Yield assessment of large-seeded lentil samples based on vegetation indices

Abstract

Background and Aim. The use of Geographic information systems (GIS) and remote sensing data has become commonplace. However, they are not actively used in all sectors of agriculture. So, due to the specifics, the breeding work is more scrupulous, which is largely due to the large amount of material under study and the small scale of experimental sites inaccessible to remote sensing. However, the improvement of unmanned aerial vehicles (UAVs) and attachments makes it possible to conduct ground monitoring at low altitudes and conduct research in breeding nurseries. In this regard, the goal was set to explore the possibility of using remote sensing data, namely vegetation indices (NDVI, GNDVI, OSAVI, NDRE), to evaluate promising samples and lines, using the example of lentils.

Materials and methods. The research was carried out in the dry steppe zone of Northern Kazakhstan, on southern, carbonate chernozems. The objects of the study are 10 samples and a standard variety of large-seeded lentils Shyrayly. The observations were carried out remotely using aerial photography, at an interval of 1 time per week, from a height of 50 m at a drone flight speed of 1.5-2 m/s and a 70 × 70% cross-overlap of images using a Phantom 4 Multispectral UAV equipped with a 5-channel multispectral 2 MP camera. The processing of aerial photographs and the creation of orthophotoplanes of the experimental site was carried out in licensed programs Agisoft Metashape Professional, DJI Terra and QGIS 3.26.
Results. The high efficiency of using vegetation indices as an additional assessment tool in the analysis of breeding crops was established. The relationship between yield and indicators of vegetation indices has been established. The period has been determined when the values of vegetation indices most objectively reflect the emerging yield.     Conclusion. It has been established that vegetation indices   and NDVI, in particular, can only be used as an additional tool on a par with other traditional methods for evaluating varietal characteristics of cultivars. Under the prevailing climatic conditions, the differences between varieties began to manifest themselves more clearly from the period of bean formation in lentils. Based on the indicators of the NDVI index and other vegetation indices, the samples "E 149", "Anfiya", "Vekhovskaya" are recommended for involvement in the breeding process as parent pairs to increase yields.