Modern weather forecasts are made by calculating solutions of the mathematical equations that express the fundamental physical principles governing the atmosphere [TM083, or search for “thatsmaths” at irishtimes.com]
The solutions are generated by complex simulation models with millions of lines of code, implemented on powerful computer equipment. The meteorologist uses the computer predictions to produce localised forecasts and guidance for specialised applications.
During World War One, long before the invention of computers, the English Quaker mathematician Lewis Fry Richardson devised a method of solving the equations and made a test forecast “by hand”. The forecast was a complete failure, giving an utterly unrealistic prediction of pressure change, but Richardson’s methodology was sound and underlies modern computer weather forecasting.
In 1922 Richardson published a remarkable book, Weather Prediction by Numerical Process. Having described his method, he presents a fantasy of a “Forecast Factory”, a building with an enormous central chamber with walls painted to form a map of the globe. There a large number of (human) computers are busy calculating the future weather.
Richardson estimated that 64,000 people would be needed to calculate weather changes as fast as they were happening. In fact, this was over-optimistic: to produce a forecast in timely fashion would require upwards of a million computers.

The upper floor. Banners identify the main characters.
The working of the forecast factory is coordinated by a Director of Operations. Standing on a central dais, he synchronises the computations by signalling with a spotlight to those who are racing ahead or lagging behind. There are striking similarities between Richardson’s forecast factory and a modern massively parallel processor.
Several artists have created illustrations of the forecast factory. One particular image has recently come to light. The painting, in ink and water colours, was made by Stephen Conlin in 1986, on the commission of Prof. John Byrne, then Head of the Department of Computer Science in Trinity College Dublin. This painting, which has gone unnoticed for many years, is a remarkable work, rich in detail and replete with hidden gems.
Conlin’s image depicts a huge building with a vast central chamber, spherical in form. On the wall of this chamber is a map with roughly half the globe visible. On an upper level sit four senior clerks. A banner on each desk identifies a major historical figure. Several scholars and savants are depicted in the painting. They include pioneers of computing like Charles Babbage, Ada Lovelace and George Boole and mathematicians like John Napier, Blaise Pascal and Gottfried Wilhelm von Leibniz.

The machine with dials is Kelvin’s tide-prediction machine. At right are Ada Lovelace and George Boole.
Communication within the factory is via pneumatic carriers, systems that propel cylindrical containers through a pipe network using compressed air. These systems were used in large retail stores, such as Clerys in Dublin, to transport documents and cash. Pneumatic networks were popular in the late 19th and early 20th centuries for transport of mail, documents or money within a building, or even across a city.
A description of Conlin’s image has recently been installed on the website of the European Meteorological Society: EMS There is a very high resolution version of the picture there, with a zoom facility. Examination of the high-res image is rewarding, and will reveal a wealth of interesting details.
Sources
Ashford, Oliver, M., 1985: Prophet – or Professor? The Life and Work of Lewis Fry Richardson. Adam Hilger, Bristol & Boston, 304pp.
Conlin, Stephen, Artist: For a range of work by the artist, see http://www.pictu.co.uk/
Lynch, Peter, 2006: The Emergence of Numerical Weather Prediction: Richardson’s Dream. Cambridge University Press, 279pp. ISBN-13: 978-0-521-857291.
Lynch, Peter, 2016: An Artist’s Impression of Richardson’s Fantastic Forecast Factory. To appear in Weather. PDF preprint
Richardson, Lewis Fry, 1922: Numerical Prediction by Numerical Process, Cambridge Univ. Press. Second Edition, Cambridge University Press 2007: ISBN: 978-0-521-68044-8.