Underwater wireless optical communication (UWOC) enables high-speed, low-latency links for the Internet of Underwater Things (IoUT), but remains highly vulnerable to complex and dynamic channel impairments. In practice, slant transmission paths frequently occur due to spatial offsets between underwater sensors and marine surface vehicles (MSVs), where link quality is jointly affected by MSV-induced misalignment, depth-dependent attenuation, and oceanic turbulence. Most existing models address these factors independently or under static assumptions, limiting their applicability. To address this challenge, this paper proposes a novel modeling and simulation framework that jointly captures the coupled effects of dynamic posture, stratified absorption and scattering, and spatially varying turbulence along slant UWOC paths. A parallelized wave optics simulation (WOS) platform is developed using Argo float data to reflect realistic ocean conditions. Leveraging this dataset, a CNN-based Fast Channel Simulation Method (CNN-FCSM) is designed, achieving over 98% prediction accuracy of received optical intensity, while reducing runtime by more than 90% compared to WOS. Comprehensive evaluations demonstrate that the proposed approach enables accurate and scalable estimation of average bit error rate (ABER) and outage probability (OP) without relying on predefined statistical fading models, offering a robust tool for dynamic UWOC analysis and IoUT system design in complex underwater environments.

In this paper, we perform the statistical modeling for turbulence-induced intensity fluctuations with optical waves propagating in turbulent ocean vertically. To accurately model the turbulence-induced optical wavefront distortion in vertical direction, we construct the wave optics simulation (WOS). In WOS procedure, we use the general oceanic turbulence spatial power-law spectrum and Argo experimental data to produce depth-dependent random phase screens. We specify the number and interval of phase screens based on Fourier optics theory so that we can determine the placement depth of phase screens reasonably. By fitting the WOS statistical results under various typical sea areas and system configurations, we propose the mixture Weibull-generalized Gamma (WGG) distribution to model oceanic turbulence-induced intensity fading. We then develop the analytical expression of the outage probability (OP) and average bit-error rate (BER) for vertical underwater wireless optical communication (UWOC) systems by use of the WGG distribution. The analytical results demonstrate the non-negligible effects of water depths on the performance of vertical UWOC systems.