APMLFeb 8, 2022

Multi-model Ensemble Analysis with Neural Network Gaussian Processes

arXiv:2202.04152v48 citations
Originality Highly original
AI Analysis

This addresses the challenge of preserving fine-scale spatial information in climate projections for researchers and policymakers, offering a novel statistical approach that rivals regional models.

The paper tackled the problem of integrating multiple climate models for projections by proposing NN-GPR, a Gaussian process regression method with a neural network covariance function, which improved accuracy in surface temperature and precipitation forecasting, particularly in high-variability regions at a 0.44°/50 km resolution without needing regional climate models.

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 models, no interpolation to a common grid, no stationarity assumptions, 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$^\circ$/50 km spatial resolution without a regional climate model (RCM). Evaluations on reanalysis data and SSP245 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.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes