Detection of Forced Change Within Combined Climate Fields Using Explainable Neural Networks.

Assessing forced climate change requires the extraction of the forced signal from the background of climate noise. Traditionally, tools for extracting forced climate change signals have focused on one atmospheric variable at a time, however, using multiple variables can reduce noise and allow for easier detection of the forced response. Following previous work, we train artificial neural networks to predict the year of single‐ and multi‐variable maps from forced climate model simulations. To perform this task, the neural networks learn patterns that allow them to discriminate between maps from different years—that is, the neural networks learn the patterns of the forced signal amidst the shroud of internal variability and climate model disagreement. When presented with combined input fields (multiple seasons, variables, or both), the neural networks are able to detect the signal of forced change earlier than when given single fields alone by utilizing complex, nonlinear relationships between multiple variables and seasons. We use layer‐wise relevance propagation, a neural network explainability tool, to identify the multivariate patterns learned by the neural networks that serve as reliable indicators of the forced response. These "indicator patterns" vary in time and between climate models, providing a template for investigating inter‐model differences in the time evolution of the forced response. This work demonstrates how neural networks and their explainability tools can be harnessed to identify patterns of the forced signal within combined fields. Plain Language Summary: Using machine learning tools called neural networks, we identify patterns of the changing climate within climate model data. Changes in the climate can be identified earlier when detecting patterns within maps of multiple variables and seasons than for single maps alone. By visualizing the patterns learned by the neural networks, we can identify which regions, variables, and seasons are most important for detecting climate change. These patterns offer insight into how climate change is represented in different climate models, and how the patterns of climate change will evolve over time. Key Points: Neural networks and their explainability tools can be harnessed to identify patterns of forced change within combined fieldsCombined fields of input allow for earlier detection of the emergence of a forced climate responseExplainable AI techniques can be used to identify patterns that describe the emergence and evolution of forced climate change [ABSTRACT FROM AUTHOR]