Scholarlink

<p>This study presents a semi-physical remote sensing approach for wheat yield estimation across Ahmedabad district, India, during the 2020–2021 Rabi season. Sentinel-2A-derived fAPAR and Land Surface Water Index (LSWI) were integrated with cumulative PAR from INSAT-3D to estimate biomass using a Radiation Use Efficiency (RUE) model. Grain yield was derived using a Harvest Index (HI) averaged from 60 crop cutting experiments. An LSWI-based scalar accounted for spatial water stress. Yield maps generated for the entire district were validated using observed data from 25 gram panchayats. The model achieved a strong correlation (R² = 0.719) and an RMSE of 335.35&#xa0;kg/ha. Higher yields were observed in regions with lower moisture stress. The results demonstrate that this approach offers an accurate, scalable, and cost-effective solution for crop yield monitoring in data-limited environments, supporting precision agriculture and regional food security planning.</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

Remote Sensing-Based Spatial Mapping of Wheat Yield Using a Semi-Physical Modeling Approach

  • 期刊论文

  • 发表日期：

    2026年5月2日

查看全文
  • Pavan Kumar Sharma,
  • Pratyush Kumar,
  • Thota Sivasankar,
  • Sukanta Kumar Saha,
  • Akshay Kuwar

摘要

This study presents a semi-physical remote sensing approach for wheat yield estimation across Ahmedabad district, India, during the 2020–2021 Rabi season. Sentinel-2A-derived fAPAR and Land Surface Water Index (LSWI) were integrated with cumulative PAR from INSAT-3D to estimate biomass using a Radiation Use Efficiency (RUE) model. Grain yield was derived using a Harvest Index (HI) averaged from 60 crop cutting experiments. An LSWI-based scalar accounted for spatial water stress. Yield maps generated for the entire district were validated using observed data from 25 gram panchayats. The model achieved a strong correlation (R² = 0.719) and an RMSE of 335.35 kg/ha. Higher yields were observed in regions with lower moisture stress. The results demonstrate that this approach offers an accurate, scalable, and cost-effective solution for crop yield monitoring in data-limited environments, supporting precision agriculture and regional food security planning.