white band: previous years yes/no/, where/when, other satellites
These are all useful avenues of investigation. The white band is not somehow 'inevitable' in C-band radar -- no one predicted its existence and no journal article describes or explains it, possibly because Ascat is seriously under-utilized and Sentinel-1AB is fairly new.
Looking back, the white band first appears in the Ascat archive at the end of August near the end of melt season at the seasonal marginal ice zone (MIZ), the transition zone between the ice pack and open Arctic. That cannot be the whole story because the Pacific half of the potential boundary zone doesn't form the white band. Some other tracker is needed there to mark the shape-shifting division between FYI and MYI ice over the freeze season.
The white band is initially far more extensive, forming a border from below the New Siberian Islands all the way around to Svalbard. Only the Laptev portion persists; the Atlantic side is still very subjected to wind effects in late December (uniq's AMSR2 gif above).
In the satellite archives we use, the visible is fairly literal (corrected reflectance) as is the infrared (offered in various false color palettes), not that different from the view out an airplane window with appropriate goggles.
Looking at Nasa's WorldView, there wasn't a clear enough day over the time and location window to determine whether or not the white band is visible at these wavelengths. The white band is not an open lead or polynya so not brighter/warmer in Suomi band15 infrared.
It is no coincidence that Ascat and Sentinel-1AB both see the white band: they observe at very similar wavelengths of 5.7cm and 5.5cm (5.255 and 5.405 GHz, C-band in radar WWII terminology). Sentinel-1AB resolves much greater detail but coverage is episodic and difficult to tile.
The question is, can the white band be seen on the ground, with instrumented airplane flyovers, or in L-band or other radar products? In other words, the band's electromagnetic reflectance spectrum is key to determining what it is (or at least ruling out what it is not).
Microwave products like Smos are brought into visible 8-bit interpreted products after extensive processing algorithms. Smos looks at a longer wavelength of 21.4cm(1.4 GHz L-band) and can measure distance through the ice (of a half meter or less thickness) to sea water. The white band is not apparent; note though the archive provides only the highly processed view.
The white band cannot be seen in AMSR2, CMEMS, Piomas, Hycom, OsiSaf, SicLeads, Smap, Smos-Smap, Jaxa, cryosat2smos or other products.
Ascat brightness diminishes rapidly with near-surface ice and atmospheric path polarity (dielectric); older ice is farther along in brine exclusion so whiter. However this is not a good fit to the white band along the former MIZ unless that ice is somehow so crumpled that brine channels have already drained. Other forms of elastic scattering can relate to surface roughness or particle size relative to incident wavelength.
Only Ascat sticks to a literal beam reflectance (grayscaled 0-1 or σ⁰ backscatter, lower left corner scale in files). It was never intended for ice; the primary instrument justification has been ocean surface wind speed measurement at mid to low altitude.
We discovered here years ago that Ascat holds hidden treasures: after a series of photoshop-type contrast enhancements, the initially bland imagery can accurately display ice movement as well as regions of brine exclusion maturity (older thicker ice) and other persistent but unexplained interior features. There is no use of Ascat enhancement in the Arctic science community.
Yesterday's before and after are shown below, with emphasis on MYI. Movement of many of the ice substructures can be tracked for months, sometimes deforming but still persisting in time series, so not processing artifact. Click to see at full scale.
There's a lot more going on with the ice pack than just one-parameter thickness.