Three-dimensional characterization and dispersion parameter estimation of SO 2 / NO x plumes from port ship emissions based on mobile monitoring and inversion

Abstract Ship-based air pollution in port regions poses a significant threat to urban air quality. We aim to analyze three-dimensional structure and assess diffusion characteristics of SO 2 / NO x ship emissions plumes in port environments, and an improved multi-stage high-accuracy estimation method based on mobile monitoring and inversion modelling (M 3 DI) is proposed. The proposed model expands the traditional Gaussian plume model by incorporating a mesoscale WRF meteorological data-driven module with dynamic assimilation of the wind field and a plume uplift correction method that equilibrates buoyant and momentum flux. It also constructs a three-dimensional inversion solution layer with Tikhonov regularization, utilizing spatiotemporal Kriging interpolation and Bayesian MCMC parameter posterior estimation methods to enable accurate estimation of the plume spatial structure, source strength, and diffusion coefficients while reducing uncertainty based on mobile monitoring paths that are ineffective. Empirical evidence shows that the model has provided 25-35% higher accuracy in estimating the volume of plumes as compared to traditional models in shipping emission datasets, with notable strengths with regard to the stability and credibility of estimating the horizontal diffusion coefficient.