|
Abstract:
|
Earth System Models (ESMs) are fundamental tools for simulating and forecasting climate variability. Different modeling centers develop different ESMs, which leads to inter-model variability and uncertainty about the future climate. Current Bayesian approaches to integrate multi-model ESM projections into a single forecast can suffer from high computational costs and limited flexibility due to strong parametric and linearity constraints. We propose an empirical Bayesian model called the Variational Target Encoder (VTE) to integrate multi-model ESM projections into a single forecast of the target climate process. The VTE parameterizes its approximate posterior and likelihood with two neural networks that respectively encode ESMs into a latent distribution and decode the latent distribution into the target forecast distribution. We show that the VTE is computationally efficient, highly flexible, and automatically regularized. On both precipitation and temperature field experiments, the VTE accurately recovered the target climate process and generated posterior credible intervals with the prescribed coverage.
|
