This study presents an experimental verification of the excess path loss for a satellite-to-Earth radio link shadowed by a single tree for elevation angles in the range from 20° to 40°. A measurement campaign was carried out in scale inside an anechoic chamber with five different species of trees at the X (8–12 GHz) and Ku (12–18 GHz) frequency bands to obtain the signal level variation as a function of the distance. It was discriminated the effect of the ground receiver terminal location with respect to the tree that affects the elevation angle value perceived by the receiver and the shadowing introduced by the tree. Results were discussed comparing with excess loss predicted by different vegetation scattering models aiming to characterise the goodness of their performance in terms of the excess signal loss predicted for short thickness of vegetation. In general, the models show a good agreement with measurements far away from the tree, however failing to describe the signal level variation if the terminal is near the tree. Nevertheless, it was determined that the models can be used to establish and upper and a lower limit to the excess signal loss.