A study of the effects of green stem syndrome on some quality parameters in soybean and the possibility of early detection

Abstract

Green stem syndrome is one of the major problems encountered in soybean production in the world because it makes harvesting with a combine harvester difficult. Although the prevalence of the green stem syndrome Turkey is unknown, in recent years it has started to be observed frequently. Leaf color characters in the growing stages of some soybean varieties have been determined according to varieties in this study. Color changes in the leaves from V3 to R8 phase were monitored using L *, a *, b * color scale. Possibility of detecting changes in leaf color before the R8 stage was studied. Some quality parameters have been evaluated in seed samples obtained from plants with and without symptoms in the R8 stage. It was determined that the germination rate of the seeds obtained from the plants with the syndrome decreased by 61.4% on average compared to those from healthy plants. Furthermore, compared to non-symptomatic seeds, symptomatic seeds were larger, had a lower fat ratio, lower palmitic and linoleic fatty acid values, and higher oleic fatty acid values. At this study was determined that the most significant difference was manifested in terms of stem moisture values during germination and harvesting. In addition, detection of green stem syndrome can be used b* color value as a marker. The hypothesis of the study is that the syndrome can be diagnosed at early stage by following color values in the soybean leaves. In the future studies the color of the leaf can also be a parameter available for the machine learning models.
 Keywords: Harvest stage, Glycine max (L.), green stem syndrome, leaf color, machine learning