This study leverages advancements in deep learning (DL) to enhance the analysis of satellite image time series (SITSs) in marine geoscience, focusing on the prediction of sea surface temperature (SST) and the detection of coastal upwelling. By employing convolutional neural networks (CNNs) and recurrent neural networks (RNNs), including long short-term memory (LSTM) networks, we introduce a novel approach utilizing convolutional LSTM (ConvLSTM) and 3-D Unet-LSTM models. These techniques provide a nuanced analysis and understanding of complex oceanographic phenomena, specifically coastal upwelling, which significantly impacts marine ecosystems and climate. The adoption of these sophisticated DL models has led to a notable improvement in predicting SST fields, achieving a reduction in root mean square error (RMSE) to 0.038 and an increase in the correlation coefficient (CC) to 0.95. This enhancement over the baseline ConvLSTM model, which had an RMSE of 0.045 and a CC of 0.92, underscores our models’ capability to accurately capture the dynamic and intricate nature of coastal upwelling. The results offer promising directions for future research in marine geoscience and remote-sensing applications, highlighting the potential of DL techniques in interpreting intricate patterns in satellite-derived data and improving predictions in environmental sciences.

中文翻译：

---

使用基于深度学习的模型对海温场中每月沿海上升流情景进行时空预测

本研究利用深度学习 (DL) 的进步来增强海洋地球科学中卫星图像时间序列 (SITS) 的分析，重点是海面温度 (SST) 的预测和沿海上升流的检测。通过采用卷积神经网络 (CNN) 和循环神经网络 (RNN)，包括长短期记忆 (LSTM) 网络，我们引入了一种利用卷积 LSTM (ConvLSTM) 和 3-D Unet-LSTM 模型的新颖方法。这些技术可以对复杂的海洋现象，特别是对海洋生态系统和气候产生重大影响的沿海上升流进行细致入微的分析和理解。这些复杂的深度学习模型的采用使得海表温度场的预测得到显着改进，将均方根误差 (RMSE) 降低到 0.038，并将相关系数 (CC) 提高到 0.95。与基线 ConvLSTM 模型（其 RMSE 为 0.045，CC 为 0.92）相比，这一增强强调了我们的模型准确捕捉沿海上升流的动态和复杂性质的能力。这些结果为海洋地球科学和遥感应用的未来研究提供了有前景的方向，凸显了深度学习技术在解释卫星数据中的复杂模式和改进环境科学预测方面的潜力。