Multivariate Temporal Self-Attention Network for Subsurface Thermohaline Structure Reconstruction

Argo observations are spatially sparse and temporally uneven, whereas satellites can provide high-resolution and continuous observations at the sea surface. The reconstruction of subsurface thermohaline structure using multisource remote sensing data is thus of great significance for investigating the ocean interior dynamics. Aiming at the existing problems of temporal feature extraction and nonlinear relationship fitting, this article proposes a multivariate temporal self-attention network (MTSAN) to effectively reconstruct the subsurface temperature anomaly (STA) and subsurface salinity anomaly (SSA) in the Pacific Ocean. The model integrates multisource remote sensing data, including sea surface temperature (SST) and sea surface salinity (SSS), wind speed, absolute dynamic topography (ADT), and significant wave height (SWH). In order to better extract the complex small- and medium-scale signals, a two-branch asymmetric residual module based on dilation causal convolution is designed to enhance the representation ability. Moreover, zonal weighted loss function with comprehensive indicators is proposed, in order to minimize the real error of grids and raise the accuracy of self-attention network. MTSAN reconstructs the STA and SSA during the El Ni
$\tilde {\mathrm {n}}\text{o}$
event, and the results show that it has good performance for spatial distribution, vertical variation, and temporal extension. The overall
$R^{2}$
and RMSE of STA are 0.536 and 0.241 °C, respectively, and the overall
$R^{2}$
and RMSE of SSA are 0.645 and 0.037 psu, respectively. In addition, the results of comparison experiments illustrate the superiority of MTSAN over other machine learning and deep learning-based methods. Overall, we provide a new temporal self-attention approach to accurately reconstruct the 3-D thermohaline structure using high-resolution quasi-real-time satellite observations.