Arctic ice: what is happening to Earth's albedo?

http://conservationatwork.wordpress.com/2010/07/20/wed-2007-melting-ice-a-hot-topic/

The depressing fact that Arctic ice levels are hitting  historic lows is so important that is is even penetrating through to the BBC news, although obviously not rated as important as the astonishing fact that Royal Personages are naked when they take their clothes off.

As part of my campaign to refute the hypothesis of the climate contrarians,  I have been trying to find a feedback value for loss of Arctic ice.

Any fule kno that loss of ice means more solar energy is absorbed, so this is a positive feedback onto earth's temperature - that is, it adds to global warming.

Climate contrarians desire and believe that net feedbacks must be negative, otherwise there would be a problem with Exxon &c warming the planet.

I was unable to find a figure for ice albedo change, but in my search I found something [apparently] much more important here at NASA's Earth Observatory published in 2005.

The Clouds and Earth’s Radiant Energy System (CERES) instrument aboard NASA’s Terra satellite has been computing Earth's total albedo.

They say:
If Earth was as white as a snowball, albedo would be 0.84 (i.e. 84% of incoming solar energy would be reflected)
If earth was covered with dark green forest, albedo would be 0.14.
The average present albedo is 0.3

A drop of as little as 0.01 in Earth’s albedo would have a major warming influence on climate—roughly equal to the effect of doubling the amount of carbon dioxide in the atmosphere, which would cause Earth to warm by about 1.2*C.

Here's the rub: Over the 4-year span (2000 through 2004), the CERES instrument measured an albedo decrease of 0.0027, which means more energy is being absorbed.

I calculated that if this rate of decrease continued for 15 years, it will eventually add 1*C to present temperatures.

Now clearly, this inference cannot be made from one reading.
There are a few questions to be answered:

• NASA say   future research will focus on comparing CERES data to data from other space-based sensors to see if there are any significant changes in Earth’s climate system during that time that could account for the change in albedo.
• Also - is it an artefact, a misreading?
• Is CERES covering the whole planet, or is it over reading areas that have decreasing albedo?
• Is this confirmed by other observations?
• Can it be explained in terms of what is known to be happening to albedo on the ground?

This last question would mean that 

• Ice albedo is less - probably true, but is how much does it account for?
• Low cloud albedo could be decreasing - is this the case?
• Forest cover could be increasing - which would be nice, but is it the case?
• Is the 2000-2004 period typical of the rest of recent history, or is it an anomaly? 

It is odd that this result turned up 7 years ago, and there is no further announcement that I could find on the NASA site. I made enquiries, and it seems that the decrease disappeared when it was integrated with other measurements. Advances in Understanding Top-of-Atmosphere
Radiation Variability from Satellite Observations
Norman G. Loeb et al, Surv Geophys (2012) 33:359–385 DOI 10.1007/s10712-012-9175-1
I'm still a bit puzzled though, because Loeb's paper deals with 20*S-20*N, and the Earth observatory result was a whole earth assessment.

Anyway, it is better news for the planet if albedo is not decreasing at the rate they reported.

2000-2004 was indeed a period of rapid warming (see below) so it could be that the albedo changes are indeed due to ice and snow retreat, accompanied by loss of cloud cover (clouds are less in warmer air). The 0.2*C increase in temperature would have resulted in a 3% decrease in cloud cover. I doubt that this would be enough to explain much of the change in albedo, but I'm not an expert.

Thanks to Robert Rohde http://globalwarmingart.org

For the tiny minority who put the interests of Exxon &c above the future of mankind, this finding, if it is accurate, is totally incompatible with their hypothesis that the net climate feedbacks will turn out to be negative.

PS Skeptical Science has a piece on albedo that says CERES 2000-4 showed a flat trend. Maybe my link is wrong. I have an email saying they're going to delete it.