Revolutionizing Convective Parameterization

The weather and climate of the tropics is dominated by cumulus clouds. These clouds are produced by vigorous convection currents within the atmosphere. The convection communicates heat and evaporation from the Earth's surface throughout the atmosphere. It is the main process controlling the change of temperature and moisture content with height in the tropical atmosphere. On the global scale cumulus clouds are responsible for the majority of the rainfall, and convection is a crucial component in the overall pattern of the Earth's atmospheric flows.

Computer modelling of the atmosphere, both for global numerical weather prediction (NWP) and for climate projection, divides the atmosphere into boxes with typical horizontal sizes of 20 and 100km respectively. This means that the models are not able to represent convective elements (such as thunderstorms) properly, because these elements are typically only around 1km in size. However, as convection has a crucial role to play in the atmosphere, it must be represented within the models. We have somehow to estimate what cumulus clouds will be present in each of the boxes and what their collective effects will be on the larger-scale atmosphere. This is known as a cumulus parameterization.

Computer modelling of the atmosphere is essential for reliable climate projections and weather forecasts. Society benefits enormously from their outputs to inform decision making on all scales from the individual member of the public to weather-sensitive business activities to the insurance sector to the emergency services to government policy on climate risks. Convection parameterization is a stubborn and difficult problem and the largest single uncertainty that we face. It is a severe and unforgiving test of just how well we understand the fundamental science of convection and its role in the atmosphere. Defects in the existing parameterizations are known to translate into serious deficiencies in weather and climate models. To give just one example, in many models the predicted convective rainfall is too frequent and too light.

RevCon is so named because the project aims at a revolution in convective parameterization by challenging many of the key assumptions that have been made within convective parameterizations for decades. Some of these assumptions are very limiting and known to be poor approximations in many circumstances. We are also convinced that they are unnecessary and that better parameterizations can be achieved without them. This project will establish what the mathematical structure of convection parameterization really should be, with that structure being informed and carefully justified through the detailed analysis of very high-resolution simulations of convective cloud systems. It is a critical contribution to a NERC / Met Office programme in this area because it will provide a necessary starting point for Phase 2 of the programme to build a new-generation parameterization for the Met Office weather forecast and climate model.

The primary impacts of this NERC / Met Office programme will be:

1. directly on the accuracy and utility of Met Office weather forecasts and climate predictions via implementation of improved convection parameterization in the Met Office's Unified Model (UM);

2. indirectly on other weather and climate prediction centres through promulgation of the methodology implemented;

3. indirectly on other weather and climate prediction centres through access to the reference data that the new methodology is based upon.

The RevCon project targets these impacts in a completely natural and straightforward way, since all of the project deliverables are important for the overall success of the programme and all map straightforwardly on to the above impacts. (Deliverables D1 and D2 in the Case for Support map to impact 3, D3 to D5 map to impacts 1 and 2, and D6 maps to impact 1).

A parameterization of convection is an essential ingredient in operational weather forecasting (including extreme events such as flooding), in seasonal weather forecasts and in regional and global climate models. A reliable and soundly-physically-based parameterization is therefore essential for the quality of weather forecasts and for reducing uncertainties in climate projections. Improved parameterization methods will ultimately impact end-users from government, industry and the public, through significant improvements to Met Office products and advice. The Met Office, in common with operational prediction centres worldwide, has made this issue a major priority. The Met Office has formal arrangements with a number of national meteorological services and research institutes around the world, concerning the use by these services of the UM. Therefore, these international partners, including Australia, Korea, new Zealand, Norway, India and South Africa, will also directly benefit from improvements to the UM.

The scientific understanding gained through this project will have similar impacts within other operational centres, and will drive forward the fundamental understanding of convection within the academic and broader WGNE (Working Group on Numerical Experimentation) community. This will be further facilitated by the availability of reference datasets from the RevCon project which will form the basis of further exploitation.

Ensuring the project's potentially very high impact requires two-way exchanges with scientists working on operational weather and climate prediction and a full engagement with the international community. We will achieve this by working with the Met Office (project partners) and through international communications and collaborations that will build new links and develop the existing links of the investigators with operational centres (eg, ECMWF, DWD, Meteo France).