Predictive Model for Tea Anthracnose in Yunnan

In a paper published in the journal Agronomy, researchers developed an accurate, non-destructive, and rapid prediction model for tea anthracnose in Yunnan. They utilized climate data from Internet of Things (IoT) devices. They applied the least absolute shrinkage and selection operator, cox proportional hazards model (LASSO-COX-NOMOGRAM), and limma difference analysis to study the disease's response to environmental changes. The model achieved high accuracy and outperformed traditional early tea disease prevention and analysis methods.

Study: Predictive Model for Tea Anthracnose in Yunnan. Image Credit: tamu1500/Shutterstock.com
Study: Predictive Model for Tea Anthracnose in Yunnan. Image Credit: tamu1500/Shutterstock.com

Related Work

Past work has shown that tea anthracnose threatens Yunnan's rich tea tree resources and severely impacts tea production. Addressing this requires timely detection and treatment, as current methods are time-consuming, laborious, and prone to misjudgment and variability.

Additionally, tea anthracnose's prevalence increases with global climate change, making early and accurate predictions essential to prevent significant damage. Furthermore, traditional prediction methods require extensive training data and are less intuitive than more modern, graphical approaches like nomograms.

Study Area Overview

The research occurred at Yuecheng Co., Ltd.'s No. 1 base in Xishuangbanna Prefecture, Yunnan Province, China, the birthplace of big-leaf tea. This area, characterized by a tropical monsoon climate, experiences a warm and humid environment with an average daily temperature of 22°C, a total effective accumulated temperature reaching 8761°C, and an annual relative humidity of 79%. The tea garden features Mengku large-leaf tea trees, approximately 10 years old, covering 90% of the strip-planted, sloped area.

Surrounded by mountains and high mountain clouds, the region has rich organic soil and strong solar radiation, creating ideal conditions for tea anthracnose to thrive year-round, particularly in autumn. Tea anthracnose is prevalent in Yunnan, affecting all stages of tea leaves, starting as dark green water-stained spots that expand into yellow-brown irregular shapes. These spots eventually develop gray-white patches with black conidia, leading to brittle, cracked leaves and potential significant leaf loss in severe cases.

Preliminary experiments at the Yunnan Organic Tea Industry Intelligent Engineering Research Center identified that a temperature of 25°C and a pH of 6 are most conducive to the disease's growth. The disease's growth is minimally affected by light conditions but thrives in high humidity and moderate temperatures.

Three hundred climate change data sets from Yunnan Province were analyzed, with 240 sets used for training and 60 for validation. The dataset was expanded using the Bootstrap method to improve prediction accuracy. Statistical analyses, including chi-square and ridit tests, were performed with a statistical package for the social sciences (SPSS) 25.0 software.

Significant variables identified from single-factor analysis were incorporated into multi-factor analysis, with LASSO-COX regression used to develop a nomogram prediction model. Model performance was evaluated through receiver operating characteristics (ROC) curve analysis, AUC values, and calibration curves. Decision curve analysis (DCA) assessed the model's prediction efficiency and applicability.

Results and Analysis

Single-factor logistic regression analysis was conducted using the Cox regression model to evaluate the influence of various environmental factors on tea anthracnose disease. This model allows for the simultaneous assessment of multiple risk factors without assuming interactions between them. Based on 30 variables, the analysis revealed that 21 factors significantly impacted the disease's progression.

Variables such as light intensity, humidity, temperature, and soil moisture were notably correlated with disease severity, with over 80% of factors showing a strong association (p < 0.01). These findings align with preliminary results and underscore the role of environmental factors as initial predictors of tea anthracnose.

The differential correlation analysis examined variations in tea anthracnose lesions and ecological factors over a week. Researchers observed significant changes in light intensity, air humidity, CO2 concentration, and soil moisture.

Increased CO2 levels and high light intensity were linked to leaf chlorophyll and cell structure alterations, affecting disease progression. Total biological quality (TBQ) radiation intensity and soil moisture also influenced disease severity, with higher soil moisture levels exacerbating the spread of disease spores. The analysis highlighted the crucial role of these environmental factors in determining disease severity, particularly the impact of TBQ radiation and soil moisture.

The team refined the predictive model using LASSO-COX-NOMOGRAM, identifying key factors like light intensity and soil moisture. Validation showed strong generalization with an AUC over 0.7, and external verification of the visualization system achieved 83.3% accuracy, outperforming traditional methods.

Conclusion 

To sum up, this study revealed significant correlations between 21 environmental factors and tea anthracnose, with 90.5% of climate change factors impacting tea growth within 7 days. Incorporating nine key factors, the LASSO-COX-NOMOGRAM model demonstrated strong prediction accuracy with AUC values of 0.745 and 0.75 and an external verification accuracy rate of 83.3%. These findings enhanced the understanding of short-term disease risks and supported eco-friendly management practices for improved tea garden health and productivity.

Journal reference:
  • Ye, R., et al. (2024). Prediction of Anthracnose Risk in Large-Leaf Tea Trees Based on the Atmospheric Environmental Changes in Yunnan Tea Gardens—Cox Regression Model and Machine Learning Model. Agronomy, 14:7, 1501. DOI: 10.3390/agronomy14071501, https://www.mdpi.com/2073-4395/14/7/1501

Article Revisions

  • Aug 15 2024 - Fixed broken journal paper link.
Silpaja Chandrasekar

Written by

Silpaja Chandrasekar

Dr. Silpaja Chandrasekar has a Ph.D. in Computer Science from Anna University, Chennai. Her research expertise lies in analyzing traffic parameters under challenging environmental conditions. Additionally, she has gained valuable exposure to diverse research areas, such as detection, tracking, classification, medical image analysis, cancer cell detection, chemistry, and Hamiltonian walks.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Chandrasekar, Silpaja. (2024, August 14). Predictive Model for Tea Anthracnose in Yunnan. AZoAi. Retrieved on December 22, 2024 from https://www.azoai.com/news/20240721/Predictive-Model-for-Tea-Anthracnose-in-Yunnan.aspx.

  • MLA

    Chandrasekar, Silpaja. "Predictive Model for Tea Anthracnose in Yunnan". AZoAi. 22 December 2024. <https://www.azoai.com/news/20240721/Predictive-Model-for-Tea-Anthracnose-in-Yunnan.aspx>.

  • Chicago

    Chandrasekar, Silpaja. "Predictive Model for Tea Anthracnose in Yunnan". AZoAi. https://www.azoai.com/news/20240721/Predictive-Model-for-Tea-Anthracnose-in-Yunnan.aspx. (accessed December 22, 2024).

  • Harvard

    Chandrasekar, Silpaja. 2024. Predictive Model for Tea Anthracnose in Yunnan. AZoAi, viewed 22 December 2024, https://www.azoai.com/news/20240721/Predictive-Model-for-Tea-Anthracnose-in-Yunnan.aspx.

Comments

The opinions expressed here are the views of the writer and do not necessarily reflect the views and opinions of AZoAi.
Post a new comment
Post

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.

You might also like...
Machine Learning Powering Breakthroughs in Climate Forecasting and Modeling