Development of an algorithm for optimizing nitrogen fertilization in wheat using GreenSeeker proximal optical sensor

Abstract Proximal plant sensing with active canopy sensors offers a leap in the non-destructive assessment of crop agronomic information. For managing fertilizer nitrogen (N), sensor readings must be translated using functional models or algorithms to fertilizer amounts. Six field experiments were conducted in three wheat seasons in the West Nile Delta in Egypt to develop and validate an algorithm based on GreenSeeker canopy reflectance sensor for field-specific fertilizer N management in wheat, which takes into account both spatial and temporal variability of N during the crop growth season. The proposed algorithm is based on the prediction of total N uptake and response index of N uptake determined from normalized difference vegetation index measured by the sensor from plots differing in yield potential as established by applying a range of fertilizer N levels in the four experiments conducted in the first two wheat seasons. The treatments in the two experiments conducted in the third wheat season were designed to define appropriate fertilizer N management prior to applying a sensor-based dose at Feekes 6 stage (jointing stage). The application of 40 and 60 kg N ha−1 at 10 and 30 days after sowing of wheat and a sensor-guided dose of N estimated by using the algorithm developed in this study resulted in yields similar to those obtained by following the general recommendation, but with an average of 66 kg N ha−1 less fertilizer N. These results were also reflected in a substantial increase in N recovery (21.9%) and agronomic (7.7 kg grain kg−1 N) efficiencies compared with the general recommendation, thereby proving the usefulness of the sensor-based algorithm in optimizing fertilizer N management in wheat.