Final MLs

This dataset comes from the second phase of MAIL methodology as described in D2.3 where all participating datasets assigned with a weighted value per pixel. This value based on a Pairwise Comparison Matrix (PCM) of ranks where the value of each indicator (i.e., slope, moisture, salinity, productivity, soil contamination, pH etc.) are grouped into 3 classes according to their physical characteristics.

The final step in the mapping of MLs, is the reclassification of the resulting product of the weighted overlay into 3 classes, depicting marginality: 1) Marginal lands with high plantation suitability, 2) Marginal lands with low plantation suitability and 3) Potentially unsuitable lands. The subdivision of these categories was carried out with three different methodological approaches;

The first is to calculate the maximum and minimum value obtained in the MLs layer and divide the difference by three. This will obtain three ranges of equal magnitude.

The second methodological approach is to calculate the 25^th and 75^th percentiles of the values obtained in the MLs layer. This approach would penalize the “Marginal lands with high plantation suitability” and the "Potentially unsuitable lands" and increasing the number of pixels belonging to the middle layer "Marginal lands with low plantation suitability".

The third one is to calculate the 33^rd and 66^th percentile of the values obtained in the layer MLs. In this case, the total number of pixels is divided equally into the three categories. But with the difference with respect to the a) method that in this case the thresholds are established with the maximum number of pixels and not with the maximum and minimum values.