Diminishing land and water resources due to increasing demands from rapid population growth calls for increasing water use efficiency of irrigated crops.  To produce more for every drop of water used in agriculture, it is important to develop location specific alternate agronomic practices vis-à-vis breeding for water use efficient (traits) crop lines. In this context, adequately calibrated and validated agricultural system models provide a fast alternative method for developing agronomic practices keeping in pace with the technological advances in limited-water agriculture.  In this study, we calibrated and validated, with reasonable accuracy (RMSE of crop yield was 982 kg/ha), the CERES-maize (DSSAT v4.) model for simulating experiments with varying irrigation levels for corn at Akron, Colorado in the semiarid Great Plains. The model was further used for developing various alternate limited-water irrigation management practices making use of 89 years of historical weather data for the location.  Irrigation strategies developed for optimizing grain yield were (1) various levels of irrigation under no-rain and with-rain scenarios with fixed planting dates, (2) various levels of irrigation when planted at fixed dates with various levels of  soil water contents  vis-à-vis no rain and with rain, (3) plant only if optimal soil water is available under rainfed conditions, and then provide various levels of irrigation, and (4) plant only if initial soil water is optimum for sowing then irrigate only if the soil water is depleted below 40% of available water to be irrigated back to various levels. Water-Nitrogen interactions under the above scenarios also investigated. Results of the study will be useful for managing irrigation water under limited-water scenarios in the Great Plains.