~A small example simply reveals a big mystery.
~ For horizon observations, sea ice must warm or cool faster than surface air, this is impossible with a standard physics interpretation.
~There must be an unknown thermal transfer factor which has huge implications with sea ice models.
Basic refraction optics are easy to understand. A road mirage exists because there is very hot air above the pavement, a mere few centimeters above, where the temperature interface has a very steep unstable adiabatic profile immediately on top of the road. The much hotter air creates an inferior mirage. Similar inferior mirages are possible mainly during Arctic fall. A lesser optical effect with a not so steep but strong adiabatic lapse rate are horizon shifts which are called dips. It was observed repeatedly that refraction horizons vary greatly over sea ice, but these variations are linked with the surface to air interface just as much. Like the hot pavement, sea ice is a thermal body, however much unlike road materials, sea ice reflects more incoming sunlight without absorbing it. But the following sequence suggests that sunlight, even a fraction of it, was very significant, in fact heated top of sea ice enough to lower the horizon dip significantly before the horizon rose again as with:
By theory, sea ice has greater thermal capacity than air, it should not change in temperature at the same time rate as air does. But this sequence suggests surface of sea ice cooled faster than the air right above. A colder than air sea ice surface causes thermal inversions, with horizontal layers readily seen in the left and center photos. Atmospheric inversions raise any object in the sky, including the horizon.
A same sequence during wide open water Arctic fall time will not repeat the same phenomena, unless there is ice present. The relation between not so varying sea surface temperature and the height of the horizon was made clear (paper almost done). There is a direct relation between sea surface and surface air temperature. The steadier sea surface temperatures revealed a robust repeatable temperature proportionality.
Current spring sea ice horizon diurnal effect is likeliest due to sea ice deeper frozen core temperature cooling the top of ice more rapidly than the usual swifter double rate of air cooling caused by less sunlight. Thermal conductivity of ice is nearly the same as soil, and is several orders of magnitude greater than air. There may be something else at play, like an inversion nearer to the camera (disproved very lately), or a diurnal temperature variation of near bottom ice sea water column (very unlikely). This may be why sea ice models can't replicate great sea ice melts. The sun and clouds are linked much more intimately with sea ice thermal exchanges. This is very well observed. The famous 2007 Arctic ocean ice melt had a lot of very thick multiyear year ice vanishing under the warm higher in altitude summer sun, the astounding speed by which very thick ice melted definitely had something to do with the sun and clear air. It seems that sea ice may be more complicated than simple, there are known biological interactions, not so known thermal features associated with structural differences as well. Here is an example, amongst hundreds, of something worth taking time to achieve a more profound study. WD April 28, 2014
Theory vs Observations:
Heat capacity of ice is twice as great as air, air heat capacity is greater than dry soil.
On a normal mid latitude night, the air cools slower than dry land, soil cools faster,
this creates often observed morning inversions. Eventually likely by direct contact conduction , soil cools surface air. Not theoretically so for sea ice. Air should cool faster than ice. But air cools uniformly, at any given time, given a starting point of sea ice temperature being the same as its surface air (the true astronomical horizon achieved), the lowering sun should create a warmer sea ice under cooler surface air. If so, the cooler surface air should be warmed by the top of sea ice. If so, the horizon should drop. But what is observed is the opposite, the horizon rises. The pictures above
strongly suggests colder sea ice than surface air. Because multiple stacked inversions reveal near surface thermal layers, of which the top layer is the most distant and last inversion. If surface air cools faster than sea ice , the layer 1 meter above cools at the same rate as the higher ones. Radiative cooling should be uniform. It is highly unlikely that the layer of air immediately on top of sea ice cools faster than the immediate air layers above especially if sea ice is warmer. WD May 4,2014.