Carbon Sink Prediction of Marine Algae and Shellfish Aquaculture Based on Multivariate Remote Sensing

In this paper, a novel carbon sink prediction model is first proposed for marine algae and shellfish based on multivariate remote sensing data to improve timeliness and scientificity. Regression models are used to evaluate the correlations between five typical ocean data and the yield of algae/shellfish from 2013 to 2020. SST and Chl-a are analyzed to select the month or monthly combination exhibiting the strongest correlation with yields. For algae, the SST in March of the current year, the Chl-a in January of the current year, and the maximum aquaculture area monitored using remote sensing images are selected to predict aquaculture yields using multiple nonlinear regression methods. The yield interannual variation results show that there is a positive correlation between algae yields and SST/Chl-a. Finally, the total carbon sink of algae and shellfish is obtained by the proportions of different algae and shellfish types, dry-wet ratio, and carbon content ratio