The complex morphology and optical anisotropy of anisotropic sea-fog particles make it difficult to accurately describe the multiple scattering behavior of polarized light. To overcome the insufficient coupling between polarization evolution and scattering direction in conventional Monte Carlo models, as well as their limited representation of particle parameters, this study proposes a polarization transmission model combining an improved Discrete Dipole Approximation with a direction-adaptive Monte Carlo method. In this model, the probability density function of scattering direction is constructed as a functional of the photon Stokes vector, enabling tight coupling between polarization state updates and photon propagation. Particle structures are refined by incorporating size, aspect ratio, Euler angles, morphological perturbations, and complex refractive indices. Meanwhile, an equivalent scattering kernel containing higher-order statistical moments is introduced to balance computational efficiency and physical fidelity. An indoor sea-fog polarization transmission platform was established to measure six incident polarization states at 450, 532, 671, and 808 nm. The results show over 80% agreement with model predictions and a root mean square error below 0.1. The study further indicates that circular polarization retains polarization better than linear polarization under high optical thickness, while longer wavelengths provide more stable polarization transmission. This framework offers theoretical support for polarization imaging and optical communication in sea-fog environments.
Zhang et al. (Mon,) studied this question.