Lagrangian Coherent Structures in High-Speed Crosflow Jets via DNS

Lagrangian Coherent Structures (LCS) are employed to interrogate transport barriers and the coherent organization of vortical interfaces in high-speed jet-in-crossflow (JICF) environments. This study utilizes high-fidelity Direct Numerical Simulations (DNS) of spatially-developing turbulent boundary layers subjected to wall-normal jet injection at sonic and supersonic conditions, with a freestream Mach number of 0.8 and jet-to-crossflow momentum flux ratios ranging from 2.8 to 10.2. A massively parallel GPU-accelerated in-house particle advection framework, Aquila-LCS, has been developed to advect on the order of billions of passive tracers through temporally resolved, 3D DNS velocity fields. LCS are extracted via Finite-Time Lyapunov Exponents (FTLE) and Finite-Size Lyapunov Exponents (FSLE), enabling the identification of attracting and repelling material manifolds that delineate jet penetration boundaries, turbulent entrainment layers, and recirculation zones dominated by intense shear and vortical coherence. A comprehensive scalability analysis of the Aquila-LCS framework on modern high-performance computing (HPC) architectures—scaling efficiently up to 8 NVIDIA H200 GPUs—demonstrates near-ideal parallel efficiency. The Lagrangian framework provides an objective, frame-invariant description of flow organization, enabling deeper insight into coherent transport phenomena in complex, multiscale, compressible jet configurations.