Guest Post by Willis Eschenbach
As always, I get distracted by the daily news. The weather news today is a lovely rainy morning here in drought-plagued California, we got just under an inch (2cm) in last night’s storm, and the outer world is green and happy. Regarding the climate news, Anthony highlighted a claimed recent darkening of the Greenland ice cap. This is said to be reducing the ice cap’s albedo, which is the percentage of sunshine reflected back to space, and thus leading to more solar absorption and more melting.
Being an inherently suspectful type oif fellow, I thought I’d take a look at the albedo and other anatomical features of Greenland. First, the big view. Let me start with a map showing the global “all-sky” albedo from the CERES satellite data. It shows the average of all satellite observations, both when the sky is clear and when it is cloudy.
Overall, the combination of the clouds and the surface reflect just under a third of all the sunlight that hits the planet. In general the albedo is smallest in the tropics and increases towards both poles. In Figure 1, you can see the inter-tropical convergence zone just above the equator. You can also see Greenland, bright red up near the north pole, with an average albedo of about 65%
The CERES data provides us with another view of the albedo, which is just the measurements taken when the sky is clear. Figure 2 shows that clear-sky albedo, the solar reflection from the surface when there are no clouds..
As you can see, without the clouds there is much less sunlight reflected from the surface. For example, the ocean reflects less than 10% of the incident sunlight … but even without clouds, Greenland still has an albedo of about 65% because like Antarctica, it has a permanent ice cap. It is the darkening of this Greenland ice cap that I set out to investigate.
Now, there’s a problem with measuring albedo near the poles. Albedo is a ratio. It is a fraction with reflected solar energy on the top and the incoming sunshine on the bottom. Most of Greenland is above the Arctic Circle. So when the sun gets to very near zero in the winter, the albedo gets very uncertain and averages get distorted. As a result, I look instead at the total amount of sunshine that is reflected from Greenland. The incoming sun is constant on an annual basis, so any change in the albedo will be reflected as a change in the total amount of sunshine reflected.
Figure 3 below shows the month-by-month changes in the all-sky reflections from Greenland. I masked out the ocean, so Figure 3 represents solar reflections of just the area of the island itself.
Figure 3. All-sky average reflected solar energy, CERES top-of-atmosphere data. Units are watts per metre squared (W/m2). Mean value is 119.8 W/m2. Top panel shows raw data. Middle panel shows the average seasonal variation in the data. Bottom panel shows the residuals, which are the raw data minus the seasonal component. Standard deviation of the residuals is indicated by the horizontal gold dashed lines.
The average amount of energy reflected by the clouds plus the surface is about 120 W/m2. There is no trend visible over the period, and the standard deviation of the residuals (bottom panel) is only about ± 2.5 W/m2.
“Ah”, I hear you thinking, “but that includes the clouds”. Indeed it does, it is not the surface albedo from the ice cap. I like to look at what is happening overall before I look at the specifics. Having seen that there is no overall albedo trend in Greenland, Figure 4 shows the Greenaland surface reflections when the sky is clear.
Figure 4. Clear-sky average reflected solar energy, CERES top-of-atmosphere data. Units are watts per metre squared (W/m2). Mean value is 115.9 W/m2. Top panel shows raw data. Middle panel shows the average seasonal variation in the data. Bottom panel shows the residuals, which are the raw data minus the seasonal component. Standard deviation of the residuals is indicated by the horizontal gold dashed lines.
I note first that the surface average reflection is about 116 W/m2, only slightly smaller than the 120 W/m2 we saw in the all-sky data in Figure 3. This shows that the albedo of the surface and the albedo of the clouds are quite similar, with the clouds reflecting slightly more than the ice cap
And just like with the all-sky data, there is no trend in the surface data either. There is no indication at all of the claimed darkening of the surface.
Finally, I was interested in what to me was the most curious feature of Figure 4. This is the large dip in surface reflection in the summer of 2012 that reaches a minimum in July. I seemed to remember some oddity that year, and a bit of searching found this from the National Snow and Ice Data Center:
An intense Greenland melt season: 2012 in review
February 5, 2013
Greenland’s surface melting in 2012 was intense, far in excess of any earlier year in the satellite record since 1979. In July 2012, a very unusual weather event occurred. For a few days, 97% of the entire ice sheet indicated surface melting.
Now, we know from Figure 2 that water has a much lower albedo than ice. So we can see that meltwater on the icecap reduced the reflection of sunlight, and led to the 2012 summer drop in reflected solar energy shown in the CERES data.
The appearance of this July 2012 event in the CERES data supports the validity of the data, and also shows that the data should be more than precise enough to show any trend in the solar reflection over the fifteen-year period of the record … and despite that, there is no such trend visible.
Go figure … I don’t know why the original researchers are claiming a darkening of Greenland, but I’m unable to find it in the CERES data.
My Usual Request: If you disagree with me or anyone, please quote the exact words you disagree with. I can defend my own words. I cannot defend someone else’s interpretation of some unidentified words of mine.
My Other Request: If you think that e.g. I’m using the wrong method on the wrong dataset, please educate me and others by demonstrating the proper use of the right method on the right dataset. Simply claiming I’m wrong doesn’t advance the discussion.