US scientist using Prague records to study probability of record temperature events

The subject of global warming seems to be rarely out of the news these days, with even George Bush - known to critics as the "Toxic Texan" - talking of "the serious challenge of global climate change" in his recent State of the Union address. One of many studies on the issue is currently being prepared by Professor Sid Redner of Boston University, who has been using data on record temperatures from Prague's Clementinum weather station. I asked him why.

"It's been extremely difficult to find long-term, reliable temperature data. It just so happened that I heard from someone who said, oh, I know someone in Prague with 230 years of data. So I asked this person for it and he sent it to me. It's sort of a serendipitous thing."

What have you used that information for?

"I'm looking at the probability of record temperature events occurring. The point is that if the temperature in the globe was not changing then the probability of having a temperature record on any given day would gradually go down with time. It's a well known fact from probability theory.

"If there is a systematic change in the temperature, such as global warming, then the frequency of records will not...obey this law that would occur if the temperature was not changing. I was trying to look for the signature of the change, of global warming trend, by looking at the frequency of record temperature events."

Photo: CTK
What did you find?

"I found that there was a reasonably visible increase in the rate of record high temperature events, which would be consistent with some systematic global warming."

How long a period do you have to survey to be sure it's not just a question of regular temperature fluctuation?

"Well, the data I had from Prague was for 231 years. And based on my theoretical analysis of the problem first of all, and given the current rate of global warming, it seemed - again you can't make a really hard and fast statement about this - but somewhere in the vicinity of 150 years or more of data was sufficient to be able to see a systematic effect on the data.

"And since Prague had 230 years it was enough to begin to see a systematic effect."

Excuse my layman's ignorance, but do you think you've proved the existence of global warming then?

"No, no, no. That's a much too strong statement. All I can say is that at this one location, Prague, there is an increase of temperature records which would be consistent with global warming.

"But if I really wanted to make a smoking gun case I should really have data from hundreds of stations around the globe for the same type of period - then one could see more cleanly.

"The thing is that climatologists who look at the weather average over the entire sees already just from the climatology data that there obviously is global warming.

"My initial interest in this problem was a purely intellectual one. We had a heat-wave in Boston and there was a record temperature set, and the weathermen said, oh, it's a record temperature, it must be global warming.

"And I just got to thinking - can you understand what is the connection between global warming and the frequency of record temperature events?"