Groundwater resources play a pivotal role in irrigation for agricultural crop production, providing ~40 percent of irrigation water supply worldwide. However, studies indicate that a substantial share of this groundwater is drawn from nonrenewable aquifers, placing into serious question the long-term ability of these systems to sustain irrigation for crop production while meeting other human and environmental water needs. The future behavior, trajectory, and sustainability of these agricultural-groundwater systems is further subject to a host of hydrologic, geologic, and socioeconomic uncertainties. We utilize a large-scale coupled farm-groundwater modeling approach for exploring future outcomes in agricultural systems that rely on groundwater for irrigation. The model combines an agent-based positive mathematical programming representation of farmer cropping and irrigation decisions with a cost-curve based method for simulating the evolution of groundwater production costs and availability. The model is deployed across the continental United States, simulating farm-groundwater outcomes across a wide range of potential future hydrologic and economic conditions. We further interrogate the sensitivity of these outcomes to key parametric uncertainties of the coupled farm-groundwater systems. Model results reveal potential hotspots of groundwater depletion, including identification of the unique combination of future conditions and coupled system characteristics that lead to outcomes of concern. We finally map out an anticipated experimental design for conducting a full global sensitivity analysis utilizing our new modeling approach.