Thanks, but I am only seeing one graph.
I look forward to the other three.
Cheers
Look again...
Thanks very much for that. It gives more insight and my guess is that we are seeing the error bars of NSIDC fighting, or at least interacting, with the error bars of PIOMAS. It doesn't help that one is partially derived from the other. Further guessing, I think it's the limits of sensors and inherent noise swamping the signal as it gets to the limits of resolution.
In defence of PIOMAS, the model has refined by comparing previous surveys and I suspect that we are just not in Kansas anymore. I keep thinking of previous expeditions that, based on satellite data, expected solid ice, only to find huge areas of slush or porous and "rotten" ice.
Another area that I find suspect is how they keep track of multi-year ice. It's a daunting task, when thickness is not a reliable indicator of age, particularly along the north of the CAA and Greenland. Again, not a criticism, just a point to consider as it applies to PIOMAS.
As for the Baffin Bay chart, I spent quite a bit of time looking at old PIOMAS data and NSIDC data. I thought it might be due to Nares staying closed but I found nothing that would cause that anomaly in the 2000's. That decade was actually pretty predictable, compared to the 80's and 90"s and the fact that it's an average makes it all the more suspect. I think it must be a data set error or a quirk in the computation.
Unfortunately, I don't have a great eureka moment to share, I just know now to take PIOMAS with a large grain of salt, whatever that means.