Given the present fascination with 'early melt ponds; and appreciating that I am a naÃ¯ve newbie, nevertheless, I wanted to reiterate the potential significance of a couple of factors affecting Arctic Sea Ice melt; Solar Irradiance, and Heat Transfer from adjacent oceans. It takes energy to melt ice. Where does this energy come from... specifically?

For the seasonal cycle - The sun!

I've been meaning to incorporate a new equation into my equation to calculate insolation, that would convert my current top of atmosphere values of insolation into values weighted by atmosphere optical thickness (Insolation = incoming solar radiation). But haven't got round to it, so what follows isn't really complete...

I have calculated PIOMAS 5 day differences in volume because my current insolation calculations are every 5 days. Then I have calculated average insolation north of 65degN (this is a very simple back of envelope type calculation that doesn't take into account area). When I get round to it I'll do the insolation calculations on a grid box basis, I did say this wasn't complete.

All this is done using the average seasonal cycle for PIOMAS daily data from 1981 to 2010..

Math types will spot that taking the difference then cumulative summing just gets you back to where you started, but I do it because it removes the overall volume from consideration.

So that graph is insolation and the seasonal cycle in volume, OK, but not very informative. But when I use the 5 day difference in volume and plot it together with the insolation things get interesting. I've inverted the volume scale so that as volume goes down the plot goes up and we can see the relationhip with insolation.

What I see there is that in the late summer as insolation declines it drags down the volume loss - the energy just isn't there to melt ice. The peak of insolation is around the summer solstice (June 21), but before that is fascinating to me, where the volume loss seems to lag the insolation. This divergence is, I suspect because of the delayed fall in albedo (how 'bright' the ice is) as the surface starts to melt in the model in May onwards.

The sun's heat is supplied not only from surface melt, but also from warming ocean water as the ice retreats, and the warming atmosphere (e.g. warm winds from warming landmasses around the Arctic). However the seasonal cycle is all driven by the sun.

Or did you mean what has caused the long term decline in volume?