The global-mean temperature at the earth's surface is estimated to have risen by 0.25 to 0.4 °C during the past 20 years. On the other hand, satellite measurements of radiances indicate that the temperature of the lower to mid-troposphere (the atmospheric layer extending from the earth's surface up to about 8 km) has exhibited a smaller rise of approximately 0.0 to 0.2 °C during this period. Estimates of the temperature trends of the same atmospheric layer based on balloon-borne observations (i.e., radiosondes) tend to agree with those inferred from the satellite observations. The panel was asked to assess whether these apparently conflicting surface and upper air temperature trends lie within the range of uncertainty inherent in the measurements and, if they are judged to lie outside that range, to identify the most probable reason(s) for the differences.
To address these questions the panel had to consider:
• the factors that contribute to uncertainties in the trends inferred from three categories of instrumental measurementsMicrowave Sounding Units (MSU) carried aboard National Oceanic and Atmospheric Administration (NOAA) satellites, radiosondes, and surface observations;
• the technical issues involved in making comparisons between global-mean temperature trends derived from measurements with differentcontinue
physical characteristics, different spatial and temporal sampling characteristics, and different error characteristics;
• the impact of the recent corrections to the algorithms for processing measurements derived from the MSU to account for satellite drifting and changes in instrument response;
• the contribution of natural climate variability to decade-to-decade climate changes, including changes in the atmosphere's vertical structure associated with natural variability;
• the changes in the atmosphere's vertical structure associated with human-induced climate changes; and
• the results of recent climate model simulations of temperature trends that take into account both natural variability and human-induced forcing.1
In the opinion of the panel, the warming trend in global-mean surface temperature observations during the past 20 years is undoubtedly real and is substantially greater than the average rate of warming during the twentieth century. The disparity between surface and upper air trends in no way invalidates the conclusion that surface temperature has been rising. The recent corrections in the MSU processing algorithms (referred to above) bring the global temperature trend derived from the satellite data into slightly closer alignment with surface temperature trends, but a substantial disparity remains. The various kinds of evidence examined by the panel suggest that the troposphere actually may have warmed much less rapidly than the surface from 1979 into the late 1990s, due both to natural causes (e.g., the sequence of volcanic eruptions that occurred within this particular 20-year period) and human activities (e.g., the cooling of the upper part of the troposphere resulting from ozone depletion in the stratosphere). Regardless of whether the disparity is real, the panel cautions that temperature trends based on data for such short periods of record, with arbitrary start and end points, are not necessarily indicative of the long-term behavior of the climate system.
Reducing uncertainties in the evaluation of the trends will require: (1) implementing an improved climate monitoring system designed to ensure the continuity and quality of critically needed measurements ofcontinue
1 A climate forcing is a perturbation to the energy balance of the earth-atmosphere system and may bring about climate change.
temperature, other climatic variables, and concentrations of aerosols and trace gases; and (2) making raw and processed atmospheric measurements accessible in a form that enables a number of different groups to replicate and experiment with the processing of the more widely disseminated data sets such as the MSU tropospheric temperature record. A number of possible research strategies for improving the understanding of uncertainties inherent in the various measurement systems and the relationship between surface and upper air temperature trends are proposed in the report.break