Scholarlink

<p>Downy mildew, caused by <i>Hyaloperonospora parasitica</i>, is an important disease affecting broccoli, leading to substantial yield losses. This study investigates the influence of weather variables, such as temperature, relative humidity, and rainfall on downy mildew severity during the 2021-22 and 2022-23. The primary aim was to identify the key environmental factors driving disease progression and to develop predictive models for accurate disease forecasting. Using random forest regression, T Min was identified as the most critical factor influencing disease severity, with warmer night conditions strongly correlating with increased disease severity. Maximum temperature and morning relative humidity were also found to contribute to disease progression, though their effects were less pronounced. The study also applied AutoRegressive Integrated Moving Average (ARIMA) models to capture the temporal dynamics of disease severity. The ARIMA (1,1,1) model showed strong predictive performance, indicating that current disease severity can be accurately forecasted based on previous weeks data. The autoregressive term was highly significant in both seasons, demonstrating stable progression pattern of disease severity once established. The combination of machine learning and time-series analysis offered an effective approach for forecasting downy mildew outbreaks in broccoli, providing growers with a valuable tool for early intervention. The findings highlight the importance of monitoring minimum temperature, particularly during the early weeks of the growing season, to anticipate disease outbreaks.</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

Weather driven prediction of downy mildew in broccoli deploying machine learning and time-series approaches

  • 期刊论文

  • 发表日期：

    2025年7月28日

查看全文
  • Ranbir Singh,
  • Diksha Loona,
  • Amoghavarsha Chittaragi,
  • Balanagouda Patil

摘要

Downy mildew, caused by Hyaloperonospora parasitica, is an important disease affecting broccoli, leading to substantial yield losses. This study investigates the influence of weather variables, such as temperature, relative humidity, and rainfall on downy mildew severity during the 2021-22 and 2022-23. The primary aim was to identify the key environmental factors driving disease progression and to develop predictive models for accurate disease forecasting. Using random forest regression, T Min was identified as the most critical factor influencing disease severity, with warmer night conditions strongly correlating with increased disease severity. Maximum temperature and morning relative humidity were also found to contribute to disease progression, though their effects were less pronounced. The study also applied AutoRegressive Integrated Moving Average (ARIMA) models to capture the temporal dynamics of disease severity. The ARIMA (1,1,1) model showed strong predictive performance, indicating that current disease severity can be accurately forecasted based on previous weeks data. The autoregressive term was highly significant in both seasons, demonstrating stable progression pattern of disease severity once established. The combination of machine learning and time-series analysis offered an effective approach for forecasting downy mildew outbreaks in broccoli, providing growers with a valuable tool for early intervention. The findings highlight the importance of monitoring minimum temperature, particularly during the early weeks of the growing season, to anticipate disease outbreaks.