CO2

[Dan Davidoff]

rt...@quads.uchicago.edu (raymond thomas pierrehumbert) writes:
:
: My suspicion is that in a zero CO2 world, even if the
: atmosphere were initially saturated the processing of air
: through the (very) cold polar regions would dry out
: the planet faster than the tropics could keep up; the
: tropics would cool, the H20 greenhouse would disappear,
: and the planet would freeze over. This is a GCM experiment
: that could be done, but I don't know that anybody has
: ever tried it. If it were done, one would have to
: be wary of moistening mechanisms (e.g. evaporation
: of hydrometeors) that aren't present in the GCM.
:

Would this also be true if we removed all CO2, but decreased the
albedo (or stoked up the sun) so that the equilibrium temperature
remained constant? That is, if there's a stabilizing property of CO2
*per se*, as opposed to its simple warming effect, could you explain
it? Or are you just saying that the CO2 provides enough of a "base"
greenhouse effect, so that the natural variations of the climate
system never result in the ice line moving far enough south to start a
run-away icing-over?

Thanks,
Dan Kirk-Davidoff
davi...@sundog.mit.ed

[William Connolley]

rt...@quads.uchicago.edu (raymond thomas pierrehumbert) writes (edited for bandwidth):
> : In a zero CO2 world, the processing of air through the (very) cold polar regions

> : would dry out the planet faster than the tropics could keep up; the
> : tropics would cool, the H20 greenhouse would disappear,
> : and the planet would freeze over. This is a GCM experiment
> : that could be done, but I don't know that anybody has ever tried it.

Manabe et al, J. Climate, 1991, does two runs, one in which CO2 increases at 1%/year
(the conventional thing) and one in which it decreases at 1%/year. After 70 years
(approx doubling/halving time) the temp goes up/down by about 2K in each case.
Coupled runs, R15.
> : If it were done, one would have to

> : be wary of moistening mechanisms (e.g. evaporation
> : of hydrometeors) that aren't present in the GCM.

Evaporation of Hyrdometeors is in all competent GCMs

- William

[raymond thomas pierrehumbert]

(concerning the climate of a zero-CO2 world)

>Would this also be true if we removed all CO2, but decreased the
>albedo (or stoked up the sun) so that the equilibrium temperature
>remained constant? That is, if there's a stabilizing property of CO2

First, I should say that my remark about surface temperature in a
no-CO2 world was pertinent to a radiative-convective model
tuned to tropical conditions; it's a somewhat problematic
thought experiment, given that one needs to make some assumptions
about heat transport out of the tropics.

As for the general stabilizing effect of CO2, my conjecture is
that CO2 is important in providing a "base greenhouse" effect
as you put it. This is important in the polar regions to
prevent a "runaway icehouse" where all your water is transported
to the polar regions where it adds to glaciers which are only
slowly recycled; it is also important in preventing hysteresis
effects in the tropics, in that enough CO2 can keep the
tropics from freezing over permanently even if there is a blip
that somehow transiently creates a dry atmosphere. As long as
the tropics is warm enough, you have at least the possibility of
moistening the rest of the planet and enhancing the CO2 warming
by the water vapor greenhouse. Exactly how this works depends
on the efficiency of meridional heat transport, which is
certainly a big issue. It's easier to keep the whole planet
cozy if you have enough stirring to keep the surface temperature
almost globally uniform.

Similarly, if the solar output went up enough, or the polar
albedo went down enough (probably not enough in itself),
it would be possible to reach a point where the role of CO2
at high latitudes was less crucial in preventing the icehouse.

I ought to remark also that if you crank down CO2 enough,
then CO2 itself becomes a condensable substance like
water vapor, at least at the poles. Then you have the
possibility of a runaway icehouse from most of the CO2 being
deposited in a (very thin!) glacier near the poles. On
Mars, CO2 is precipitable, but not all the CO2 goes out
because the planet reaches an equilibrium where the
saturation vapor pressure of CO2 near the poles, at the
temperature corresponding to the equilibrium atmos.
temp, equals the atmospheric pressure.

All this is rather idle speculation, which needs to
have some firm numbers attached. It would be
interesting to consider in more detail what would really
happen in a zero-CO2 earth. One needs to model the
meridional heat and moisture fluxes to seriously
consider this problem, but I doubt that one needs
to go as far as a GCM to get something interesting.
It sounds like a good Masters thesis topic to me.