You are here:
MERIS Retrieval of Water Quality Components in the Turbid Albemarle-Pamlico Sound Estuary, USA
Citation:
Sokoletsky, L. G., R. S. LUNETTA, M. S. Wetz, AND H. W. Paerl. MERIS Retrieval of Water Quality Components in the Turbid Albemarle-Pamlico Sound Estuary, USA . Remote Sensing. MDPI, Basel, Switzerland, 3(4):684-707, (2011).
Impact/Purpose:
The decomposition of the reflectance spectra into inherent optical properties (IOPs), including absorption (a) and backscattering (bb) coefficients, is a very useful conceptual tool for estimating water quality components (WQCs) in natural waters [1-7]. These WQCs include chlorophyll a (Chl), volatile (organic) suspended solids (VSS), fixed (inorganic) suspended solids (FSS), total suspended solids (TSS), and chromophoric dissolved organic matter (CDOM) absorption (aCDOM) . Decomposition from reflectance to IOPs spectra is important because reflectance data may be obtained using either in situ and/or remote-sensing (RS) measurements. IOPs can be directly related to Chl and suspended particles concentrations and aCDOM, and these relationships render the decomposition of reflectance to IOPs process as is critically important for solving various problems arising in hydrologic optics and hydrologic remote sensing.
Description:
Biological, geophysical and optical field observations carried out in the Neuse River Estuary, North Carolina, USA were used to develop a semi-empirical optical algorithm for assessing inherent optical properties associated with water quality components (WQCs). Three wavelengths (560, 665 and 709 nm) were explored for algorithm development. WQCs included chlorophyll a (Chl), volatile suspended solids (VSS), fixed suspended solids (FSS), total suspended solids (TSS) and absorption of chromophoric dissolved organic matter (aCDOM) CDOM). The relationships between the remote-sensing reflectance and (1) the reflectance factor and spherical albedo, (2) the Gordon’s parameter, (3) the plane albedo, (4) the particulate absorption coefficient at 665 nm (aTSS, red), (5) the backscattering coefficient, and (6) the WQCs, were derived. We simulated and analyzed the following three factors: (i) multiple scattering, (ii) CDOM absorption in the red and near infrared spectral domains, and (iii) scattering phase function to estimate an accuracy of aTSS, red prediction. The algorithm was validated by (a) visual comparison of Chl dynamics between experimental and predicted values; (b) numerical comparison between measured and modeled Chl values; and (c) an error analysis. The numerical comparison yielded the highest correlation between predicted and measured WQCs for Chl (R2= 0.89) and the lowest for FSS (R2= 0.00), while the mean-normalized (to the mean) root-mean-squares errors obtained for aCDOM(412.5) and FSS were 35% and 59%, respectively. WQCs retrieval accuracy was typically significantly better at values of aTSS,red > 0.5 m-1.