Integrating Management Zones and Canopy Sensing for Improved Nitrogen Recommendation Algorithms.

Active crop canopy sensors have been studied as a tool to direct spatially variable nitrogen (N) fertilizer applications, with the goal of increasing the synchrony between N supply and crop demand and thus improving N use efficiency (NUE). However, N recommendation algorithms have often proven inaccurate in certain subfield regions due to local spatial variability. Modifying these algorithms by integrating soil-based management zones (MZ) may improve their accuracy by allowing sensors to accommodate the entire spectrum of field conditions. The objective of this study was to use soil properties and maize response to N to delineate field-specific MZ and then evaluate whether these MZ can effectively identify areas with differential response to N fertilizer. This integrated approach was compared to sensor-based application alone and to uniform N management. Experiments were conducted in 2016 and 2017 on 8 producers’ irrigated fields in south central Nebraska, USA, each differing greatly in local topography and soil properties. Soil apparent electrical conductivity (EC_a), reflectance, and landscape position data were collected with a Veris^® MSP3 on-the-go soil sensing platform. Field-length treatment strips were then located in regions of greatest spatial variability. 10 to 16 N response blocks (45m x 12m) were placed end to end in the strip. Blocks consisted of six smaller plots arranged in a 2x3 randomized complete block design. Six N rates ranged from 0 to 280 kg ha^-1, with increments of 56 kg ha^-1. For two sites in 2016, soil EC_a was significantly correlated to both mid-season NDRE and grain yield (P<.0001). However, delineated MZ showed differential yield response to N at only one site in 2016. These preliminary results suggest algorithm refinement may require soil or field specificity. Additional information from 2017 sites will be presented.