Lagrangian Fluid Transport Simulation Using Mobile Devices

Simulating the transport of plastics, nutrients and pollutants in the ocean is important to address societal questions in environmental management and sustainability. Many Lagrangian transport simulators have been introduced in the past decades, yet they all share two distinct limitations, namely, long compute times and reliance on large compute equipment. These limitations hinder a transient incorporation of outdoor observations for Lagrangian physical simulations. This paper introduces a novel particle advection approach for Lagrangian ocean transport simulations, specifically designed for mobile devices and field use. A proof-of-concept for particle advection via 4th order Runge-Kutta time integration is presented and validated across different datasets. The approach is parallelized for SIMD architectures on mobile platforms. The results demonstrate a time-integration of 500,000 particles in approximately 10.2 ms per timestep, enabling an interactive co-visualization of the simulation. Achieved runtimes on mobile devices are within the same order of magnitude as established non-portable Lagrangian ocean-transport simulators, such as TRACMASS and OceanParcels, with comparable scalability. Consequently, this novel simulation approach opens new possibilities for field-conducted simulations in the future.