Book of the week 14: The Weather Machine by Andrew Blum
I was intrigued by news reports that the Met Office was planning to drop more than a billion pounds on a new supercomputer, and wondered what it was that these clever weather forecasters did with all that silicon. So I picked up Andrew Blum’s recent book, The Weather Machine.
Blum starts with the weather map – and John Ruskin’s metaphor of the “weather machine”, transcending the local observations of an individual forecaster and linking together what James Gleick calls “local surprises” into a larger map. After all, one part of the weather forecasting game is straightforward: if it’s raining to the west of you and the wind is blowing from the west, you can expect rain soon. Weather forecasts begin with weather observations: the more observations, the better.
In the 1850s, the Smithsonian Institution in Washington DC used reports from telegraph operators to patch together those “local surprises” into a national weather map. This map was based purely on observations, but it was still a useful starting point before we had either the scientific understanding or the computational power necessary to make a reliable forecast.
The scientific understanding began to dawn in 1904, when Norwegian mathematician Vilhelm Bjerknes published “The problem of weather prediction”, an academic paper describing the circulation of masses of air. If you knew the density, pressure, temperature, humidity and the velocity of the air in three dimensions, and plugged the results into Bjerknes’s formulas, you would be on the way to a respectable weather forecast – if only you could solve those computationally-demanding equations.
The British mathematician Lewis Fry Richardson attempted just that, attempting to predict the weather of 20th May 1910 given the starting conditions. Alas, it was 1922 before he finished the sums – despite continuing to calculate in the evenings after long days as an ambulance driver during the war. Nor did the equations accurately describe the weather that day, 12 years earlier. Still: one must start somewhere.
Fry Richardson dreamed of a forecasting factory, a stadium filled with 64,000 human computers, conducted by lights and other signals as they furiously calculated the weather equations faster than the weather itself could evolve. It was a remarkable vision: modern weather forecasting works much as Fry anticipated, except that there is no need to perform the calculations by hand – or 128,000 hands. The Met Office’s billion-and-a-half dollars of silicon will do the job nicely.
Blum’s narrative ranges widely and finishes at the European Centre for Medium Range Weather Forecasts HQ in Reading. (My understanding is that this arrangement will survive Brexit, partly because the EMRWF is a separate organisation from the EU. But don’t quote me on that.) The EMRWF, says Blum, are the elite among meteorologists, and he spends some time exploring why they are so admired. Part of the secret is a way of working that can be split into modules and relentlessly tested, experimented with, and improved.
I strongly recommend the book, which is a fascinating glimpse of a mysterious world.