Conventional nitrogen (N) management for potato production in the Upper Midwest, USA relies on using split-applications of N fertilizer or a controlled release N product. Using remote sensing to adaptively manage N applications has the potential to improve N use efficiency and reduce losses of nitrate to groundwater, which are important regional concerns. A two-year plot-scale experiment was established to evaluate adaptive N-management using remote sensing compared to conventional practices for Russet Burbank variety potatoes grown on an irrigated, coarse-textured soil in Becker, MN. Nitrogen treatments included (Control) a 45 kg N/ha control treatment, (270 Split) a splitapplied urea treatments of 270 kg N/ha, (270 CR) a controlledrelease polymer coated urea [PCU] treatments of 270 kg N/ha, and (VR Split) a variablerate splitapplied urea treatment based on remote sensing observations using the MERIS Terrestrial Chlorophyll Index [MTCI] interpreted using the Nitrogen Sufficiency Index [NSI]. Using the CROPSCAN MSR-16R ground-based multispectral radiometer, spectral reflectance measurements were collected weekly during the growing season to monitor crop N status in the variable rate treatment (VR Split) and calculate NSI with 270 CR as the “well-fertilized” reference. If the NSI value immediately prior to scheduled fertilizer application was less than 0.95, then N-fertilizer was applied to VR Split at a rate of 22 kg N/ha. The variable-rate treatment (VR Split) received 248 and 226 kg N/ha in 2016 and 2017 respectively, which is 22 and 44 kg N/ha less than the conventional N management practices (270 Split, 270 CR), and there were no significant differences in the quantity or quality of tuber yield between these treatments. This study demonstrates that adaptive N management using remote sensing is a promising method to optimize N rate and timing to account for spatial and temporal variability in crop N status for potato production.