Simulating Argo Float Trajectories and Along-track Physical and Biogeochemical Variability in the California Current System

Xiao Liu ^1* John Patrick Dunne ¹ Elizabeth Joan Drenkard ¹ Gregory C Johnson ²

• ¹ NOAA Geophysical Fluid Dynamics Laboratory, Princeton, United States
• ² Pacific Marine Environmental Laboratory, National Oceanic and Atmospheric Administration (NOAA), Seattle, Washington, United States

Trajectories of >1,600 virtual Argo profiling floats and their sampled variability in key ocean physical and biogeochemical variables are simulated using a 0.125° global ocean physical-biogeochemical model (NOAA GFDL's MOM6-SIS2-COBALTv2) and an offline Lagrangian particle tracking algorithm. Virtual floats are deployed at 92 locations within 26-50°N, 114-132°W in the California Current System (CCS) during the summers and winters of 2008-2012 with varying sampling strategies adopted (e.g., floats are set to park and drift at different depths, and to profile at different intervals). The overall direction and spatial spreads of simulated float trajectories depend on the latitudes of deployment locations with the largest area and variability sampled by floats deployed in the central CCS. Floats drifting at shallower depths (200 m and 500 m) tend to sample larger variability associated with larger sampled area, while those drifting at 1000 m show the strongest association with eddy-like ocean features. Sensitivity experiments with varying sampling intervals suggest that spatiotemporal variability in float observables are adequately sampled with a typical 5-day or 10-day interval. Furthermore, simulated float trajectories and sampled variability are compared against 3 real float trajectories and along-track observations. Results suggest that the fidelity of both our model simulations and the prevalent Argo float sampling design are generally satisfactory in characterizing interior ocean biogeochemical variability. This study provides new insights to inform optimal float deployment planning, sampling strategies, and data interpretation.