|Franzen, D - NORTH DAKOTA STATE UNIVERSITY|
|Schepers, James - RETIRED ARS EMPLOYEE|
|Holland, K - HOLLAND SCIENTIFIC|
|Raun, W - OKLAHOMA STATE UNIVERSITY|
Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/22/2016
Publication Date: 7/14/2016
Citation: Franzen, D.W., Kitchen, N.R., Schepers, J.A., Holland, K.H., Raun, W.R. 2016. Algorithms for in-season nutrient management in cereals. Agronomy Journal. 108(5):1775-1781. doi: 10.2134/agronj2016.01.0041.
Interpretive Summary: Many tools are now available for cereal producers when making nitrogen (N) fertilizer recommendations. Some tools rely on measurements of the soil, while others are crop growth modeling tools that integrate weather, soil, and crop genetics. Over the last several decades, light reflectance canopy sensors have also been employed as a technology on which to base variable-rate N applications in cereals. It has been shown that use of crop sensing can improve N use efficiency (NUE) and cereal grain yields. The demand for improved decision making products for cereal production systems has placed added emphasis on using plant sensors in-season that directly address site-specific growing environments. The objective of this work was to provide a review of in-season sensor-based decision rules (often called algorithms) presently being used in cereal grain production systems. The reviewed algorithms primarily come from multiple land-grant University and government programs, though one commercial algorithm was also included. A common thread in this review is the present use of active sensors for quantifying differences in fertilized and non-fertilized areas within a specific cropping season. In-season prediction of yield potential and crop N response over different sites and years has been demonstrated using the sensor reflectance information, but typically other information is also needed (e.g., planting date, sensing date, cumulative growing degree days, and rainfall). While the development and promotion of other approaches for nutrient management continues, algorithms using active sensors for in-season N management are viable, affordable, and can be modified by producers as deemed appropriate. Grain producers will benefit by using canopy sensing technology for N fertilizer applications because they can reduce excess N applications, which will save them money. If fertilizer can be better matched with crop need, N loss to lakes and streams will also be reduced and the environment will be improved.
Technical Abstract: The demand for improved decision making products for cereal production systems has placed added emphasis on using plant sensors in-season, and that incorporate real-time, site specific, growing environments. The objective of this work was to describe validated in-season sensor based algorithms presently being used in cereal grain production systems for improving nitrogen use efficiency (NUE) and cereal grain yields. A review of research programs in the Central Great Plains that have developed sensor-based N recommendations for cereal crops was performed. These embodied algorithms generated within multiple land-grant University and government programs date back to the first passive sensor algorithm tested. A common thread in this review is the present use of active sensors, particularly those using the Normalized Difference Vegetation Index (NDVI) for quantifying differences in fertilized and non-fertilized areas, within a specific cropping season. In-season prediction of yield potential over different sites and years was possible using NDVI, planting date, sensing date, cumulative growing degree days (GDD), and rainfall. Other in-season environment-specific inputs have also been used. The earliest formal report using a passive estimate of NDVI was in 1974, and that has advanced to by-plant N fertilization using NDVI and by-plant statistical properties. Most recently, sensor-based algorithm research has focused on the development of generalized mathematical models for determining optimal crop N application. Algorithms of this type rely on an a priori understanding of crop nutrient use related to crop growth and yield. While the development and promotion of fee-based modelling approaches for nutrient management continues, algorithms using active sensors for in-season N management are viable, affordable and that can be modified by users as deemed appropriate.