Turbidity currents are density-driven flows transporting particles within fluids of small density contrasts, playing a critical role in sediment deposition on seabeds. Understanding these flows is crucial for strategic geological insights, particularly in oil exploration. In this study, we simulate turbidity currents using a stabilized finite element formulation in an Eulerian-Eulerian framework, which provides high-fidelity predictions at significant computational costs. To address these costs, we employ a surrogate modeling approach that integrates Proper Orthogonal Decomposition (POD) and Autoencoders [1,2] for dimensionality reduction. This surrogate model, trained on high-fidelity simulation data, enables efficient predictions of flow dynamics across unseen parametric configurations. Building on recent advances in generative AI, we leverage Denoising Diffusion Probabilistic Models (DDPMs) [3], a state-of-the-art approach in super-resolution tasks, for further enhancement of surrogate predictions. Unlike traditional super-resolution methods, DDPMs require only high-resolution data during training, eliminating the need for paired low- and high-resolution datasets. Inspired by their success in flow field reconstruction, we utilize DDPMs to refine the surrogate's predictions. The proposed workflow combines high-fidelity parametric simulation data with diffusion-based [4] super-resolution techniques to improve the accuracy of surrogate predictions in unexplored regions of the parameter space. This hybrid approach demonstrates the potential of integrating advanced generative AI techniques into surrogate modeling, offering a novel path forward for computational fluid dynamics and geoscientific applications. [1] - M. Cracco et al., “Deep learning-based reduced-order methods for fast transient dynamics”, Arxiv Preprint 2212.07737, 2022. [2] - S. Fresca and A. Manzoni, “POD-DL-ROM: Enhancing deep learning-based reduced order models for nonlinear parametrized PDEs by proper orthogonal decomposition”, Comput. Methods Appl. Mech. Engrg., 2022. [3] - J. Ho, A. Jain and P. Abbeel, “Denoising Diffusion Probabilistic Models”, arXiv: 2006.11239, 2020. [4] - D. Shu, Z. Li, A. Barati Farimani, “A physics-informed diffusion model for high-fidelity flow field reconstruction”, Journal of Computational Physics, Volume 478, 2023.