Numerical Weather Forecasting is younger than rock ‘n roll!!
Before 1960, the idea that a reasonably accurate two-day weather forecast could be made routinely was a pipe dream – now it’s a routine reality. In fact, it was not until just after World War I that a theory concerning the structure, life cycle and precipitation distribution associated with mid-latitude cyclones, the weather systems that bring snow and rain to Madison, was first proposed by a group of Norwegian scientists, led by Vilhelm Bjerknes, an ambitious but professionally frustrated physicist.
Bjerknes was the first to suggest that weather forecasting might proceed from the equations derived from physics that describe the behavior of the atmosphere. This possibility caught the attention of Lewis F. Richardson, a British scientist who set out to prove that forecasting the weather by means of calculation could, indeed, be accomplished. Richardson spent six weeks making the calculations necessary to make an eight-hour forecast of sea-level atmospheric pressure over Europe — a relatively simple problem by today’s standards. He failed miserably — but published his account anyway, in his 1922 book, Weather Prediction by Numerical Process.
Richardson’s book collected dust for nearly 30 years before the earliest computers were let loose on the problem in the early 1950s. Enough progress was made that the first daily computer-based forecasts began to be issued beginning May 6, 1955. Numerical weather prediction has subsequently matured into what is arguably the most under-appreciated scientific advance of the last 50 years.