2022 Volume 32 Issue S Pages 1-5
Authors have been working on vegetation experiment using Greening Soil Material (GSM) to establish measures for combating desertification in Southern Mongolia. GSM can control moisture and nutrient retention capacity in shallow part of ground for growing plants healthily by only mixing local sand and dried animal feces. Each material can be collected in the vicinity of the target site. GSM is effective when buried at a depth of 0.05-0.1 m below ground level. Therefore, using GSM can be achieved a low-cost and environmental-friendly method. A targeting plant, licorice (Glycyrrhiza uralensis Fisch.), is one of the most valuable medicinal plant, but its distribution area has been shrinking due to soil deterioration. Therefore, planting licorice by using GSM could be sustainable and high value-added revegetation which can be accomplished revegetation with cultivation. In this paper, as one of the methods to evaluate the survival rate of licorice from the installation conditions and initial physical and chemical environment of GSM. A machine-learning model was employed for the logistic regression analysis, and the proportion of appropriate training data to improve the prediction accuracy was examined. In addition, the obtained regression equation was used to predict the survival rate of licorice from GSM validation data which were not used in the model.
