A Regional Investigation of In-season Nitrogen Requirement for Maize Using Model and Sensor-based Recommendation Approaches

Authors

• Laura J. Stevens, University of Nebraska-LincolnFollow

Date of this Version

5-2014

Document Type

Thesis

Citation

Stevens, Laura J. 2014. A regional investigation of in-season nitrogen requirement for maize using model and sensor-based recommendation approaches. MS thesis, University of Nebraska-Lincoln.

Comments

A Thesis Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Master of Science, Major: Agronomy, Under the Supervision of Professor Richard B. Ferguson. Lincoln, Nebraska: May, 2014

Copyright (c) 2014 Laura Joy Stevens

Abstract

N management for corn can be improved by applying a portion of the total N during the growing season, allowing for adjustments responsive to actual field conditions. This study was conducted to evaluate two approaches for determining in-season N rates: Maize-N model and active crop canopy sensor. Various sensor algorithms designed for making in-season N recommendations from crop canopy sensor data were evaluated. The effects of corn hybrid and planting population on recommendations with these two approaches were considered. In a 2-yr study, a total of twelve sites were evaluated over a 3-state region, including sites in Missouri, Nebraska, and North Dakota. In-season N recommendations were generally lower when using the sensor-based approach with Holland and Schepers (2012) algorithm than the model-based approach. This resulted in observed trends of higher partial factor productivity of N and agronomic efficiency for the sensor-based treatments. At specific sites, conditions leading to high levels of mineralized N becoming available to the crop during the growing season increased environmental and economic benefit of the sensor-based approach. The optimum N rate was estimated using a linear-plateau model. Compared to the sensor-based approach with the Holland and Schepers algorithm, the model-based approach more closely estimated the optimum N rate and erred by over-recommending N. Profit loss from the sensor with Holland and Schepers algorithm was greater when considering all sites collectively due to the greater cost of lost yield when N was under-applied, versus the lower cost of excess N when N was over-applied.

Adviser: Richard B. Ferguson