"Agriculture uses more water than any other human activity in the world. Projections of increasing demands on agriculture along with climatic change and variability necessitate evaluating the current and potential contributions of irrigation water to agriculture, and initiatives to improve production while reducing our demands on surface and groundwater systems. This thesis develops and demonstrates new procedures to determine appropriate estimates of agricultural production and water use resulting from proposed changes in irrigation water use. This research improves on previous efforts to evaluate the increasing or decreasing of full irrigation at regional-scale, and presents the first broadly-applicable methodology for evaluating supplemental irrigation at nested spatiotemporal scales.Irrigation has generally been evaluated at its extremes, defining areas exclusively as either fully irrigated or non-irrigated. Previous efforts to estimate the contributions of irrigation to current yields and evaluate potential changes in production have determined generalizations relating irrigated and non-irrigated yields from temporally and spatially limited statistical data. Such efforts have admittedly limited their investigations by evaluating only the extremes of irrigation and further by correlating the separate contributions of surface water and groundwater to production. This research first improves on estimating the potential of the extremes, that is increasing or decreasing full irrigation, by evaluating changes in irrigated area as they relate specifically to irrigation-source. Further, this research develops and demonstrates the first broadly-applicable methodologies for evaluating the in-betweens of full and non-irrigation, namely supplemental irrigation. Supplemental irrigation holds significant opportunity to increase agricultural and water productivity, and reduce water use. By recognising a system of irrigation different than the extremes, including potentially reducing water use on irrigated fields and increasing limited water use on non-irrigated fields at appropriate and opportune times, the in-betweens of irrigation allow for the benefits of supporting production while not necessarily further appropriating water for irrigation.This thesis first evaluates the potential changes in agricultural production and water use resulting from the complete expansion of full irrigation under two scenarios related to the adoption of specific irrigation sources, demonstrated for rice and wheat in India. The results show a potential increase in production of 14-25% for rice and 3% for wheat, with a 31% and 3% increase in water use respectively. Similarly, the study investigates the decreases in agricultural production from stopping irrigation, and we estimate that rice would be at 60% of current production, and wheat at 51%. Together, these two evaluations showcase the end ranges of the relationship between agricultural production and irrigation water use. Specifically, this is achieved by partitioning region-specific rice and wheat production into that related to irrigated and non-irrigated areas, and further by irrigation source. This partitioning of production by irrigation source, to the best of our knowledge, is novel. The partitions are used to estimate potential increases in agricultural production, and evaluating such increases as explicitly related to irrigation source, is similarly novel. " --