Presentation 2-5, 15:25~16:05.
Speaker
Won Chang (Seoul National University)
Title
A Likelihood-Free Approach to Ice Model Emulation and Calibration Using Deep Diffusion Models
Abstract
Rapid changes in the cryosphere can affect climate change, such as global sea-level rise. Computer models are useful for understanding the behavior of Antarctic ice sheets and can be used to study their impact on rising sea levels. However, uncertainty quantification of model parameters is challenging because the model outputs and observational data are high-dimensional and spatially correlated. Furthermore, they are semicontinuous with an excess of zeros. To address these challenges, we propose a diffusion model-based emulator that can accurately generate the pseudodata across various parameter settings. Since the resulting likelihood function from the emulator is intractable, we propose an approximate Bayesian computation method with a Siamese network. The Siamese network is trained to determine whether images generated by the emulator with proposed parameters closely resemble observational data based on the similarity of their image features. We apply our method to calibrate the computer model for the West Antarctic Ice sheet data, where the current approaches are infeasible due to the aforementioned challenges. The proposed approach can be used to generate future projections of sea level rise based on modern ice sheet observations through the calibration of the PSU-3D ice model.