A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science, School of Environment, Resources and Development

Thesis (M.Sc.) - Asian Institute of Technology, 1999

The target area was Cai Lay district, which is the most intensive rice cultivation in Mekong River Delta. The study was based on primary data collected through farmer survey. These data were applied for farming system analyses and also for model validation. Experimental data were collected from experimental stations and provincial offices and applied for model validation. The CERES-rice model were then applied for optimization the intensified cropping system in the Mekong River Delta. The results indicate that farmers used too much nitrogen causing losses and low nitrogen recovery fraction. Farmers 'yield declined at nitrogen application above 50 kg N/ha. At nitrogen level up to 80 kg N/ha, there was a good agreement between observed and simulated data. Above 80 kg N, the simulated results were above the observed ones. This is due to frequent logging. Based upon model validation, the CERES-rice model can be applied for prediction of nitrogen and rice cultivation management in alluvial soils in Mekong River Delta. By simulation, it was found that to obtain highest yield and high nitrogen recovery fraction, at high rates, nitrogen should be applied into three splits. Optimum nitrogen fertilizer can be replaced by 80% of the available soil nitrogen. Simulated results vary between 67-185 kg N/ha according to the amount of available soil nitrogen before sowing. Using 25-year simulated weather condition and present price variation and the Gini-dominance approach, the optimum was found to be 100 kg N/ha.