for my Master-Thesis in marine ecology I will map changes in mangrove-distribution over a period of 30 years in BCS, Mexico, using Landsat MSS, TM and ETM+ data provided by GLCF. To compare the images of different time periods I need to correct them radiometrically. My question now is: is it necessary to do the calibration using this algorithm, which involves gain, bias, sun elevation,...? Because I cant find this information in the Metadata of the images of 1990. I read papers, where there was simply done radiometric correction by doing Normalization processes through linear regression. Would that technique be enough for my purposes? I am not a remote sensing student, but a biologist, so I try to keep my analysis simple but - of course- appropriate. Thank you very much for your help.
To compare your images, you will need to correct them radiometrically and also for atmospheric effects. The radiometric correction can be done in certain softwares which access the signal gain and offset data over the web and make the appropriate adjustments for you.
The atmospheric corrections are more difficult and require the isolation of the atmospheric spectrum and its removal by a dark-subtraction techinque. The atmospheric spectrum can be found in pixels over deep water. Finally, the normalisation process you refer to is still required to convert your data to pseudo-reflectance.
I am happy to correspond further should you need more help.