Lewis Fry Richardson
Lewis Fry Richardson was a Quaker mathematician, regarded as the father of modern weather forecasting. His work was often well ahead of its time, only verified and made practicable decades after Richardson published it.
Richardson was the youngest of seven children born to a Quaker family of tanners and leather-makers in Newcastle. He went to Bootham School in York, then Durham College of Science, before reading Natural Sciences at King’s College, Cambridge, where he graduated with a first class degree in 1903.
In 1912, after the sinking of the Titanic, he conducted experiments using a horn and an umbrella to test whether ships could use sound to detect icebergs in fog, and registered a patent for acoustic echolocation in air.
In 1913, he became superintendent of the British Meteorological Office’s research laboratory at Eskdalemuir in Scotland, and began work on mathematical methods of weather forecasting. Up till then, meteorologists had focused on explaining the past: Richardson aimed to predict future weather.
When in August 1914, the First World War broke out, Richardson was refused leave of absence to join the ambulance corps. He quit his job in 1916 and joined the Friends Ambulance Unit, attached to the 16th French Infantry Division. Extraordinarily, he continued his research between shifts in the trenches, determined to demonstrate his method by creating a real forecast.
His fellow meteorologist, Vilhelm Bjerknes had published detailed data for the weather over Central Europe on the 20th May 1910. Using the data for 7am, Richardson aimed to ‘forecast’ the weather six hours later and compare his ‘prediction’ to Bjerknes’ actual data for 1 pm.
Continuous changes over time in atmospheric conditions could be modelled as complex differential equations, but these were impossible to solve. Richardson made the problem manageable, though still complex, by dividing the region into 25 cells, around 125 miles square, each of five layers. In alternating cells, he either recorded atmospheric pressure, moisture and temperature, or calculated windspeed and direction. This produced discrete packages of data, or ‘finite differences’. He then made laborious calculations, predicting how these cells would change over the following six hours, in effect ‘forecasting’ the weather.
The result, though apparently a failure, led to further ideas, especially about atmospheric turbulence. In Richardson’s published account, Weather Prediction by Numerical Process, he postulated that the fundamental driving mechanism of the atmosphere was ‘turbulent cascades’, or as he put it in rhyming verse:
In 2004, Irish Meteorologist Peter Lynch demonstrated that the problem with Richardson’s original study lay, not with his methods, but with very short-term variations in Bjerknes’ pressure data.
Indeed Richardson’s numerical methods still underpin all modern weather forecasting. However, it took the advent of computers to do the job quickly enough to predict the next day’s weather before it arrived – something Richardson estimated would require 60,000 people working with slide rules!
Richardson rejoined the Met Office after the war, but resigned as a matter of conscience when it became part of the Air Ministry in 1920. He was suffering from ‘shell shock’ (as PTSD was then termed). He began to study psychology and statistical studies of collective violence. He defined a ‘deadly quarrel’ as any event that resulted in at least one death and applied a logarithmic scale to the number of deaths they resulted in. Using collected data since 1920, he looked for causes and correlations, but with few clear results. The results were published posthumously in 1960, in The Statistics of Deadly Quarrels.
This work led indirectly to another area where Richardson’s approach proved ahead of its time. One factor he considered was the length of shared borders between countries. This led to his discovery of the Coastline Paradox – the observation that the length of a coastline increases with the fineness of the scale in which it is measured, something now known as Richardson’s Law. This work presaged Benoit Mandelbrot’s description of fractals, and was quoted by Mandelbrot in his seminal 1967 paper "How Long Is the Coast of Britain?”
Richardson died in 1953, a year before the first BBC weather forecast.
Since 1997, the European Geosciences Union has awarded the Lewis Fry Richardson Medal for "exceptional contributions to nonlinear geophysics.”