Hurricanes and engines

[David Eric Smith]

Hi Nick and all,

Figured all lurk and no pay makes Eric a parasite.  

I think Steve is very right about the engine interpretation, and would
offer a few details to it.  I haven't done the math for this, but
suspect I can correctly guess qualitative features.

A secondary point of quantification: SFI had a visitor from NOAA about
a year ago (I forget her name; she was interested in medium-range
weather forecasting and how well it stood up under onset of chaotic
dynamics) who told me that the existence of the cyclonic storm seasons
raises the polar temeratures by about 3 degrees in early winter
relative to what it would be without them.  So Steve is right about
transport of heat from bottom to top of the atmosphere, but one should
also appreciate the importance of equator-to-pole transport.  I
suspect this is part of the reason for the tropic-to-temperate paths
taken by storms in the Atlantic (haven't ever looked at pacific
storms to see if they do the same thing).

Second point to appreciate is the extreme importance of water in
distinguishing hurricanes, and this adds to Steve's point, and changes
it perhaps a bit.  It is not just transport of heat from lower to
upper atmosphere.  The thing hurricanes do systematically is draw in
cool, relatively dry air from mid to upper atmosphere hundreds of
miles away, dragging it over the ocean where it can absorb warm water
which is then convected upward.  The water carries ocean heat into the
atmosphere through the specific heat of vaporization, I suspect in far
larger degree than through any temperature change.  As the convected
water vapor reaches the upper atmosphere it condenses, releasing the
latent heat to the air where it can be radiated off to space, and
releasing the now-cooled water to fall as rain, some of which
re-evaporates on the way down but much of which falls back into the
ocean whence it came, replacing the warm water with cold.  The air
transported up in the convective column flows out in dehydrated form
at high atmospheric levels, to replace the air drawn in at the base of
the system to sweep over the ocean.

My guess is that if one compared hurricane convection to disordered
Benard convection or thunderstorm systems covering the same area, one
would find that the hurricane cycles a vastly greater volume of
atmosphere laterally across the ocean surface, and with that cycles a
vastly greater volume of ocean water through the
evaporation/condensation cycle, and that through that mechanism it
transports more heat both vertically, and up-lattitude as a
second-order effect, than disordered storms could.  The extent to
which it seems relevant to recognize that as a function is the extent
to which the function, particularly, is relief of a thermal stress.
The large-scale relief of stress seems to be the "force" (to use a
word badly) that favors the emergence and stabilization of
nonequilibrium channels through which the stress is relieved.

Regarding "individuality" at the level of named storms: It is
interesting that, if we regard the reliability of hurricane form as
empirical evidence, the most efficient such engine the atmosphere
finds is one in which the tens of thousands of square miles of shear
flow are subordinated to the angularly ordered convective system, most
notable near the eyewall.  In that sense each storm creates conditions
that essentially preclude the concurrent formation of other eyewall
systems in proximity competing for (or driven into existence by) the
same sources of energy.  The exception, of course, being the eyewall
replacement, in which the instability of extreme angular momenta
around a very small eye apparently weaken its ability to constrain all
the surrounding lateral convection, and the conditions that led to the
original eyewall do lead to the formation of a new secondary eyewall
concentric with the destabilizing one that is on the way out.

I would readily identify this regularity of form with the regularity
of form of the chemistry in lightning strikes, and probably with the
regularity of form of certain biochemical pathways.  The individual
who emerges in biology, however, probably does so in response to
rather different sequences of pressures, even though once emerged, he
shares certain "Darwinian" aspects with the individual hurricane.  

I have thought for years it would be wonderful to see a book in which
a comprehensive suite of mathematical models of hurricanes was
reviewed, from abstract to quantitatively accurate.  If anybody knows
of a compact and well-written book, I would enjoy reading it.

Eric