Scholarlink

<p>The Lijiang River, a UNESCO-listed karst waterway, requires accurate riverbed depth monitoring to support sustainable management. However, efficient techniques for rapid depth measurement remain lacking in this Karst Basin. This study develops and validates remote sensing-based depth inversion models to address this technological gap, utilizing both hyperspectral and multispectral data sources. Two inversion models were established through MATLAB: (1) A hyperspectral model based on ISI921VF field spectroradiometer data (350–1050&#xa0;nm spectral range), and (2) A multispectral model utilizing QuickBird-2 (QB-2) satellite data (B1-B4 bands) was further tested along the Jingpingshan Bridge-Guilin Hydrological Station reach. The optimal hyperspectral model achieved a 0.9 correlation coefficient (binary quadratic polynomial) with 0.49&#xa0;m standard error; The QB-2 cubic polynomial model (B2-B4 bands) showed superior performance (R²=0.93, SE = 0.15&#xa0;m); Field validation revealed 70% of QB-2 model residuals were &lt; 0.5&#xa0;m (-0.734&#xa0;m to 0.710&#xa0;m range), with t-test-confirmed elevation accuracy (<i>p</i> &lt; 0.05). Remote sensing depth inversion is viable for Lijiang River management when using quality multispectral data. This approach provides critical geomorphological insights for basin-scale conservation.</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

Reach on Remote Sensing Retrieval of Water Depth in the Lijiang River

  • 期刊论文

  • 发表日期：

    2025年12月23日

查看全文
  • Min Hao,
  • ZhiQiang Jia,
  • Hong Wu,
  • DaNing Huang,
  • Xiao Wo

摘要

The Lijiang River, a UNESCO-listed karst waterway, requires accurate riverbed depth monitoring to support sustainable management. However, efficient techniques for rapid depth measurement remain lacking in this Karst Basin. This study develops and validates remote sensing-based depth inversion models to address this technological gap, utilizing both hyperspectral and multispectral data sources. Two inversion models were established through MATLAB: (1) A hyperspectral model based on ISI921VF field spectroradiometer data (350–1050 nm spectral range), and (2) A multispectral model utilizing QuickBird-2 (QB-2) satellite data (B1-B4 bands) was further tested along the Jingpingshan Bridge-Guilin Hydrological Station reach. The optimal hyperspectral model achieved a 0.9 correlation coefficient (binary quadratic polynomial) with 0.49 m standard error; The QB-2 cubic polynomial model (B2-B4 bands) showed superior performance (R²=0.93, SE = 0.15 m); Field validation revealed 70% of QB-2 model residuals were < 0.5 m (-0.734 m to 0.710 m range), with t-test-confirmed elevation accuracy (p < 0.05). Remote sensing depth inversion is viable for Lijiang River management when using quality multispectral data. This approach provides critical geomorphological insights for basin-scale conservation.