In images retrieved from spectrometers such as MERIS, an increase in the radiances, especially in the near infrared bands, can often be observed over water and in the vicinity of vegetated coasts. The affected area can extend over 10km or even more. This increase is not (always) related to an increase in the aerosol optical depth but is caused by the so-called adjacency effect. This effect results when photons are reflected and scattered towards the sensor and where a substantial contrast exits between the target and its surrounding. During processing of such measurements the correction schemes need to take this effect into account, otherwise the increased radiance is erroneously associated with other physical processes, and the derived geophysical quantities have an increased error.
Observations over waters near to land surfaces, both for coastal and inland waters, are affected because of the large contrast in the red and infrared part of the spectrum, where water is almost black and vegetated areas are very bright. The effect is easier to illustrate in coastal waters because it decreases from coastline to off-shore. This effect can best be studied in MERIS Full Resolution images, but even in reduced resolution the adjacency effect can also clearly be seen.
In order to build up a reliable tool for the correction of the adjacency effect, the ICOL (= Improve Contrast over Ocean and Land) prototype processor was developed as BEAM plugin. Although this prototype (ICOL1.0) has proven to retrieve results as expected from theory, it has some well-known limitations:
To overcome these limitations and to be able to apply the AE correction within operational processing chains, an improved BEAM plugin (ICOL+) has been developed. With this processor, the AE correction is applicable anywhere over ocean as well as over land and considers case 2 waters, clouds and sea ice. An improved convolution scheme (the core mathematical part of the AE corection) has been integrated to significantly speed up the processor. Moreover, it is possible with ICOL+ to apply the AE correction on Landsat TM data. This feature mainly serves as a demonstration for the general portability of the AE correction scheme to other instruments within future projects (e.g. related to the Sentinel missions). However, the full correction scheme has been implemented as for MERIS and has shown to provide results as expected ([RD-2]).
This manual basically describes and illustrates how to use the application. For a detailed description of the underlying algorithms see [RD-1], the technical realisation of the software package is outlined in [RD-3].