Fertilizer application for crop production under variable climate condition is not well studied in Ethiopia. Models can be used to develop best recommendation based on weather; soil and experimental data in predicting yield and providing agronomic recommendation with climate scenarios. The objective of the study is to develop agronomic recommendation for maize production based on the virtual climate variability with a tool DSSAT model seasonal analysis at Melkassa, Ethiopia. A field experiment was undertaken at Melkassa, to study a combined agronomic strategy with three maize cultivars from Short Season (SS); Medium Season (MS) and Long Season (LS); two irrigation level and eight rates of urea fertilizer treatments were used and all other crop management activities applied uniformly with standard plot arrangement. All measured data on phenology, grain yield and biomass from the field experiment were used for model simulations. These weather scenarios were used in the seasonal analysis program to run each treatment combination with 30-year data. The results of both biophysical and economic analyses of the Seasonal Analysis program predicted an application of medium season cultivar with irrigation water as moisture source and 400 kg/ha urea with 200 kg/ha crop residues as the most dominant management option for maize production at Melkassa Ethiopia. The present study revealed that the generated future weather data were reliable and DSSAT could successfully use it to predict the future crop yields under different management practices and select the best one for sustainable production of maize crops using DSSAT crop models.