Varying thermal fluxes as portrayed by sea ice horizons.

~A few examples.
~Clouds help discern 2 distinct albedo reactions.

At stake is this graph done by very capable mathematician Tamino, essentially portraying total sky albedos given various weather possibilities:

At latitude 75 degrees North we will deal with zenith angles 70 to 90 degrees. Essentially the 2
blue lines. Knowing Tamino's thorough laser dedication to exactitude, this graph likely represents the standard widely used Albedo reference numbers. Note the "clear sky over bright sea ice" vs "cloud over bright sea ice" small 10% difference. We deal here with these 2 features. Keep in mind the actual horizontal view largely contradicts this. If there is albedo reflecting clouds at Local Apparent Noon, the horizon view is dramatically different than with "clear sky albedo" consistently and repeatedly, incoming sun rays become very deflected, leaving a dramatic difference in horizon heights, either at Local Apparent Noon or especially seen easier later as the sun lowers to the horizon. A complete cloud cover leaves the horizon altitude in a steady state of flux, as opposed to "clear sky albedo" which enables the observer to witness various thermal effects un-impeded. This implies that back radiation from bottom of clouds leave sea ice in neither extremes of cooling or warming. Therefore heat which could be gained from direct sunshine is lost. During the dark Arctic long night, clouds do just the opposite, a lot of heat is not lost to space, while during the long Arctic midnight sun days, persistent strong albedo clouds prevent a great deal of melting.

The integrated albedos (graph above) perspective may be an erroneous concept. Albedo layers should be calculated individually layer by layer following the sun ray path. The idea of merging albedo layers is good, but 50% seems too low given the greater cloud cover when sea ice cools the warming summer surface.

During the Arctic sea ice melt season there can be up to 4 stratocumulus decks stacked on top of each other. It is the most common Arctic cloud, with small water droplets 10 micrometers in diameter, there is also often wide ranging fog banks becoming stratus and reverting back to fog in long lasting cycles. Summer time viewing of sea ice from the vantage point of High Resolution satellite pictures is almost always a tasking job:

"Cloud albedo varies from less than 10% to more than 90% and depends on drop sizes, liquid water or ice content, thickness of the cloud, and the sun's zenith angle. The smaller the drops and the greater the liquid water content, the greater the cloud albedo, if all other factors are the same.

Low, thick clouds (such as stratocumulus) primarily reflect incoming solar radiation, causing it to have a high albedo, whereas high, thin clouds (such as Cirrus) tend to transmit it to the surface but then trap outgoing infrared radiation, causing it to have low albedo. It contributes to the greenhouse effect.[1][2]" wikipedia

The presence of extensive multi layered cloud spreads, often covering the entire Arctic Ocean for weeks, certainly brings up the cloud albedo value well above the 50% mark. On occasions, such as during summer 2007, with persistent anticyclones, next to land and moving northwards, shatter clouds important ice protective vail, plummeting the ice pack to melt rapidly.

One primary reference for 50% integrated albedo was the Sheba project, although the vanishing albedo is correct interpretation of latest melt trends, the nature of sea ice albedo is less variable than cloud cover.

An annual cycle of Arctic surface cloud forcing at SHEBA , JOURNAL OF GEOPHYSICAL RESEARCH: SUBMITTED JANUARY 19, 2001

Sheba Albedo graph has been cited in other journals namely: http://www.pnas.org/content/111/9/3322.full

While the data proved interesting, the said research ship location in summer 1998, Beaufort Chukchi seas, which by coincidence marked the beginning of the greater melting summers to come:

The open water footprint amongst sea ice at minima in 1998 suggests there was a great deal of insolation leaving mostly atmospheric absorption and sea ice albedo to reduce the melt otherwise with the full force of nearly direct sunlight and of course lesser total albedo.

Calculation using separate step by step not integrated albedos:

Entire Arctic ice melt calculation using best albedo data available separately, suggests a correct interpretation of the physics, in a simple equation I call ASIMP, Arctic Sea Ice Melt Potential

ASIMP = Area of entire Arctic Ocean X TOA TSI for Arctic Ocean 6 months from spring equinox to fall equinox X average summer sea ice albedo X Atmospheric Absorption factor X Average cloud albedo / sea ice latent heat of fusion.

Using

Arctic Ocean sea ice area: 14E6 km2
Top Of Atmosphere average TSI for same area for Equinox to Equinox 70 to 90 N= ~~257~~ 247 W/m2
(equinox calculation closely matching known TSI graph median)
Clear Sky Albedo (sea ice albedo) = 50%
Atmospheric Absorption = 23%
Average cloud albedo over the summer = 80 %
(1) Sea ice latent heat of fusion =   2.95e17 j/km3 U.N. fisheries and agriculture department
(with density 0.89 t/m3 U.N, 1% salinity at -2 C according to Doctor Ono's chart)

Gives ASIMP = 15,081 km3 (calculated mar 7, 2017) close to the actual POMAS latest summer melt 17,500 km3.   (77% cloud albedo is needed to make ASIMP equal to 2016 summer melt). These variables were at first taken from different sources, I did not try to fit with known melts, and they need be perfected, leaving the greatest yearly variation to cloud albedo. This reformulation calculation gives a fairly interesting estimate.

----------Arctic "clear sky albedo" vs cloud albedo effects:

Series pictures displaying different sky conditions and resulting thermal flux action at the sea ice horizon. If horizon outgoing thermal heat is equal to incoming Astronomical Horizon is achieved.
All pictures mostly above A.H. indicating net thermal loss from the sea, those at A.H. will be indicated. None can be below A.H,.

On all pictures, top left: date time, top right: general sky condition, lower left: temperature wind speed with narration, lower right: sea ice horizon elevation (2.6' is astronomical horizon) below is sun elevation in degrees.

Dominant Low cloud albedo