Scripps Argo Trajectory-Based Velocity Product: Global Estimates of Absolute Velocity Derived from Core, Biogeochemical, and Deep Argo Float Trajectories at Parking Depth

Global estimates of absolute velocities can be derived from Argo float trajectories during drift at parking depth. A new velocity dataset developed , maintained at Scripps Institution of Oceanography is presented based on all Core, Biogeochemical , Deep Argo float trajectories collected between 2001 and 2020. Discrepancies between velocity estimates from the Scripps dataset and other existing products including YoMaHa and ANDRO are associated with quality control criteria, as well as selected parking depth and cycle time. In the Scripps product, over 1.3 million velocity estimates are used to reconstruct a time-mean velocity field for the 800-1200 dbar layer at 1 degrees horizontal resolution. This dataset pro-vides a benchmark to evaluate the veracity of the BRAN2020 reanalysis in representing the observed variability of abso-lute velocities and offers a compelling opportunity for improved characterization and representation in forecast and reanalysis systems.SIGNIFICANCE STATEMENT: The aim of this study is to provide observation-based estimates of the large-scale, subsurface ocean circulation. We exploit the drift of autonomous profiling floats to carefully isolate the inferred circula-tion at the parking depth, and combine observations from over 11000 floats, sampling between 2001 and 2020, to deliver a new dataset with unprecedented accuracy. The new estimates of subsurface currents are suitable for assessing global models, reanalyses, and forecasts, and for constraining ocean circulation in data-assimilating models.