Comparing high-fidelity LES of early contrail formation with ground-based images

Contrails have been identified as a principal contributor to aviation’s impact on the atmospheric radiative balance. However, considerable uncertainties remain regarding their effects, highlighting the need for deeper scientific understanding. Consequently, the development of reliable and validated contrail models is critical for accurate prediction and effective mitigation of their climate impact.

This study leverages ground-based camera imagery to identify and track early-stage contrails, providing benchmark data for a high-fidelity solver that simulates initial contrail formation. The solver employs a three-dimensional large-eddy simulation (LES) framework, incorporating an Eulerian–Lagrangian approach for two-phase compressible flow, enabling the simulation of the contrail's jet and vortex regimes over short timescales (<2 minutes). Images from Reuniwatt all-sky cameras capture a 5600 km² region at 30-second intervals, enabling observation of the vortex and initial dissipation regime.

Contrails are manually identified and labeled in the images, while flight trajectories are overlaid using ADS-B information. Atmospheric conditions are derived from reanalysis datasets at flight altitude, and aircraft/engine parameters are estimated from publicly available sources. These inputs are integrated into the simulations to replicate real-world conditions for each case.

By comparing the simulation results with observational data, this study aims to evaluate the reliability and applicability of the model, as well as to shed light on contrail formation mechanisms. Particular attention is given to scenarios where other theoretical approaches and lower-fidelity models have historically been less accurate.