Journal of Integrative Agriculture ›› 2024, Vol. 23 ›› Issue (10): 3576-3591.DOI: 10.1016/j.jia.2024.08.001

• • 上一篇    下一篇

气候变化条件下马铃薯孢囊线虫分布北移的机器学习集成模型预测

  

  • 收稿日期:2023-12-24 接受日期:2024-02-28 出版日期:2024-10-20 发布日期:2024-09-11

Machine learning ensemble model prediction of northward shift in potato cyst nematodes (Globodera rostochiensis and G. pallida) distribution under climate change conditions

Yitong He1, 2, Guanjin Wang2, 3, Yonglin Ren2, Shan Gao4, Dong Chu1#, Simon J. Mckirdy2#   

  1. 1 Shandong Engineering Research Centre for Environment-friendly Agricultural Pest Management, College of Plant Health and Medicine, Qingdao Agricultural University, Qingdao 266109, China 
    2 Harry Butler Institute, Murdoch University, Murdoch, WA 6150, Australia
    3 School of Information Technology, Murdoch University, Perth, WA 6150, Australia
    4 Hebei Dahaituo National Nature Reserve Management Centre, Chicheng 075500, China

  • Received:2023-12-24 Accepted:2024-02-28 Online:2024-10-20 Published:2024-09-11
  • About author:Yitong He, E-mail: he_yitong@hotmail.com; #Correspondence Dong Chu, E-mail: chinachudong@qau.edu.cn; Simon J. McKirdy, E-mail: s.mckirdy@murdoch.edu.au
  • Supported by:
    This research was funded by the National Key R&D Program of China (2021YFD1400200) and the Taishan Scholar Constructive Engineering Foundation of Shandong, China (tstp20221135).

PDF

摘要：

马铃薯孢囊线虫（PCNs）是马铃薯生产的重大威胁，在许多国家造成了严重危害。鉴于气候变化对害虫入侵和分布模式的深远影响，预测PCNs在未来气候条件下的分布对于实施有效的生物安全战略至关重要。机器学习（ML），特别是集成模型，凭借其从复杂数据集中学习和预测的能力，已成为预测物种分布的强大工具。本研究旨在利用机器学习技术预测气候变化背景下的PCNs分布，为其入侵风险评估提供科学依据。为确保预测的准确性，我们首先利用全球气候模型生成一致的气候预测因子，以消除因预测因子差异带来的不确定性。然后，使用5个机器学习算法分别构建单算法集成模型（ESA）和多算法集成模型（EMA），并对模型进行训练和评价。模型评价结果表明：EMA模型并非总是优于ESA模型；人工神经网络算法构建的ESA模型在节约算力的同时，获得最优的预测效果。模型预测结果表明：热带地区PCNs的分布范围呈北移趋势，热带地区面积减少，北纬地区面积增加。尽管全球适生区域的总面积变化不大，约占陆地总面积的16-20%（目前为18%），但这一分布变化仍可能对马铃薯生产产生重大影响。因此，生产者和管理者需要密切关注这一趋势，并采取相应措施来应对潜在的生物安全挑战。本研究不仅为评估PCNs侵入新地区的风险提供了科学依据，还为跟踪其他入侵物种分布变化提供了参考模型。利用机器学习技术预测物种分布变化，可以更好地了解气候变化对物种分布的影响，从而制定更有效的生物安全控制计划，有助于应对未来气候变化带来的挑战。

Abstract:

Potato cyst nematodes (PCNs) are a significant threat to potato production, having caused substantial damage in many countries.  Predicting the future distribution of PCN species is crucial to implementing effective biosecurity strategies, especially given the impact of climate change on pest species invasion and distribution.  Machine learning (ML), specifically ensemble models, has emerged as a powerful tool in predicting species distributions due to its ability to learn and make predictions based on complex data sets.  Thus, this research utilised advanced machine learning techniques to predict the distribution of PCN species under climate change conditions, providing the initial element for invasion risk assessment.  We first used Global Climate Models to generate homogeneous climate predictors to mitigate the variation among predictors.  Then, five machine learning models were employed to build two groups of ensembles, single-algorithm ensembles (ESA) and multi-algorithm ensembles (EMA), and compared their performances.  In this research, the EMA did not always perform better than the ESA, and the ESA of Artificial Neural Network gave the highest performance while being cost-effective.  Prediction results indicated that the distribution range of PCNs would shift northward with a decrease in tropical zones and an increase in northern latitudes.  However, the total area of suitable regions will not change significantly, occupying 16–20% of the total land surface (18% under current conditions).  This research alerts policymakers and practitioners to the risk of PCNs’ incursion into new regions.  Additionally, this ML process offers the capability to track changes in the distribution of various species and provides scientifically grounded evidence for formulating long-term biosecurity plans for their control. 

Key words: invasive species distribution , Future climates ,  Homogeneous climate predictors , single-algorithm ensembles , multi-algorithm ensembles , artificial neural network