I would like some clarification on tech tip 4114. It describes the calibration formulae that ENVI applies to convert raw DN values to physical radiance values for ASTER 1A and 1B data. I see that for ASTER 1A calibration, the formula c1 * DN / c2 + c0 in effect applies 4100 different calibration formulae to the data – one for each column of pixels (and derived from the RadiometricCorrTable). For ASTER 1B, however, the data has already been calibrated to radiance, so I can’t see why it would be necessary to apply the unit conversion coefficient (UCC), particularly as the formula includes the DN-1 term [(DN - 1) * UCC]. Perhaps it is necessary to “scale the data back to floating point radiances”, as stated in the tech tip, because the ASTER 1B data has been supplied with the inverse of this UCC scaling applied, in combination with the 1A formulae? A number of writers stress the necessity of applying the UCC, but it’s not clear to me whether this is an alternative to the ASTER 1A calibration formula, or whether it should be applied to the results of this. In other words, do I need to apply the UCC formula to my 1A calibrated radiance data, or can I use it as is for my next pre-processing steps? Any thoughts most welcome, Harry PS My data was processed in ENVI 4.6.