2
$\begingroup$

I want to rotate a vector from the inertial J2000 frame into ITRF2008. I am using the NASA SPICE library, which provides a rotation matrix from J2000 to ITRF93 (they plan to upgrade the library to a modern ITRF in the near future but so far only ITRF93 is supported).

So my task is to then rotate from ITRF93 to ITRF2008. Given my vector $V_{J2000}$, I can easily compute $$ V_{ITRF93} = R_{ITRF93}^{J2000} V_{J2000} $$ where the rotation matrix $R_{ITRF93}^{J2000}$ is given by the NASA SPICE library. Now, according to http://itrf.ign.fr/doc_ITRF/Transfo-ITRF2008_ITRFs.txt, the transformation from ITRF93 to ITRF2008 is given as follows: $$ V_{ITRF93} = V_{ITRF2008} + T + M V_{ITRF2008} $$ where the vector $T$ and matrix $M$ are given in the link above. Rearranging this gives: $$ V_{ITRF2008} = (I + M)^{-1} (V_{ITRF93} - T) $$ so this amounts to a translation by $T$ followed by a matrix multiplication. However, one issue is that the matrix $I+M$ is not orthogonal (according to the parameters in the link above). So this is not really a rotation, but a transformation followed by a scaling.

It seems to me there should be an orthogonal rotation matrix allowing me to go from ITRF93 to ITRF2008: $$ V_{ITRF2008} = R_{ITRF2008}^{ITRF93} V_{ITRF93} $$ Does anyone know how to compute this rotation matrix?

My ultimate goal is to compute the spherical coordinate components of $V_{ITRF2008}$ (i.e. geocentric latitude, longitude and radius).

$\endgroup$
2
$\begingroup$

Such a rotation matrix does not exist. In Vallado's book I found the following: "Because these stations [used to derive ITRF] are affected by plate tectonic motion ... conversions between ITRF solutions involves translation, scale, and rotation"

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.