Climate Metric Reality Check #3 – Evidence For A Lack Of Water Vapor Feedback On The Regional Scale

An essential component of the IPCC perspective of global warming is that atmospheric water vapor must increase in order to amplify the radiative warming effect of carbon dioxide. Without this amplification, the global warming that would be due to just carbon dioxide would be quite modest. The multi-decadal global models predict such an amplification, with the claim that the relative humidity remains nearly constant as the atmosphere warms. The atmospheric depth total column water vapor (called “precipitable water”) is a useful metric for this purpose.

A valuable summary of this issue is in the paper by Kevin E. Trenberth, John Fasullo, and Lesley Smith entitled

Trends and variability in column-integrated atmospheric water vapor, 2005, Climate Dynamics 24: 741–758 DOI 10.1007/s00382-005-0017-4

where among the conclusions they estimate an increase of precipitable water of

“…. the trend is 1.3±0.3% per decade”

but that while

“Precipitable water is a very important variable for climate, as discussed in the introduction, but the quality of most of the global analyses of this quantity leaves [a] great deal to be desired.”

They also found that

“Precipitable water variability for 1988–2001 is dominated by the evolution of
ENSO and especially the structures that occurred during and following the 1997–98 El Nino event.”

They also found large regional variations in long-term trend of precipitable water even with what they conclude is the most accurate measurement approach (SSM/I) as shown in Figure 9 in the Trenberth et al. paper.

To improve our understanding of this issue, we have investigated this question over North America using the North American Regional Reanalysis (NARR). This reanalysis, which is in a data rich area with radiosondes, should eliminate the concerns that were expressed with respect to the global reanalysis reported on in the Trenberth et al. paper. The NARR, therefore, should be the most accurate description of atmospheric variables such as temperature and humidity over this region. Our paper on this approach, currently under review, is

Wang, J.-W., K. Wang, R.A. Pielke, J.C. Lin, and T. Matsui, 2007: Does an atmospheric warming trend lead to a moistening trend over North America? Geophys. Res. Letts., submitted,

with the abstract

“An increase in the atmospheric moist content has been generally assumed when the lower-tropospheric temperature increases, with relative humidity holding steady. Rather than using simple linear regression, we propose a more rigorous trend detection method that considers time series memory. The autoregressive moving-average (ARMA) parameters for the time series of lower-tropospheric temperature (Tcol), precipitable water vapor (PWAV), and total precipitable water content (PWAT) from the North American Regional Reanalysis data were first computed. We then applied Monte Carlo method to replicate ARMA time series and collected samples for estimating the variances of their OLS trends. Student’s t tests showed that the Tcol from 1979 to 2006 was significant and positive; however, the PWAV and PWAT were not. This suggests that atmospheric temperature and water vapor trends do not follow the conjecture of constant relative humidity. We thus urge further evaluations of lower-tropospheric temperature and water vapor trends for the globe.”

Other sources of data confirm the lack of increase in water vapor content. A particularly valuable set of information is the total column water vapor time series available at the website Sun and Sky Science.

This analysis for 4 Feb 1990 to 16 Dec 2007 covers 17.9 years. Plotted are only days in which the sun was not obscured by obvious clouds.

The trend line, added using Excel, yields a linear fit (y = a + bx) of y = 3.5925 -1.4968x.

Solving for precipitable water (i.e., y in the regression equation) at the beginning and end of the series gives 3.0999 and 3.00228 cm, respectively, for a decline of 0.09762 cm [the mean value over this time period is 3.06cm]. This yields a decline per decade of 0.0545 cm. (This is rounded to -0.055 cm/decade in the chart.)


Figure caption: “These include some of the time series of data collected at Geronimo Creek Observatory [Texas] at or near local solar noon when the Sun was unobscured by clouds. All data shown here have been previously published or discussed in peer-reviewed papers with the exception that these data are updated through 2007″ [from Sun and Sky Data].

These nearly 18 years of data were acquired by Forrest M. Mims III in an ongoing program using the measurement approach described in

F. M. Mims III, An inexpensive and stable LED Sun photometer for measuring the water vapor column over South Texas from 1990 to 2001, Geophysical Research Letters 29, 20-1 to 20-4, 2002.

This data and instrumentation comes from the measurement approach also described in the papers

F. M. Mims III, Sun Photometer with Light-Emitting Diodes as Spectrally Selective Detectors, Applied Optics, 31, 33, 6965-6967, 1992


David R. Brooks, Forrest M. Mims III and Richard Roettger, Inexpensive Near-IR Sun Photometer for Measuring Total Column Water Vapor, Journal of Atmospheric and Oceanic Technology 24, 1268-1276, July 2007

with the abstract of the third paper

“An inexpensive two-channel near-IR sun photometer for measuring total atmospheric column water vapor (precipitable water) has been developed for use by the Global Learning and Observations to Benefit the Environment (GLOBE) environmental science and education program and other nonspecialists. This instrument detects sunlight in the 940-nm water vapor absorption band with a filtered photodiode and at 825 nm with a near-IR light-emitting diode (LED). The ratio of outputs of these two detectors is related to total column water vapor in the atmosphere. Reference instruments can be calibrated against column atmospheric water vapor data derived from delays in radio signals received at global positioning satellite (GPS) receiver sites and other independent sources. For additional instruments that are optically and physically identical to reference instruments, a single-parameter calibration can be determined by making simultaneous measurements with a reference instrument and forcing the derived precipitable water values to agree. Although the concept of near-IR detection of precipitable water is not new, this paper describes a first attempt at developing a protocol for calibrating large numbers of inexpensive instruments suitable for use by teachers, students, and other nonspecialists.”

While the analysis at a location or even over North America needs to be expanded globally, the lack of a long-term increase in precipitable water even on a large regional scale or point location should be disquieting to the global climate modelers, as the water vapor in a column is a result of water vapor transport over large distances.

This lack of regional increase in precipitable water, and the conclusion in the Trenberth paper that the global value is closely linked to large El Nino events, suggests that the atmosphere may not be maintaining a near constant relative humidity when the atmosphere warms over decadal time periods.

Also, the highly stable, inexpensive instrument developed by Mims that is being used to measure the long-term time series of precipitable water reported in the Brooks et al. 2007 paper should be made an integral part of the global climate monitoring system.

Leave a comment

Filed under Climate Change Metrics

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.