This paper reports the variations in crop production in the Midwestern United States resulting from climate change associated with a hypothetical regional nuclear conflict. We used Agro-IBIS, a dynamic crop growth model to simulate the response of maize and soybean cropsto cooler, drier, and darker conditions resulting from the nuclear war related smoke. Weemployed observed climate conditions for regions in Iowa, Illinois, Indiana and Missouri and perturbed these observations using the output from a general circulation model simulation of a scenario in which 5 Tg of elemental carbon was injected into the upper troposphere. The results show notable reductions in both maize and soybean production for a decade after the hypothetical war, which would have important consequences for food production. Maize yields would be expected to decline between 10 and 40 percent while soybean production would be reduced by 2 to 20 percent. Temporal variation in the magnitude of yield changes for both crops generally follows the temporal variation in climatic anomalies, exhibiting the largest decline in the first five years following the 5 Tg event and then gradually leveling out. The drivers of yielddecline for both crops appear to be primarily caused by changes in the length of the growing period and, to a lesser extent, by reduced precipitation and changes in maximum daily temperature during the growing period. The seasonal average of daily maximum temperature difference between the control and nuclear scenarios has a quadratic relationship to yield differences where extreme changes lead to increased loss of yield but mean changes have a neutral effect.