Score-based generative emulation of impact-relevant Earth system model outputs

2025-10-05

AI Summary

Climate policy evolves faster than the release cycle of Coupled Model Intercomparison Project (CMIP) simulations, leading impact assessments to rely on outdated climate projections. To address this, we propose a spherical diffusion generative model tailored for rapid scenario analysis. First, we adapt the score-based diffusion framework to spherical grids, enabling joint probabilistic modeling of multiple climate variables. Second, we design a lightweight spherical neural network architecture that enables efficient training and sampling on a single mid-tier GPU. Third, we establish a comprehensive diagnostic suite evaluating probabilistic calibration, inter-variable dependencies, and extreme-event representation. Validation across multiple Earth System Models (ESMs) and emission scenarios demonstrates that generated climate fields match original ESM outputs in statistical fidelity and physical response consistency, with errors below internal variability levels. The approach significantly reduces computational cost and latency for high-resolution climate field generation.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Generation
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Web Mining and Content Analysis: Web data generation and simulation
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Policy targets evolve faster than the Couple Model Intercomparison Project cycles, complicating adaptation and mitigation planning that must often contend with outdated projections. Climate model output emulators address this gap by offering inexpensive surrogates that can rapidly explore alternative futures while staying close to Earth System Model (ESM) behavior. We focus on emulators designed to provide inputs to impact models. Using monthly ESM fields of near-surface temperature, precipitation, relative humidity, and wind speed, we show that deep generative models have the potential to model jointly the distribution of variables relevant for impacts. The specific model we propose uses score-based diffusion on a spherical mesh and runs on a single mid-range graphical processing unit. We introduce a thorough suite of diagnostics to compare emulator outputs with their parent ESMs, including their probability densities, cross-variable correlations, time of emergence, or tail behavior. We evaluate performance across three distinct ESMs in both pre-industrial and forced regimes. The results show that the emulator produces distributions that closely match the ESM outputs and captures key forced responses. They also reveal important failure cases, notably for variables with a strong regime shift in the seasonal cycle. Although not a perfect match to the ESM, the inaccuracies of the emulator are small relative to the scale of internal variability in ESM projections. We therefore argue that it shows potential to be useful in supporting impact assessment. We discuss priorities for future development toward daily resolution, finer spatial scales, and bias-aware training. Code is made available at https://github.com/shahineb/climemu.

Problem

• Emulating Earth system model outputs for rapid climate projections
• Providing impact-relevant climate variables through generative modeling
• Addressing outdated climate projections with efficient emulator diagnostics

Innovation

• Score-based diffusion model on spherical mesh
• Deep generative modeling of climate variable distributions
• Single GPU emulation for Earth system model outputs