A high resolution (0.833 degrees latitude by 1.25 degrees longitude) AMIP (
Atmospheric Model Intercomparison Project) integration is compared with a c
ontrol integration at standard resolution (2.5 degrees by 3.75 degrees). Bo
th integrations use HadAM2b, a recent version of the Hadley Centre atmosphe
ric general circulation model. ECMWF reanalysis data for the AMIP period (1
979-1988), together with other climatologies, are used to evaluate the resu
lts. An additional integration at standard resolution using the high resolu
tion shorter time step is used to help distinguish between changes due to m
odel resolution and those which are due to time step dependencies in the ph
ysical parametrizations. Enhanced resolution increases the vertical motion,
intensifies the hydrological cycle, reduces slightly the model's cold bias
in the troposphere, shifts the westerly jets poleward and tends to increas
e the eddy kinetic energy and variability of the model. The high resolution
simulation has less mid-latitude cloud, so altering the radiation balance.
There is no evidence to suggest that increasing resolution has an impact o
n the model's response to SST forcing.