Skots said (two years ago):
I think I understand it now, Rog,
When you say that "the air doesn't warm the ocean", what you really mean is " the air doesn't warm the ocean to any significant degree compared to the heat directly from the sun".
I guess my next question is, what do you mean when you say, "the air doesn't warm the ocean to any significant degree compared to the heat directly from the sun"? :)
Do you mean that any long wave generated heat transfer into the ocean is insignificant, without comparing it to the sun? That it can't or doesn't penetrate below the boundary layer? By the way what is significant? Two watts per sq. meter?, over fifty years? Two hundred years? Six hundred years?
Considering that the atmosphere is responsible for keeping the planet 33 degree C or 91 degree F warmer than it would otherwise be, ( You do accept that basic physics, don't you?), and assuming the full 33 degree C is due to long wave, would you say that the an atmosphere that warmed the earth some 30 degree C, instead of 33C, would affect average atmospheric temps. more than average ocean temps.? And over several millennium?
I think I have a better developed and more complete answer for Skots than the one I provided back then.
Using Trenberth's figures:
‘Local’ Energy Equilibrium of the ocean (W/m^2):
Solar in = evaporation out + conduction/convection out + net radiative flux out = input to atmosphere
170 = 78 + 24 + ~66-70 = ~170
‘Local’ Energy Equilibrium of the atmosphere (W/m^2):
Input from the ocean below + Direct absorption from the Sun above = radiation out to space
170 + 67 = ~235-239
The upward and downward components of the radiative flux in the atmosphere cause convection within the atmosphere which inflates it and supports the adiabatic lapse rate. This is partly why the surface air temperature and ocean surface are at a temperature ~33C warmer than an atmosphere devoid of radiatively active gases would be (There would still be some adiabatic lapse due to the thermal difference between equator and poles which would drive a reduced atmospheric circulation, according to some).
Surface conduction, pressure, humidity, convection, and the adiabatic lapse rate supported by the radiative flux, over the course of several billion years, has kept ocean and atmosphere in a dynamic thermal equilibrium whereby the ocean surface is slightly warmer than the surface air, to enable the rate of energy loss the ocean needs to maintain in order to remain in local energy equilibrium.
Both the ocean surface and surface air are warmer than the upper atmosphere where radiation to space takes place at the rate demanded by solar input and the Stefan-Boltzmann Black Body equation because of the dynamic equilibrium demanded by adiabatic lapse. The radiative flux plays an important role in that equilibrium, but small changes in the balance of the components of the flux will not measurably change the bulk temperature of the ocean at the multi-decadal scale because the thermal inertia of the ocean and its evaporation rate lies behind the longwave radiative flux component’s differential of ~66-70W/m^2. The net radiative balance is involved in the differential between air and ocean surface temperature, but small changes in the net balance of the IR flux can only change the bulk temperature of the ocean very, very, very slowly; like, thousands of years for a measurable difference.
The tail does not wag the dog.
Infra-red longwave radiation downwelling from the atmosphere towards the ocean surface can't do anything significant to change the temperature of the ocean directly, because the ocean is opaque to IR, absorbing it in the first 0.05mm, the ocean 'skin' temperature is cooler on the surface than in the next mm so conduction downwards generally won't occur, and surface eddies are not significant enough in average open ocean conditions to turbulently convect heat downwards either.
On the other hand, the Sun's short wave radiation warms the oceans top layers very effectively and directly, penetrating many metres into the sea, and the ocean’s response time to solar input is much quicker, on the order of hours. It is therefore reasonable to assume that the empirically measured reduction in low tropical cloud cover 1980-1998 (ISCCP data) is far more likely responsible for the increase in ocean heat content and sea surface temperature (and marine air temperature) measured by buoys, satellites and engine coolant intakes over the same period.