Assessment of plant nutrition parameters in cereals by PlanetScope and Sentinel-2 multispectral data

Abstract

The use of satellite data with high spatial and temporal resolution is a key approach for assessing the nutritional status of field crop stands during the growing season in precision agriculture. This study focuses on the use of multispectral data from the PlanetScope (PlanetLabs) and Sentinel-2 satellite for assessment of the nutritional status and monitoring the development of field crops during the growing season 2022 and 2023. An field experiment with the area of 29 ha was conducted in the South Moravia region (Czech Republic; N 48o58.03', E 16o38.39') by the support of the research projects TA CR TQ03000882 and MENDELU AF-IGA2023-IP-036. The crop survey was realized in winter wheat (2022) and winter barley (2023) between BBCH 29 and BBCH 50 by plant sampling with total number of 20 samples accross the field. The plant samples were analysed for Nitrogen content and aboveground biomass, from which Nitrogen uptake (Nupt) and Nitrogen Nutrition Index (NNI) were calculated. Various vegetation indices were calculated from Sentinel-2 and PlanetScope data including EVI (Enhanced Vegetation Index), GNDVI (Green Normalized Difference Vegetation Index), NDRE (Normalized Difference Red Edge Index), NDVI (Normalized Difference Vegetation Index), SAVI (Soil Adjusted Vegetation Index), SRI (Simple Ratio Index), TCARI (Transformed Chlorophyll Absorption Ratio Index). Correlation analyses and regression modelling were performed between the vegetation index values and crop parameters obtained by traditional plant sampling. The results from PlanetScope satellite data showed that in 2022, in the early stages of vegetation (BBCH 30), the vegetation indices EVI (R = 0.91), GNDVI (R = 0.889), NDVI (R = 0.886), SAVI (R = 0.886) and SRI (R = 0.886) achieved the highest correlation values compared to the measured crop parameters. In the second half of the crop growing season (BBCH 50), the vegetation indices NDRE (R = 0.826) and GNDVI (R = 0.792) reached high correlation values compared to the stand parameter fresh biomass. In 2023, almost all vegetation indices reached high correlation values with the biomass, NNI and Nupt in BBCH 29-30, except TCARI and EVI. In the second term, only the vegetation indices GNDVI and NDRE compared to biomass and nitrogen uptake per hectare reached statistically significant correlation values. In the last observation period (BBCH 50), there was an increase in the frequency of vegetation indices that reached statistically significant correlations, especially in comparison with stand parameters such as Nupt and NNI. The results of the study suggest that vegetation indices calculated from PlanetScope EVI, NDVI and SAVI are more appropriate to use in the early growth stages of vegetation. In the later growth stages, it is more effective to use the vegetation indices such as NDRE and GNDVI, which can better highlight differences in nitrogen status for optimization of the fertilization strategy by site specific crop management practices in precision agriculture.