| description abstract | A comparison of near-global monthly mean surface temperature anomalies to those of global Microwave Sounding Unit (MSU) 2R temperatures for 1979?95 reveals differences in global annual mean trends that are shown to be largely attributable to important physical differences in the quantities that are measured. Maps of standard deviations of the monthly mean anomalies, which can be viewed as mostly measuring the size of the climate signal, reveal pronounced differences regionally in each dataset. At the surface, the variability of temperatures is relatively small over the oceans but large over land, whereas in the MSU record the signal is much more zonally symmetric. The largest differences are found over the North Pacific and North Atlantic Oceans where the monthly standard deviations of the MSU temperatures are larger by more than a factor of 2. Locally over land, the variance of the surface record is larger than that of the MSU. In addition to differential responses to forcings from the El Niño-Southern Oscillation phenomenon and volcanic eruptions, these characteristics are indicative of differences of the response to physical processes arising from the relative importance of advection versus surface interactions and the different heat capacities of land and ocean. The result is that the regions contributing to hemispheric or global mean anomalies differ substantially between the two temperature datasets. This helps to account for the observed differences in decadal trends where the surface record shows a warming trend since 1979 of 0.18°C per decade, relative to the MSU record. While a common perception from this result is that the MSU and surface measurements of global temperature change are inconsistent, the issue should not be about which record is better, but rather that both give a different perspective on the same events. | |
