Comparative analysis of the performance of the GOFS, PSY4 and AMSEAS ocean model frameworks in the Virgin Islands and Puerto Rico coastal ocean

Abstract

ABSTRACT This work compares the performance of three ocean model frameworks that currently produce outputs of the ocean properties specific to the US Caribbean ocean; the Global Ocean Forecast System (GOFS), US Navy Coastal Ocean Model for the American Seas (AMSEAS) and the Daily Global Physical Bulletin (PSY4). Separate comparisons are done for the ocean properties in the open ocean and nearshore regions. For the open ocean, the model outputs are compared with the AVISO satellite altimetry data for the sea-surface height anomaly (SSHA), the OSCAR data for surface current velocities and the G1SST satellite data for sea-surface temperature (SST). For the nearshore analysis, the model outputs are compared with in-situ buoy measurements and HOBO logger data in the nearshore regions. Our analysis shows that the PSY4 produces the most realistic outputs of SSHA and surface current velocities in the open ocean, whereas all the models produce a strong correlation in terms of the seasonal variability of the surface temperature when compared to the G1SST data. The AMSEAS model, despite being a fine resolution regional model, underperforms in terms of the surface current velocity outputs in the open ocean due to the influence of the simulated submesoscale turbulence on the mesoscale variability. In the nearshore regions, none of the models produce agreeable outputs on the SSHA and current velocities. These findings provide useful insight on the applicability of the model outputs for various operations that require oceanographic data specific to the US Caribbean ocean.