Sir, - Brendan McWilliams in his article "Those other forecasters" (January 22nd) cites a number of reasons why weather and economic computer forecasts may fail. What he does not do, however, is state the reason why the forecasts must fail - if not sooner, then later. The reason is summed up by the picturesque phenomenon of "The Butterfly Effect" - the notion that a butterfly stirring the air today in Peking can transform storm systems next month in New York (Chaos by James Gleick).
The Butterfly Effect (or sensitive dependence on initial conditions) arises from the (necessarily) non linear equations used to model the weather or economic system. It ultimately dooms all numerical forecasts to inaccuracy - even those based on numerical models running on the Cray supercomputers of the European Centre for Medium Range Forecasting. - Yours, etc.,
Knocklyon Road, Templeogue,
Dublin 16.