Multi-model Ensemble Analysis with Neural Network Gaussian Processes

Presented by Trevor Harris, Texas A&M University, Department of Statistics

Tuesday, January 16 2024
3:30 PM-4:30 PM ET
AUST 105
Webex Meeting Link
Coffee will be served at 3:00 pm in the Noether Lounge (AUST 326)

Multi-model ensemble analysis integrates information from multiple climate models into a unified projection. However, existing integration approaches based on model averaging can dilute fine-scale spatial information and incur bias from rescaling low-resolution climate models. We propose a statistical approach, called NN-GPR, using Gaussian process regression (GPR) with an infinitely wide deep neural network based covariance function. NN-GPR requires no assumptions about the relationships between climate models, no interpolation to a common grid, and automatically downscales as part of its prediction algorithm. Model experiments show that NN-GPR can be highly skillful at surface temperature and precipitation forecasting by preserving geospatial signals at multiple scales and capturing inter-annual variability. Our projections particularly show improved accuracy and uncertainty quantification skill in regions of high variability, which allows us to cheaply assess tail behavior at a 0.44 degree/50 km spatial resolution without a regional climate model (RCM). Evaluations on reanalysis data and SSP2-4.5 forced climate models show that NN-GPR produces similar, overall climatologies to the model ensemble while better capturing fine scale spatial patterns. Finally, we compare NN-GPR's regional predictions against two RCMs and show that NN-GPR can rival the performance of RCMs using only global model data as input.

Speaker Bio:

I am an Assistant Professor at Texas A&M University in the Department of Statistics. My current research is primarily on the application of deep learning to problems in Climatology and Epidemiology and on developing robust tools for applying deep learning models in scientific contexts. Ongoing work develops analyzes and post-processes climate model output with deep neural networks, tools for granger causality and policy evaluation with differentiable models, forecasts West Nile virus with graph neural networks, and more. Past research includes work in functional data analysis, anomaly detection, change point detection, and robust nonparametric hypothesis testing.