摘要：Sensor-based precision fertilizer technologies are being developed and researched by production scientists. One such technology uses normalized difference vegetation index (NDVI) reflectance measurements of growing winter wheat plants and a nitrogen fertilizer optimization algorithm (NFOA) to determine nitrogen requirement necessary for plants to reach their yield plateau. A number of precision fertilizer application systems that use this technology are considered in this paper. A linear response stochastic plateau wheat yield function conditional on NDVI reflectance measurements is estimated and used within an expected profit-maximization framework to estimate upper bounds on the returns from the precision nitrogen application systems. The on-the-go precision system that assumes perfect information was approximately $7 per acre more profitable than the convention of applying 80 pounds of nitrogen prior to planting in the fall. The whole-field precision system was break-even with conventional methods.
