Relative atmospheric correction (AC) methods avoid the evaluation of atmospheric components of any kind. They rely on the fact that, for one sensor channel, the relation between the radiances at Top Of Atmopshere and at ground level follows a linear trend for the variety of Earth features present in the image. This linear relation is in fact an approximation of reality, but for practical purposes it is precise enough when there are other, more important sources of error.
The relative AC methods are Two reflectance measurements and Two reference surfaces.
The output of this method is an absolute atmospherically corrected image, so it can be used on an individual basis for multi-temporal comparison or parameter evolution and also for flux quantification. “Absolute” means that the image output has physical units and that the calculated ground radiances are compatible with the actual atmospheric constituents. The application of this method requires the use of a portable radiometer able to measure in the same wavelength range as the image band to be corrected, preferably at the time of the satellite overapass. If many bands are to be corrected, then the radiometer should have filters that allow measurement in all these individual bands separately.
The output of this method is an image that matches a reflectance that is compatible with the atmosphere of a similar image taken on a previous date. No absolute values of radiances are obtained in any of the two images, only allowing comparative results. This method works on an individual band/channel basis and is valid for establishing a basis for a uniform comparison to study, for example, the evolution of non-flux related parameters such as indexes, or when certain convenient land properties can be derived directly or indirectly from the normalized radiance values in a band.
The method relies on the existence of at least one dark and one bright invariant area. Normally, a sizable area should avoid mixed pixels (mixed land cover). As rule of thumb it should be a minimum of 2 or 3 times larger than the image spatial resolution. Reflective invariant areas are considered to retain their reflective properties over time. Deep reservoir lakes, sandy beaches or deserts, open quarries, large asphalted areas, and large salt deposits are examples of areas that are reflectively invariant. The supposition is that, for these pixels, the reflectance should always be the same since the reflective properties of the materials of which they are composed do not vary with time. If a difference in reflectance occurs for the reflective invariant area in the two date images, it can only be attributed to the different state of the atmosphere on those dates. The atmospheric composition is unknown in the two images, but its influence is measurable by analysing the change in radiance for the reflective invariant areas for the two dates.