Wip,

Do you perchance have pixel counts for your 16 thickness bins? If those numbers were available for each member of a time series, it would be neat to make one of those wiggle line graphics (area ordinate, thickness abcissa) to see how the proportioning of volume across thickness classes is changing. This assumes that the Piomas grid is more or less equal-area. This would be similar but different from how NSIDC does their ice age classes (as a normalized stack).

It may be that daily numbers are too noisy and some sort of rolling or weekly average would be necessary. It might be best just to address the Arctic Ocean basin.

I did not have such data readily available, also the PIOMAS grid is far from equal area. On the other hand I know the area of each grid cell.

Anyway with the data that I have I may be able to do something of interest. The PIOMAS algorithm calculates at every time-step a thickness distribution for each grid cell. The data is available as gice. Gice gives 12 sub grid cell thickness "bins":

[0.00, 0.26, 0.71, 1.46, 2.61, 4.23, 6.39, 9.10, 12.39, 16.24, 20.62, 25.49]

The lowest bin is the fraction with "no ice", second bin is the fraction with thickness 0<h<=0.26m and so fort.

So to calculate the area of all ice belonging to a particular bin, I take each cell, multiply the gice[bin] with the cell's area and sum it over the PIOMAS domain.

In the attach graph I compare on day 151 (31st May) of each year the area per thickness categories. Here I sum the bins, starting with the highest thickness to get lines indicating ice thicker than 16.24, thicker than 12.39 all the way to ice thicker than 0 .

Now the line indicating thicker than 1.46m (fourth from the top) is kind of interesting. The 1.46m is close to what (in the PIOMAS universe) can be melted down to zero within a single melting season. This is shown by plotting the 10th September area (bin area of all bins except the first) (dash-dotted line) which is near the ">1.46 31st May" line, bit above before 2007, below after that.

The "volume gap" is clearly visible and suggests a possible record low area. Note that the 2012 record low is not "foreseen", that could happen again (in the opposite direction).