Agrology (Nov 2021)
Seasonal dynamics of normalized difference vegetation index in some winter and spring crops in the South of Ukraine
Abstract
Spatial crop monitoring using vegetation indices is one of the most promising technologies for crop mapping and remote phenological observations. The aim of the study was to determine the patterns of seasonal dynamics of the spatial normalized difference vegetation index for the main crops grown in the south of Ukraine and to connect it to their phenology. Remote sensing data provided by the OneSoil AI platform, which uses Sentinel-1 and Sentinel-2 imagery as a basis, was used to derive the monthly index values for the 2016–2021 growing season for nine selected crops grown in the experimental fields at the NAAS Institute of Irrigated Agriculture, Kherson, Ukraine. The fallow field was also included in the study to determine the cutoff values of the vegetation index, which are not representative of any healthy vegetation. It was determined that each crop has its unique pattern of the dynamics of the vegetation index, except for winter wheat and winter barley, which demonstrated quite similar models. The peak values of the vegetation index were observed in May for winter crops (wheat, barley, rapeseed) and early-spring crops (chickpea, peas), while the late-spring crops (grain corn, grain sorghum, soybeans, sunflower) reached the peak values in July. It is possible to suggest that the highest demand for mineral nutrition and watering will fall in the mentioned time periods of late spring and midsummer. Phenological monitoring revealed that the highest values of the spatial normalized difference vegetation index were observed in the following stages of crop growth, namely: winter wheat, winter barley – stem elongation; winter rapeseed – flowering; chickpea – branching; peas – budding and flowering; sunflower – stem growth; soybeans - pod formation; grain sorghum – panicle ejection and flowering; grain corn – panicle ejection and flowering. The results provide novel information for further implementation in the mathematical models for automation of crops recognition, mapping, and phenological observations based on the remote sensing data. Further scientific research in this direction will be aimed at increasing the spectrum of crops studied and a detailed investigation of the relationship between the value of the normalized difference vegetation index and their phenology.
Keywords
