A Lagrangian Vertical Coordinate Dynamical Core for Global Atmospheric Modelling

Lead Research Organisation: University of Exeter
Department Name: Engineering Computer Science and Maths

Abstract

The weather forecasts and climate predictions produced by the Met Office are based on a sophisticated computer model that numerically solves the equations of atmospheric dynamics and thermodynamics. The equations are solved by dividing the Earth's atmosphere up horizontally into cells based on a latitude-longitude grid and vertically into a number of layers or levels. The current Met Office model, and its successor (currently under development, known as ENDGame), define these levels to be at specified heights above the Earth's surface. An alternative would be to use so-called Lagrangian vertical levels, which, by definition move up and down with the fluid. There is some evidence from experiments with computer models around the world that using Lagrangian vertical levels could better capture air mass properties and the transport of moisture and other constituents, and improve the representation of the budget of energy and other important thermodynamic properties such as entropy. This, in turn, could lead to more accurate weather forecasts and more realistic climate simulations and predictions. With this motivation, the aim of the project is to develop a version of ENDGame that uses Lagrangian vertical levels, and carefully assess the extent to which any of the potential benefits have been realized. One of the disadvantages of Lagrangian vertical levels is that, as they move with the fluid, they eventually tilt and fold until they can no longer represent the state of the atmosphere accurately. This folding can take many hours or days high up in the atmosphere but can be much quicker near steep mountains or areas of thunderstorm activity. Therefore, to be used in a computer model, Lagrangian levels must be periodically re-initialized and the model winds and thermodynamic fields remapped to the re-initialized levels. A major part of the proposed work will involve investigating the best way to re-initialize and remap, so that the benefits of using the Lagrangian levels are not degraded. The culmination of the project will be a careful comparison of the transport and conservation properties of the standard ENDGame model using its height-based levels and the new version using Lagrangian levels. Apart from the different vertical levels used, these two models will be as similar as possible; the comparison will therefore be a clean test of the benefits (or otherwise) of the use of Lagrangian levels. Such a clean comparison has not been carried out before. The results will provide valuable information to the Met Office and to other groups around the world developing computer models of the atmosphere. Close contact with the Met Office throughout this project will ensure that, if the Lagrangian vertical coordinate is successful, then it can be readily implemented in the operational version of ENDGame.

Publications

10 25 50
 
Description We have developed a computer model of the atmosphere whose vertical levels move with the flow. The expectation is that this will improve the conservation properties of the model.
Exploitation Route Several groups around the world are developing atmospheric models for research and forecasting. Our findings can inform their developments.
Sectors Environment

 
Description The findings are influencing the development of new atmospheric models for research and forecasting.
First Year Of Impact 2014
Sector Environment
Impact Types Economic

 
Description A scalable dynamical core for next generation weather and climate prediction - Phase 2
Amount £260,000 (GBP)
Funding ID NE/K006762/1 
Organisation Natural Environment Research Council 
Sector Public
Country United Kingdom
Start 02/2013 
End 01/2016
 
Title Lagrangian vertical coordinate ENDGame 
Description We have developed a version of the ENDGame atmospheric model dynamical core that uses a Lagrangian vertical coordinate. 
Type Of Material Computer model/algorithm 
Provided To Others? No  
Impact We are currently investigating whether the Lagrangian vertical coordinate version confers any benefits in terms of conservation properties, other measures of accuracy, or computational cost compared to the original height coordinate version. If there are benefits then this is likely to influence the design of future atmospheric model dynamical cores.