Northern Gulf of Mexico estuarine coloured dissolved organic matter derived from MODIS data

Abstract

Coloured dissolved organic matter (CDOM) is relevant for water quality management and may become an important measure to complement future water quality assessment programmes. An approach to derive CDOM using the Moderate Resolution Imaging Spectroradiometer (MODIS) was developed that may be more accessible to water quality managers by selecting an off-the-shelf software and algorithm with standard atmospheric correction. This study focused on demonstrating the transferability of a remote-sensing reflectance (R_rs) band ratio algorithm, R_rs(667)/R_rs(488), previously developed to derive CDOM absorption (a_g(λ)) at multiple MODIS wavebands in open ocean and coastal waters to the United States northern Gulf of Mexico estuaries. In situ calibration regressions at 412, 443, 469, and 488 nm had coefficients of determination (R²) of 0.76, 0.71, 0.69, and 0.57, respectively. Waveband calibrations at 531, 547, and 555 nm were below R² = 0.50, and therefore were not considered further. MODIS R_rs, from the standard atmospheric correction, followed nearly identical spectral shapes to the in situ HyperSAS R_rs, but were on average 0.002 ± 0.0004 sr⁻¹ less. A satellite to in situ validation match-up window of ≤1 hour was selected with an R² = 0.82 and root mean square error (RMSE = 1.79) at 412 nm. An in situ water quality mooring demonstrated that the overall response and range of MODIS a_g(412) were similar, with relative mean error from –32% to 42%. The advantage to managers was synoptic coverage across multiple estuaries and the ability to provide estimates of derived water quality parameters between the water quality assessment programme sample collection periods, which could offer more holistic assessment.

Acknowledgements

The information in the article has been subjected to review by the National Exposure Research Laboratory and approved for publication. Mention of trade names or commercial products does not constitute endorsement or recommendation for use. The views expressed in this article are those of the authors and do not necessarily reflect the views or policies of the US Environmental Protection Agency. We acknowledge Kathleen Daniels, David Beddick, Brandon Jarvis, John Lehrter, and James Hagy for assistance in sample collection, data processing, and data management.

Additional information

Funding

The information in this document has been funded in part by the US Environmental Protection Agency and NASA Grant # NNH08ZDA001N-DECISIONS.