The increase in atmospheric concentrations of water vapor with global warming is a large positive feedback in the climate system. Thus, even relatively small errors in its magnitude can lead to large uncertainties in predicting climate response to anthropogenic forcing. Here we use 8 years of global, vertically resolved water vapor observations from the Atmospheric Infrared Sounder (AIRS) instrument from 2002 to 2009 in combination with a radiative transfer scheme to derive a water vapor feedback resulting from climate variability over this time period. We estimate this short-term water vapor feedback to be 2.2 +/-0.4Wm-2K-1. We then obtain water vapor feedbacks for different record lengths over the twentieth century from simulations of 14 climate models available as part of the Coupled Model Intercomparison Project version 3(CMIP3). Based on the relationship in CMIP3 models between feedbacks derived over short and long timescales, we estimate a range of likely values for the long-term twentieth century water vapor feedback of 1.9 to 2.8 Wm-2K-1, which is larger than many previous estimates.

Finally, we use the twentieth century simulations to determine the record length necessary for the short-term feedback to approach the long-term value. In most of the climate models we analyze, the short-term feedback converges to within 15% of its long-term value after 25 years. The major implication of this result is that to accurately estimate a water vapor feedback on long-term transient warming, a longer observational record is necessary than is currently available. Observational estimates of this feedback from current, short-term records are thus not necessarily representative of its long-term value.