Comparison of the 200 hPa circulation in CSM and CCM3 simulations and NCEP and211 ERA reanalysis: seasonal cycle and interannual variation

摘要

In this paper the monthly mean vorticity and divergence at 200 hPa are compared211u001efrom four data sources: The NCEP/NCAR reanalyses 1958 through 1994, the ECMWF 211u001e(ERA) reanalyses, 1979 through 1994, a NCAR CCM3 integration using prescribed 211u001eSSTs from 1979 through 1993, and the NCAR CSM 300 year integration. The NCEP, ERA 211u001eand CCM3 all provide data for the period 1979 through 1993. The timescales 211u001eexamined are the annual cycle and interannual variations. The annual mean 211u001evorticity of the ERA and NCEP match very closely. The annual cycle is likewise 211u001eclose except in the eastern equatorial Pacific and Indian Ocean. Compared to the 211u001ereanalyses, the models have adequate annual means but suffer in the depiction of 211u001ethe annual cycle in the regions of the jet maxima and in some regions of the 211u001eTropics. The CSM appears to inherit errors from the CCM3 and apparently add some 211u001enew ones. The annual mean divergence shows a much larger difference between the 211u001ereanalyses. This is most pronounced in the Tropics especially over the African 211u001eand South American land masses. The models also show large differences, with the 211u001eCSM being an outlier in the tropical Pacific. For many tropical and extratropical 211u001elocations even the annual cycle is not well defined between the NCEP and ERA 211u001ereanalysis. The NCEP, ERA, CCM3 and CSM agree with respect to the variance of the 211u001emonthly mean vorticity. The variance for low pass filtered data is too large in 211u001ethe ENSO regions for the CCM3, but too small for the CSM. Both models tend to 211u001eunderestimate the low frequency variance in midlatitudes. The ERA has 211u001esubstantially more monthly variance in the divergence than the NCEP data, 211u001eespecially over the tropical South America and Africa and the dateline. Both 211u001emodels have variance more on the order of the ERA, and have an anomalous maximum 211u001ein the eastern Indian Ocean, the CSM much more so. The CSM shifts the maxima in 211u001ethe equatorial Pacific from 180 seen in the reanalyses to 150E. If anything the 211u001eCCM 3 appears to be too sensitive to SST anomalies, which exacerbates the poor 211u001eocean simulation in the tropical Pacific. There are errors in the CCM3 211u001eintegration which foreshadow, deficiencies in the CSM integration, so the ocean 211u001eis not solely at fault.