Sustainable and environmentally benign switchgrass production systems need to be developed for switchgrass to become a large-scale dedicated energy crop. An experiment was conducted in California from 2009 to 2011 to determine the sustainability of low-and high-input irrigated switchgrass systems as a function of yield, irrigation requirement, crop N removal, N translocation from aboveground (AG) to belowground (BG) biomass during senescence, and fertilizer 15 N recovery (FNR) in the AG and BG biomass (0-300 cm), and soil (0-300 cm). The low-input system consisted of a single-harvest (mid-fall) irrigated until flowering (early summer), while the high-input system consisted of a two-harvest system (early summer and mid-fall) irrigated throughout the growing season. Three N fertilization rates (0, 100, and 200 kg N ha À1 yr
À1) were applied as subtreatments in a single application in the spring of each year. A single pulse of 15 N enriched fertilizer was applied in the first year of the study to micro-plots within the 100 kg N ha À1 subplots. Average yields across years under optimal N rates (100 and 200 kg ha À1 yr À1 for low-and high-input systems, respectively) were 20.7 and 24.8 Mg ha À1 .However, the low input (372 ha mm) required 47% less irrigation than the high-input system (705 ha mm) and achieved higher irrigation use efficiency. In addition, the low-input system had 46% lower crop N removal, 53% higher N stored in BG biomass, and a positive N balance, presumably due to 49% of 15 N translocation from AG to BG biomass during senescence. Furthermore, at the end of 3 years, the low-input system had lower fertilizer 15 N removed by harvest (26%) and higher FNR remaining in the system in BG biomass plus soil (31%) than the high-input system (45% and 21%, respectively). Based on these findings, low-input systems are more sustainable than high-input systems in irrigated Mediterranean climates.