Search (over 200 pages above & below)

Crop Agriculture

Image-Based Atmospheric Correction & Scatter Background

(See Landsat & Sentinel-2 Conversion to SR Tutorials drop-down menu for step-by-step conversion to surface reflectance based on image-based atmospheric correction)

Image-based atmospheric correction removes/deducts atmospheric scatter (haze) reflectance from Top of Atmosphere (TOA) reflectance (TOA reflectance is represented by the total amount of light measured by the satellite in simplified diagram below). Wavelengths larger than visible scatter, but not nearly as much as visible (as is listed on the Relative Scatter Calculator page). The process of removing scatter is based on the premise that in a satellite image scene containing millions of pixels there should be pixel reflectance values that are virtually zero; the reason there is not for certain bands, is due to atmospheric scatter, which erroneously increases reflectance to higher values (shifting the entire histogram to the right).

Atmospheric scatter diagram for satellite imagery atmospheric correction and surface reflectance


A reflectance amount should be deducted from all values in the scene to account for this increase. The image-based method to account for the increase is called dark object subtraction (DOS) and was developed by Chavez (1988). DOS does not consider secondary scattering into shadowed areas (Chavez, 1996). Chavez (1996) then deducted .01 from the established scatter reflectance value (so that the established low value had a surface reflectance of .01) because of "the fact that very few targets on the Earth's surface are absolute black, so an assumed one-percent minimum reflectance is better than zero percent" (Chavez, 1996).

Visible light striking off particles in the atmosphere prior to striking the surface (then transmitting back to the satellite sensor, erroneously increasing sensor values) is the main source of scatter, and is the type accounted for in image-based atmospheric correction (illustrated in the simplified atmospheric scatter diagram above). Blue scatters the most, followed by green, red, then NIR (scatter amounts for bandwidths larger than NIR are negligible at most, and do not need to be deducted). A clearer atmosphere has less overall scatter and a higher ratio of scatter between smaller and longer wavelengths than a hazier atmosphere; conversely, a hazier atmosphere has more overall scatter and more equal scatter between bands (Chavez, 1988). There is a negative power relationship between band center wavelength and scatter (Chavez, 1988). Scatter can also result in light being redirected to the sensor from a different path after it has reflected off the surface, or being redirected back down to the surface (downwelling irradiance; neither are shown in diagram).



Chavez, P.S., Jr. 1996. Image-based atmospheric corrections–revisited and improved. Photogrammetric Engineering and Remote Sensing 62(9): pp.1025-1036.

Chavez, P.S., Jr. 1988. An improved dark-object subtraction technique for atmospheric scattering correction of multispectral data. Remote Sensing of Environment 24: pp.459-479.