Scholarlink

In Himalayan ecosystem, forest fires is a major problem which affects biodiversity, environmental sustainability, and local communities. Therefore, there is a need of timely and accurate detection in complex mountainous area for environmental computing and ICT-driven disaster management. This research paper presents an integrated, daily fire detection and validation framework that integrates three satellite datasets Sentinel-2, MODIS, and VIIRS, within Google Earth Engine. Google Earth Engine (GEE) plays a central role in the ICT framework, offering cloud-powered tools for geospatial analysis, daily fire tracking, and seamless integration of satellite data. Sentinel-2 provides high-resolution spectral data for fire detection using refined thresholds across SWIR, NDVI, NBR, and dNBR bands (based on Hu et al. model). MODIS thermal anomalies act as a confidence booster, which validates Sentinel-2 spectral fire detections with real-time heat signatures. MODIS works as a confidence booster, which validates Sentinel-2 detections through thermal anomaly overlaps. VIIRS is used as a ground-truth validator, which computes daily metrics including Overall Accuracy (OA), Precision, Recall, ROC and AUC graphs. This research shows that Sentinel-2 can detect active fires in selected ROI even when MODIS is silent, and VIIRS validation checks their reliability. This adaptable framework is using ICT technology which supports daily fire monitoring in difficult terrain and can help in forest administration, climate resilience and advance precision agriculture.

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

Environmental Computing Framework for Daily Fire Detection in the Himalayan Foothills Using Sentinel-2, MODIS, and VIIRS

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Jyoti Ramola,
  • Rishi Prakash,
  • Anurag Vidyarthi

摘要

In Himalayan ecosystem, forest fires is a major problem which affects biodiversity, environmental sustainability, and local communities. Therefore, there is a need of timely and accurate detection in complex mountainous area for environmental computing and ICT-driven disaster management. This research paper presents an integrated, daily fire detection and validation framework that integrates three satellite datasets Sentinel-2, MODIS, and VIIRS, within Google Earth Engine. Google Earth Engine (GEE) plays a central role in the ICT framework, offering cloud-powered tools for geospatial analysis, daily fire tracking, and seamless integration of satellite data. Sentinel-2 provides high-resolution spectral data for fire detection using refined thresholds across SWIR, NDVI, NBR, and dNBR bands (based on Hu et al. model). MODIS thermal anomalies act as a confidence booster, which validates Sentinel-2 spectral fire detections with real-time heat signatures. MODIS works as a confidence booster, which validates Sentinel-2 detections through thermal anomaly overlaps. VIIRS is used as a ground-truth validator, which computes daily metrics including Overall Accuracy (OA), Precision, Recall, ROC and AUC graphs. This research shows that Sentinel-2 can detect active fires in selected ROI even when MODIS is silent, and VIIRS validation checks their reliability. This adaptable framework is using ICT technology which supports daily fire monitoring in difficult terrain and can help in forest administration, climate resilience and advance precision agriculture.