First derivative indices for the remote sensing of inland water quality using high spectral resolution reflectance

Abstract

Remote sensing is a technique with potential for monitoring the surface water quality of lakes and rivers. However, new technologies are required in order to overcome the poor spatial and spectral resolutions of conventional satellite sensors. This paper presents the results of novel derivative analyses applied to both high spectral resolution subsurface reflectance and similar data obtained at 450 m above lakes in the Netherlands. Several first derivatives, at 670, 722, and 840 nm wavelengths, showed good potential for the prediction of surface suspended matter concentrations. These were similar in strenght to correlations found between reflectance in a single band and suspended sediment. First derivatives of reflectance, at 620, 638, and 661 nm in particular, showed higher correlations with chlorophyll a concentrations than compared with individual reflectance wavebands. The results suggest that derivatives of reflectance, calculated using high spectral resolution reflectance, may be used as the basis for indices for the detection of inland water quality and warrant further study.