Assessing the Agronomic Subfield Variability by Sentinel-2 NDVI Time-Series and Landscape Position

Abstract

Optimizing crop yield is one of the main focuses of precision farming. Variability in crop within a field can be influenced by many factors and it is necessary to better understand their interrelationships before precision management methods can be successfully used to optimize yield and quality. In this study, NDVI time-series from Sentinel-2 imagery and the effects of landscape position, topographic features, and weather conditions on agronomic spatial variability of crop yields and yield quality were analyzed. Landscape position allowed the identification of three areas with different topographic characteristics. Subfield A performed the best in terms of grain yield, with a mean yield value 10% higher than subfield B and 35% higher than subfield C, and the protein content was significantly higher in area A. The NDVI derived from the Sentinel-2 data confirms the higher values of area A, compared to subfields B and C, and provides useful information about the lower NDVI cluster in the marginal areas of the field that are more exposed to water flow in the spring season and drought stress in the summer season. Landscape position analysis and Sentinel-2 data can be used to identify high, medium, and low NDVI values differentiated for each subfield area and associated with specific agronomic traits. In a climate change scenario, NDVI time-series and landscape position can improve the agronomic management of the fields.