Advances in the use of satellite and radar data led by Reading mathematicians are helping agencies across the world, from the UK Met Office to NASA, to make better weather forecasts.

Weather services are of benefit across all sectors of society – from farming to flying. In the UK alone, the benefits of the public weather service are valued at around £1.5 billion per annum. Furthermore, lives can depend on accurate forecasts of extreme winds, snow, floods or drought.

Forecasts are made by putting together computer predictions with real-life weather observations, for example from satellites or radar. The effective use of these observations of the atmosphere, snow and ice, oceans and land surface depends upon a quantitative description of our confidence in the observation accuracy. But observational data are expensive to gather, and errors in the data are difficult to analyse.

The research of Professors Sarah Dance and Nancy Nichols has helped meteorologists worldwide to make more accurate weather forecasts. They have come up with better ways to estimate correlations, or statistical patterns, in observation errors to improve weather predictions. Thanks to this work, expensive data from satellites and radar have become more cost-effective for forecasting agencies to use, and meteorologists have a better understanding of how observational uncertainty affects forecasts.

Agencies across the world now use the Reading methodology, including the UK Met Office, European Centre for Medium-Range Weather Forecasts, the US Naval Research Laboratory, and NASA. The work of Professors Dance and Nichols has influenced weather forecasting strategy, diagnostic activities, and training for the next generation of meteorologists. As ever more satellite and radar data become available, the research continues to have a far-reaching influence on the best use of weather observations across the globe.


Find out more

Project website: DARE (Data Assimilation for the Resilient City)

Data Assimilation: The secret to better weather forecasts (YouTube video)

How accurate are our atmospheric observations? (research paper summary)

High speed mathematics: reducing the computation time for weather forecasting (research paper summary)

Project website: Flooding from Intense Rainfall

Making better use of weather observations (summary from Flooding from Intense Rainfall research programme, 2018) –

View the full impact case study on the REF 2021 website: Improved accuracy of weather forecasting internationally through advanced use of observational data