研究目的
To assess the effectiveness of using freely available Copernicus Sentinel-1 SAR data for operational agricultural flood monitoring in the United States.
研究成果
Sentinel-1 SAR is effective for operational agricultural flood monitoring, providing accurate, timely, and affordable estimates with over 95% accuracy, enabling near real-time assessment and decision-making for disaster response.
研究不足
The study relies on manual thresholding which may not be fully automated; some fields may have been harvested before the flood event, potentially affecting accuracy; the method is specific to C-band SAR and may not generalize to other wavelengths or regions without adaptation.
1：Experimental Design and Method Selection:
The study uses Sentinel-1 SAR data for flood detection, employing manual thresholding and change detection methods. It integrates with NASS Cultivated Layer and Cropland Data Layers for agricultural specificity.
2：Sample Selection and Data Sources:
Data from Hurricane Harvey event in Texas and Louisiana, U.S., in 2017, with pre-flood and post-flood Sentinel-1A imagery, NASS layers, and ground reference data from optical sensors.
3：List of Experimental Equipment and Materials:
Sentinel-1A satellite data, software tools including SNAP toolbox, ERDAS Imagine, and ArcGIS.
4：Experimental Procedures and Operational Workflow:
Preprocessing of SAR images (calibration, terrain correction, de-speckling), mosaicking, log transformation, thresholding for water detection, change assessment between pre- and post-flood images, intersection with agricultural layers for acreage estimation.
5：Data Analysis Methods:
Manual thresholding, pixel counting for area estimation, validation using ground reference data with accuracy assessment.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
