Water, Water Everywhere but a Little Less in the Stratosphere
by Bill Chameides | February 4th, 2010
posted by Erica Rowell (Editor)
A new paper suggests stratospheric water vapor may have slowed global warming in the naughties.
The long-term trend in global temperatures is pretty clear — it’s upward. With the exception of the 1950s, every decade of the last 100 years has been warmer, on average, than the preceding one. And despite lots of specious pronouncements that global warming stalled in the decade of the naughties (2000–2009) — which I’ll refer to as the ’00s or the 2000s from here on out — global average temperatures during that time period were about 0.3 degrees Fahrenheit warmer than they were in the ‘90s.
Trend in average global temperature anomalies since 1900. Data obtained from the U.S. National Oceanic and Atmospheric Administration’s National Climatic Data Center. The raw data and full methodology used to compute the global averages are available here: http://www.ncdc.noaa.gov/oa/climate/research/anomalies/index.html#mean.
Still, temperature variations during the ’00s are puzzling. If you take a close look at the graphic, you will see that much of the temperature increase between the 2000s and the ‘90s occurred near the transition between the two decades. Temperatures during the 2000s were relatively flat despite the fact that greenhouse gases during the decade continued to increase. (It’s important to note that such behavior is not anomalous — compare the ‘80s to the ‘70s and the ‘90s to the ‘80s.)
The puzzle is this: If greenhouse gas concentrations increased each year during the decade, increasing the radiative forcing that drives global warming, why didn’t temperatures also increase each year?
Sussing Out Short-term Climate Variability from Longer Term Trends
And that brings us into the not particularly well-understood realm of short-term climate variability. The fact is that greenhouse gases are not the only drivers of climate change, and so we should not expect to see a perfect correlation between greenhouse gas concentrations and global temperatures from year to year.
Fine, you might say, but what are those other climate drivers, and was their behavior consistent with the temperature trend in the ’00s? Well, we have discussed two in past posts:
(i) solar variations and
(ii) the El Nino/Southern Oscillation (ENSO).
And it turns out that both of these acted during the ’00s in a way that would slow global warming and thus provide at least a partial explanation for the flat temperatures of the decade.
While the last solar maximum around the year 2000 likely provided a little oomph to 1998′s record-making warmth, the period of declining solar activity that followed probably helped contribute to cooler temps during the rest of the decade.
With regard to solar variations: The Sun reached the maximum of its 11-year cycle around 2000, and its activity was in decline for the rest of the decade.
With regard to ENSO: For much of the later part of the decade, the South Pacific Ocean was stuck in either a mild or negative (la Nina) phase of ENSO the later generally favoring cooler global temperatures.
New Paper Shows Water Vapor as a Significant Driver in Last Decade’s Climate
Last week in a paper published in the journal Science, Susan Solomon of the Earth System Research Laboratory at the National Oceanic and Atmospheric Administration and colleagues provided an additional explanation for the decade’s climate.
The authors noted that water vapor concentrations in the stratosphere were observed to have declined in the beginning of the decade, and because water vapor is a greenhouse gas, this would have had a cooling effect on the climate. (The decline in stratospheric water vapor should not be confused with the trend in water vapor in the lower atmosphere or troposphere, which has been positive because of the link between tropospheric water vapor and atmospheric temperatures.)
Solomon and her colleagues estimate that the decrease in stratospheric water vapor would have canceled out about 25 percent of the warming from the increase in greenhouse gases alone over the decade, thus providing a partial explanation for the flat temperatures of the 2000s.
Of course Solomon et al.’s work begs the question: what caused water vapor concentrations in the stratosphere to decrease? The short answer is that we are not sure.
Water vapor gets into the stratosphere from the breakdown of methane and by large (primarily tropical) storms that extend into the stratosphere. There are no mechanisms that could explain the water vapor change from methane, so more than likely the water vapor change is linked to a change in the intensity of storms.
Some Answers, Yes, but Still Lots of Questions
But what could have caused the change in storms? Was the change a response to climate change (and therefore a climate feedback)? Or was it unrelated to warming (and therefore a manifestation of internal climate variability)? These are good questions but not ones we are in a position to answer yet.
The Solomon et al. paper is an important contribution — it adds to our understanding of why global temperatures fluctuate (or not) from year to year. It also aptly illustrates that all of climate science is most definitely not “settled.” We have an awful lot to learn.
But nothing in the Solomon et al. paper or related works on temperature trends during the past decade undermines the fundamental fact that on decadal time scales global temperatures are on the rise nor does it undermine the multiple lines of evidence linking warming to emissions of greenhouse gases.
filed under: climate change, faculty, global warming, temperaturesand: climate, research, solar cycle, solar maximum, solar minimum, solar variation, storms, sunspots, Susan Solomon, water vapor
6 Comments
All comments are moderated and limited to 275 words. Your e-mail address is never displayed. Read our Comment Guidelines for more details.
Connect With Us
about TheGreenGrok
In The Green Grok, Dean Bill Chameides elucidates causes and potential remedies for environmental change and identifies pathways towards a more sustainable future.
Featured Videos
EVs from Chevy and Tesla
Sustainable Food Shopping
For every mol of CO2 released by the combustion of fossil fuels a mol of H2O is also released. The oxidation of organic carbon (“CH2O”) uses one mol of O2 to yield one of CO2 and one of water… basically respiration, the reverse of photosynthesis. Modern jet traffic in the upper troposphere is not insignificant (or maybe it is?)… watch this 2-minute video… http://www.youtube.com/watch?v=z1US_4uf4YE This atmospheric oxidation is releasing both CO2 and water vapor well away from the Earth’s surface, and most of it in the northern hemisphere. Has this been factored into the models?
Ken, These have been factored into the models. The biggest impact of jet traffic is the warming effect from the generation of contrails.
In a study published in 2005, James Hansen of NASA’s Goddard Institute for Space Studies in New York, and colleagues ran models that increased the contrail coverage in Minnis’s study by a factor of five. Even with this significant increase, Hansen’s team found that global mean temperature change was in the neighborhood of 0.03°C (0.05°F)âa minute amount. “Aircraft are likely to be a significant factor in future climate, but probably not via their contrails,” Hansen told me. “I think our main concern about aircraft will be their CO2 emissions, not contrails, which are a pretty small climate forcing.” http://www.pbs.org/wgbh/nova/sun/contrail.html
The NOAA.gov data referenced are anything but “raw”. Read the Full methodology. Raw temperatures are those actually measured directly, not those that have been gridded, boxed, weighted, homogenized and winsorized. The temperature anomaly chart, like most, shows no zero-level mean annual temperature associated with it. NASA-GISS and HadCRUT use 14°C (57.2°F) whereas NOAA uses 15°C (59°F). Most of the literature prior to the ’80s used 15°C and the decade from 1931-1940 was an above average warming decade. The anomaly charts back before the 80s showed temperature anomalies for that decade as high as +0.6 = 15.6°C. See, e.g, Budyko, Fig. 1 (1969) Mitchell Fig. 1 (1975), NAS Report, Fig. A6 (1975)….all summarized here: copy and paste: http://revolution2.us/content/docs/global_cooling/614-615.html The plus 0.6 anomaly around 1940 is missing from the chart figured (as it is from most others today) and the anomalies prior to ~1930 are all shown as negative. Mitchell’s Fig. 1 shows all of these anomalies as positive while the Natl. Geo chart shows them as near normal and similar to the 70s. Question: What was the peak absolute mean temperature in the 1931-1940 decade? Was it 14°C plus 0.6 or was it 15°C plus 0.6? Either way, before all of the “methodology” it WAS essentially the same as the last decade.
Ken, 1. By following the links provided in the post you can get pretty close to the actual raw data. For example the GHCN link gets you to a century-scale dataset consisting of monthly surface observations from ~7,000 stations from around the world. http://www.ncdc.noaa.gov/oa/climate/ghcn-monthly/index.php 2. The temperature anomalies in the chart were taken from the NCDC datasets as indicated, and if you had gone to the NCDC Web site, you would learn that the baseline was the 20th century. 3. Yes, you are right that different groups use different assumptions to calculate their long-term trends which explains the variation in the mean temperature anomaly. In spite of this, there is good agreement between agencies highlighting the robustness of the trends. http://www.wmo.int/pages/mediacentre/press_releases/images/NewImage.PNG 4. The anomaly charts you reference are for the Northern Hemisphere; you can’t compare them to global data.
“The anomaly charts you reference are for the Northern Hemisphere; you can’t compare them to global data.” The three charts I referenced are all NH charts. We should be able to compare any NH anomaly charts directly. NASA-GISS provides no NH base temperatures. They do not do absolute temperature analyses, only anomalies. NCDC-NOAA doesn’t use any hemisphere averages either. Phil Jones (HadCRUT) uses 14.0°C for the globe and 14.6 for the NH. His NH chart is here: http://cdiac.ornl.gov/trends/temp/jonescru/graphics/glnhsh.png Comparing charts using the Jones NH mean and the data available in the mid 70s…. 1975 NH charts. Base period unknown. NH mean (“normal”) temperature: 15°C (59°F). 1910 is “normal” at 15°C. 1938 is plus 0.6 at 15.6°C. 2008 NH chart. Base period ’61-’90. NH mean (“normal”) temperature: 14.6°C (58.3°F). 1910 is minus 0.5 at 14.1°C. 1938 is plus 0.2 at 14.8°C. 2005 is plus 0.6 at 15.2°C Quite different charts. Conclusions drawn in 1976? The early 20th century was about normal before warming steadily into the late 30s and then cooling back to near normal in 1975. Warming predicted into the 90s. Conclusions we draw in 2008? The early 20th century was well below normal, warming steadily to slightly above normal in the late 30s and then cooling below normal into 1975 with warming to above normal into 2000 and beyond. Two very different stories. One suggests that the last decade was about the same as the decade of the 30s. The other says that the last decade was the highest in the last 150 years.