Technically this is possible, but in practice I am not so sure. You could script the generation of numerous Observation Areas, that are all offset by 1 mesh point. This would capture the slices of the volume in different observation areas, which you could then export the fields using the scripting language as well.
However, if you are measuring the absorption of a waveguide you would probably be fine with only having a couple observation areas and interpolating the data in between the areas. If you are exciting propagating modes then the field distribution should be changing very slowly over the length of the waveguide.
Damian