.

BROOKE: The initial reports put the death toll at 1900, that quickly rose to 5000 and more than 7500. Now some experts say that the toll will top 10000 and Max Weiss who's a respected analyst of seismic hazards recently calculated that the eventual death toll could top 57000, How do we reckon with that range?

HOUGH: So in this case the media is watching science in action. Watching research play out and results sort of fly around, that may not reflect a consensus of the scientific community. Normally that plays out behind the curtain, but when an earthquake happens, all of a sudden the process can splash into the media.

BROOKE: And you've suggested that pumping the numbers up, prematurely and then knocking them down later could create a kind of complacency?

HOUGH: It isn't like somebody out there is pumping them up on purpose. The death toll really was quite a bit lower than we would have expected. You know we know that Nepal and Kathmandu have been enormously vulnerable. And now we've seen a big one if not the big one and the death toll is going to be on the order of 10,000 and not 100s of thousands as people have feared. So there is a concern that people might get the message that Oh well, the big one there or maybe anywhere else isn't going to be as bad as the experts are saying. And that would be the wrong message to take away from this because it does appear that this is a very unusual earthquake for the kind of shaking it generated. There's just no reason to believe that future earthquakes they're going to be as kind.

BROOKE: Past performance is no guarantee of future results.

HOUGH: The magnitude of an earthquake doesn't tell the whole story in terms of the actual shaking that it generates.

BROOKE: All right, what was so special about the way it shook in nepal.

HOUGH: That I think is going to be the most important science story. It was what we call very long period motion. If you look at videos from Kathmandu, in the strongest motion, the ground is going back and forth over the course of about 5 seconds. It's sort of like a very long swell in the ocean. So you have buildings that are quite short. If you imagine little ships on very big swells in the ocean, those ships may be able to ride out the waves without too much damage.

BROOKE: Mhm. And in the coverage of earthquakes, you have noticed an interesting side story, the prediction of the next quake.

HOUGH: We can't predict earthquakes and so you're not seeing reliable predictions. But predictions come out of the woodwork after a big earthquake after it happened. I heard an esteemed colleague asked after the Northridge earthquake in 94 if anyone had predicted the earthquake and he said not yet. People will come out and they will say either we predicted it, or look here's some signal that happened before the earthquake and the earthquake could have been predicted. That's what people want. They want to know is the big one gonna hit tomorrow or 50 years from now. And that's what we just can't tell them.

BROOKE: Have you ever been in one?

HOUGH: Oh yeah. There was the Northridge earthquake, magnitude 6.7. I was sound asleep and it's like somebody comes along and with absolutely no warning, picks up your house and starts shaking it back and forth as hard as they can. It was incredibly loud, that was one of the surprises to me. There were transformers exploding, there were car alarms going off, there was the house creaking. you know I'm an earthquake professional, but they are scary.

BROOKE: Susan thank you very much.

HOUGH: Oh you're welcome, it was nice talking to you.

BROOKE: Susan Hough is a seismologist at the US geological survey. You can find the one-page printable version of the Breaking  News Consumer’s Handbook, Tectonic Edition, at onthemedia.org