LAUSR

Optical remotely sensed data often have data gaps due to cloud coverage, which hinders their full potential in many environmental applications. The question of how to accurately and effectively reconstruct the measurements in the data gaps remains a challenge. In this study, we developed a bidirectional long short-term memory (BiLSTM) model with a novel and adaptable custom temporal penalty layer for spatiotemporal gap filling. The model was trained and tested in time-series daily nighttime sea surface temperature (SST) images acquired by the Himawari-8 satellite. The modeling results showed strong performance, accurately reconstructing spatial and temporal features of cloud-affected SST data. Our neural network achieved a per-image MAE of 0.1193 °C and per-image root mean square error (RMSE) of 0.1293 °C. The model was also able to produce realistic SST time-series predictions that were consistent with the expected seasonal variables. Importantly, the BiLSTM model outperformed the previous state-of-the-art simple spatial gap-filling processor (SSGP) algorithm in terms of error metrics, structural similarity (SSIM), and computational efficiency. Overall, the results of this study indicate that the BiLSTM neural network model we developed is highly suitable for spatiotemporal gap filling of SST and other remotely sensed data.