Multispectral remote-sensing algorithms for particulate organic carbon (POC): The Gulf of Mexico

Abstract

To greatly increase the spatial and temporal resolution for studying carbon dynamics in the marine environment, we have developed remote sensing algorithms for particulate organic carbon (POC) by matching in-situ POC measurements in the Gulf of Mexico with matching SeaWiFS remote sensing reflectance. Data on total particulate matter (PM) as well as POC collected during nine cruises in spring, summer and early winter from 1997-2000 as part of the Northeastern Gulf of Mexico (NEGOM) study were used to test algorithms across a range of environments from low %POC coastal waters to high %POC open ocean waters. Finding that the remote-sensing reflectance clearly exhibited a peak shift from blue to green wavelengths with increasing POC concentration, we developed a maximum normalized difference carbon index (MNDCI) algorithm which uses the maximum band ratio of all available blue-to-green wavelengths, and provides a very robust estimate over a wide range of POC and PM concentrations (R 2 =0.99, N=58). The algorithm can be extrapolated throughout the region of shipboard sampling for more detailed coverage and analysis.

Mapping the Changjiang Diluted Water in the East China Sea during summer over a 10-year period using GOCI satellite sensor data

“…The difference between the in situ and estimated salinity is greater in low-salinity waters, which occur in optically complex ocean environments under particle-laden conditions and/or under the influence of CDOM (Baker and Lavelle, 1984;Siegel et al, 2005a;Ahn et al, 2008;Son et al, 2009b;Son et al, 2012). Although CDOM was inversely calculated from ocean color parameters, the correlation with CDOM was not relevant to the present study owing to the lack of in situ CDOM data.…”

“…The multiple band ratio provides a robust means for distinguishing bio-optical regimes according to the concentration range over which each band is dominant (Son et al, 2009b;Son et al, 2012). In this study, the values were directly (but not linearly) proportional to beam attenuation and were fitted with a cubic polynomial (Figure 4B, F = 61.4, p< 0.01, R 2 = 0.93, N = 391).…”

“…The correlations between a proxy of particle concentration (c) and salinity suggested that the concentration of particle-laden water could provide a useful estimate of salinity from space (Kim et al, 2009b;Son et al, 2012). Particle components (PM, POC, and chlorophyll-a) are determined by applying empirical, semianalytical, and/or modeling approaches to ocean color products (Mishonov et al, 2003;Gardner et al, 2006;Son et al, 2009a;Son et al, 2009b). It is reasonable to expect c and salinity to vary widely in the area of the ECS (Figures 3, 4) because the Changjiang River delivers a large mass of particles to the shelf environment, and particulate plumes are widely dispersed along and across the shelf (Moh et al, 2018).…”

“…Figure 4B shows the results of a regression between Rrs band ratios versus c, which used a multiple-wavelength approach to produce the MNDCI (Son et al, 2009b;Son et al, 2012, Eq. 2).…”

Mapping the Changjiang Diluted Water in the East China Sea during summer over a 10-year period using GOCI satellite sensor data

“…The difference between the in situ and estimated salinity is greater in low-salinity waters, which occur in optically complex ocean environments under particle-laden conditions and/or under the influence of CDOM (Baker and Lavelle, 1984;Siegel et al, 2005a;Ahn et al, 2008;Son et al, 2009b;Son et al, 2012). Although CDOM was inversely calculated from ocean color parameters, the correlation with CDOM was not relevant to the present study owing to the lack of in situ CDOM data.…”

“…The multiple band ratio provides a robust means for distinguishing bio-optical regimes according to the concentration range over which each band is dominant (Son et al, 2009b;Son et al, 2012). In this study, the values were directly (but not linearly) proportional to beam attenuation and were fitted with a cubic polynomial (Figure 4B, F = 61.4, p< 0.01, R 2 = 0.93, N = 391).…”

“…The correlations between a proxy of particle concentration (c) and salinity suggested that the concentration of particle-laden water could provide a useful estimate of salinity from space (Kim et al, 2009b;Son et al, 2012). Particle components (PM, POC, and chlorophyll-a) are determined by applying empirical, semianalytical, and/or modeling approaches to ocean color products (Mishonov et al, 2003;Gardner et al, 2006;Son et al, 2009a;Son et al, 2009b). It is reasonable to expect c and salinity to vary widely in the area of the ECS (Figures 3, 4) because the Changjiang River delivers a large mass of particles to the shelf environment, and particulate plumes are widely dispersed along and across the shelf (Moh et al, 2018).…”

“…Figure 4B shows the results of a regression between Rrs band ratios versus c, which used a multiple-wavelength approach to produce the MNDCI (Son et al, 2009b;Son et al, 2012, Eq. 2).…”

Estimating Organic and Inorganic Part of Suspended Solids from Sentinel 2 in Different Inland Waters

“…Several bio-optical algorithms have been created to estimate the Particulate Organic Carbon (POC) concentration in oceanic layers [66][67][68][69][70] and eutrophic lakes [71]. These algorithms were obtained for band ratios corresponding to red-to-blue (R 665 /R 490 ), red-to-green (R 665 /R 560 ), or blue-to-red (R 490 /R 665 ) for ocean studies and for blue-to-green (R 443 /R 560 and R 490 /R 560 ) ratios for lake studies.…”

Mapping the Changjiang Diluted Water in the East China Sea during summer over a 10-year period using GOCI satellite sensor data

“…The difference between the in situ and estimated salinity is greater in low-salinity waters, which occur in optically complex ocean environments under particle-laden conditions and/or under the influence of CDOM (Baker and Lavelle, 1984;Siegel et al, 2005a;Ahn et al, 2008;Son et al, 2009b;Son et al, 2012). Although CDOM was inversely calculated from ocean color parameters, the correlation with CDOM was not relevant to the present study owing to the lack of in situ CDOM data.…”

“…The multiple band ratio provides a robust means for distinguishing bio-optical regimes according to the concentration range over which each band is dominant (Son et al, 2009b;Son et al, 2012). In this study, the values were directly (but not linearly) proportional to beam attenuation and were fitted with a cubic polynomial (Figure 4B, F = 61.4, p< 0.01, R 2 = 0.93, N = 391).…”

“…The correlations between a proxy of particle concentration (c) and salinity suggested that the concentration of particle-laden water could provide a useful estimate of salinity from space (Kim et al, 2009b;Son et al, 2012). Particle components (PM, POC, and chlorophyll-a) are determined by applying empirical, semianalytical, and/or modeling approaches to ocean color products (Mishonov et al, 2003;Gardner et al, 2006;Son et al, 2009a;Son et al, 2009b). It is reasonable to expect c and salinity to vary widely in the area of the ECS (Figures 3, 4) because the Changjiang River delivers a large mass of particles to the shelf environment, and particulate plumes are widely dispersed along and across the shelf (Moh et al, 2018).…”

“…Figure 4B shows the results of a regression between Rrs band ratios versus c, which used a multiple-wavelength approach to produce the MNDCI (Son et al, 2009b;Son et al, 2012, Eq. 2).…”

Estimating Organic and Inorganic Part of Suspended Solids from Sentinel 2 in Different Inland Waters

“…Several bio-optical algorithms have been created to estimate the Particulate Organic Carbon (POC) concentration in oceanic layers [66][67][68][69][70] and eutrophic lakes [71]. These algorithms were obtained for band ratios corresponding to red-to-blue (R 665 /R 490 ), red-to-green (R 665 /R 560 ), or blue-to-red (R 490 /R 665 ) for ocean studies and for blue-to-green (R 443 /R 560 and R 490 /R 560 ) ratios for lake studies.…”

Mapping the Changjiang Diluted Water in the East China Sea during summer over a 10-year period using GOCI satellite sensor data

“…The difference between the in situ and estimated salinity is greater in low-salinity waters, which occur in optically complex ocean environments under particle-laden conditions and/or under the influence of CDOM (Baker and Lavelle, 1984;Siegel et al, 2005a;Ahn et al, 2008;Son et al, 2009b;Son et al, 2012). Although CDOM was inversely calculated from ocean color parameters, the correlation with CDOM was not relevant to the present study owing to the lack of in situ CDOM data.…”

“…The multiple band ratio provides a robust means for distinguishing bio-optical regimes according to the concentration range over which each band is dominant (Son et al, 2009b;Son et al, 2012). In this study, the values were directly (but not linearly) proportional to beam attenuation and were fitted with a cubic polynomial (Figure 4B, F = 61.4, p< 0.01, R 2 = 0.93, N = 391).…”

“…The correlations between a proxy of particle concentration (c) and salinity suggested that the concentration of particle-laden water could provide a useful estimate of salinity from space (Kim et al, 2009b;Son et al, 2012). Particle components (PM, POC, and chlorophyll-a) are determined by applying empirical, semianalytical, and/or modeling approaches to ocean color products (Mishonov et al, 2003;Gardner et al, 2006;Son et al, 2009a;Son et al, 2009b). It is reasonable to expect c and salinity to vary widely in the area of the ECS (Figures 3, 4) because the Changjiang River delivers a large mass of particles to the shelf environment, and particulate plumes are widely dispersed along and across the shelf (Moh et al, 2018).…”

“…Figure 4B shows the results of a regression between Rrs band ratios versus c, which used a multiple-wavelength approach to produce the MNDCI (Son et al, 2009b;Son et al, 2012, Eq. 2).…”

Estimating Organic and Inorganic Part of Suspended Solids from Sentinel 2 in Different Inland Waters

“…Several bio-optical algorithms have been created to estimate the Particulate Organic Carbon (POC) concentration in oceanic layers [66][67][68][69][70] and eutrophic lakes [71]. These algorithms were obtained for band ratios corresponding to red-to-blue (R 665 /R 490 ), red-to-green (R 665 /R 560 ), or blue-to-red (R 490 /R 665 ) for ocean studies and for blue-to-green (R 443 /R 560 and R 490 /R 560 ) ratios for lake studies.…”