USING CROP SIMULATION-MODELS AND GIS FOR REGIONAL PRODUCTIVITY ANALYSIS

The scope of applicability of site-specific models can be extended to regional planning and productivity analysis by combining their capabil ities with a Geographic Information System (GIS). In this study, regio nal productivity analyses for three sites in western Puerto Rico were carried out using the DSSAT-BEANGRO V1.01, the dry bean (Phaseolus vul garis L.) model and AEGIS. DSSAT is an integrated decision support sys tem that contains several crop models with standardized input and outp ut. AEGIS is a regional planning decision support system that uses DSS AT capabilities within a GIS for regional productivity analysis. The a nalysis indicated that a considerable soil and weather variability exi sts within the three study sites. The optimum management factors such as cultivar selection, planting date and irrigation strategy would sig nificantly differ from one site to another. For the Mayaguez and Isabe la areas, the long season cultivar 'Porrillo Sintetico' would always b e a better choice. It would consistently out perform the short season cultivar ''Cuarentena. '' On the other hand, in the semiarid environme nt of Magueyes, 'Porrillo Sintetico' would perform better than ''Cuare ntena'' in good rainfall years, and the reverse would happen in bad ra infall years. However, in the long run, 'Porrillo Sintetico' would out yield, 'Cuarentena' by 20% even in this area. The simulated ''best'' planting dates for rainfed condition matched fairly well with the reco mmended ''best'' dates for two study sites. However, for the third sit e (Mayaguez), these dates varied considerably for both rainfed and irr igated conditions. The cumulative probability distribution analysis fo r these dates showed that the simulated ''best'' dates would produce h igher yields for all simulated years. The environmental diversity (soi l and weather) of the study sites makes the results of this study indi cative of several other locations in the Caribbean basin and other tro pical regions of the world.