Electronics360: "Testing the Theory of Relativity: Atomic Clocks"

Are we making certain assumptions that may be failing us? If a slightly erratic earth orbit contributes to some orbital variation, wouldn't it be possible for other physical influences to make the measurements inaccurate? Electromotive forces, gravity or undetected mass? There seems to be no compensation for any of those.