Marine Fog Prediction at the Naval Research Laboratory—Factors Affecting the Model Skill

A reduction of visibility by fog over bodies of water and coastal areas remains a formidable challenge for maritime operations. The ability to accurately forecast spatial and temporal characteristics of fog and related visibility is critical. Accuracy in fog prediction is a challenge for current-generation Numerical Weather Prediction (NWP) models in part because of the varied spatial scales and complex interaction of the governing processes in space and time.

Maritime fog is inherently an air-sea interaction problem. The formation, duration, and dissipation of fog, and its spatial extent, are closely linked to the processes at the air-sea interface. The state-of-the-science fully coupled air-sea system (COAMPS®) is a unique tool capable of representing the processes at the air-sea interface. Based on the International Comprehensive Ocean-Atmosphere Data Set (ICOADS) we identified regions with the highest occurrence of the maritime fog. We have performed deterministic hindcasts and evaluated COAMPS regarding the horizontal grid spacing, lead time, visibility calculation method, and the applicability of the method to the environmental conditions conducive to maritime fog formation.

Bimodal distribution in histograms of visibility binned in categories used in ICOADS cannot always be reproduced by COAMPS. We assess the performance of COAMPS and offer both an insight into its shortcomings and a way to improve its skill.