Distinguishing areas of organic, biodynamic and conventional farming by means of multispectral images. A pilot study

Abstract

The aim of the present investigation was to develop a basic methodological scheme for distinguishing organic, biodynamic and conventional areas of winter wheat cultivation, based on the satellite images from Sentinel-2. The target areas are located in the experimental field of the Institute of Agriculture in Karnobat, Bulgaria. The plant condition was determined on a three-point rating scale according to the average plant height, percentage of projective cover of the aboveground plant parts and intensity of the green color of leaves during the end of tillering to complete ripeness. To determine the homogeneity of the target polygons, we used a small drone DJI Phantom 4, with 1/2.3 CMOS camera and Dronedeploy to compile flight plans for shooting and photogrammetric processing of the images. ArcGIS was used in GIS procedures and analyses. Microsoft Excel, with Solver ADD-IN was used to clusterize the target polygons in order to distinguish the polygons of organic, biodynamic and conventional farming by the multispectral indices (CRI 1, MSAVI, NDVI, NDWI, PSSRa, PSSRc, WDRVI). All selected indices were calculated in GIS environment. Ground observations showed that there were differences in the condition of plants in different cultivation systems during the selected observation phases. The obtained results reveal that the differences in dynamics and development during the growth and development stages of the wheat in the organic, biodynamic and conventional fields can be traced by spectral characteristics from satellite images and they can be used to distinguish them.