MULTI-AGENT SIMULATION APPROACH ON THE IMPACT OF AGRICULTURAL LAND-USE CHANGE ADAPTATION STRATEGY (FARM CREDIT) ON FARM HOUSEHOLD LIVELIHOOD IN SEMI-ARID GHANA

摘要

This study applied multi agent simulation to investigate the impact of farm credit (an agricultural land-use change adaptation strategy) on farm household livelihood. General household and land use data were obtained from 186 sampled household. Two household types were identified; the first household type was better-off in terms of land area cultivated and income generated from rain-fed rice. On the other hand the second household type was better-off in terms of land area cultivated and income generated from maize. Determinant of crop choices of each household types were generated via logistic regression using household and plot characteristics as predictors. Crop choice model and agricultural yield dynamic model was programmed in Netlogo 5.0.5. Using a process based decision, household choice of maize adoption with respect to maize credit (maize credit scenario) was simulated, and its impact on farm household livelihood was compared with baseline scenario.