Determining Variable Rate Fertilizer Dosage in Forage Maize Farm Using Multispectral UAV Imagery

Abstract

This research aimed to evaluate the capability of the combination of aerial UAV multispectral imagery and an equation-oriented approach for monitoring nitrogen status and variable-rate nitrogen fertilizer management in forage maize farms. To achieve this goal, four levels of nitrogen fertilizer were applied in a randomized complete block design (0, 50, 100, and 150%) in eight-leaf and tasseling growth stages. A method based on the biomass of aerial organs and leaf nitrogen content was used to estimate variable rate nitrogen application. Among vegetative indices extracted from aerial images, the correlation between the Normalized Difference Vegetation Index (r = 0.77, P ≤ 0.01), Nitrogen Reflectance Index (r = 0.70, P ≤ 0.01) and Modified Triangular Vegetation Index2 (r = 0.67, P ≤ 0.01) with leaf nitrogen content were positive and significant at the eight-leaf growth stage. Similarly, the Normalized Difference Vegetation Index (r = 0.77, P ≤ 0.01), Nitrogen Reflectance Index (r = 0.87, P ≤ 0.01) and Modified Triangular Vegetation Index2 (r = 0.66, P ≤ 0.01) had a high correlation with the leaf nitrogen content at the tasseling growth stage. Based on the obtained results, a total of 223, 192, 173, and 100 kg/ha urea fertilizer were estimated to be applied in 0, 50, 100, and 150% nitrogen fertilizer plots, respectively. Findings suggested that the nitrogen changes and nitrogen rate needed to apply were detected by aerial multispectral imagery with good accuracy.