The isobaric (and other options) are not truly "iso --" but quasi
because the motion is only approximated by computing the velocity
required to maintain the parcel on the iso surface. When the next time
period's meteorological data are read in, a parcel may find itself on
another surface. So the departure from "iso" is one measure of the
precision of that particular trajectory (precision not accuracy, that is
another issue). As to the vertical motion plot, you must be careful because
the vertical scale is height above model ground level, not pressure level.
So that as you are traveling along a pressure surface the distance between
the ground and the trajectory endpoint will change (decrease with
higher elevation and increase with lower elevation).
I hope that answers your questions.
Glenn Rolph and Roland Draxler
NOAA ARL