The Role of Deep Mesoscale Eddies in Ensemble Forecast Performance

Abstract

Present forecasting efforts rely on assimilation techniques that adjust the model basic state, meaning that profiles of temperature and salinity are used as measured or converted to temperature and salinity through statistical relationships. This information influences the upper ocean ( $< 1000$ m depth), while minimally influencing the deep ocean. Nevertheless, development of the full water column circulation critically depends upon the dynamical interactions between upper and deep fields. A review of ensemble forecasts in the Gulf of Mexico demonstrates the importance of the initial deep ocean features in the evolution of the surface field. Initial conditions throughout the full water column that agree with observations are needed to improve the forecast predictions. Here, best and worst ensemble members in two 92-day forecasts are identified and contrasted in order to determine how the deep ocean features differ between these groups. The forecasts cover the duration of the Loop Current Eddy Thor separation event, which coincides with available deep observations. Model member performance is assessed with a newly developed ranking method, demonstrated with surface variables against verifying analysis and satellite altimeter data during the forecast time-period. Deep cyclonic and anticyclonic features are reviewed, and compared against deep observations, indicating subtle differences in locations of deep eddies at relevant times. These results highlight both the importance of deep circulation dynamics of the Loop Current system and more broadly motivate efforts to assimilate deep observations to better constrain the deep initial fields and improve surface and sub-surface predictions.