Table of Contents
The standard L2 products which can be derived using the MEPIX tool in QWG mode are:
The algorithms have been implemented on the basis of the recent DPM and IODD versions (i80, i8r0). One major difference to the IPF implementation is that the pressure products are calculated for all surfaces. This has been done in order to be able to assess later the possibility of using the pressure products for pixel classification. However, over cloud and cloud free land pixels, respectively, the cloud top pressure and land surface pressure products should be identical between MEPIX and the standard processing.
The cloud top pressure (CTP) is basically derived with an algorithm as described for the standard Level 2 processing ([1], section 4.5.3). The main characteristics of this algorithm are summarized in the following table.
Table 3.1. Cloud Top Pressure derival
| MEPIX CTP algorithm characteristics | Comparison of MEPIX and IPF/megs products |
|---|---|
| To retrieve the Cloud Top Pressure,a neural net (NN) approach is used. The MERIS signals in channel 10, 11, the surface albedo and the geometry (sun zenith angle, viewing zenith angle and azimuth angle) are used as input of the Neural Network. The net produces the cloud top pressure. Depending on the surface albedo two different neural nets are used (one for surface albedo equal to zero, one for nonzero surface albedo). Neural Nets are selected according to spectral shift index. | Implementation is the same. Differences in results less than 1 percent for pixels flagged as cloudy (likely due to truncation errors) |
The surface pressure is also derived with an algorithm as described for the standard Level 2 processing ([1], section 4.5.2). The main characteristics of this algorithm are summarized in the following table.
Table 3.2. Surface Pressure derival
| MEPIX surface pressure algorithm characteristics | Differences to standard L2 algorithm |
|---|---|
| To retrieve the surface pressure, a polynominal algorithm is used, as described in detail in [1]. | Implementation is the same. Differences in results less than 1 percent for pixels over land and not flagged as cloudy (likely due to truncation errors). |
The cloud flags are also derived with an algorithm as described for the standard Level 2 processing ([1], section 5.5). The main characteristics of this algorithm are summarized in the following table.
Table 3.3. Cloud Flags derival
| MEPIX cloud flag determination characteristics | Comparison of MEPIX and IPF/megs products |
|---|---|
| For pixels identified as LAND, tests are performed on the ratio of Rayleigh-corrected reflectance at several wavelengths. The coarse Rayleigh correction uses an algorithm described in [1], section 5.5.6, to compute the reflectance due to Rayleigh scattering. Finally, a set of Boolean parameters are used to index a decision table which provides the CLOUD_F flag (see [1], step 5.5.1 for details). | No differences. Almost full agreement in results. |