Ocean-Atmosphere interactions in Northwest African and Gulf of Guinea Coastal Upwelling Systems

Abstract

The Ocean and Atmosphere interaction is a complex system that requires particular attention in coastal upwelling systems due to their nutrient supply, biological productivity, influence on local weather patterns and regional climate variability, and socio-economic importance. The present study investigates the behaviour of ocean, atmosphere, and their interactions in the coastal upwelling systems of Northwest Africa and Gulf of Guinea during ocean cooling and warming episodes, with a focus on the dynamical and interannual variability of Senegal-Mauritania and Gulf of Guinea upwelling regions. Multiple datasets including different physical and biogeochemical variables, several upwelling indices and statistical tools were used. Results highlighted the seasonal dynamic and interannual variability of two upwelling systems modulated by both local and remote forcings such as wind stress-driven Ekman dynamics, mesoscale and cyclonic eddies, regional ocean ocean currents, Equatorial dynamics including Kelvin and Rossby waves and coastal trapped waves, and large-scale climate drivers. The results also underlined the influence of other geographical and environmental factors such as coastal shape and orientation, the geographical position of upwelling system, the motion of Intertropical convergence zone, and Azores and Saint Helena high pression systems that modulated pression and temperature gradients. The study identified marine heat waves, which are characterized by weakened Ekman transport and suppressed vertical mixing, as emerging stressors of coastal upwelling dynamics. These events significantly impact coastal marine ecosystem productivity, surface air temperature, and precipitation, underscoring the importance of addressing them in local weather predictions. The findings also demonstrate the ongoing limitations of the latest generations of the Coupled Model Intercomparison Project Phases 5 and 6 (CMIP5 and CMIP6) in capturing coastal processes and upwelling features, and call for the use of more powerful artificial intelligence predictive models, such as long short-term memory and convolutional neural networks (LSTM-CNN), to improve weather predictions and reduce uncertainties in climate model simulations in coastal upwelling regions.