Stratospheric water vapor decline credited with slowing global warming

After a steep rise in global average temperatures in the 1990s, the 2000s have seen relatively flat temperatures, despite increasing levels of CO2 emissions by humans. This reduced warming may be partially due to a sharp decrease in stratospheric water vapor that began after 2000, according to research published yesterday in Science by a team of researchers led by Dr. Susan Solomon of NOAA's Earth System Research Laboratory in Boulder. Water vapor is a potent greenhouse gas capable of significantly warming the planet, and its potency is much higher when it is located in the lower stratosphere where temperatures are extremely cold. Greenhouse gases located in cold regions of the atmosphere are more effective at heating the planet because they absorb heat radiation from the Earth's relatively warm surface, but then re-emit energy at a much colder temperature, resulting in less heat energy lost to space. Even though stratospheric water vapor can exist at concentrations more than 100 times lower than at the surface, the 10% drop in stratospheric water vapor since 2000 noted by Solomon et al. acted to slow down global warming by 25% between 2000 - 2009, compared to that which would have occurred due only to carbon dioxide and other greenhouse gases.


Figure 1. Stratospheric water vapor in the tropics, between 5°S - 5°N, as measured by the HALOE instrument on the Upper Atmosphere Research Satellite (UARS), between 1993 - 2005. The bottom portion of the plot shows the lower stratosphere, just above where tall thunderstorms are able to transport water vapor into the stratosphere. A strong yearly cycle is evident in the water vapor, due to the seasonal variation in heavy thunderstorms over the tropics. Once in the lower stratosphere, the waver vapor takes about 1.2 years to travel to the upper stratosphere, as seen in the bending of the contours to the right with height. Note that beginning in 2001, very low water vapor concentrations less than 2.2 parts per million by volume (ppmv) began appearing in the lower stratosphere, due to substantial cooling. Image credit: Rosenlof, K. H., and G. C. Reid (2008), Trends in the temperature and water vapor content of the tropical lower stratosphere: Sea surface connection, J. Geophys. Res., 113, D06107, doi:10.1029/2007JD009109.

The observations
We haven't been able to observe water vapor in the stratosphere very long--accurate global measurements only go back to 1991, when the HALOE instrument aboard the Upper Atmosphere Research Satellite (UARS) began taking data (Figure 1). Stratospheric water vapor showed an increase of about 0.5 parts per million by volume (ppmv) during the 1990s. But after 2000, a sudden drop of 0.4 ppmv was observed, and this decrease has persisted into 2009. To see how these changes impacted the amount of global warming, Solomon et al. fed the observations into a specialized high-resolution model that computed the change in heat from the fluctuating water vapor levels. They found that the increase in stratospheric water vapor in the 1990s led to about a 30% increase in the amount of global warming observed during that decade, and the decrease of 0.4 ppmv since 2000 led to a 25% reduction between 2000 - 2009.

How water vapor gets into the stratosphere
The stratosphere has two main sources of water vapor: transport from the lower atmosphere (the troposphere) via tall thunderstorms, and the chemical breakdown of methane gas into water vapor and carbon dioxide. With regard to greenhouse effect warming, transport of water vapor by thunderstorms is the most important source, since this mechanism delivers water vapor to the lowest part of the stratosphere, where temperatures are coldest and greenhouse gases are more effective at warming the climate. There is a limit as to how much water vapor that can enter the stratosphere via thunderstorms, though. Temperature decreases with altitude from the surface to the bottom of the stratosphere, where they begin to rise with height due to the solar energy-absorbing effect of the stratospheric ozone layer. As moisture-laden air rises in thunderstorms towards the lower stratosphere, it encounters the atmosphere's "cold point"--the coldest point in the lower atmosphere, at the base of the stratosphere. Since the amount of water vapor that can be present in the atmosphere decreases as the temperature gets colder, and moisture being transported to the stratosphere must traverse through the "cold point" of the atmosphere, the air gets "freeze dried" and loses most of its moisture.


Figure 2. The departure from average of tropopause temperature (dark line) and Sea Surface Temperature (light dashed line) for the tropical Pacific Ocean between 10°S - 10°N, from 1981 - 2007. The tropopause is the bottom boundary of the stratosphere. The SST data is for 139°W - 171°W longitude, and is from the NOAA Optimal Interpolation v2 data set. The tropopause data is from balloon soundings, for the region 171°W - 200°W. The SST is plotted so that the anomalies increase as one looks down. Note that prior to about 2000, tropopause temperatures and SSTs increased and decreased together, but that beginning in 2000 - 2001, a sharp climate shift occurred, and the two quantities became anti-correlated. The sudden drop in tropopause temperature in 2000 - 2001 caused a sharp drop in stratospheric water vapor. Image credit: Rosenlof, K. H., and G. C. Reid (2008), Trends in the temperature and water vapor content of the tropical lower stratosphere: Sea surface connection, J. Geophys. Res., 113, D06107, doi:10.1029/2007JD009109.

Why did stratospheric water vapor drop in 2000?
Tall thunderstorms capable of delivering water vapor into the stratosphere occur primarily in the tropics, particularly over the Western Pacific, where a huge warm pool with very high Sea Surface Temperatures (SSTs) exists. In 2000, this region experienced a sharp increase in SST of 0.25°C, which has remained consistent though the 2000s (Figure 2). Coincident with this increase in SST came a sharp drop in the "cold point" temperature of the tropopause--the lower boundary of the stratosphere. This reduction in "cold point" temperature meant that less water vapor could make it into the stratosphere over the Tropical Pacific, since more thunderstorm water was getting "freeze dried" out. Did global warming trigger this increase in Pacific SST, resulting in cooling of the "cold point" and less water vapor in the stratosphere? Or was it random variation due to some decades-long natural cycle? This key question was left unanswered by the Solomon et al. study, and observations of stratospheric water vapor don't go back far enough to offer a reasonable guess. One factor arguing against global warming having triggered a negative feedback of this nature is that prior to 2000, increases in Western Pacific SST caused increases in "cold point" temperatures--behavior opposite of what has been seen since 2000.

If global warming has triggered the decrease in stratospheric water vapor seen since 2000, it could mean that the climate models have predicted too much global warming, since they don't predict that such a negative feedback exists. On the other hand, if this is a natural cycle, we can expect the recent flattening in global temperatures to average out in the long run, with a return to steeper increases in temperature in the coming decades. Climate models currently do a poor job modeling the complex dynamics of water vapor in the stratosphere, and are not much help figuring out what's going on. Complicating the issue is the fact that about 15% of all thunderstorms capable of delivering water vapor into the stratosphere are generated by tropical cyclones (Rosenlof and Reid, 2008), and tropical cyclones are not well-treated by climate models. We also have to factor in the impact of stratospheric ozone loss, which acts to cool the lower stratosphere. This effect should gradually decrease in future decades as CFC levels decline, though. The stratosphere is a devilishly complicated place that can have a significant impact on global climate change, and we are many years from understanding what is going on there.

References
Romps, D.M., and Z. Kuang, "Overshooting convection in tropical cyclones", Geophysical Research Letters, 2009; 36 (9): L09804 DOI: 10.1029/2009GL037396

Rosenlof, K. H., and G. C. Reid (2008), Trends in the temperature and water vapor content of the tropical lower stratosphere: Sea surface connection, J. Geophys. Res., 113, D06107, doi:10.1029/2007JD009109.

Portlight Haiti update
Paul Timmons, who directs the Portlight.org disaster-relief charity that has sprung up from the hard work and dedication of many members of the wunderground.com community, was interviewed by NBC yesterday. The reporter doing the story is planning to follow the Portlight-donated goods to Haiti and interview the people with disabilities that receive the donations. It is uncertain when the story will be aired, but I'll try to give everyone a heads-up.

Next post
My next post will probably be Tuesday (Groundhog's Day), when I plan to discuss Phil's forecast for the rest of winter. I'll throw in my two cents worth, too.

Jeff Masters