That depends on what your resulting quantity is and how it is calculated from the measured values. Typically a satellite will measure some kind of radiance with a camera and here the calibration uncertainty plays a big role, but those values probably has be to translated to some kind of vegetation index and here the main uncertainty will probably come from other sources: atmospheric changes, uniformity of the target, accuracy of the reflectivity estimates etc. My quick guess is that the calibration uncertainty is insignificant compared to other sources of uncertainty when relating the NDVI to actual vegetation.

Normalized Difference in vegetation Index (NDVI) results should be ascertain with ground truthing. The similarity in vegetation types, radiometric caliberations, cloud haziness, smoke blockage in big industrial zones, atmospheric filtering, spatial resolution of sensor, dynamic range (bit size) may contribute in uncertainities, thus the satellite images processing results would be ambiguous.