Each of the defined land qualities with their associated attribute was digitally encoded in GIS
database to create the nine thematic layers. The diagnostic factors of each layer were assigned with
factor rating values (S1=1.0 for highly suitable, S2 = 0.8 for moderately suitable, S3 = 0.4 for
marginally suitable and N = 0.1 for unsuitable).
The evaluation model for each crop was given using the values of the factor rating as equation (1)
These layers were integrated by spatially overlaying each with the suitability model of the
defined nine layers, which yielded 4 classes according to the results shown in table 1.
The reliability or accuracy of the suitability map for each crop was assessed, based on the
suitability maps conducted by LDD (2005) for rubber tree and ground truth surveys for rice, sugar-caneand cassava. The number of the ground truth locations surveyed for rice, sugar-cane and
cassava accounted for 66, 23 and 18 locational samples respectively .