EXAMINATION OF GLOBAL ATMOSPHERIC-TEMPERATURE MONITORING WITH SATELLITE MICROWAVE MEASUREMENTS .1. THEORETICAL CONSIDERATIONS

In recent studies (Spencer and Christy, 1990; and Spencer et al., 1990 ) it is suggested that observations at 53.74 GHz made by the Microwave Sounding Unit (MSU), flown on NOAA operational weather satellites, ca n yield a precise estimate of global mean temperature and its change a s a function of time. Hansen and Wilson (1993) question their interpre tation of temporal changes on the grounds that the microwave observati ons could be influenced by the opacity of the variable constituents in the atmosphere. This issue has broad interest because of the importan ce of detection of global climatic change. In order to help resolve th is issue, in this study we utilize a radiative transfer model to simul ate: (a) the observations of MSU Channel 1 (Ch. 1) at 50.3 GHz, in the weakly absorbing region of the 60 GHz molecular oxygen absorption ban d; and (b) the observations of MSU Channel 2 (Ch. 2) at 53.74 GHz, in the moderately strong absorption region of the same band. This radiati ve transfer model includes extinction due to clouds and rain in additi on to absorption due to molecular oxygen and water vapor. The model si mulations show that, over the oceans, extinction due to rain and cloud s in Ch. 1 causes an increase in brightness temperature, while in Ch. 2 it causes a decrease. Over the land, however, both Ch. 1 and Ch. 2 s how a decrease in brightness temperature due to rain and cloud extinct ion. These theoretical results are consistent with simultaneous observ ations in Ch. 1 and Ch. 2 made by MSU. Based on theory and observation s we infer that a substantial number of the MSU observations at 53.74 GHz used by Spencer et al. contain rain and cloud contamination. As a result, their MSU derived global mean temperatures and long term trend is questionable.