« PreviousContinue »
(Hurrell and Trenberth, 1997; Christy et al., 1997). There are two main issues raised by all of these measurements: 1) are the satellite and balloon measurements of slight cooling incompatible with global warming predictions?, and 2) is the disagreement between surface and deep layer temperatures incompatible with the physics contained in current GCM's? I will deal mainly with the latter question.
Even the seemingly small level of disagreement between surface and deep layer temperature trends can not be reproduced by current GCM's, which slave globally-averaged deep tropospheric temperatures to any surface temperature fluctuations (e.g. Hurrell and Trenberth, 1997). However, we know that convective air currents and the air-conditioning effects of water cause the atmospheric temperature profile to change by a much larger amount, about 80 deg. C, between the surface and 10 km altitude when compared to a hypothetical Earth without these processes (Manabe and Strickler, 1964). Thus, it is critical to validate whether the parameterizations of moist convective processes in GCM's are realistic enough to capture the ways in which the tropospheric temperature profile fluctuates naturally in response to this transport of heat from the surface to the upper troposphere.
summarize one of our research efforts in which we are analyzing new satellite measurements of middle and upper tropospheric water vapor from the SSM/T-2 instrument flying on the Defense Meteorological Satellite Program satellites.
The Role of Water Vapor in Global Warming
The water vapor in GCM's increases in response to the warming produced by doubling of carbon dioxide, causing a "positive water vapor feedback" on temperature. All GCM's produce values of water vapor feedback that are so similar to one another that the modelers view this agreement as evidence for its validity. This view has been further bolstered by several observational studies that have proported to validate the positive water vapor feedback paradigm
(Rind et al., 1991; Ramanathan and Collins, 1991; Soden and Fu, 1995). Unfortunately, these studies neglected the role that tropospheric circulation systems have in redistributing water vapor spatially. Water vapor feedback can only be meaningfully evaluated over entire circulation systems, such as the entire tropical half of the Earth, or preferably over the whole Earth (Lau et al., 1996; Spencer and Braswell, 1997). Thus, those studies did not validate positive water vapor feedback. They merely observed that the warm, ascending branches of tropospheric circulation systems have more water vapor than the cooler, descending branches. This will always be true, even in the case of negative water vapor feedback.
It is my contention that the physics contained in GCM's are still insufficient to have much confidence in the predicted magnitude of global warming, partially because the cloud microphysical processes which help determine water vapor feedback are still poorly understood. A statement from the body of the 1995 Intergovernmental Panel on Climate Change (IPCC) report also supports this view:
"Feedback from the redistribution of water vapour remains a substantial
To appreciate the water vapor feedback problem, it is useful to recognize that we generally associate a warmer surface with higher water vapor contents of the air above that surface. This positive relationship typically occurs in the turbulent boundary layer, which occupies only the bottom 1-2 km of the troposphere. However, it is the humidity structure of the entire depth of the atmosphere that determines water vapor feedback, and thus the magnitude of global warming. Above the boundary layer, in the free troposphere, other less well understood processes control the humidity of the air. The primary moistening process in this layer is evaporation of clouds (Fig. 2).
In the tropics, and at higher latitudes during the warm season, these clouds are usually deep convective in nature, with strong updrafts of warm moist air in their cores. These cloud processes are not explicitly represented in GCM's, due to their very small spatial scale compared to the scales resolved in the models. Instead, their effects on the atmosphere are parameterized. The same model limitations hold true for the smaller stratus and cumulus clouds that form near the top of the boundary layer (Fig. 2).
What is the process that controls how much water vapor is supplied to the environment from rainfall systems? The answer is cloud water. Some portion of all of the cloud water
Fig. 2. Schematic of a typical warm season tropospheric circulation system, with its cycling of water through evaporation, cloud formation, rainfall, and cloud re-evaporation. These rainfall systems control the supply of water vapor to the cloud-free air, whose efficiency at cooling the Earth increases nonlinearly as the humidity decreases.
contained in a rain system will fall back to the surface as precipitation, while the rest evaporates into the air. This evaporated cloud condensate constitutes the primary humidity source for the free troposphere. The fraction of the total condensed water that falls out as rain is called the "precipitation efficiency (e)" of the rain system. If all water falls out as rain, & =1. If none of it falls out as rain, 8=0. A typical rain system has an efficiency near 0.5. However, these cloud formation, rain-out, and dissipation processes are only crudely represented in GCM's at the present time. It has been shown theoretically that high rainfall efficiency leads to a cool and dry climate, while low rainfall efficiency leads to a warm, moist climate (Renno et al., 1994), through its control of humidity.
The important question for water vapor feedback and thus climate change is not so much what the current rainfall efficiency is, but how it changes as greenhouse gas concentrations increase. If the efficiency decreases during warming, then we have a potential negative water vapor feedback, which could mitigate, rather than enhance, the small amount of direct warming due to carbon dioxide increases. There is some evidence that precipitation efficiency increases with temperature. For instance, we know that warm tropical precipitation systems are more efficient than cold, high latitude systems. The observation that heavy rain can fail out of rather shallow tropical clouds is evidence for this. Other processes controlling precipitation efficiency are not well understood, however. We have embarked on an investigation of these issues with a high-resolution cloudresolving model run in a climate mode in order to address some of the deficiencies in the GCM's.
Satellite Measurements of Upper Tropospheric Humidity
New water vapor measurements of the middle and upper troposphere are now being derived from the SSM/T-2 water vapor sounder on the DMSP satellites. These microwave measurements have the advantage of being able to sense through most cirrus clouds. These new measurements reveal that vast stretches of the tropical upper troposphere (averaged over the 6-12 km altitude range) are very dry, with half of the tropics being covered by relative humidities below 20% on a daily basis (Spencer and Braswell, 1997). Individual layer-average humidities fall as low as 2%. These low humidities are extremely important, for they provide radiative exhaust ports through which the Earth can lose radiant (infrared) energy most efficiently. Furthermore, their ability to lose energy increases nonlinearly as the humidity falls (Lindzen, 1995). Pierrehumbert (1995) argues that these dry zones are the controlling influence on the average climate of the tropics.
As discussed above, rainfall systems have an important influence on the humidity of tropospheric dry zones, with a small change in humidity causing a potentially large change in the trapping of infrared energy. If the average temperature of the tropics is indeed sensitive to small humidity fluctuations, why does the tropical atmosphere remain at so stable a temperature? I continue to believe the basic reason is that feedbacks in the climate system are predominately negative, causing any perturbation from the average state to be damped out, and restoring the system back to a stable temperature. In this context, it is interesting to note that the average value of positive cloud feedback in 19 GCM's fell by 71% between 1990 and 1995 (Cess et al., 1996; IPCC, 1995). As the physical processes in GCM's continue to be improved, and as faster computers allow more detailed physics to be incorporated, I expect that the predicted magnitude of global warming will continue to fall, as it has since the first IPCC assessment in 1990. In the mean time, observations from NASA's Earth Observing System AM and PM satellites will provide new information on upper tropospheric vapor, cloud, and radiation processes that will be critical for advancing our understanding of climate variability and improving the physical processes contained in the GCM's.
Cess, R.D., and 33 others, 1996: Cloud feedback in atmospheric general circulation models: An update. J. Geophys. Res., 101, 12791-12794.
Christy, J.R., R.W. Spencer, and W.D. Braswell, 1997: How accurate
"thermometers"? Nature, 25 September.
Hurrell, J.K., and K.E. Trenberth, 1997: Spurious trends in the satellite MSU temperature record arising from merging different satellite records. Nature, 13 March
Intergovernmental Panel on Climate Change, 1995: Climate Change 1995: The Second IPCC Assessment. University Press, Cambridge, Great Britain, 572 pp.
Lau, K.-M., C.-H. Ho, and M.-D. Chou, 1996: Water vapor and cloud feedback over the tropical oceans: Can we use ENSO as a surrogate for climate change? Geophys. Res. Lett., 23, 2971-2974.
Lindzen, R.S., 1995: The importance and nature of the water vapor budget in nature and models. In Climate Sensitivity to Radiative Perturbations: Physical Mechanisms and their Validation, NATO ASI Series 1: Global Environmental Change, 34, H. Le Treut (editor), Springer-Verlag, Deidelberg, 331pp.
Manabe, S., and R.F. Strickler, 1964: Thermal equilibrium of the atmosphere with a convective adjustment. J. Atmos. Sci., 21, 361-385.
Pierrehumbert, R.T., 1995: Thermostats, radiator fins, and the local runaway greenhouse. J.
Atmos. Sci., 52, 1784-1806.
Ramanathan, V., and W. Collins, 1991: Thermodynamic regulation of ocean warming by cirrus clouds deduced from observations of the 1987 El Nino. Nature, 351, 27-32. Renno, N. O., K.A. Emanuel, and P. H. Stone, 1994: Radiative-convective model with an explicit hydrologic cycle 1. Formulation and sensitivity to model parameters. J. Geophys. Res., 99, 14,429-14,441.
Rind, D., E.W. Chiou, W. Chu, J. Larsen, S. Oltmans, J. Lerner, M.P. McCormick, and L. McMaster, 1991: Positive water vapour feedback confirmed in by satellite data. Nature, 349, 500-503.
Soden, B. J., and R. Fu, 1995: A satellite analysis of deep convection, upper tropospheric humidity, and the greenhouse effect. J. Climate, 8, 2333-2351.
Spencer, R.W., and J.R. Christy, 1992: Precision and radiosonde validation of satellite gridpoint
temperature anomalies, Part I: MSU channel 2. J. Climate, 5, 847-857.
Spencer, R.W., and W.D. Braswell, 1997: How dry is the tropical troposphere? Implications for
global warming theory. Bull. Amer. Meteor. Soc., 78, 1097-1106.
Sun, D.-Z. and R.S. Lindzen, 1993: Distribution of tropical tropospheric water vapor. J. Atmos.
Sci., 50, 1643-1660.