Machine learning unveils the role of biochar application in enhancing tea yield by mitigating soil acidification in tea plantations.

Journal: The Science of the total environment
PMID:

Abstract

Biochar, a widely utilized soil amendment in environmental applications, has been employed to enhance tea cultivation. This study utilized three machine learning models to investigate the effects of biochar on tea growth and yield, with the random forest (RF) model demonstrating superior performance (R = 0.8768, Root Mean Square Error = 6.1537). Feature importance analysis revealed that biochar characteristics and experimental conditions constitute critical factors exerting an impact on the output, accounting for 39.2 % and 38.6 %, respectively. Specifically, the Ca content of biochar (weight 0.274), the quantity of biochar applied (weight 0.206), and the calcium (Ca) content of soil (weight 0.120) emerged as the three most significant factors affecting tea yield. In conclusion, the machine learning models developed in this study elucidate the multifactorial impact of biochar application on tea yield, providing theoretical and methodological support for practical biochar application strategies in tea production.

Authors

  • Rongxiu Yin
    Tea Research Institute, Guizhou Provincial Academy of Agricultural Sciences, Guiyang 550006, China.
  • Xin Li
    Veterinary Diagnostic Center, Shanghai Animal Disease Control Center, Shanghai, China.
  • Yating Ning
    Tea Research Institute, Chinese Academy of Agricultural Sciences, Hangzhou 310008, China.
  • Qiang Hu
    School of Information Science and Technology, Qingdao University of Science and Technology, Qicngdao 266061, China.
  • Yihu Mao
    Tea Research Institute, Chinese Academy of Agricultural Sciences, Hangzhou 310008, China.
  • Xiaoqin Zhang
    Department of Radiology, The First Affiliated Hospital of Wenzhou Medical University, PR China.
  • Xinzhong Zhang
    Tea Research Institute, Chinese Academy of Agricultural Sciences, Hangzhou 310008, China. Electronic address: zhangxinzhong@tricaas.com.