Surely the graph Wipneus produced is the area flux of ice across some line in Fram Strait connecting Svalbard and Greenland multiplied by the average thickness of the ice. So ice melting before it reaches that line, or ice exported through the gap between Svalbard and FJL etc., is not counted in this graph. And ice melting in the Greenland Sea after the line is not relevant to this figure.
And since the graph is in absolute numbers, it doesn't show the export as a percentage of melt.
And the graph doesn't show the volume of just old ice being exported.
All of which are quite interesting, but are more difficult (some impossible?) to calculate.
Just as an example, The Lincoln Sea at the entrance to Nares Strait is currently exporting very young thin ice on a southbound surface current. As soon as the ice gets flushed, the polynya freezes again, and within some days the new ice is exported again. Over several months this might show as significant exported volume but is totally sustainable. OTOH, if the "arch" in the Lincoln Sea were to crumble, we could see export of very old and thick ice over some weeks, which would be a major blow to the arctic sea ice even if total exported volume would be similar.
So my "wishful thinking" would be to calculate the net effect of export minus the ice created behind in the exporting source zones, which also depends on low temperatures able to refreeze the opening leads and polynyas.