The relationship between dissolved organic matter absorption and dissolved organic carbon in reservoirs along a temperate to tropical gradient

• Published in: Remote sensing of environment 2015-01, Vol.156, p.395-402
• Main Authors: Hestir, Erin L., Brando, Vittorio, Campbell, Glenn, Dekker, Arnold, Malthus, Tim
• Format: Article
• Language: English
• Subjects: Absorption coefficient; Carbon; CDOM; Concentration (composition); Dissolved organic carbon; Dissolved organic matter; Lakes; Landsat; Remote sensing; Reservoirs; Sentinel-2; Slopes; Spectra; Water quality
• ISSN: 0034-4257; 1879-0704
• DOI: 10.1016/j.rse.2014.09.022

Recent and upcoming launches of new satellite sensors will provide the spatial, spectral and radiometric resolution to globally assess freshwater chromophoric dissolved organic matter (CDOM), and thus estimate dissolved organic carbon (DOC) concentration. However, estimating DOC from optical remote sensing requires a robust relationship between CDOM and DOC. This is particularly problematic for reservoirs because they have variable dissolved organic matter composition that complicates the CDOM–DOC relationship. We investigated six manmade reservoirs along a temperate to tropical gradient that represent a range of reservoir types and watershed conditions to determine whether a linear relation between CDOM and DOC could be established. We measured CDOM absorption and DOC concentration during the wet and dry seasons in the six reservoirs. We found the CDOM absorption coefficient and CDOM spectral slope were uncorrelated due to exogenous DOC inputs from multiple sources. Alone, the absorption coefficient of CDOM was a poor predictor of DOC concentration. Including both CDOM absorption coefficient and spectral slope in a multiple regression accounted for both composition and concentration, significantly improving the regression r2. By using both CDOM absorption coefficient and spectral slope, we identify a framework for a potential solution to overcome the influence of dissolved organic matter source and transformation history on the CDOM–DOC relationship. We conclude that local variability, seasonality and optical complexity should be considered in remote sensing based approaches for global freshwater DOC estimation. •Reservoir hydrodynamics lead to optically complex CDOM absorption.•R2 improves when regressions for DOC use CDOM absorption coefficient and spectral slope.•Local variability and optical complexity may affect remote sensing of DOC.