Determination of sufficiency values of canopy reflectance vegetation indices for maximum growth and yield of cucumber

Large nitrogen (N) fertilizer applications are a feature of intensive vegetable production systems, and optimal N management is required to maximize N use efficiency and minimize N losses. Vegetation indices (VIs) of canopy reflectance, measured with proximal sensors, are generally strongly related to crop N status. For practical application, sufficiency values that distinguish between N deficiency and sufficiency are required. In this work, sufficiency values of Vls for maximum crop growth and for yield were determined for two cucumber crops grown in contrasting seasons (Autumn and Spring). Canopy reflectance was measured with a Crop Circle ACS-470 sensor. Sufficiency values for maximum growth were based on the relationships of VIs with the Nitrogen Nutrition Index (NNI), i.e. the ratio between actual and critical crop N contents. Sufficiency values for maximum yield were based on linear-plateau relationships of yield with VIs. Strong relationships were obtained between all VIs and both NNI and yield for most of the weekly measurements during both crops. For NNI, best-fit relationships were linear, quadratic, power or exponential, and had coefficients of determination (R-2) of 0.61-0.98. For yield, most linear-plateau relationships between yield and Vls had R-2 values of 0.47-0.89. VIs based on reflectance in green and red edge had slightly better relationships with NNI and yield than VIs in the red, with the Green Normalized Difference Vegetation Index (GNDVI) and the Green Ratio Vegetation Index (GRVI) being the most sensitive and consistent indices for estimating both crop NNI and yield. Relationships between VIs and NNI and yield for all weekly measurements of each crop, and for the two crops combined, were also analyzed to provide unique sufficiency values for maximum growth and yield that applied to the entire crop cycle of each crop and of both crops considered together. Overall, there were slight differences between sufficiency values for maximum growth and for maximum yield and the unique sufficiency values were generally intermediate to the weekly sufficiency values. This study demonstrated the potential for using VIs for monitoring crop N nutrition and yield in cucumber. The calculated sufficiency values of VIs may facilitate the use of proximal optical sensors in farming practice for optimal N management through periodic monitoring for deviation from sufficiency values. (C) 2016 Elsevier B.V. All rights reserved.