Multifidelity Computer Model Emulation with High-Dimensional Output: An Application to Storm Surge.

Hurricane-driven storm surgeis one of the most deadly and costly natural disasters, making precise quantification of the surgehazard of great importance. Inference of such systems is done through physics-based computer modelsof the process. Such surgesimulators can be implemented with a wide range of fidelitylevels, with computational burdens varying by several orders of magnitude due to the nature of the system. The danger posed by surgemakes greater fidelityhighly desirable, however such models and their high-volume output tend to come at great computational cost, which can make detailed study of coastal floodhazards prohibitive. These needs make the development of an emulator combining high-dimensional output from multiple complex computer modelswith different fidelitylevels important. We propose a parallel partial autoregressive cokriging model to predict highly-accurate storm surgesin a computationally efficient way over a large spatial domain. This emulator has the capability of predicting storm surgesas accurately as a high- fidelity computer modelgiven any stormcharacteristics and allows accurate assessment of the hazards from storm surgesover a large spatial domain. R codes are provided to show the numerical results through an R package called ARCokrig.