An assessment of remote sensing algorithms for colored dissolved organic matter in complex freshwater environments.

This study evaluated fifteen algorithms representing four major categories of retrieval algorithms for aquatic colored dissolved organic matter (CDOM): empirical, semi-analytical, optimization, and matrix inversion methods. The specific goal here was to evaluate (and understand) the strengths and limits of these algorithms in predicting CDOM dynamics along a gradient of varying water quality in a large, freshwater ecosystem. The data were collected in May and October of 2012 from the estuarine areas of the Kawkawlin and Saginaw Rivers, and Lake Huron. Algorithms were evaluated through comparisons to in-situ CDOM measurements, such that the analysis of these field measurements showed that CDOM levels in these areas displayed a range of CDOM absorption coefficients a_CDOM(440) (0.1-8.5m^-1). In general, the majority of the algorithms underestimated high CDOM waters (a_CDOM(440)>2m^-1) and overestimated low CDOM scenarios (<0.5m^-1). Six algorithms that performed consistently better compared with the other models (overall RMSE of <0.45) in estimating in-situ CDOM levels were three empirical, two semi-analytical, and one MIM algorithms. Our analysis identified a set of parameters for the matrix inversion methods (MIM) that allow them to work effectively across a broad range of CDOM levels. Analysis of our results indicated that the most effective wavelengths/band locations for estimating CDOM could vary depending on the levels of spectral interference from high concentrations of particulate matter in the water column. In addition, our results suggest that including wavelengths>600nm in the algorithms improves CDOM estimation accuracy significantly, particularly for complex freshwater environments.