Remote Sensing Monitoring of Colored Dissolved Organic Matter and Dissolved Organic Carbon in the Arctic Ocean

With the help of MODIS satellite data, the QAA (Quasi-Analytical Algorithm) algorithm was used to obtain the absorption coefficient of colored dissolved organic matter (CDOM) in the surface waters of the Arctic Ocean during the summer of 2010 and 2011. CDOM and DOC were classified according to the field measured reflectance data, and the correlation differences between CDOM and DOC in different bands (412, 443, 490, 532, 555, 667 nm ) and different categories were studied. Finally, a boosted regression tree model was developed to analyze the main driving factors of DOC concentration changes. The results show that the inversion of CDOM absorption coefficients in surface waters of the Arctic Ocean using the QAA algorithm is highly feasible. In addition, classification based on peak reflectance could improve the inversion accuracy. When the peak reflectance located in 555 nm, the correlation between CDOM and DOC (R²=0.95) was higher than overall correlation (R²=0.94), and the inversion accuracy of CDOM was relatively high (R²=0.88, MRE=28.6%). In comparison, when the peak reflectance located in 490 or 465 nm, the correlation between CDOM and DOC (R²=0.76) was weak, and the inversion accuracy of CDOM was relatively low (R²=0.37, MRE=36.4%). Analysis of influencing factors shows that CDOM, suspended particulate matter, and chlorophyll-a were the main drivers of DOC concentration changes in the Arctic Ocean during summer.