Re: Visible moisture, below freezing temperatures, but no icing. How is this possible
Well the point I was making was technically icing could occur at just about any temperature and similarly, just because it's 0C outside and there's visible moisture in the air, doesnt mean icing will occur.
Its like those first few flakes of snow that never stick... The surface needs to cool sufficiently to support ice.
In most planes, this doesnt take much on the surfaces as the specific heat of aluminum is 897 Joules/kg so it takes approximately of 0.00025 kilowatts of energy per hour to heat or cool 1kg of aluminum 1C in 1 hour.
(896 joules/kg * 1 watt-second/joule * 1 watt-minute/60 watt-second * 1 watt-hour/60 watt-minute * 1 kilowatt-hour/1000 watt-hour) = 0.00025kWhr
Compared to the 2000 to 4000 joules/kg required to do the same to water or ice at low temperatures.
A single kilo of 0C water is encountered in 0C air every 263 cubic meters (dry air at STP has a density of 1.27kg/cubic meter). 1 Kilo of 0C water can reduce a full kilo of aluminum by 2-4C before the water warms 1 degree. More when you consider wind also pulls heat away from a plane.
At an approximation of empty weight, a C172 at 500 kilos and ambient temperature of 16C on the ground need only come in contact with roughly 1000 kilos of 0C water in the air to be cooled sufficiently to set the stage for icing. The first 125 kilos of water warms to 8C as it cools the plane to equilibrium at 8C, the second does the same down to 4C. Below 4C the water takes less energy to warm so it takes 250kilos to cool down to 2C, 250kg down to 1C, 250kg down to 0.5C. for a total of 1000 kilos.
The cross-sectional area of a C172 is approximately 57.6 square meters (as calculated based on coefficients of drag and the NACA published coefficients for airfoil on which the C172 is based... I did not actually do this calculation but you can find it here:
http://www.temporal.com.au/c172.pdf on page 8, for the egg-heads and those interested in this sort of thing, I recommend reading the full 20 pages as it shows the math behind the performance data for a C172 and if you know the math you can apply it to any plane) at that cross-sectional, it would take 4566m to encounter a sufficient volume of water equal to 1000 kilos (based on the earlier chart that says there is 3.8g of water per kilo of air at 0C at sea level) which computes to a distance of approximately 2.5 nautical miles.
*The above equation is still far from accurate and may in fact be wrong; it uses rough numbers and rough calculations of entropy to provide an approximation of otherwise much more complex and long forgotten equations from thermodynamics.*
Its not exact because Im not calculating the actual weight of the aluminum which is going to be below the empty weight, the actual cross section of the plane, how much air volume it actually flies through in that distance or how much water is in that volume of air, plus things like fuel or air pockets inside the wings and wind across the wings also adds/removes heat to/from the equation but its close enough to give an approximation. These uncalculated factors are also a portion of the reason why control surfaces and leading edges are among the first thing to ice over (they receive the most accumulation of water per second, have the most cooling due to wind and they have little direct exposure to warmer substances that can add heat to the equation).
A composite aircraft with surfaces more resistant to changes in surface temp or an aluminum aircraft with full tanks of fuel may be more resistant to icing initially and may be able to fly through an icing level with less risk if the pilot understands icing isn't immediate as the temperature of the wing and the water and the air must all be within a certain envelope... So you could theoretically take off at 16C, quickly climb to 8000ft and fly in icing conditions for 2.5 NM before ice begins to form but once the surfaces cool and the conditions are set for icing, the build up of ice and the impact it has on lift, stall speed and the control of the airplane (ice in general impacts control but ice can also freeze control surfaces in place) grows exponentially every second you remain in that condition.
Understanding the envelope at which the conditions of the plane, the air, the water and its associated temperatures add up to icing can make a significant determination to a go/no go decision or initial climb out decision if your planned altitude is FL100 (10,000ft because FL100 doesnt technically exist) and there is icing between FL070 (7,000 ft) and FL090 (9,000ft) and you have 16C temps on the ground, you can push through the icing layer before ice can accumulate if you dont dilly-dally in the climb both before, during and after the icing layer. Same for the descent... your plane will already be at temps in which ice can start to form but cold water will actually heat your wing initially so if you descend quickly through that icing layer you can get to the other side and while you may have some ice build up, the warmer temps on the other side should melt off the small accumulation quickly... Dickering around in that icing layer though will get you killed.
Of course this understanding does not make up for a good deicing system nor should it be relied on in a manner similar to relying on a deicing system and I certainly dont recommend you go out flying in ice to test my equations or because I suggested by my equations its doable... Ice should be avoided and for good reason but this understanding can make a difference to your considerations because sometimes you have to go through that icing layer, particularly when its unforecasted or unknown and quick decisive actions in that layer can mean the difference between no ice, a small amount of ice and enough ice to cause insufficient lift and CFIT or total loss of control and UnFIT.
Ohh! Okay, okay, okay, I got it now! Thank you! 
