Scholarlink

This research integrates Land Use and Land Cover (LULC) analysis with rice crop phenology monitoring, leveraging multi-source remote sensing data. The study commenced by collecting Landsat 7 ETM + datasets for 2001 and 2011, and Landsat 8 OLI/TIRS data for 2023, to investigate LULC changes over two decades. A Random Forest classification model was applied to categorize the region into distinct land classes: Built up, Waterbody, Riverbed, Vegetation, and Fallow/Barren Land, achieving classification accuracies of 90, 84, and 92% for the respective years. The results revealed a substantial increase in built-up areas (435 km2) and water bodies (271 km2), minor variations in riverbeds (17 km2), and a significant decline in vegetation, accompanied by an expansion of fallow lands. Change detection highlighted the dynamic evolution of the region, stressing the importance of sustainable land management. Parallel to this, the research focuses on enhancing rice crop phenology monitoring. Sentinel-2 and Landsat data were collected for the Vuyyuru region, Krishna District, during the rice growing season, with Normalized Difference Vegetation Index (NDVI) values calculated to track the growing stages. NDVI data allowed for the generation of phenological curves that captured key growth stages from planting to harvesting. Metrics such as season start, peak NDVI, and growing season length were extracted and analyzed to visualize crop health, evaluate weather impacts, and suggest potential improvements for crop management. This integrated approach contributes to both LULC dynamics understanding and agricultural decision-making.

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

Comprehensive Time Series Analysis and Rice Crop Phenology Monitoring in Krishna District Using Google Earth Engine (2001–2023)

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Kuppa Datta Sai,
  • Sikha Nissy,
  • Manne Suneetha

摘要

This research integrates Land Use and Land Cover (LULC) analysis with rice crop phenology monitoring, leveraging multi-source remote sensing data. The study commenced by collecting Landsat 7 ETM + datasets for 2001 and 2011, and Landsat 8 OLI/TIRS data for 2023, to investigate LULC changes over two decades. A Random Forest classification model was applied to categorize the region into distinct land classes: Built up, Waterbody, Riverbed, Vegetation, and Fallow/Barren Land, achieving classification accuracies of 90, 84, and 92% for the respective years. The results revealed a substantial increase in built-up areas (435 km2) and water bodies (271 km2), minor variations in riverbeds (17 km2), and a significant decline in vegetation, accompanied by an expansion of fallow lands. Change detection highlighted the dynamic evolution of the region, stressing the importance of sustainable land management. Parallel to this, the research focuses on enhancing rice crop phenology monitoring. Sentinel-2 and Landsat data were collected for the Vuyyuru region, Krishna District, during the rice growing season, with Normalized Difference Vegetation Index (NDVI) values calculated to track the growing stages. NDVI data allowed for the generation of phenological curves that captured key growth stages from planting to harvesting. Metrics such as season start, peak NDVI, and growing season length were extracted and analyzed to visualize crop health, evaluate weather impacts, and suggest potential improvements for crop management. This integrated approach contributes to both LULC dynamics understanding and agricultural decision-making.