I see this paper cited a lot in the climate literature, because it was one of the early papers that established the idea that the response of global mean hydrologic cycle is constrained by the net radiation at the surface.
The figure below shows the change in precipitation and evaporation (which have to balance globally) against the global mean surface temperature change. I’m not sure if this was the first paper to show this type of figure, but it has been reproduced many times since, and the general result is very robust. The change in the amount of tropospheric water vapor can be understood from the Clausius–Clapeyron equation, which suggests that the holding capacity of the atmosphere should increase about 7% per degree Kelvin of global mean surface temperature. The main message of the figure below is that the precipitation and evaporation increase at a much lower rate than Clausius–Clapeyron would predict, about 2-4% per degree K. This means that the strength of the hydrologic cycle increases with global warming, although in a somewhat muted fashion.
It’s interesting to note that while the result of a muted precipitation response to global warming appears to be theoretically sound and robust across models, satellite observations suggest that precipitation is increasing closer to 7% per K, instead of 3% per K (Wentz et al 2007). Issac Held suggests on his blog that in spite of this result there probably is a difference between the response of precipitation and water vapor, because otherwise it would have certain consequences for the climate sensitivity. Personally, I would not be surprised if the satellite precipitation estimates were off in this regard, since this is a difficult thing to measure. Several conversations with people who work on satellite retrieval algorithms have bolstered this sentiment.
In this paper, Boer mentions the “whitehouse” effect, which is similar to the “greenhouse” effect. Apparently this term never caught on, since I’ve never heard or read anyone else using it. This is probably a good thing, since the “greenhouse effect” doesn’t work like a greenhouse (or a blanket) at all! This is because greenhouses (and blankets) work primarily by physically inhibiting convection, whereas the “greenhouse effect” only describes a radiative mechanism. I don’t want to think about how people might mis-understand something called the “whitehouse effect”!
The whitehouse effect is the shortwave analog to the greenhouse effect, but it’s really about reflection rather than absorption and emission. To clarify, consider the change in the net radiative balance at the surface as having a shortwave and longwave component
The longwave part of the net radiative change includes the “greenhouse effect” (i.e. the atmosphere radiating energy downward) and the longwave feedback (i.e. warmer things radiate more energy away).
We can sum up the shortwave into three mechanisms of reflecting, or limiting the surface absorption of solar radiation, namely the land surface (including snow/ice), clouds, and the clear-sky atmosphere. Boer uses the “whitehouse effect” to describe the combination of shortwave feedbacks from clouds and clear-sky atmosphere.
The clear-sky shortwave effects of CO2 are generally too small to affect much of anything, but the shortwave cloud feedbacks end up being useful for describing the strength of the hydrologic cycle. In the figure above, you’ll notice that the CCC model that Boer is focusing on has the weakest response of precipitation compared to the other models. Here is a summary of his explanation:
- global precipitation is constrained by radiative heating of the surface
- energetically, heating by precipitation has to balance cooling by evaporation
- evaporative cooling at the surface has to balance warming by sensible heat flux and radiation
- sensible heating of the surface is small, and consistent between models
- Radiative heating of the surface is smaller in the CCC model due to the shortwave feedbacks
- cloud amount generally decreases (warming)
- cloud optical thickness increases in the tropics (cooling)
- surface snow/ice retreats in the polar regions, lowering the albedo (warming)
- the effect of thicker clouds overwhelms the shortwave warming effects from reduced cloud cover and reduced snow/ice at the surface
- Other models show a net warming of the surface by shortwave feedbacks
(see table below)
Overall, the main novelty of this paper was to show that the change of the hydrologic cycle with global warming was not simply dependent on the surface temperature, but rather a more complicated relationship with all the processes that control the energy input to the surface.