Seasonal circulation fields in the northern Gulf of Mexico calculated by assimilating current meter, shipboard ADCP, and drifter data simultaneously with the shallow water equations

Velocity measurements from current meter moorings, shipboard acoustic Doppler current profilers (ADCPs), and satellite-tracked drifters are used to estimate the seasonally averaged barotropic circulation throughout the continental shelf in the northern Gulf of Mexico in two regions: the Northeast Gulf of Mexico (NEGOM) shelf and the Louisiana-Texas (LATEX) shelf. A variational assimilation approach (weighted least squares) is used to combine the data simultaneously with a system of dynamical equations (time-independent shallow water equations), where the weights are based on the expected errors in the dynamical equations and measurements. Smoothing and boundary constraints are also included in this system to impose a spatial decorrelation scale and to restrict the component of flow that is perpendicular to the coastline, respectively. Seasonal solutions are therefore best fits to the weighted measurements, dynamics, and constraints, and thus do not fit any component exactly. Analysis of the solution deviations to the dynamical equations reveal where and during which seasons the dynamics and data are in disagreement. The largest deviations from the proposed dynamics occur at the head of the DeSoto Canyon on the NEGOM shelf and at the southwest corner of the LATEX shelf. Therefore, in these areas, there are processes occurring that are not described by the simplified barotropic dynamics. At the head of the DeSoto Canyon, the large source of error is believed to be due to deep-ocean water penetrating across the shelfbreak, whereas the process that occurs at the southwest corner of the LATEX shelf is proposed to be a result of a convergence zone caused by loop current eddies. Error analysis reveals that current meter data are more capable of constraining the solution than the other two data types due to its ability to more accurately observe barotropic currents. Separate experiments are conducted using the shipboard ADCP and drifter data sets individually to determine the relative observation capability. These experiments reveal that both data types provide reasonable estimations of seasonal flow fields. Results on the LATEX shelf suggest that shipboard ADCP data is more capable of observing the seasonal barotropic flow than drifter data. Whereas, on the NEGOM shelf results imply that drifter data are more capable. This result, however, may be skewed due to the shipboard ADCP and drifter data sets being collected during different years.