Yorkshire Philosophical Society

Event Information

The Herbert Read Lecture

Why do good weather forecasts go bad?- How forecasts are made and what weather forecasters are doing to make them better
Dr David Schultz,
Centre for Atmospheric Science, School for Earth, Atmospheric and Environmental Sciences, University of Manchester

In principle, five relatively simple relationships are all that are needed to forecast the weather. If it is that easy, why do forecasts sometimes fail? This talk will present a brief history of weather prognostication, the process of modern computer weather prediction, and how meteorologists are working to improve forecasting for the future.

Report
By Jim Matthew
It was interesting to learn that the modern approach to weather prediction came from a Yorkshireman educated at Bootham School called Lewis Fry Richardson: in 1922 he envisaged a forecast factory of 64,000 human computers combining accurate meteorological data on a grid with the basic laws of fluid dynamics to predict the weather ahead. Alas it was into post-war computer development before his ideas could be implemented. Dr. Schultz gave a splendid account of how prediction has improved steadily over the last 50 years or so with four-day predictions in 2010 equivalent in precision to one-day predictions in 1967. But things still go wrong from time to time. The observational data is imperfect and the global grid on which calculations are made needs to be improved. There are bits of important physics missing from the model (e.g. complexities in radiation from clouds). Finally there is the phenomenon of chaos, the butterfly effect, where tiny changes in the initial conditions can cause radically different outcomes in the weather, inhibiting long range prediction. The audience emerged enlightened and well entertained.