Scholarlink

<p>Leaf wetness duration (LWD) plays a crucial role in plant health, crop development, and agricultural decision-making. However, direct LWD measurements remain scarce, and existing methods often rely on indirect estimates based on relative humidity (RH) or dewpoint depression (DP) thresholds. This study aimed to (i) evaluate the performance of RH- and DP-based methods for estimating daily non-rainy LWD, (ii) develop an hourly LWD prediction model for spatial and temporal applications across diverse climates, and (iii) integrate this model with easily accessible databases to generate LWD climatology and link it to weather forecasting systems for improved predictive capabilities in the study region. Field experiments were conducted across nine eucalypt plantation regions in Brazil, covering a range of climatic conditions. Leaf wetness sensors and standard meteorological instruments were deployed to collect data in situ. Among the tested RH and DP thresholds, 86% RH and 3&#xa0;°C DP provided the most reliable daily LWD estimates. A non-linear model incorporating DP and month as predictors was developed for hourly LWD estimation, achieving a high accuracy (adjusted R² = 0.95, performance index = 0.87). This model was then applied to spatially map the climatology of monthly average hourly LWD across study area, using NASA-POWER data. Additionally, integration with the European Centre for Medium-Range Weather Forecasts (ECMWF) system enabled seven-day LWD forecasts. The proposed approach offers a cost-effective, scalable, and reliable tool for precision agriculture, disease risk management, and sustainable crop planning, enhancing decision-making in forestry and agricultural systems based on LWD monitoring and prediction.</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

Spatial data science applied on forecasting of leaf wetness duration in tropical and subtropical climates

  • 期刊论文

  • 发表日期：

    2025年5月1日

查看全文
  • Clayton Alcarde Alvares,
  • Ítalo Ramos Cegatta,
  • Everton Pires Soliman,
  • Edival Angelo Valverde Zauza,
  • Maria Carolina Cunha Zonete,
  • José Luiz Stape

摘要

Leaf wetness duration (LWD) plays a crucial role in plant health, crop development, and agricultural decision-making. However, direct LWD measurements remain scarce, and existing methods often rely on indirect estimates based on relative humidity (RH) or dewpoint depression (DP) thresholds. This study aimed to (i) evaluate the performance of RH- and DP-based methods for estimating daily non-rainy LWD, (ii) develop an hourly LWD prediction model for spatial and temporal applications across diverse climates, and (iii) integrate this model with easily accessible databases to generate LWD climatology and link it to weather forecasting systems for improved predictive capabilities in the study region. Field experiments were conducted across nine eucalypt plantation regions in Brazil, covering a range of climatic conditions. Leaf wetness sensors and standard meteorological instruments were deployed to collect data in situ. Among the tested RH and DP thresholds, 86% RH and 3 °C DP provided the most reliable daily LWD estimates. A non-linear model incorporating DP and month as predictors was developed for hourly LWD estimation, achieving a high accuracy (adjusted R² = 0.95, performance index = 0.87). This model was then applied to spatially map the climatology of monthly average hourly LWD across study area, using NASA-POWER data. Additionally, integration with the European Centre for Medium-Range Weather Forecasts (ECMWF) system enabled seven-day LWD forecasts. The proposed approach offers a cost-effective, scalable, and reliable tool for precision agriculture, disease risk management, and sustainable crop planning, enhancing decision-making in forestry and agricultural systems based on LWD monitoring and prediction.