Towards Causal Representations of Climate Model Data

TL;DR

This paper explores the use of causal representation learning, specifically CDSD, to improve the interpretability and robustness of machine learning emulators for climate models, evaluated on key climate datasets.

Contribution

It introduces the CDSD method for causal discovery in climate data and evaluates its potential to enhance emulator interpretability and robustness.

Findings

CDSD shows promise in capturing causal structures in climate data.

Challenges remain in generalizability and accuracy of causal emulators.

The work highlights limitations and future directions for causal climate modeling.

Abstract

Climate models, such as Earth system models (ESMs), are crucial for simulating future climate change based on projected Shared Socioeconomic Pathways (SSP) greenhouse gas emissions scenarios. While ESMs are sophisticated and invaluable, machine learning-based emulators trained on existing simulation data can project additional climate scenarios much faster and are computationally efficient. However, they often lack generalizability and interpretability. This work delves into the potential of causal representation learning, specifically the \emph{Causal Discovery with Single-parent Decoding} (CDSD) method, which could render climate model emulation efficient \textit{and} interpretable. We evaluate CDSD on multiple climate datasets, focusing on emissions, temperature, and precipitation. Our findings shed light on the challenges, limitations, and promise of using CDSD as a stepping stone towards more interpretable and robust climate model emulation.