I think the first, most conspicuous feature of the global mean energy flow system is the appearence of small integer ratios. Some relationships between certain energy flow components seem well-established: for example, the correlation between the net radiation at the surface and half of the outgoing radiation in the clear-sky is very well documented, both in the history of radiation transfer theory, both in different observations.
A very specific but easily demonstrable connection stands between the total radiative energy absorrbed by the surface and twice the outgoing longwave radiation, still in the clear-sky. This is a feature of the simplest geometric greenhouse model, but surprisingly accurate in the observations.
The all-sky versions of this pair of clear-sky equations is easily generated, by separating atmospheric transfer of radiation from the longwave effect of clouds; thus, LWCRE appears in these equations.
The two all-sky equations are satisfied by a difference of 0.1 Wm-2 on the global energy and water cycle assessments of NEWS and GEWEX, and the four equations together verified by 22 years of observations in the NASA CERES datasets within an unimaginably small difference of 0.0007 Wm-2.
But the fact that the unit flux (the building block, or the pacer, or pace-maker of the energy flow system) is LWCRE, and that it equals 1/51 of the total solar irradiance, is mind-blowing.
But the other, very accurately verified fact that both in observations and in the climate models, solar reflection at the top of the atmosphere, both in clear-sky and all-sky, occupies integer position (8 units and 15  units on the intercepting cross-section disk to incoming solar radiation, thus 2 units and 15/4 units after spherical weighting by a factor of 4), without any reference to surface reflective properties, or to cloudy reflective properties or to atmospheric aerosols, is so implausible that I wouldn't dare to mention them if they were not soo accurately verified in the datasets (53.36 Wm-2 in the integer ratio system, 53 Wm-2 in every model in the clear-sky, and 100.05 Wm-2 in the arithmetic ratio system, and 100 Wm-2 in the several IPCC data sources; similar to solar absorbed in the atmosphere: 73.37 and 80.04 Wm-2 in the clear-and all-integer system, 73 and 80 Wm-2 in the observations and models). — But, after all, science is not about plausibility but validity, isn't it?
These recognitions certainly point beyond our current understanding of solar reflection in Earth's physical system, and calls for some kind of new physical theory, a total determination with quantized energy flows in the annual global planetary mean. Within this context, understanding of our observed climatic processes: fluctuations around, or permanent (systematic) changes and deviations from the integer positions, and their causes, is more urgent than ever.
A physics tutor at a History and Philosophy of Science course at the university told: "The physics of the 21st century will be as novel and incomprehensible compared to the physics of the 20th century as it was novel and incomprehensible compared to the classical physics of the 19th century."

Welcome in the 21st century?