Scholarlink

<p>Hydrological assessments in data-scarce regions remain challenging due to the limited availability of gauging stations, often hindering accurate evaluation of reservoir dynamics. Reservoirs, being vital sources of freshwater, typically require conventional bathymetric surveys that are time-intensive and cost-prohibitive. This study introduces an integrated and cost-effective framework leveraging multi-source satellite datasets and advanced techniques to estimate reservoir bathymetry of the Bhadra Reservoir, India. Temporal variations in Water Spread Area (WSA) were derived using the Normalized Difference Water Index (NDWI) and Modified NDWI (MNDWI) from Sentinel-2A/B and Landsat-8/9 imagery, respectively, utilizing the Google Earth Engine (GEE) platform. Satellite altimetry-derived Water Levels (WL) from Sentinel-3B and ICESat-2, exhibiting a high correlation of 0.99, were incorporated into machine learning regression models to develop accurate WSA–WL relationships. Model performance was assessed using statistical metrics, and the best-fit equations were selected. The resulting WSA–WL relationships, along with reservoir capacity estimates, were integrated with Surface Water Occurrence (SWO) maps to generate high-resolution bathymetric reconstruction. Results indicate a gross capacity loss of 7.97% since 1964, with an estimated siltation rate of 3.63 × 10<sup>3</sup>&#xa0;m<sup>3</sup>/km<sup>2</sup>/year. The bathymetry map of the Bhadra Reservoir, developed with 2-m interval contours, indicates that over 45% of the reservoir area lies within the first deepest contour level, reflecting its proximity to the deepest foundation zone. This scalable approach offers a practical tool for reservoir monitoring, flood forecasting, and sedimentation management in ungauged or poorly gauged basins.</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

Synergistic use of multi-sensor altimetry and multispectral data for bathymetric mapping of Bhadra Reservoir

  • 期刊论文

  • 发表日期：

    2026年3月26日

查看全文
  • Ahalya Nalluri,
  • H. Ramesh

摘要

Hydrological assessments in data-scarce regions remain challenging due to the limited availability of gauging stations, often hindering accurate evaluation of reservoir dynamics. Reservoirs, being vital sources of freshwater, typically require conventional bathymetric surveys that are time-intensive and cost-prohibitive. This study introduces an integrated and cost-effective framework leveraging multi-source satellite datasets and advanced techniques to estimate reservoir bathymetry of the Bhadra Reservoir, India. Temporal variations in Water Spread Area (WSA) were derived using the Normalized Difference Water Index (NDWI) and Modified NDWI (MNDWI) from Sentinel-2A/B and Landsat-8/9 imagery, respectively, utilizing the Google Earth Engine (GEE) platform. Satellite altimetry-derived Water Levels (WL) from Sentinel-3B and ICESat-2, exhibiting a high correlation of 0.99, were incorporated into machine learning regression models to develop accurate WSA–WL relationships. Model performance was assessed using statistical metrics, and the best-fit equations were selected. The resulting WSA–WL relationships, along with reservoir capacity estimates, were integrated with Surface Water Occurrence (SWO) maps to generate high-resolution bathymetric reconstruction. Results indicate a gross capacity loss of 7.97% since 1964, with an estimated siltation rate of 3.63 × 103 m3/km2/year. The bathymetry map of the Bhadra Reservoir, developed with 2-m interval contours, indicates that over 45% of the reservoir area lies within the first deepest contour level, reflecting its proximity to the deepest foundation zone. This scalable approach offers a practical tool for reservoir monitoring, flood forecasting, and sedimentation management in ungauged or poorly gauged basins.