Back to Search Start Over

Spectral range within global aCDOM(440) algorithms for oceanic, coastal, and inland waters with application to airborne measurements.

Authors :
Houskeeper, Henry F.
Hooker, Stanford B.
Kudela, Raphael M.
Source :
Remote Sensing of Environment. Feb2021, Vol. 253, pN.PAG-N.PAG. 1p.
Publication Year :
2021

Abstract

The optically active component of dissolved organic material in aquatic ecosystems, or colored dissolved organic matter (CDOM), is represented by the coefficient of absorption due to the dissolved aquatic constituents at 440 nm, a CDOM (440). Remote sensing of a CDOM (440) enables characterization of ecosystem processes and aids in retrieval of chlorophyll a , a proxy for phytoplankton biomass. Spectrally adjacent band-ratio domains, e.g., blue to green, have previously been applied for remote sensing of a CDOM (440) in coastal and oceanic waters with similar results compared to more complex semi-analytical algorithms. Estimation of a CDOM (440) from ratios of the most spectrally separated ocean color wavebands (end members), e.g., ultraviolet (UV) to near-infrared (NIR), termed end-member analysis (EMA), has previously been shown to increase the accuracy of global a CDOM (440) retrievals from in-water observations of diffuse attenuation and to enable a unified algorithmic perspective without requiring regional adjustment of internal bio-optical parameters. EMA of above-water observations is evaluated herein, with a focus on coastal and inland waters in which increasing optical complexity and likelihood of bottom reflectance challenge the oceanic algorithms developed for deep and optically simple (case-1) waters. Analysis herein of three independent, in situ , bio-optical datasets indicates significant correlation between a CDOM (440) and end-member band ratios (next-generation 320 and 780 nm or legacy 412 and 670 nm ratios) with a coefficient of determination, R 2, of 0.87 (log-scale) or higher based on a dataset spanning the dynamic range of global, conservative water bodies. For applicable wavelengths, EMA algorithms are shown to agree with case-1 relationships and to produce consistent log-scale uncertainties across more than three orders of magnitude in a CDOM (440) values (0.001–2.305 m−1). EMA using UV and NIR wavelengths (320 and 780 nm) is applied to low-altitude airborne observations and satisfies 25% uncertainty based on unbiased percent differences (UPDs) within each of three dissimilar match-up sites ranging in a CDOM (440) from 0.02–0.57 m−1. Results demonstrate that EMA is a useful and robust approach for the remote sensing of a CDOM (440) in coastal and inland waters, which are generally shallower, contain more optically complex environments, and span a greater range in a CDOM (440) than oceanic waters. • Spectral separation increases dynamic range of aCDOM band-ratio algorithms. • UV-NIR algorithms mitigate bottom reflectance effects in shallow waters. • UV-NIR aCDOM algorithms satisfy proposed uncertainty criteria for global waters. • UV-NIR aCDOM algorithms produce consistent global log-scale aCDOM uncertainties. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
00344257
Volume :
253
Database :
Academic Search Index
Journal :
Remote Sensing of Environment
Publication Type :
Academic Journal
Accession number :
147734206
Full Text :
https://doi.org/10.1016/j.rse.2020.112155