A peer-reviewed study by a team of climate modelers led by Ben Santer published in June compared model reconstructions of the past few decades to observations in the tropical troposphere. Aha, you say, the famous missing hotspot. Yes indeed, and the authors of this study acknowledge upfront that people have been looking for the missing warming for 20 years now. The contribution of this new paper was to examine how warming in models correlates with increasing levels of water vapor, compared to how warming and water vapor correlate in the real world. The result was that for the observed amount of extra water vapor the models say there ought to have been a lot more warming. In response to which the UK Daily Mail declared “Satellites may have been underestimating global warming for the last 40 YEARS, scientists warn“, in other words if the models don’t fit the data, the data must be wrong. But the study authors themselves were more circumspect. Appropriately so, cautions satellite and climate expert Roy Spencer, since we don’t really know how water vapor responds to warming, and the measurements of humidity aren’t very reliable anyway.
Spencer also noted that the satellite-based warming measurements are confirmed by independent weather balloon series, so it doesn’t work to simply assume they should be higher.
One of the points the study makes is that models exhibit a tight, linear relationship between atmospheric warming and changes in a particular measure of water vapor. And reality doesn’t line up with that line. But Spencer explains in a lengthy blog post that while water vapor changes near the surface should be tightly coupled to Sea Surface Temperatures, how it couples to air temperatures way up in the troposphere is not well known. Furthermore, satellites have trouble measuring water vapor in the mid-troposphere because they mainly look at a frequency dominated by near-surface observations. So if anything in the data is amiss, it’s more likely to be the water vapor measure than the temperatures.
Spencer reviews published evidence explaining why water vapor changes are hard to model and to measure. His bottom line is as follows:
The existing observations as presented in the Santer et al. study are largely consistent with the view that global warming is proceeding at a significantly lower rate than is predicted by the latest climate models, and that much of the disagreement between models and observations can be traced to improper assumptions in those models.