Applying Spectral Unmixing to Determine Surface Water Parameters in a Mining Environment

摘要: Compared to natural waters, mine waters represent an extreme water type that is frequently heavily polluted. Although they have been traditionally monitored by in situ measurements of point samples taken at regular intervals, the emergence a new generation multispectral and hyperspectral (HS) sensors means image spectroscopy has potential become modern method for monitoring polluted surface waters. This paper describes approach employing linear Spectral Unmixing (LSU) analysis data map relative abundances components (dissolved Fe—Fediss, dissolved organic carbon—DOC, undissolved particles). The ground truth (8 ponds) were used validate results spectral mapping. same applied HS was tested using resampled WorldView2 (WV2) resolution. A key aspect processing define proper pure end members fundamental types. highest correlations detected between studied parameters fractional images HyMap WV2 data, respectively, were: Fe (R2 = 0.74 R2vw2 0.6), particles 0.57 0.49) DOC 0.42 < 0.40). These further classified create semi-quantitative maps. In conclusion, classification still benefited from higher resolution data; however reflectance can be suitable mapping specific inherent optical properties (SIOPs), which significantly differ one another view (e.g., mineral suspension, phytoplankton), but it seems difficult differentiate among diverse suspension particles, especially when more complex together with tripton or other etc.).