Deep Learning for Chlorophyll-a Concentration Retrieval: A Case Study for the Pearl River Estuary

Abstract

The abundance of phytoplankton is generally estimated by measuring the chlorophyll-a concentration (Cchla), which is an important factor in photosynthesis and can be used to analyze the density and biomass of phytoplankton in the ecosystem. The band-ratio-based empirical or semi-analytical algorithms are operationally applied to retrieve Cchla in global oceans, which generally experience difficulties from the diversity of optical properties and the complexity of the radiative transfer equations in analytical analyses, respectively. With an attempt to develop an accurate Cchla retrieval model for the optically complex coastal and estuarine waters, this study aimed to explore the deep learning (DL) methods in satellite retrieval of Cchla. A two-stage convolutional neural network (CNN), named Cchla-Net, was proposed, which utilized the spectral information of remote sensing reflectances at MODIS/Aqua’s visible bands. In the first-stage phase, the Cchla-Net was pretrained by a set of remote sensing patches, in which the Cchla was generated from an existing model (OC3M). The pretrained results were than used as the initial values to refine the network with the synthetic oversampled in-situ dataset in the second-stage training phase. Using in-situ samples for training with the new initial values has a higher probability to reach the global optimum. The quantitative analyses showed that the two-stage training was more likely to achieve a global optimum in the optimization than the one-stage training. Matchups of the in-situ Cchla measurements were used to evaluate the retrieval models. Results showed that the proposed Cchla-Net produced obvious better performance than the empirical and semi-analytical algorithms, implying the DL method was more effective for optically complex waters with extremely high Cchla. This study provided an applicable method for remote sensing retrieval of Cchla, which should be helpful for studying the spatial distribution and temporal variability in the productive Pearl River estuary (PRE) waters.