AN EVALUATION OF MONTHLY MEAN MSU AND ECMWF GLOBAL ATMOSPHERIC TEMPERATURES FOR MONITORING CLIMATE

Monthly mean brightness temperature anomalies derived from channel 2 o f the microwave sounding units (MSUs) on board NOAA satellites over th e past decade are examined and compared with both weighted and pressur e-level ECMWF monthly mean temperatures for the 96 months of 1982-89. Very good agreement between the MSU and channel 2 weighted ECMWF anoma lies is found over most of the globe with correlation coefficients ove r 0.9, but the agreement falls off over the tropics, the South Atlanti c, and high latitudes of the Southern Hemisphere. The ECMWF analyses a gree best with the MSU data in regions of good radiosonde coverage, wh ile lower correlations are found in regions where the analyses depend more heavily on satellite data. Systematic errors introduced into the analyses by the retrieval techniques applied to the radiance data larg ely explain this apparent contradiction. Additionally, changes to the analysis-forecast system at ECMWF over the decade appear as apparent c hanges in climate, and these discontinuities most strongly affect the tropics and are evident in regions of fewer observations. To the exten t that the weighted ECMWF data agree with the MSU brightness temperatu res the vertical dependence of the MSU data can be examined. Correlati ons of the MSU data with ECMWF temperature anomalies at individual pre ssure levels are highest at 300 mb over the globe, a level apparently least affected by the frequent changes and improvements at ECMWF. Over regions of good data coverage, such as the Northern Hemisphere landma sses and Australia, the MSU anomalies correlate very highly with all l evels of the troposphere up to 200 mb. Thus, the MSUs appear to be an extremely useful tool for measuring global tropospheric temperature fl uctuations on a monthly and longer time scale. Problems in the ECMWF t emperature record since 1982 are examined in detail for the tropics. I n September 1982 the introduction of diabatic nonlinear normal-mode in itialization resulted in significant temperature increases in the trop ical middle troposphere, especially at 500 mb. In May 1985 tropical te mperatures at 700 mb (850 mb) increased (decreased) after the implemen tation of the T106 spectral model with major accompanying changes to p hysical parameterizations. Tropical temperatures near the tropopause d ecreased substantially after the May 1986 enhancement of the vertical resolution of the model from 16 to 19 levels with 3 new stratospheric levels. Problems at 1000 mb are present throughout the 96-month study period and are directly related to the manner in which analyzed temper atures are obtained at ECMWF. Comparisons with temperatures obtained f rom radiosonde stations in the tropics show that the ECMWF analyses ha ve clearly improved with time, especially after May 1985 after which t he 1000-200-mb temperatures show much greater coherence. These results show the importance of realizing the inherent problems with operation ally based gridded datasets, and they strongly support the need for re analysis of all data using a state-of-the-art four-dimensional data as similation system.