Guest Post by Willis Eschenbach
The CERES data has its problems, because the three datasets (incoming solar, outgoing longwave, and reflected shortwave) don’t add up to anything near zero. So the keepers of the keys adjusted them to an artificial imbalance of +0.85 W/m2 (warming). Despite that lack of accuracy, however, the CERES data is very precise and sensitive.
As an example of what that sensitivity can reveal about the climate system, consider Figure 1, which shows the upwelling (outgoing) longwave (LW) and reflected solar shortwave (SW), month by month, for 13 years (N=156). Since these are individual CERES datasets, their trends and values should be valid.
Figure 1. Upwelling longwave (shades of blue) and upwelling reflected shortwave (shades of red) for the globe as well as the two hemispheres separately. Cyclical seasonal variations have been removed.
Now, there are several very curious aspects to this figure. The first and most surprising issue is that the hemispheric values for shortwave, and also the hemispheric values for longwave, are nearly identical from hemisphere to hemisphere. Why should that be so? There is much more ocean in the southern hemisphere, for example. There is solid land at the South Pole rather than ocean. In addition, the underlying surface albedos of the two hemispheres are quite different, by about 4 watts per square metre. Also, the southern hemisphere gets more sunlight than the northern hemisphere, because the earth’s orbit is elliptical.
So given all these differences … why should the longwave and shortwave in the two hemispheres be the same?
The next thing of interest is the stability of the system. The trends in all six of the measurements are so tiny I’ve expressed them in W/m2 per century so that their small size can be appreciated … if the trends continue, in a century they may change by a watt or two. Note that despite the small spread of the measurements, none of the trends are significant.
The next thing of interest is that in addition to the values being similar in both hemispheres, the trends are also quite similar. All of the trends are very slightly negative.
Finally, despite the great difference in the size of the LW and SW signals (240 vs 100 W/m2, Figure 1), the size of the variations in the two signals are quite similar. Here is a boxplot of the three pairwise comparisons—the anomaly variations in global, and northern and southern hemisphere.
Figure 2. Boxplots of the variations in the longwave and reflected shortwave shown in Figure 1, for the globe (left panel), the northern hemisphere (center panel) and the southern hemisphere (right panel).
Since these are boxplots, we know that half of the data lies inside the colored boxes. This means that half of the time, the longwave and the shortwave are within ± one-half watt of the seasonal value. Plus or minus one-half watt half the time, and within a watt and a half for 95% of the time, for a total of 156 months … this to me is amazing stability.
Given the myriad differences between the northern and southern hemispheres, my explanation of this amazing stability is that a) the temperature of the planet is regulated by a variety of threshold-based processes, and b) the set-point of that regulation is controlled by globally consistent values for the physics of wind, water, and cloud formation.
Now, there certainly may be some other explanation for this amazing stability and symmetry of the climate despite the large differences in the geometry and composition of the two hemispheres. That’s my explanation. If you have a better one … bring it on.
Best regards to all,
NOTE ON DATA AND CODE: I’ve turned over a new leaf, and I’ve cleaned up my R computer code. I’ve put all the relevant functions into one file, called “CERES Functions.R”. That file of functions, plus the data, plus the code for this post, are all that are required to duplicate the figures above. I just checked, it’s all turnkey.
DATA: CERES 13 year (220 Mbytes, has all the CERES data in R format.)
FUNCTIONS: CERES Functions.R (Has all the functions used to analyze the data.)
CODE FOR THIS POST: Amazing Stability CERES (Has the code to create the figures and calculations used above.)