Benefit of high resolution models for climate predictions.

Atmospheric climate models have increased in resolution over several decades since their inception and now typically run at 50-100km resolution globally. However, the representation of observed climate has not yet fully converged and errors remain. Several fluid dynamical mechanisms have been proposed for what missing effects remain at current model resolutions and this is an active research area as computing power grows and permits yet higher resolution.

One such process is ‘eddy feedback’, whereby small scale eddies and waves feedback positively on large-scale features in the atmospheric circulation. Such a feedback is thought to be important in maintaining the Atlantic jet stream, for example. This process occurs through eddy fluxes of momentum and vorticity in the atmosphere, but recent studies suggest it is too weak in current general circulation models when run at climate resolution. This project will quantify these deficits using a range of climate model resolutions and investigate its implications for long-range forecasting and climate change. This project will test for a greater climate response to imposed conditions in the sea surface temperature and sea ice cover when climate model resolution is increased. The project will investigate eddy feedback in regional climate simulations using our latest high-resolution global models (~12km resolution). It will first make use of existing simulations, but there is also scope for the student to perform new simulations to further test ideas in collaboration with CASE partner scientists at the Met Office. Measures for determining the strength of eddy feedback will be calculated and examined, providing an opportunity to gain expertise in this important and poorly understood area of climate dynamics. It is anticipated that the project will focus particularly on the importance of model resolution for simulating the effects of sea surface temperature and sea ice anomalies on the Atlantic jet stream. The potential implications of this work are large because seasonal, decadal and perhaps even longer climate predictions could all be affected by a lack of eddy feedback in current climate models. With support from the supervisors, the candidate will have the opportunity to shape the project design to focus on aspects of the problem of most interest to them. One area of potential study could be the atmospheric response to Arctic sea-ice loss. The project could perform very high-resolution simulations with perturbed sea ice and compare these to ‘standard’ resolution experiments being run elsewhere as part of the international Polar Amplification Model Intercomparison Project.

Grant reference
2415542
Funder
Natural Environment Research Council
Total awarded
£0 GBP
Start date
30 Sep 2020
Duration
3 years 6 months 1 day
End date
31 Mar 2024
Status
Active