Overlooked Current Estimation Biases Arising from the Lagrangian Argo Trajectory Derivation Method

Abstract

An Argo simulation system is used to provide synthetic Lagrangian trajectories based on the Estimating the Circulation and Climate of the Ocean model, Phase II (ECCO2). In combination with ambient Eulerian velocity at the reference layer (1000 m) from the model, quantitative metrics of the Lagrangian trajectory-derived velocities are computed. The result indicates that the biases induced by the derivation algorithm are strongly linked with ocean dynamics. In low latitudes, Ekman currents and vertically sheared geostrophic currents influence both the magnitude and the direction of the derivation velocity vectors. The maximal shear-induced biases exist near the equator with the amplitudes reaching up to about 1.2 cm s^-1. The angles of the shear biases are pronounced in the low latitude oceans, ranging from -8° to 8°. Specifically, the study shows an overlooked bias from the float drifting motions that mainly occurs in the western boundary current and Antarctic circumpolar current (ACC) regions. In these regions, a recently reported horizontal acceleration measured via Lagrangian floats is significantly associated with the strong eddy-jet interactions. The acceleration could induce an overestimation of Eulerian current velocity magnitudes. For the common Argo floats with a 9-day float parking period, the derivation speed biases induced by velocity acceleration would be as large as 3 cm s^-1, approximately 12% of the ambient velocity. It might have implications to map the mean mid-depth ocean currents from Argo trajectories, as well as understand the dynamics of eddy-jet interactions in the ocean. An Argo simulation system is used to provide synthetic Lagrangian trajectories based on the Estimating the Circulation and Climate of the Ocean model, Phase II (ECCO2). In combination with ambient Eulerian velocity at the reference layer (1000 m) from the model, quantitative metrics of the Lagrangian trajectory-derived velocities are computed. The result indicates that the biases induced by the derivation algorithm are strongly linked with ocean dynamics. In low latitudes, Ekman currents and vertically sheared geostrophic currents influence both the magnitude and the direction of the derivation velocity vectors. The maximal shear-induced biases exist near the equator with the amplitudes reaching up to about 1.2 cm s^-1. The angles of the shear biases are pronounced in the low latitude oceans, ranging from -8° to 8°. Specifically, the study shows an overlooked bias from the float drifting motions that mainly occurs in the western boundary current and Antarctic circumpolar current (ACC) regions. In these regions, a recently reported horizontal acceleration measured via Lagrangian floats is significantly associated with the strong eddy-jet interactions. The acceleration could induce an overestimation of Eulerian current velocity magnitudes. For the common Argo floats with a 9-day float parking period, the derivation speed biases induced by velocity acceleration would be as large as 3 cm s^-1, approximately 12% of the ambient velocity. It might have implications to map the mean mid-depth ocean currents from Argo trajectories, as well as understand the dynamics of eddy-jet interactions in the ocean.