Relating new theories of extratropical cyclone development to the present and future atmosphere

Cyclones have a major impact on people and the economies of countries in the mid-latitudes. Damaging impacts include severe gales and flooding associated with heavy rainfall. These weather systems also have less dramatic but nevertheless crucial impacts on temperature, cloudiness and rainfall. For example, autumn 2000 brought UK rainfall of unprecedented extent and duration, and prompted speculation about its relationship with climate change. In response, DEFRA commissioned a report (FD2304) to examine the link. The rainfall was associated with the unusually persistent and repeated passage of frontal depressions, rather than especially extreme cyclones. Such seasonal rainfall extremes were found to be more frequent in simulations of the end of the century by the Hadley Centre climate model, although the natural variability is so large that the autumn 2000 rainfall cannot be attributed to human influence on climate. There are many uncertainties associated with the representation of the Earth system by climate models, especially considering the behaviour of storms. However, the physical mechanisms behind changes in storm frequency and intensity are not sufficiently well understood from a theoretical perspective to determine whether the results of the simulations are physically reasonable. A weather forecast for the UK is hardly ever made without reference to a cyclone and its associated fronts that are with us or due to arrive soon. Numerical weather prediction models generally provide good forecasts of cyclone occurrence out to more than 5 days ahead by solving numerically the 'primitive equations' of motion and thermodynamics on 'resolved scales' and parameterising the effects of smaller scales. Although the theories of large-scale dynamics involve balanced flow governed by the evolution of single quantity called potential vorticity, such balanced models are not sufficiently accurate to be used in an operational forecasting context. Furthermore, theories of cyclone development assume that they grow from a 'background state' that does not vary around latitude circles, but the atmosphere never passes through such a state. Therefore, there is a mismatch between the capabilities of theory and the description of the atmosphere by forecast models. Nevertheless, the theory does provide a framework for understanding the evolution of forecasts and their errors and there is scope for improving theory to the point where it can distinguish quantitatively between potential mechanisms of growth. Theory has a greater role to play in robust arguments for changes in storm behaviour in the future. For example, policy makers and many other sectors of society would like to know whether 'extreme storms' will become more frequent in a warmer world resulting from greenhouse gas forcing - if these changes do occur they will have major economic impacts. The aim of this project is to bring some of the latest theories of mid-latitude cyclones to bear on analyses of the actual atmosphere in an attempt to identify robust predictions for the dependence of cyclone properties, such as surface wind strength, on background state properties, such as the pole to equator temperature gradient. By establishing quantitative relationships between background states (climate) and storm behaviour, this project will produce a framework for analysing changes to extreme storms under different climate scenarios. If the changes to storms simulated by a climate model accord with the expectations of sound physical reasoning, grounded in theory, it would be possible to place greater confidence in the climate projections.