Many soil properties influence nitrous oxide production in a highly non-linear manner, and most of these soil properties vary significantly within and between fields. The same soil properties influence soil fertility and N fertilizer requirements of crops. Application of fertilizer N at rates in excess of crop requirements increases the risk of nitrous oxide emission. Crop yield response to applied fertilizer N is also highly non-linear. Thus, developing management strategies to maximize economic yield response to applied N fertilizer, while minimizing the risk of nitrous oxide emissions remains a challenge, particularily in fields with significant soil spatial variability. A field study was carried out to measure the effects of N fertilizer application rate on nitrous oxide emission. The influence of within field spatial variability of soil biophysical properties on the relationship between N fertilizer rate and nitrous oxide emission was also assessed, and the spatial variability of crop yield response to N fertilizer measured. Results of the study indicate the presence of significant field scale spatial covariance of parameters for empirical functions describing cumulative nitrous oxide emissions and crop yield, as a function of applied N fertilizer rate. A combined stochastic-empirical model was developed based on the measured spatial covariance relationships and the relationship of the parameters to selected soil biophysical properties. Implications of the nature of the spatial covariance relationship to optimizing field scale crop yield response to applied N fertilizer while minimizing nitrous oxide emissions, for fields with variable soil biophysical properties is presented.