Eenglish-tests.com

C2 Text 25 · Interview · 760 words

Chaos and Prediction

An interview with a weather scientist about chaos theory, the butterfly effect and why some things can never be predicted far ahead.

In 1814 the French mathematician Pierre-Simon Laplace imagined an intellect so vast that, knowing the position and motion of every particle in the universe, it could calculate the entire future as easily as the past. Two centuries later, we have computers Laplace could not have dreamed of, yet a reliable weather forecast still rarely stretches beyond a week or so. To find out why, our science writer, Daniel Price, spoke to Dr Amara Lindqvist, who builds weather prediction models at a research centre in Exeter.

Price: Let's start with the famous phrase. What is the butterfly effect? Lindqvist: It goes back to Edward Lorenz, an American meteorologist. In the early 1960s he was running a simple weather model on a computer. Wanting to look at one run again, he typed in a number from a printout, which had been rounded to three decimal places instead of six. He expected the new run to track the old one closely. Instead, after a while, the two forecasts diverged completely. A difference of less than one part in a thousand had grown until it swamped everything else. Years later, he gave a talk asking whether the flap of a butterfly's wings in Brazil could set off a tornado in Texas. The image stuck.

Price: So the butterfly causes the tornado? Lindqvist: That's the popular reading, and it's slightly misleading. The point isn't that butterflies are secretly in charge of the weather. It's that in certain systems, tiny differences in the starting conditions grow so rapidly that, beyond a certain time, prediction becomes impossible in practice. We call this sensitive dependence on initial conditions. The unnerving part is that the system is completely deterministic: the equations contain no randomness at all. Every step follows strictly from the one before. Yet the outcome is, after a while, as unpredictable as if it were random.

Price: That sounds like a contradiction. Lindqvist: It feels like one, which is why chaos was such a shock. People had assumed that if you knew the rules, you could, at least in principle, know the future. Chaos showed that knowing the rules isn't enough; you'd also need to know the present with impossible precision. And you never can. Every measurement has some error, and there are only so many weather stations, balloons and satellites. Laplace's intellect would need infinite accuracy, and nothing in the real world has that.

Price: Then how do you forecast at all? Lindqvist: By being honest about what we don't know. Rather than running a model once, we run it many times, fifty or so, each with slightly different starting conditions, all consistent with our measurements. That's called an ensemble. If the runs stay close together, we can be fairly confident. If they fan out, we know the atmosphere is in a touchy mood, and we say so. That's where those percentages come from. Over the past few decades, better data and faster computers have pushed useful forecasts further ahead by roughly a day each decade, which is a quiet triumph. But there is a ceiling. Most researchers think day-to-day weather can't be predicted in detail beyond about two weeks, however good our machines become.

Price: Doesn't that make climate projections, decades ahead, seem rather presumptuous? Lindqvist: It's a fair question, and the answer lies in what you're predicting. I can't tell you whether it will rain in Exeter on the fourteenth of July in thirty years' time. But I can say a good deal about the average summer then, just as I can't predict a single roll of a dice, yet I can tell you that, over thousands of rolls, each number will turn up about a sixth of the time. Weather is the single roll; climate is the long-run statistics. Chaos limits the first far more than the second.

Price: Does chaos appear outside the atmosphere? Lindqvist: Everywhere, once you start looking. A double pendulum, just two rods joined together, swings erratically enough to mesmerise a room full of students. Some heart rhythms, animal populations and even the long-term orbits of planets show chaotic behaviour. Not everything is chaotic, of course. We can predict eclipses centuries ahead, because that system is far more forgiving.

Price: Finally, does chaos make you feel humble or frustrated? Lindqvist: Humble, mostly. There's something rather beautiful about it. The universe follows rules, but it doesn't hand us the future on a plate. We have to earn every extra day of forecast, and we've learned to say 'probably' and mean it precisely. I'd take that over false certainty any day.

Words in colour are explained in the glossary below.

Glossary

WordMeaning
intellect/ˈɪntəlekt/a mind that can think and understand, especially a very clever one
track/træk/to follow the same path as something
diverged/daɪˈvɜːdʒd/went in different directions, becoming more and more different
swamped/swɒmpt/became so large that it covered or overwhelmed everything
stuck/stʌk/stayed in people's minds and continued to be used
unnerving/ʌnˈnɜːvɪŋ/making you feel nervous or uneasy
deterministic/dɪˌtɜːmɪˈnɪstɪk/where every result is fixed completely by what came before, with no chance
fan out/ˌfæn ˈaʊt/to spread out from one point in different directions
touchy/ˈtʌtʃi/easily upset; here, easily changed by small things
ceiling/ˈsiːlɪŋ/an upper limit that cannot be passed
presumptuous/prɪˈzʌmptʃuəs/too confident; claiming more than you have the right to
erratically/ɪˈrætɪkli/in an irregular way that is hard to predict
mesmerise/ˈmezməraɪz/to hold someone's attention completely, as if by magic
forgiving/fəˈɡɪvɪŋ/not punishing small mistakes; here, not badly affected by small errors
on a plate/ˌɒn ə ˈpleɪt/very easily, without any effort from you

Chaos and Prediction: tests

Found this text useful? Share it with a friend or your class.