Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

PACER: Physics Informed and Uncertainty Aware Climate Emulator

Hira Saleem, Flora Salim, Cormac Purcell

TL;DR

An auto-regressive ODE-SDE framework for climate emulation that integrates the fundamental physical law of advection, while being trained under a negative log-likelihood objective to enable principled uncertainty quantification of stochastic variability is proposed.

Abstract

Physics based numerical climate models serve as critical tools for evaluating the effects of climate change and projecting future climate scenarios. However, the reliance on numerical simulations of physical equations renders them computationally intensive and inefficient. While deep learning methodologies have made significant progress in weather forecasting, they are still unstable for longer roll-out climate emulation task. Here, we propose PACER, a relatively lightweight 2.1M parameter Physics Informed Uncertainty Aware Climate EmulatoR. PACER is trained across is trained across varying spatial resolutions and physics based climate models, enabling faithful and stable emulation of temperature fields at multiple surface levels over a 10 year horizon. We propose an auto-regressive ODE-SDE framework for climate emulation that integrates the fundamental physical law of advection, while being trained under a negative log-likelihood objective to enable principled uncertainty quantification of stochastic variability. We show PACER's emulation performance across 20 climate models outperforming relevant baselines and advancing towards explicit physics infusion in ML emulator.

PACER: Physics Informed and Uncertainty Aware Climate Emulator

TL;DR

An auto-regressive ODE-SDE framework for climate emulation that integrates the fundamental physical law of advection, while being trained under a negative log-likelihood objective to enable principled uncertainty quantification of stochastic variability is proposed.

Abstract

Physics based numerical climate models serve as critical tools for evaluating the effects of climate change and projecting future climate scenarios. However, the reliance on numerical simulations of physical equations renders them computationally intensive and inefficient. While deep learning methodologies have made significant progress in weather forecasting, they are still unstable for longer roll-out climate emulation task. Here, we propose PACER, a relatively lightweight 2.1M parameter Physics Informed Uncertainty Aware Climate EmulatoR. PACER is trained across is trained across varying spatial resolutions and physics based climate models, enabling faithful and stable emulation of temperature fields at multiple surface levels over a 10 year horizon. We propose an auto-regressive ODE-SDE framework for climate emulation that integrates the fundamental physical law of advection, while being trained under a negative log-likelihood objective to enable principled uncertainty quantification of stochastic variability. We show PACER's emulation performance across 20 climate models outperforming relevant baselines and advancing towards explicit physics infusion in ML emulator.

Paper Structure

This paper contains 32 sections, 23 equations, 4 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Machine Learning (ML) and Physics based Climate Emulators:
  4. Background
  5. Neural SDE:
  6. Methodology
  7. Problem Statement
  8. Advection Process
  9. Velocity Inference
  10. Velocity initialization from past states.
  11. Velocity refinement:
  12. Neural SDE for Trajectory Refinement
  13. Ensemble refinement and uncertainty.
  14. Uncertainty Quantification
  15. Experiments and Results
  16. ...and 17 more sections

Figures (4)

  • Figure 1: Complete architectural pipeline of PACER.
  • Figure 2: Time–mean maps for air temperature fields across vertical levels (2,3,4,5). Both the quantitative results in table \ref{['Fv3']} and visual results indicate stronger fidelity at higher levels (L4–L5) than at lower tropospheric levels (L2–L3), with reduced discretizations.
  • Figure 3: Physics vs. no-physics across four CMIP6 climate models
  • Figure 4: Numerical discretization effect on emulation of Temperature.