The simulation of the Asian summer monsoon, and its sensitivity to horizontal resolution, in the UK Meteorological Office Unified Model

The quality of the simulation of the Asian summer monsoon in the climate version of the UK Meteorological Office's Unified Model, and the impact upon this of increased horizontal resolution, is investigated using two atmosphere-only model runs forced with observed sea surface temperatures (SSTs) and sea ice extents. The runs each cover the period from 1979 to 1988, but have different horizontal resolutions, with one at climate resolution (2.5 degrees latitude by 3.75 degrees longitude, about 300 km at midlatitudes) and the other at global forecast model resolution (0.833 degrees latitude by 1.25 degrees longitude, approximately 100 km at midlatitudes). The characteristic monsoon circulation and the spatial distribution of precipitation are in good agreement with observations. However, the model has a tendency to overestimate the strength of the monsoon, and also exhibits an early monsoon onset. The large-scale interannual variations in circulation appear to be simulated reasonably well (as far as can be determined using this short dataset), although the magnitude of the interannual variability of precipitation is overestimated. However, the regional circulation and precipitation changes between El Nino and La Nina years show some significant differences between the model and the observations. The dominant mode of intraseasonal variability seen in both model simulations is, in agreement with observations, associated with the active/break cycle of the monsoon (although this only explains about 10% of the total variance in both simulations). There is some evidence that the SST changes associated with El Nino may produce a coherent forcing of the secondary (east-west) mode of intraseasonal variability during the onset phase of the monsoon in the model. However, comparison with observations suggests that this may not be representative of what occurs in the real atmosphere. There is no evidence that the SST variations are causing the system to prefer either the active or the break monsoon phase, as was suggested by Palmer. With increased model horizontal resolution, extra detail is provided in the precipitation distribution, but the mean monsoon simulation is scarcely altered and the systematic errors remain. The interannual variations in circulation and precipitation appear not to be greatly altered, and the overall pattern of intraseasonal variability is also unaffected. This study suggests that the systematic errors in the monsoon simulation are not a result of poor horizontal resolution, but may be due to problems with the model physics.