A knowledge of spatio-temporal variability in potential yield is essential for site-specific nutrient management in crop production. The objectives of this project were to develop a model for photosynthetically active radiation (PAR) intercepted by almond and walnut trees based on data obtained from respective tree(s) and estimate potential crop yield in individual trees or in blocks of five trees. This project uses proximally sensed PAR interception data measured using a lightbar mounted on a mobile platform and a crop growth model to estimate potential yields of almond and walnut trees. An analytical model was developed to estimate PAR intercepted by the tree in which tree canopy was assumed to be spherical in shape. PAR intercepted by a tree was estimated taking into account the effect of row spacing, tree spacing within the row, latitude and longitude of the orchard, day of the year and row orientation. Scans were collected at solar noon in almond and walnut orchards during the 2012 and 2013 growing seasons. Diurnal scans were also collected during the 2012 season and were used to validate the model. Estimated versus measured data of PAR interception in almond and walnut trees had coefficient of determination of 0.86 and 0.94, respectively. The coefficient of determination for the relationship between actual yield and absolute midday PAR intercepted was 0.81 and 0.63 for almond and walnut trees, respectively. The coefficient of determination for the relationship between actual and potential yield was 0.80 and 0.59 for almond and walnut crops, respectively. Actual yield from those trees with lower values of midday PAR interception was found to be closer to their respective potential yield than those trees with higher values of midday PAR interception. The results suggest that there is a potential to use spatially variable PAR interception data to implement site-specific input management and enhance production.