The effect of nitrogen (N) fertilizer on switchgrass biomass production has been evaluated in a number of locations on a small-plot scale; however, field-scale information regarding switchgrass response to N and N use efficiency (NUE) in different regions of the USA is limited. Switchgrass was planted in South Dakota (SD), New York (NY), Oklahoma (OK), and Virginia (VA) in 2008 and in Iowa (IA) in 2009. Three N levels (0, 56, and 112 kg ha(-1)) were applied to 0.4-0.8 ha plots at each location beginning in spring the year after planting. Biomass production, N removal, apparent N recovery (ANR), and NUE were determined at all locations. Biomass yield response to N varied among locations and varied according to initial soil N concentration. Low initial soil N concentration increased biomass yield response to N fertilization, while high initial soil N concentration reduced this response. High amounts of initial soil N caused fertilizer N removal to be low. Fertilizer N uptake in switchgrass might be inhibited by competition from initial soil N. Seasonal temperature and precipitation were not strongly correlated with biomass yield and N-use of switchgrass at the studied locations. In this study, ANR was below 10% at all locations and years. Nitrogen-use efficiency was significantly related to initial soil N. High NUE was observed at locations where initial soil N was low. These data suggest that NUE depends on site-specific N management strategies that are responsive to soil N supply and plant N status. (C) 2013 Elsevier Ltd. All rights reserved.