As a key component of the 3D circulation, the mid-layer (500-1500 m depth) circulation in the South China Sea (SCS) plays a pivotal role in energy transfer and water mass exchange. However, accurately characterizing its circulation structure remains challenging, mainly stemming from the paucity of observational data and the limited accuracy of numerical models. In recent years, Argo trajectory-derived velocities have offered valuable insights for evaluating and refining numerical models. To alleviate systematic biases in simulating the SCS mid-layer circulation, a multi-source observational data assimilation system based on the ROMS I4D-Var framework is developed. Within this system, Argo trajectory-derived velocities were assimilated alongside satellite observations and in-situ profiles (referred to as EXP3) in May 2022. Results indicated that assimilating Argo trajectory-derived velocities significantly improved the simulation of the SCS mid-layer circulation by reducing random errors. The root mean square error (RMSE) of absolute velocity decreased from 5 cm/s to 3 cm/s. Notably, a significant improvement was achieved in current direction accuracy, with the proportion of points with well-simulated directions increasing from 26% to 66%. A comprehensive analysis of velocity and direction revealed that the proportion of well-modeled points (57%) in EXP3 was approximately 4.4 times that in the control experiment (without data assimilation, 13%). Furthermore, EXP3 successfully corrected the vertical current anomaly of the Luzon Strait, reproducing the "sandwich" pattern of water exchange between the Pacific Ocean and the SCS. These findings underscore the significance of Argo trajectories for advancing our understanding of the 3D circulation in the SCS.
Liu et al. (2026) studied this question.