Global ocean dynamics involve the large-scale flow and circulation of the world’s oceans, which are essential for regulating heat distribution, sequestering carbon, recycling nutrients, and influencing global weather patterns. Of particular interest in this review paper is the interaction between counterflowing currents, which frequently occur in estuaries and deep-sea environments, and have significant implications for oceanic and estuarine dynamics. When counterflowing currents, including but not limited to gravity currents, develop within the same region and encounter each other, these interactions lead to complex flow dynamics characterized by fronts, eddies, and intense turbulence, which in turn initiate vertical nutrient transport and mixing processes. In addition to affecting both regional and global thermohaline circulation, these interactions play an important role in turbulent mixing, sediment dynamics, and biological productivity. By synthesizing existing studies on counterflowing currents interactions, this paper explains the complex processes that control these phenomena and their significance in oceanic and estuarine environments. We highlight relevant gaps in existing knowledge and suggest future research approaches to advance the understanding of these important oceanic processes.

Coastal flood models for storm surge often face considerable computational challenges due to the large number of computational cells in integrated sea–land scenarios and the need for extremely small time steps in certain regions, which are typically constrained by the globally established minimum time step. Aiming to address these limitations, this study presents an efficient integrated sea–land flood model for storm surge disaster prediction in coastal urban areas with dense buildings. The model uses a graphics processing unit (GPU)-accelerated framework to handle large computational grids and incorporates a local time step (LTS) approach to mitigate restrictions from locally refined grids, extremely small time steps, and flow condition disparities between sea and land. A GPU-optimized parallel algorithm enhances computational performance by refining the numerical framework, optimizing kernel functions, and improving memory utilization, demonstrating a seamless integration with the LTS approach. The efficiency of the model is demonstrated through storm surge and flood simulations in Macau, China, covering the sea, straits, and densely built coastal land areas. Results show that LTS scheme reduces computation time by approximately 40 times, markedly improving computational efficiency across different mess configurations. Owing to GPU and LTS acceleration, the model provides a powerful tool for real-time coastal flood forecasting.