Scholarly record

ESTIMATION OF WINTER WHEAT PARAMETERS FOR SITE SPECIFIC CROP MANAGEMENT BY UNMANNED AERIAL MULTISPECTRAL IMAGING

Vojtěch Lukas

Abstract

Assessment of cereal plant status and other biophysical parameters, such as nitrogen uptake, nitrogen content and total amount of aboveground biomass, is important information for decision support of agronomists for correct treatment of crop management practices. Especially in in precision farming, where whole area mapping is required, this information could be provided by remote sensing as an efficient method for the monitoring of crop within field heterogeneity. The aim of this study was to verify the use of unmanned aerial multispectral imaging for estimation of agronomic relevant crop parameters of winter wheat. For this purpose, an field experiment was conducted in 2018 in locality Kojcice (Czech Republic) with two fields (10.2 and 26.2 ha) sown by winter wheat. The aboveground biomass of canopy (dry), nitrogen concentration in plants and nitrogen uptake were estimated by plant sampling realized in two important vegetation stages ? stem elongation (BBCH 33) and heading (BBCH 51). Plant samples were taken from 23 (Field Kazy) and 33 samples (Field Vrcha). Simultaneously with plant sampling, an unmanned aerial mission was carried out by Micasense Red Edge multispectral camera, which records images in visible (RGB), red-edge (RE) and near-infrared (NIR) spectrum. Set of nine broadband vegetation indices was calculated: NDVI, GNDVI, EVI2, MSAVI2, NDRE, NRERI, RENDVI, SRI, etc. The results of correlation analysis showed that relationship between nitrogen uptake and vegetation indices differs in both vegetation stages and each field. In stem elongation, the correlation with N uptake was lower (max. r = 0.665, NDRE) in comparison to heading stage (max. r = 0.856, MSAVI2). Also sensitivity of indices to the vegetation indices varied - red-edge indices (NRERI, NDRE) proved positive correlation in stem elongation. In phenology stage BBCH 51 statistically significant positive correlation to the N uptake was reached (r = 0.685 ? 0.856) for all computed vegetation indices. As the conclusion could be stated that multispectral imaging is appropriated for identification of nitrogen uptake within the fields by remotely piloted aerial systems. However the relationship between crop parameters and vegetation indices was not consistent in vegetation stages and also in both fields, thus calculation of set of vegetation indices is useful, at least from the red-edge part of electromagnetic spectrum.

Publication Impact Profile

• Citations
• Scopus - Citation Indexes: 2

• Captures
• Mendeley - Readers: 8

PLUMX see details

DOI resolver: doi.org/10.5593/sgem2019/2.2/s10.065

Publication details

Back to publication list