There actually is a decent test for whether PIOMAS is biased towards the high side.
Let's assume that Extent/Area are decent measures. If PIOMAS were wrong, then you would expect-
-Extent/Area have larger declines than PIOMAS.
-More late summer PIOMAS decline as PIOMAS is forced to catch up with open water.
-Higher thickness values as stated by PIOMAS since the area is not something that can be distorted.
None of these have happened. Extent and Area have also shown flattening in the last few years. PIOMAS has actually shown less late summer volume decline. PIOMAS has had very low thickness numbers the last couple years.
The most likely hypothesis is that PIOMAS is reasonably accurate and recently the ice has actually not melted as much as people thought it would.
i find yours a good post.
that said, i don't think they're accurate but for now i think you have good arguments that they are.
i once said that it will be a big "ahaaa...." experience once the flaw will be found.
one argument, while i think it's correct, i'd like to supplement. it's not the melting that i for example think is faster than they say but the freezing i believe is slower than before.
further the state of the ice allows for quite some distribution that would remain above the 15% even though there's less ice-mass. this would as well explain why the number are eventually going down parallely in late summer even though they could both be either incorrect and/or be distracting.
i mean if we have a 100% ice cover and a 20% ice cover that is a huge difference and still they would both still show full extent. as to area i don't thing that current resolutions of several kilometers always allow to measure small the many smaller gaps between smaller ice floes correctly.
what i suspect (dunno) is a sudden drop, once the ever warmer winters, ever warmer waters and ever warmer air will have reduced the remaining ice to a thickness and fragmentation that it will drop below the 15% in masse as well as the gaps between floes will become big enough to be always or better detected by sensors.