A spatiotemporal attention-augmented ConvLSTM model for ocean remote sensing reflectance prediction

Remote sensing reflectance ( R rs ) is an essential parameter in ocean color remote sensing and a fundamental input for the estimation of ocean color elements. Predicting R rs has the potential to enable simultaneous prediction of multiple marine environmental parameters, facilitating multi -perspective analysis of marine environmental changes. This paper proposes a spatiotemporal attention-augmented ConvLSTM-based model for ocean R rs prediction. The developed model can predict R rs for up to seven days by simultaneously learning spatiotemporal features from time series R rs and auxiliary environmental variables. According to the experiments, the proposed model achieves optimal performances on R rs predictions at 443, 488, and 555 nm, with Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE) for the first four prediction days less than 5.6*10 -4 sr -1 and 8.6 %, respectively, which are better than the performance of the convolutional neural network (CNN), the LSTM, CNN-LSTM, and the ConvLSTM. The spatial and temporal variations of R rs are also compared to evaluate the effectiveness of the model, presenting a consistent spatiotemporal pattern between predicted and observed R rs . We also found that integrating sea surface temperature (SST), photosynthetically available radiation (PAR), and aerosol optical thickness at 869 nm (AOT 869 ) into the model can improve the prediction accuracy in various degrees. This work suggests the proposed deep learning model can predict R rs for 7 days with a convincing performance, providing critical data and technical support for ocean-related applications, such as algae bloom monitoring.