The reflectance were measured for each sample filter to

The remote sensing reflectance of the MS dataset wasacquired in the range of 400–900 nm with a sampling interval of 0.3 nm bydeploying a dual sensor system with two inter-calibrated ocean opticsspectroradiometers (Ocean Optics Inc., Dunedin, FL, USA). The in situ Rrs(?) from LT and LE were measured with a hand-held ASDspectroradiometer (Analytical Spectral Device, Inc.Boulder, Colorado, U.S.A.

), from 350 to 1050 nm at 1.4 nm intervals. 2.1.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

3. Absorption CoefficientsSurface water samples were collected and filtered immediatelyafter the Rrs(?) measurements and analyzed on the sameday in the laboratory. The particulate absorption coefficient was quantifiedutilizing the transmittance or transmittance/reflectance method of Tassan andFerrari 22 along with the NASA Ocean Optics Protocols, Revision 4, Volume IVprotocol 23. Percent transmittance and reflectance were measured for eachsample filter to calculate the particulate absorption spectrum. Filters werethen bleached, and the transmittance and reflectance measured to calculate theabsorption coefficients of non-algal particles. The phytoplankton absorptioncoefficients (aph(?)) at each station were calculated bysubtraction of non-algal particle absorption coefficients from total particleabsorption coefficients.

The Gaussian peak heights (aGau(?)) were obtained from aph(?) using the Gaussian decomposition scheme described in Wang etal. 9, which used a least-square curve fitting technique in Matlab to obtainthe aGau(?). These aGau(?) wereused as ground truth to validate the outputs from Rrs(?)inversions.For the MS dataset, a Perkin Elmer Lambda 850Spectrophotometer (Perkin Elmer Inc.

, Waltham, MA, USA) was used to measure theabsorption coefficient of phytoplankton, detrital matter, and gelbstoff in the380–750 nm range at 1 nm spectral resolution. Detailed information regardingthe environmental characteristics and measurement methods can be found in Mishraet al. 18. A Shimadzu UV-2401 Spectrophotometer was used for LT. Moreinformation about LT and details about water sample collection, measurementprotocols, and processing methods can be found in Duan et al. 24 and Ma etal. 25. Samples from LE were measured with a Perkin-Elmer Lambda 35 UV/VisDual-beam Spectrophotometer from 300–800 nm at 1 nm resolution as described in Mouwet al.

26. 2.1.4. Pigment Concentrations and Group CompositionWater samples for Chl-a and PC concentrationfrom Lake Erie were filtered through 0.7 and 0.2 ?m glass fiber filters(Whatman GF/F, 25-mm and 47-mm diameter), respectively.

The Chl-a was estimatedusing a Turner Designs 10-AU Fluorometer in the laboratory following the NASAOcean Optics Protocols, Revision 5, Volume V 23. Following NASA protocols 27,filters for phycocyanin concentration determination were extracted in phosphatebuffer (Ricca Chemical, Arlington, Texas, U.S.A. pH 6.8) usingtwo freeze-thaw cycles, followed by sonication. Then the PC concentration wasestimated from the relative fluorescence measured using a Turner Aquafluor fluorometer21.

Phytoplankton populations for the Lake Erie stations in 2013 were alsoused in this study which were identified and counted using standard lightmicroscopy 21.2.2. Satellite ImageryThe Hyperspectral Imager for the Coastal Ocean (HICO) wasthe first space-borne hyperspectral imaging spectrometer designed to sample thecoastal ocean. HICO covered selected coastal regions at 90 m spatial resolutionwith full spectral coverage (380 to 960 nm sampled at 5.

73 nm intervals). HICOimagery of the same time frame as in situ measurements in Lake Erie in 2014 wasdownloaded from the Oregon State University website (http://hico.coas.oregonstate.edu/), andfollowing the steps provided on the website the level 1B (L1B) imagery wasatmospherically corrected to level 2 (L2) data using the online version ofTafkaa_6s model, followed by georectification.

The HICO Rrs(?) data haveabout the same spectral resolution as the in situ measured reflectance (5.73 nmfor HICO and 5 nm for original Rrs(?)). These data were used as an examplefor MuPI application in hyperspectral satellite imagery.TheModerate Resolution Imaging Spectroradiometer onboard the Aqua satellite(MODIS-Aqua) launched in 2002 is one of thecontemporary satellite ocean color missions. The L1B imagery with 1 km spatialresolution was downloaded from the National Aeronautics and SpaceAdministration (NASA) website (https://ladsweb.nascom.

nasa.gov/).The L2 imagery (with 748 nm included in Rrs(?)) from the same time frame as in situmeasurements in Lake Erie in 2014 was obtained using the data processing L2genfrom the SeaWiFS Data Analysis System (SeaDAS) software using the standard nearinfrared (NIR) scheme 28–30 for atmospheric correction.

MODIS imagery wasused as an example of MuPI application to multi-spectral satellite remotesensing data.MEdium Resolution Imaging Spectrometer (MERIS) on board theEuropean Space Agency’s Envisat platform, was launched in 2002 and ended itsmission in May 2012. The L3 seasonal composed imagery and L2 reduced resolutionimagery (1 km spatial resolution) was obtained from NASA ocean color website (https://oceancolor.

gsfc.nasa.gov/).Lacking coincidence with in situ measurements, the same day imagery of MODISand MERIS on 3 September 2011 was used for the comparison of the retrievingresults. The MERIS seasonal composed imagery in Lake Erie was used to obtain ageneral seasonal variation of Gaussian absorption coefficients.

 2.3. The aGau(?)Estimation2.3.

1. MuPI ModelTo obtain aGau(?) from remote sensing reflectance, amulti-pigment inversion method 9 was used. The corresponding functions usedin this method are included in Table 1. In this method, the Gaussian scheme,which uses 13 Gaussian curves to reconstruct the spectrum of phytoplanktonabsorption coefficient, was incorporated into the relationship in which Rrs(?) is a function of total spectral absorption (a(?)) and backscattering(bb(?)) coefficients 31,32. With an Rrs(?) spectrumas the only input, the function was solved with a spectral optimization methodwhich returns the values of unknowns (aGau(?), absorption of non-algal particlesand gelbstoff: adg(?), and backscattering coefficient ofparticles, bbp(?)), that minimizes the differencebetween the estimated and input Rrs(?) spectra.

A summarized flowchart ofthe method is shown in Figure 2. The parameters used for the 13 Gaussian curvesin this study are shown in Table 2, which includes the peak locations, widthsand empirical relationships between the peak heights.