Scholarlink

<p>Field-based hyperspectral and laboratory analyses were conducted across nine major iron mines in the Keonjhar–Sundargarh Iron Ore Belt, Odisha, India, to characterize iron-bearing minerals and establish predictive relationships between spectral and geochemical parameters. Reflectance spectra (350–2500&#xa0;nm) exhibit prominent Fe³⁺ absorption features between 700 and 1250&#xa0;nm, diagnostic of hematite, goethite, and limonite. Key spectral attributes—including band-depth maxima, absorption area, full width at half maximum (FWHM), and skewness—were quantified using continuum-removal and derivative-based analyses. The spectra reveal systematic variability linked to mineral composition, grain size, and aluminium substitution, the latter producing a consistent red-shift of band-depth maxima. Chemometric modelling shows strong exponential relationships between total Fe content and both band depth (R² = 0.91) and absorption curve area (R² = 0.86), with low RMSE (5.6–8.6%) and high modelling efficiency (0.79–0.89). Second-derivative reflectance also serves as a robust linear indicator of ferric iron abundance. These results demonstrate that integrating hyperspectral and geochemical data provides a rapid, non-destructive framework for quantitative ore-grade estimation and mineralogical discrimination within Banded Iron Formations. The approach strengthens the geochemical interpretation of spectral absorption parameters and offers a practical tool for assessing compositional variability in iron-oxide-dominated lithologies.</p> Graphical Abstract <p></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

Spectro-Mineralogical and Chemometric Characterization of Banded Iron Formations from the Keonjhar–Sundargarh Iron Ore Belt, Odisha, India: Implications for Geochemical and Spectral Indicators of Ore Grade

  • 期刊论文

  • 发表日期：

    2026年4月29日

查看全文
  • Dibyendu Dutta,
  • Wasim Akram,
  • Sourav Samanta,
  • Libeesh Lukose

摘要

Field-based hyperspectral and laboratory analyses were conducted across nine major iron mines in the Keonjhar–Sundargarh Iron Ore Belt, Odisha, India, to characterize iron-bearing minerals and establish predictive relationships between spectral and geochemical parameters. Reflectance spectra (350–2500 nm) exhibit prominent Fe³⁺ absorption features between 700 and 1250 nm, diagnostic of hematite, goethite, and limonite. Key spectral attributes—including band-depth maxima, absorption area, full width at half maximum (FWHM), and skewness—were quantified using continuum-removal and derivative-based analyses. The spectra reveal systematic variability linked to mineral composition, grain size, and aluminium substitution, the latter producing a consistent red-shift of band-depth maxima. Chemometric modelling shows strong exponential relationships between total Fe content and both band depth (R² = 0.91) and absorption curve area (R² = 0.86), with low RMSE (5.6–8.6%) and high modelling efficiency (0.79–0.89). Second-derivative reflectance also serves as a robust linear indicator of ferric iron abundance. These results demonstrate that integrating hyperspectral and geochemical data provides a rapid, non-destructive framework for quantitative ore-grade estimation and mineralogical discrimination within Banded Iron Formations. The approach strengthens the geochemical interpretation of spectral absorption parameters and offers a practical tool for assessing compositional variability in iron-oxide-dominated lithologies.

Graphical Abstract