Scholarlink

<p>The two-spotted spider mite (<i>Tetranychus urticae</i> Koch) is a major pest in protected vegetable production systems. This study investigates the within-canopy distribution and seasonal incidence of <i>T. urticae</i> on cucumber (<i>Cucumis sativus</i> L.) in polyhouses in temperate Kashmir, India. Mite density was quantified across different canopy strata and leaf surfaces throughout the cropping season. Findings revealed a significant preference for the middle canopy layer, accounting for nearly 50% of total mite populations during the peak infestation period. Mite abundance was consistently higher on the abaxial (lower) leaf surface compared to the adaxial, with lower coefficients of variation, indicating consistent colonization. Seasonal monitoring showed that mite incidence began from the 20th Standard Meteorological Week (SMW), peaked at the 27th SMW, and declined thereafter. Beta regression confirmed significant effects of sampling date on infestation levels. Machine learning models, including random forest (<i>R</i><sup>2</sup> = 0.93) and support vector machines (<i>R</i><sup>2</sup> = 0.92), were applied to forecast mite population dynamics using concurrent and lagged weather parameters. To enable model-agnostic interpretation, SHAP (SHapley Additive exPlanations) values were computed. Temperature and relative humidity were key predictors, and models with a one-week lag retained strong predictive power (<i>R</i><sup>2</sup> = 0.84). These results highlight the critical influence of canopy position, leaf surface, and microclimatic conditions on <i>T. urticae</i> distribution and demonstrate the utility of predictive modeling for timely and localized pest management strategies under protected cultivation in temperate regions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Canopy stratification, seasonal dynamics, and machine learning-based forecasting of Tetranychus urticae Koch on cucumber under polyhouse conditions

  • 期刊论文

  • 发表日期：

    2025年8月11日

查看全文
  • Hafsa Ajaz Tramboo,
  • Asma Sherwani,
  • Malik Mukhtar Ahmad,
  • Akhtar Ali Khan,
  • Ajaz Ahmad Malik,
  • Ronit Jaiswal,
  • Mohd Abas Shah

摘要

The two-spotted spider mite (Tetranychus urticae Koch) is a major pest in protected vegetable production systems. This study investigates the within-canopy distribution and seasonal incidence of T. urticae on cucumber (Cucumis sativus L.) in polyhouses in temperate Kashmir, India. Mite density was quantified across different canopy strata and leaf surfaces throughout the cropping season. Findings revealed a significant preference for the middle canopy layer, accounting for nearly 50% of total mite populations during the peak infestation period. Mite abundance was consistently higher on the abaxial (lower) leaf surface compared to the adaxial, with lower coefficients of variation, indicating consistent colonization. Seasonal monitoring showed that mite incidence began from the 20th Standard Meteorological Week (SMW), peaked at the 27th SMW, and declined thereafter. Beta regression confirmed significant effects of sampling date on infestation levels. Machine learning models, including random forest (R2 = 0.93) and support vector machines (R2 = 0.92), were applied to forecast mite population dynamics using concurrent and lagged weather parameters. To enable model-agnostic interpretation, SHAP (SHapley Additive exPlanations) values were computed. Temperature and relative humidity were key predictors, and models with a one-week lag retained strong predictive power (R2 = 0.84). These results highlight the critical influence of canopy position, leaf surface, and microclimatic conditions on T. urticae distribution and demonstrate the utility of predictive modeling for timely and localized pest management strategies under protected cultivation in temperate regions.