Arctic Ice Polar Gridded Data
Posted by Jeff Condon on January 31, 2009
While working on the half dozen other datasets, I’ve also been continuing my now famous work on sea ice. After some small battles and study, I am now able to process the gridded data from NSIDC. This allows me to calculate a reasonable ice area variation. Below are two plots of the gridded data and masks as presented for the bootstrap algorithm. All data were provided by the NSIDC – Sea Ice Concentrations from Nimbus-7 SMMR and DMSP SSM/I Passive Microwave Data.
There is so little difference in the graphs above it’s amazing. The light blue spot in the middle is the data which satellites don’t measure, you can see the change from the old satellites to the new which cover a greater area. From the oldest year available to the most recent (lowest ice level January on record) Well now I can do a new trick, let’s subtract the two plots and see what we get.
All the red areas above are regions where ice level was reduced the green are increased ice. Since these are both winter January times how can ice be lost or gained near the pole? It must be a very fractional percentage since my plot doesn’t discern magnitude of loss very well. The next plot shows only the negative changes.
Well it will be fun over the next few days to dig into this a bit more. I’m still working on the frequency response of the tropospheric temps in the meantime. Looking at the graphs above, I wonder how much ice loss there is if we ignore the losses and gains near pole. Do they have a big or small effect on the trend results? On my last trip overseas, I got to fly over the Hudson bay, it was amazing. Totally covered with ice in December for certain. There were cracks but only due to shifting. If I look at the 2008 data in December, it should confirm my observation.