Clouds and Climate Change: The Thick and Thin of It
December 2000
The addition of heat-trapping greenhouse gases to the atmosphere by humans is certain to induce changes in Earth's climate. The extent and pace of climate change depends in part on the sensitivity of the climate to these perturbations. What makes this difficult to estimate is that as the climate begins to warm, the atmosphere reorganizes itself in ways that could either amplify or mitigate the original input of energy that initiated the climate change.
Clouds play a leading role in this real-life mystery. Clouds both reflect sunlight, which cools the Earth, and trap heat in the same way as greenhouse gases, thus warming the Earth. Different types of clouds do more of one than the other. The net effect of clouds on climate change depends on which cloud types change, and whether they become more or less abundant, thicker or thinner, and higher or lower in altitude.
Many people assume that since more water will evaporate from the oceans as the climate warms, it will be cloudier, with thicker and denser clouds. However, a warmer atmosphere needs more water vapor molecules to become saturated and to condense into clouds, so it is hard to anticipate exactly how clouds respond to human-induced climate perturbations. For example, although summer is warmer than winter, and the humidity is usually higher in summer, nevertheless the sky is not noticeably cloudier on average in summer than in winter.
Stratus and stratocumulus decks that make for gray, overcast days.
Can observations of the current climate give us clues about how clouds will change in the future? We can combine satellite observations, which look down on the atmosphere from above, with observations from surface-based instruments that look up at the sky from below, to see both sides of the cloud story. A few years ago, George Tselioudis, William Rossow and David Rind of GISS examined satellite data measuring the amount of sunlight reflected by low-level clouds, i.e., the stratus and stratocumulus decks that produce gray, overcast days and the puffy cotton-ball cumulus clouds that dot the sky on clearer days. Until then, people had assumed that these low clouds would be brighter and reflect more sunlight wherever the air was warmer, because they would be thicker and contain more liquid water. On the contrary, Tselioudis et al. found that over more than half the Earth, and especially in the warmer places (both in the tropics and in midlatitude summertime), these clouds were actually brighter when the air was cold than when it was warm.
Many people remained skeptical of this result because satellites observe the Earth from afar, and the interpretation of satellite data is not always straightforward. Recently, however, Anthony Del Genio and Audrey Wolf took a different approach, using Department of Energy data from a site in Oklahoma that observed the clouds by remotely sensing them from below and by sending weather balloons through them to make some direct measurements. They recorded the amount of liquid water in the clouds, the locations of their tops and bases, and examined how these vary in response to changes in atmospheric temperature and humidity.
Although we saw essentially the same behavior as the satellite, our additional data enabled us to understand how warm-air clouds can be less bright than their cold-air counterparts. The atmosphere above cold ground tends to remain calm, like a pot of water on a stove before the flame is turned on. Water vapor molecules evaporated into such air, or blown in from other places, can build up and form a thick layer of clouds that reflects sunlight well. However, the air above warm ground becomes turbulent, like a pot of hot or boiling water. The warm currents of rising air (the kind that gliders use to stay afloat) carry water vapor molecules up and away, preventing clouds from forming close to the ground. Instead, only a thin layer of not very reflective clouds develops a kilometer or two above the ground, at the altitude where rising air currents terminate.
What does this mean for climate change? Mao-Sung Yao, Tselioudis, Del Genio, and William Kovari used the GISS global climate model to predict changes in different types of clouds, and the sensitivity of the climate, to a doubling of carbon dioxide concentration. They found that low-level clouds in the model behaved much the same as anticipated from satellite and surface data. At midlatitudes, clouds became a bit thinner and less reflective in the simulated warmer climate. They became less reflective in the tropics as well, but for a different reason: clouds in the warmer climate lost more of their water due to greater rainfall. But other types of clouds did not behave in the same way. For example, the big "anvil" clouds that accompany thunderstorms at high altitudes became more extensive and brighter in the warmer climate, instead.
Since the changes in low- and high-level clouds mostly cancelled each other out, the net global effect of the clouds did not differ very much in the warmer climate scenario from that in today's climate. This scenario differs considerably from what many climate scientists had been assuming in the 1990s. It had been thought that brighter clouds would partly "save" us from significant global warming, by reflecting more energy into space. Instead, these results suggest that clouds are not necessarily the white knight that will rescue us from climate change. Therefore, our society should seriously consider reasonable steps to limit future emissions of greenhouse gases and soot aerosols as part of an overall strategy to reduce air pollution.
Reference
Del Genio, A.D., and A.B. Wolf 2000 The temperature dependence of the liquid water path of low clouds in the southern Great Plains. J. Climate 13, 3465-3486.
Yao, M.-S., and A.D. Del Genio 1999. Effects of cloud parameterization on the simulation of climate changes in the GISS GCM. J. Climate 12, 761-779.
Tselioudis, G., A.D. Del Genio, W. Kovari, and M.-S. Yao 1998. Temperature dependence of low cloud optical thickness in the GISS GCM: Contributing mechanisms and climate implications. J. Climate 11, 3268-3281. |
