A Numerical Model Study of Coastal Submesoscale Dynamics Observed with Unmanned Aerial Vehicle

-
Mixed Online/In-Person

Sin-Young Kim
Ph.D. Candidate
Department of Oceanography
Pusan National University

Friday, September 12, 2025, 3:10pm
Chase 105
 

Abstract

Submesoscale oceanic motions with spatial scales of 0.1–10 km are prevalent throughout the upper ocean, bridging the gap between mesoscale dynamics and small-scale turbulence. A complex Rias coast characterizes the southern coast of Korea with a tide-dominated flow environment. Vivid pictures of turbulence unveiling the complexity of coastal dynamics at multiple scales are painted by a topographic interaction. However, observing the dynamical characteristics and evolution of these submesoscale processes has remained challenging due to their fine spatial and temporal variability. In this study, we combined high-resolution Unmanned Aerial Vehicle (UAV)-based observations with numerical simulations to investigate the structure, dynamics, and generation mechanisms of submesoscale features in  the study region, leveraging the complementary strengths of both approaches. First, we introduce a framework for decomposing and estimating submesoscale surface wave and current from UAV imagery using the Multidimensional Empirical Mode Decomposition, 2D Fourier Transform, and the Optical Flow method. Second, we reproduced the hydrodynamics of observed submesoscale processes using the Delft3D-FLOW and Regional Ocean Modeling System (ROMS) and diagnosed the dynamical properties of turbulent flow in terms of eddy interactions. 

Categories