Improving the Performance of Active-Optical Reflectance Sensor Algorithms Using Soil and Weather Information.

Active-optical reflectance sensors (AORS) use corn (Zea mays L.) plant tissue as a bioassay of crop N status to determine future N requirements. However, studies have shown AORS algorithms used for making N fertilizer recommendations are not consistently accurate. Thus, AORS algorithm improvements should be explored. The objective for this research was to determine if soil and weather information could be utilized with the University of Missouri AORS algorithm (ALG_MU) to improve in-season (~ V9 growth stage) N fertilizer recommendations. Nitrogen response trials were conducted across eight states over three growing seasons, totaling 49 sites with soils ranging in productivity. Nitrogen fertilizer recommendations from the ALG_MU were related to the end-of-season calculated economic optimal N rate (EONR). The RMSE of the ALG_MU over all sites was 81 and 74 kg N ha^-1 with 0 and 45 kg N ha^-1 applied at planting, respectively. When adjusted using weather (total and distribution of precipitation) and soil properties (clay and soil organic matter), RMSE values improved by 26 and 22 kg N ha^-1 for 0 and 45 kg N ha^-1 applied at planting, respectively. Without adjustment, 22 and 29% of sites were within 34 kg N ha^-1 of EONR with 0 and 45 kg N ha^-1 at planting, respectively. With soil and weather adjustment, 45 and 51% of sites were within 34 kg N ha^-1 of EONR for corn that received 0 and 45 kg N ha^-1 at planting, respectively. Thus, site-specific weather and soil information could be used to improve the performance of the ALG_MU.