[Spice_discussion] SPICE Geomagnetic kernel

[Bogdan Nicula]

Hello SPICE experts.

It was bugging me for a while that, as it stays written in pck0000x.tpc:

     The SPICE Toolkit doesn't currently contain software to model the
     earth's north geomagnetic centered dipole as a function of time.

One can write a text frame kernel capturing the orientation at a certain fixed epoch, that’s certain.

Also one can write software like:
http://hesperia.gsfc.nasa.gov/ssw/stereo/gen/idl/spice/stereo_igrf_pole.pro
calculating on the fly the requested frame. This approach has the disadvantage that functions like pxform() cannot be used directly or that one cannot use the GF subsystem.

The following sounds perhaps somewhat crazy: I had the idea of creating a C-kernel with the orientation of the MAG frame.

I interpolated the geomagnetic harmonic coefficients for each day between 1900-2015 of the epochs of IGRF11. In order to be able to interact with the CK subsystem, I also created a “spacecraft clock” kernel with rate 1 with respect with TDB and an offset to allow negative “clock ticks”. Then wrote the orientation in a type 3 CK (linear interpolation), the idea being that the rate of change is very low and the errors during a day of interpolation would be insignificant.

I think it turned up quite OK, therefore I would like to share with you the results:
http://proba2.oma.be/aux/data/spice/kernels/ck/geomagnetic.bc
http://proba2.oma.be/aux/data/spice/kernels/fk/geomagnetic.tf
http://proba2.oma.be/aux/data/spice/kernels/sclk/geomagnetic.tsc

Would appreciate your thoughts, advice, rotten tomatoes…

Cheers,
Bogdan