Proper nitrogen status monitoring is essential to achieving a healthy potato crop and to maximizing yield. Petiole collection and analysis is the most common way of monitoring crop nitrogen status. A possible alternative to this method is the utilization of hyperspectral remote sensing and computer-assisted modeling. Biophysical parameters, including nitrogen status, can be related to the spectral reflectance of the crop canopy, which is captured in hyperspectral imagery. Collected spectral data can be applied to statistical models to relate the observed wavelengths to a desired biophysical quality of the crop. In addition to predicting nitrogen, crop yield can also be predicted using similar methods, allowing growers to better understand the growth of their crop in response to different inputs and management strategies. One of the most promising methods of creating predictive models from hyperspectral datasets is to use partial least-squares regression (PLSR), a computational modeling method which utilizes weighted predictor components to model the relationship between predictor variables and a response variable. This project aimed to capture hyperspectral images as well as petiole nitrogen status and tuber yield values from research plots consisting of two potato varieties (Russet Burbank and Soraya) grown under four nitrogen application schemes. Ground truthing data was collected weekly during the tuber bulking stage over two growing seasons. These data were combined with the extracted spectral data to be used with PLSR modeling, to develop decision support models for predicting crop nitrogen status and final yield at harvest.