Using ACCESS-rAM3 to investigate the impact of model resolution and key parameters on profound marine low cloud and top of atmosphere upward shortwave radiation bias in BARRA-C2.
Scientific justification:
We found profound marine low cloud and top of atmosphere upward shortwave radiation bias in BARRA-C2 (4.4km, RAL3.2) throughout the year, month, and daytime hours coompared to satellite observations. The bias undermines our confidence in high-resolution climate model simulations, which have been thought to be a promising avenue to reduce the uncertainty of climate sensitivity with better simulations of marine low clouds. While the convection is explicitly resolved rather than parameterised, other parameterisations, e.g. boundary layer, cloud microphysics, and a bimodal cloud scheme, may be linked to the model bias. Here we would like to run a set of sensitivity tests on the model resolution and key parameters to investigate the underlying mechanisms of the model bias.
How model design will address scientific justification:
We are planning three sensitivity tests first to investigate the role of model resolution. As shown in the attached figure, we will use ERA5 to drive a simulation over the domain d12km with a horizontal grid spacing of 12 km, which drives three simulations over the domain d4p4km, d4p4kms, and d1p1km. It will inform us the role of model resolution on the model bias.
Design:
Model: ACCESS-rAM3
Configuration: https://code.metoffice.gov.uk/svn/roses-u/d/q/7/0/0
Initial conditions: ERA5
Run plan: 3 simulations, each for 2 days
Total KSUs required: 3 x 2d x 10 KSU/d = 60 KSU
Time required: 3 * 2d * 6 hours/d = 36 hours
Total storage required: 3 x 2d x 750Gb/d = 4.4TB
