Multifidelity Computer Model Emulation with High-Dimensional Output.

Hurricane-driven storm surgeis one of the most deadly and costly natural disasters, making precise quantification of the surgehazard of great importance. Physics-based computermodels of storm surgecan be implemented with a wide range of fidelitydue to the nature of the system, though the danger posed by surgemakes greater fidelityhighly desirable. However, such models and their high-dimensional outputs tend to come at great computational cost, which can make highly detailed studies prohibitive. These needs make the development of an emulator combining high-dimensional output from multiple complex computer models with different fidelitylevels important. We propose a parallel partial autoregressive cokriging model that is able to address these issues. Based upon the data-augmentation technique, model parameters are estimated via Monte Carlo expectation-maximization algorithmand prediction is made in a computationally efficient way when input designs across different fidelitylevels are not nested. This methodology is applied to emulate high- fidelity storm surgesin Southwestern Florida from two different surgemodels. R codes are provided to show the numerical demonstration through an R package called ARCokrig.