The relative cost of Nitrogen (N) fertilisers in a cropping input budget, the 33% Nitrogen use efficiency (NUE) seen in global cereal grain production and the potential environmental costs of over-application are leading to changes in the application rates and timing of N fertiliser. Precision agriculture (PA) provides tools for producers to achieve greater synchrony between N supply and crop N demand. To help achieve these goals this research has explored the use of management classes derived from historic field data and in-season crop reflectance sensors in an attempt to quantify, and manage the effects of, spatial and temporal variation in N uptake. This simple study combines the two techniques to try and quantify in-season variation in N requirements, and furthermore attempts to improve the predictive ability of in-season yield prediction functions through the inclusion of historic soil and yield data sets. Experiments from two example fields are used to quantify seasonal variations in N using in-season reflectance data. A process was designed to build site-specific N requirement algorithms from reflectance and historic input data. The variation in historic yields and current season reflectance indices across potential management classes indicates that the magnitude of variation in plant N requirements is sufficient to implement management classes in conjunction with in-season crop reflectance sensors. Furthermore the development of modified site-specific yield prediction functions according to management classes built from soil ECa data, previous yield observations and calibrated yield prediction functions significantly enhanced yield prediction accuracy. These improved in-season yield predictions were used to construct N application strategies that proved more cost effective than traditional approaches. The combination of site-specific historic data and in-season reflectance information shows promise for the development of N application decision support to improve NUE in both economic and environmental terms.