TY - JOUR
T1 - Assessing Chromophoric Dissolved Organic Matter (CDOM) Distribution, Stocks, and Fluxes in Apalachicola Bay Using Combined Field, VIIRS Ocean Color, and Model Observations
AU - Joshi, Ishan D.
AU - D'Sa, Eurico J.
AU - Osburn, Christopher L.
AU - Bianchi, Thomas S.
AU - Ko, Dong S.
AU - Oviedo-Vargas, Diana
AU - Arellano, Ana
AU - Ward, Nicholas D.
PY - 2017/1/1
Y1 - 2017/1/1
N2 - Understanding the role of estuarine-carbon fluxes is essential to improve estimates of the global carbon budget . Dissolved organic matter (DOM) plays an important role in aquatic carbon cycling. The chromophoric fraction of DOM (CDOM) can be readily detected via in situ and remotely-sensed optical measurements . DOM properties, including CDOM absorption coefficient at 412 nm ( a g 412 ) and dissolved organic carbon (DOC) concentrations were examined in Apalachicola Bay, a national estuarine research reserve located in the northeast Gulf of Mexico , using in situ and satellite observations during the spring and fall of 2015. Synoptic and accurate representation of estuarine-scale processes using satellite ocean color imagery necessitates the removal of atmospheric contribution (~ 90%) to signals received by satellite sensors to successfully link to in situ observations. Three atmospheric correction schemes ( e . g ., Standard NIR correction, Iterative NIR correction, and SWIR correction) were tested first to find a suitable correction scheme for the VIIRS imagery in low to moderately turbid Apalachicola Bay. The iterative NIR correction performed well, and validation showed high correlation (R 2 = 0.95, N = 25) against in situ light measurements. A VIIRS-based CDOM algorithm was developed (R 2 = 0.87, N = 9) and validated (R 2 = 0.76, N = 20, RMSE = 0.29 m − 1 ) against in situ observations. Subsequently, a g 412 was used as a proxy of DOC in March (DOC = 1.08 + 0.94 × a g 412 , R 2 = 0.88, N = 13) and in November (DOC = 1.61 + 1.33 × a g 412 , R 2 = 0.83, N = 24) to derive DOC maps that provided synoptic views of DOC distribution, sources, and their transport to the coastal waters during the wet and dry seasons. The estimated DOC stocks were ~ 3.71 × 10 6 kg C in March and ~ 4.07 × 10 6 kg C in November over an area of ~ 560 km 2 . Volume flux (out of the bay) almost doubled for March 24 (735 m 3 s − 1 ) relative to November 4 (378 m 3 s − 1 ). However, estimates of DOC fluxes exported out of the bay from model-derived currents and satellite-derived DOC were only marginally greater in March (0.163 × 10 6 kg C d − 1 ) than in November (0.124 × 10 6 kg C d − 1 ) and reflected greater DOC stocks in the fall. The combination of satellite-, field-, and model-based observations revealed the strong linkage between the Apalachicola River plume , a major source of DOM, and the overall hydrodynamic forcing that controlled distributions of CDOM abundance, DOC concentration, stocks, and fluxes in the bay.
AB - Understanding the role of estuarine-carbon fluxes is essential to improve estimates of the global carbon budget . Dissolved organic matter (DOM) plays an important role in aquatic carbon cycling. The chromophoric fraction of DOM (CDOM) can be readily detected via in situ and remotely-sensed optical measurements . DOM properties, including CDOM absorption coefficient at 412 nm ( a g 412 ) and dissolved organic carbon (DOC) concentrations were examined in Apalachicola Bay, a national estuarine research reserve located in the northeast Gulf of Mexico , using in situ and satellite observations during the spring and fall of 2015. Synoptic and accurate representation of estuarine-scale processes using satellite ocean color imagery necessitates the removal of atmospheric contribution (~ 90%) to signals received by satellite sensors to successfully link to in situ observations. Three atmospheric correction schemes ( e . g ., Standard NIR correction, Iterative NIR correction, and SWIR correction) were tested first to find a suitable correction scheme for the VIIRS imagery in low to moderately turbid Apalachicola Bay. The iterative NIR correction performed well, and validation showed high correlation (R 2 = 0.95, N = 25) against in situ light measurements. A VIIRS-based CDOM algorithm was developed (R 2 = 0.87, N = 9) and validated (R 2 = 0.76, N = 20, RMSE = 0.29 m − 1 ) against in situ observations. Subsequently, a g 412 was used as a proxy of DOC in March (DOC = 1.08 + 0.94 × a g 412 , R 2 = 0.88, N = 13) and in November (DOC = 1.61 + 1.33 × a g 412 , R 2 = 0.83, N = 24) to derive DOC maps that provided synoptic views of DOC distribution, sources, and their transport to the coastal waters during the wet and dry seasons. The estimated DOC stocks were ~ 3.71 × 10 6 kg C in March and ~ 4.07 × 10 6 kg C in November over an area of ~ 560 km 2 . Volume flux (out of the bay) almost doubled for March 24 (735 m 3 s − 1 ) relative to November 4 (378 m 3 s − 1 ). However, estimates of DOC fluxes exported out of the bay from model-derived currents and satellite-derived DOC were only marginally greater in March (0.163 × 10 6 kg C d − 1 ) than in November (0.124 × 10 6 kg C d − 1 ) and reflected greater DOC stocks in the fall. The combination of satellite-, field-, and model-based observations revealed the strong linkage between the Apalachicola River plume , a major source of DOM, and the overall hydrodynamic forcing that controlled distributions of CDOM abundance, DOC concentration, stocks, and fluxes in the bay.
KW - Apalachicola Bay
KW - Atmospheric-correction
KW - Carbon stocks fluxes
KW - CDOM algorithm
KW - DOC
KW - VIIRS
UR - https://digitalcommons.usf.edu/msc_facpub/1440
UR - https://doi.org/10.1016/j.rse.2017.01.039
U2 - 10.1016/j.rse.2017.01.039
DO - 10.1016/j.rse.2017.01.039
M3 - Article
VL - 191
JO - Remote Sensing of Environment
JF - Remote Sensing of Environment
ER -