Which active optical sensor vegetation index is best for nitrogen assessment in irrigated cotton?

Authors: Bronson, Kevin; Conley, Matthew; Hunsaker, Douglas - Doug; Thorp, Kelly; French, Andrew; BARNES, ED - Cotton, Inc

Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/1/2020
Publication Date: 4/5/2020
Citation: Bronson, K.F., Conley, M.M., Hunsaker, D.J., Thorp, K.R., French, A.N., Barnes, E.M. 2020. Which active optical sensor vegetation index is best for nitrogen assessment in irrigated cotton? Agronomy Journal. 112(3):2205-2218. https://doi.org/10.1002/agj2.20120.
DOI: https://doi.org/10.1002/agj2.20120

Interpretive Summary: Active optical sensors (AOS) use for nitrogen (N) management of row crops continues to grow. Since early studies in the mid-1990s, several commercial AOS are now available. Typically, canopy reflectance in red and near infrared bands are used to calculate the normalized difference vegetation index (NDVI). More recently, commercially available AOS include a third, red-edge band that allows the calculation of the normalized difference red edge (NDRE) vegetation index. We present five site-years of N management studies in Maricopa, AZ with cotton (Gossypium hirsutum L.) where we calculated 12 vegetation indices (VI’s) using canopy reflectance data from shortly after plant emergence to mid bloom. The study objective was to assess the ability of VI’s to detect N deficiency among various N management treatments. Results show that in four of five site-years, NDRE detected N deficiency in reduced N-reflectance-based treatments compared to well-fertilized soil test-based N treatments. Nitrogen deficiency in reflectance plots was observed in three of five site-years for NDVI amber (NDVIA), physiological reflectance index (PRI), and chlorophyll index vegetation index (CIRE), and in only one site-year with typical NDVI red (NDVIR). Petiole NO3 detected N deficiencies in reflectance treatment plots 7 to 33 d earlier than NDVIA or NDRE. Considering that AOS with an amber waveband are no longer available for purchase, we recommend the use of AOS with NIR and red-edge bands and the calculation of NDRE for use in-season N management of irrigated cotton.

Technical Abstract: Use of active optical sensors (AOS) in nitrogen (N) management of row crops continues to grow. Since the first studies in the mid-1990s, several commercial AOS are now available. Typically, canopy reflectance in red and near infrared (NIR) bands are used to calculate the normalized difference vegetation index (NDVI). More recently, commercially available AOS include a third, red-edge band that allows the calculation of additional vegetation indices (VIs). We present two studies of five site-years of N management studies in Maricopa, AZ on a Casa Grande sandy loam with cotton (Gossypium hirsutum L.). The 2014-2015 study was conducted under an overhead sprinkler irrigation system (OSI), and the 2016-2018 study was in subsurface drip irrigation (SDI). The study objective was to compare the ability of 12 different VIs to detect N deficiency among N treatments from shortly after emergence to mid-bloom. In the OSI study, which showed delayed, small N treatment effects, the normalized difference red edge index (NDRE) and the chlorophyll index using red edge (CIRE) detected N deficiency in zero-N and in reduced N-reflectance-based plots 7-23 days before other VIs did. With SDI, the choice of VI was less critical as several VIs could distinguish N deficiency in zero-N and in reflectance plots. The commonly used NDVI red (NDVIR) only detected N deficiency in reflectance plots in one of five site-years. In conclusion, we recommend the use of AOS with NIR and red-edge bands and the calculation of NDRE or CIRE to guide AOS-based in-season N management of irrigated cotton.