In this article, a computationally efficient pseudo-2D model for real-time dynamic simulations of polymer electrolyte membrane fuel cells (PEMFCs) is developed with a specific focus on water and thermal management. The model accounts for temperature dynamics, two-phase flow and flooding in the diffusion media, and membrane water crossover as well as absorption and desorption processes. Computational efficiency is achieved by leveraging the disparate time scales within the system dynamics, in addition to exploiting the large aspect ratio of the cell layers to create a spatio-temporal decoupling. Taking advantage of such decoupling, the model yields a computationally efficient solution while providing detailed information about the state of water and temperature throughout the cell. Through this approach, the current implementation of the model is found to be about twice faster than real time. Moreover, a case study is carried out where different mechanisms contributing to overall water balance in the cell are investigated. The results are shown to be in qualitative agreement with published experimental data, thereby providing a preliminary validation of the modeling approach. Finally, using the modeling results, an equivalent electrical circuit model is proposed to help elucidate water transport inside various cell layers. Real-time estimation, prediction, and control of cell hydration and temperature distribution is essential for optimizing the performance of polymer electrolyte membrane fuel cells (PEMFCs), as well as avoiding critical conditions and mitigating cell degradation. These applications necessitate mathematical models that not only run in real time, but also incorporate the important physical phenomena related to water transport and thermal management. However, including such phenomena comes at the cost of higher computational requirements, resulting in a trade-off between model accuracy and computational speed, which must be carefully balanced based on the desired application. As a result of these competing requirements, developing mathematical models that achieve a balance between the needs for high fidelity and low computational demand remains an active area of research.Within the context of this paper, a fuel cell model is considered to have high fidelity if it incorporates the following phenomena: i) 3-D effects including anisotropic material properties 1-3 to resolve transport phenomena in all physical directions; ii) transient behavior; iii) detailed multistep hydrogen oxidation reaction (HOR) and oxygen reduction reaction (ORR) kinetics; 4,5 iv) multiphase flow in gas channels and porous media; v) non-isothermal effects; 6 and vi) multicomponent diffusion.7,8 A more detailed explanation of these considerations follows.In terms of dimensionality, 3-D models are of highest fidelity, because they are capable of capturing transport in both through-themembrane and along-the-channel directions and also account for the channel-land effects in the third dimension. Moreover, these models can easily in...