Leaderboard

Gotcha. Unless you are an atmospheric scientist, or have a true grasp on atmospheric science, I would not attempt trying to program complex physical based models such as 6S or MODTRAN. Eric Vermote has dedicated his research career to 6S. Instead, I would recommend something like a scene-based model like cos(t), Chavez 1996. Easily programmable and implemented, and produces good results. Calibrate to TOA radiance, then to planetary reflectance, then apply cos(t). http://info.asprs.org/publications/pers/96journal/september/1996_sep_1025-1036.pdf http://earth.gis.usu.edu/imagestd/ (example code with equations).

Yes, the MTL.txt file contains all of the metadata that you need to duplicate the examples of cos(t) that I provided. You will find both gain and offset calibration factors to convert DN to either TOA radiance or TOA reflectance. You will also fin Sun Elevation to calculate the sun angle and you will find the Julian day to calculate Sun-Earth distance (d). Follow this user guide (second section) to understand how to convert data to TOA reflectance. http://landsat.usgs.gov/Landsat8_Using_Product.php