16 January 2024

Origin of the "butterfly effect" theorem


I learned this today while reading about chaos theory and its manifestations in our modern world:
Edward Lorenz was a weatherman during World War II, tasked with forecasting cloud cover before American bombing raids in the Pacific. But meteorology in those days was largely guesswork and produced only crude predictions. After the war ended, Lorenz decided to try to unlock the secrets of the weather using more sophisticated methods and harnessing the nascent power of computing. He created a simplified, miniature world on his LGP-30 computer: Instead of the millions of different variables that affect weather systems in the real world, his model had just 12 variables.

One day, Lorenz decided to rerun a simulation he’d done earlier. To save time, he decided to start midway through, plugging in the data points from the prior snapshot. He figured that so long as he set the variables at the same levels, the weather patterns would be repeated just as they were before: same conditions, same outcomes.

But something strange happened instead. The weather in his rerun simulation was different in every way. After a lot of scowling over the data, Lorenz realized what had happened. His computer printouts had rounded data to three decimal places. If, for example, the exact wind speed was 3.506127 miles an hour, the printout displayed it as 3.506 miles an hour. When he plugged the slightly truncated values from the printouts back into the simulation, he was always off by a tiny amount (in this case, just 0.000127 miles an hour). These seemingly meaningless alterations—these tiny rounding errors—were producing major changes.

That observation led Lorenz to a breakthrough discovery. Minuscule changes could make enormous differences: Raising the temperature one-millionth of a degree could morph the weather two months later from clear blue skies into a torrential downpour, even a hurricane. Lorenz’s findings were the origin of the “butterfly effect” concept—the notion that a butterfly flapping its wings in Brazil could trigger a tornado in Texas—and, ultimately, of chaos theory. They also explain why meteorologists are still unable to forecast the weather beyond a short time frame with much accuracy; if any calculation is off by a tiny amount, the longer-term forecast will be useless.
The article goes on to explain why "humans are the puppets of small, seemingly arbitrary or accidental events" (the bombing of Hiroshima instead of Kyoto, for example).

Embedded image from the Wikipedia entry on this topic. 
"One meteorologist remarked that if the theory were correct, one flap of a sea gull's wings would be enough to alter the course of the weather forever. The controversy has not yet been settled, but the most recent evidence seems to favor the sea gulls."
Related (and the source article that set off my inquiry this morning):
"I know it seems counter-intuitive but one of the most glaring signs of our warming climate is that of our coldest temperatures."

2 comments:

  1. I don't think the location of that butterfly has been precisely defined? If it was, then all of the modelling would be much more accurate.

    ReplyDelete
  2. May be too technical for some, but a good read for others: https://www.quantamagazine.org/the-quest-to-decode-the-mandelbrot-set-maths-famed-fractal-20240126/ The Quest to Decode the Mandelbrot Set, Math’s Famed Fractal

    ReplyDelete