Estimating Numerical Model Skill and Representativeness of Kinematic Properties of the Loop Current System in the Gulf of Mexico

Abstract Ocean circulation eddy-resolving numerical models are crucial tools widely used for understanding the dynamics of the ocean and forecasting its state. As part of the UGOS program, several such state-of-the art models are focused on simulating the Loop Current System circulation in the Gulf of Mexico (GOM). Thus, there is a need to evaluate their performance in the context of the Oil & Gas (O&G) Industry operational requirements. For the GOM, models often fail to accurately represent the Loop Current (LC) and Loop Current Eddies (LCEs) in terms of feature location and velocity distribution. Our approach is to separate the assessment of model skill in terms of the position of current fronts and the associated current magnitude. To quantitatively evaluate model ability to resolve the position of LC/LCE fronts at the water surface, a set of distance metrics were tested and created. These metrics include the Hausdorff Distance, the Site Eddy Impact Index, and the Feature Similarity Index, all of which are ways to define the difference between the observation-based 1.5-knot Frontal Analysis Isotach and equivalent 1.5-knot isotach position from model runs. As a first step in assessing model skill in representing the magnitude and shape of the current field at depth, a comprehensive dataset of Vessel Mounted Acoustic Doppler Current Profiler records is used and a characteristic shape of the cross section of the leading front of LCEs is defined. Another model performance assessment framework used is based on a proprietary dataset of Lagrangian drifters (Far Horizon Drifters) spanning from 2003 to 2024. The model current velocity associated with the trajectory of each drifter is directly compared to drifter velocities and other methods of comparison are explored. This paper describes the comparison methodologies that have been developed to date, which are intended to be used later in the UGOS program to assess model forecast skill consistent with O&G Industry requirements.