The Offset Tracking tool of SNAP is based on intensities. You could, technically, split, calibrate and deburst SLC data but you will then end up with nearly the same information than already provided in the GRD. I see no large advantage in using SLC for this kind of analysis.
using SLC makes only sense if you make use of the phase, but there is no automated way of doing this in SNAP.
Otherwise, you can use GRD as well for common intensity tracking (included in SNAP)
if there is an interferometric approach for offset tracking, and you find a way to implement it, you can use the SLC products, of course.
Edit: some were published but I cannot tell if they are good and transferable:
Thx a lot. I have seen the tutorial about the offset tracking based on GRD product in Amundsen Antarctic, which talked about that GRD has been detected, multi-looked and so on. So that’s to say I don’t need to do calibration, multi-looking, right?
and I still have a question. when I exported output as GeoTIFF format to ARCGIS, it could’t generate pyramid and can’t project to Polar Lambert Projection. it had extreme distortions.