A technical comment: Analysis of ‘examination of “global atmospheric temperature monitoring with satellite microwave measurements”’

The potential for residual hydrometeor contamination effects in the global temperature time series produced by Spencer and Christy from MSU channel 2 (MSU2) data has been addressed by Prabhakara et al. (1995, 1996). They use tropical oceanic MSU channel l (MSU1) data to estimate the hydrometeor effects on MSU2. We present several lines of evidence to show that their technique greatly overestimates the hydrometeor effects on MSU2. This overestimation is due to the faulty assumption that the hydrometeors that cause MSU1 warming are the same as (or always exist with) the hydrometeors that cause cooling in MSU2. Instead, the hydrometeors responsible for MSU1 warming are liquid phase, while those responsible for MSU2 cooling are large ice particles. Because liquid phase clouds are much more widespread than the large-ice portions of deep convective systems, their method greatly overestimates the areal coverage of contaminated tropical MSU2 data. In addition, we show that the convective screening procedure of Spencer and Christy removes the negative correlation between MSU1 and MSU2 their conclusions rest upon. Radiosonde validation of monthly tropical MSU2 anomalies over the tropical West Pacific also support these conclusions.