Northen hemisphere ice and snow area, first look
Hemisphere area coverage of snow and ice for land and ocean. Early data has been excluded after discovering an anomaly with early daily sea ice data, excluded is early part which is every other day data.
This is an experimental first attempt at creating a hemisphere ice coverage dataset.
Why? I don’t like hidden information, want to see for myself and perhaps shed more light on whatever is going on.
As with sea ice data there is a distinct increase in the amplitude of the annual cycle in recent years. No attempt has been made to produce a more regular annual cycle exclusion.
How was this done?
Best time resolution data is weekly for Rutgers and daily for g02315 NOAA sea ice.
Combining these two poses a problem.
- land data is area
- ocean data is extent only for daily data
I failed to find a computed law between extent and area so I took a pragmatic route. Noting the monthly g02135 datasets does give area and extent, and the difference between them (reported here) I told software to produce a crude model of the difference (easy for me to do).
Experimenting with the monthly data gave a good enough result
Not perfect but reasonable. As it turns out ridiculously simple to do, literally the product of the model output and extent data. (was a bit of a huh! moment but if it works…)
Playing a party trick, all I have to do is change the model output to daily and it is ready to use with daily sea ice extent data.
As a final step I decided to output the result on the weekly data points rather than interpolate the land data to daily. Plots are figure 1.
Detailed work would probably refine the result slightly. Keep in mind the work might be a mistake and plain wrong.
Data. I’ve not provided anything, takes time and effort. If you want data, discuss.
1. Sea ice every other day seems to be all there is. Published by NOAA seems to be a version of the earlier NASA STEREO dataset. It’s possible the distinct change when the data moved to literally daily is real but coherent with a sampling change is notable. Not been investigated, particularly what precisely changed.