An official website of the United States government
Abstract Many problems in climate science require the identification of signals obscured by both the “noise” of internal climate variability and differences across models. Following previous work, we train an artificial neural network (ANN) to predict the year of a given map of annual‐mean temperature (or precipitation) from forced climate model simulations. This prediction task requires the ANN to learn forced patterns of change amidst a background of climate noise and model differences. We then apply a neural network visualization technique (layerwise relevance propagation) to visualize the spatial patterns that lead the ANN to successfully predict the year. These spatial patterns thus serve as “reliable indicators” of the forced change. The architecture of the ANN is chosen such that these indicators vary in time, thus capturing the evolving nature of regional signals of change. Results are compared to those of more standard approaches like signal‐to‐noise ratios and multilinear regression in order to gain intuition about the reliable indicators identified by the ANN. We then apply an additional visualization tool (backward optimization) to highlight where disagreements in simulated and observed patterns of change are most important for the prediction of the year. This work demonstrates that ANNs and their visualization tools make a powerful pair for extracting climate patterns of forced change.
more »
« less
Viewing Forced Climate Patterns Through an AI Lens
Abstract Many problems in climate science require extracting forced signals from a background of internal climate variability. We demonstrate that artificial neural networks (ANNs) are a useful addition to the climate science “toolbox” for this purpose. Specifically, forced patterns are detected by an ANN trained on climate model simulations under historical and future climate scenarios. By identifying spatial patterns that serve as indicators of change in surface temperature and precipitation, the ANN can determine the approximate year from which the simulations came without first explicitly separating the forced signal from the noise of both internal climate variability and model uncertainty. Thus, the ANN indicator patterns are complex, nonlinear combinations of signal and noise and are identified from the 1960s onward in simulated and observed surface temperature maps. This approach suggests that viewing climate patterns through an artificial intelligence (AI) lens has the power to uncover new insights into climate variability and change.
more »
« less
- Award ID(s):
- 1749261
- PAR ID:
- 10448231
- Publisher / Repository:
- DOI PREFIX: 10.1029
- Date Published:
- Journal Name:
- Geophysical Research Letters
- Volume:
- 46
- Issue:
- 22
- ISSN:
- 0094-8276
- Page Range / eLocation ID:
- p. 13389-13398
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract 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.more » « less
-
Abstract We use an ensemble of simulations of a coupled model (NCAR Community Earth System Model) driven by natural radiative forcing estimates over the pre‐industrial past millennium to test the efficacy of methods designed to remove forced variability from proxy‐based climate reconstructions and estimate residual internal variability (e.g., a putative “Atlantic Multidecadal Oscillation”). Within the framework of these experiments, the forced component of surface temperature change can be estimated accurately from the ensemble mean, and the internal variability of each of the independent realizations can be accurately assessed by subtracting off that estimate. We show in this case, where the true internal variability is known, that regression‐based methods of removing the forced component from proxy reconstructions will, in the presence of uncertainties in the underlying natural radiative forcing, fail to yield accurate estimates thereof, incorrectly attributing unresolved forced features (and multidecadal spectral peaks associated with them) to internal variability.more » « less
-
Abstract Separating the climate response to external forcing from internal climate variability is a key challenge. While most previous studies have focused on surface responses, here we examine zonal‐mean patterns of North Pacific subsurface temperature responses. In particular, the changes since 1950 driven by anthropogenic aerosol emissions are found by using a pattern recognition method. Based on the single‐forcing large‐ensemble simulations from two models, we show that aerosol forcing caused a nonmonotonic temporal response and a characteristic zonal‐mean pattern within North Pacific, which is distinct from the pattern associated with internal variability. The aerosol‐forced pattern with the nonmonotonic temporal feature shows a substantial temperature change in subpolar regions and a reversed change on the southern flank of the subtropical gyre. A similar characteristic pattern and nonmonotonic time evolution are extracted from the subsurface observations, which likely reflect the subsurface responses to the aerosol forcing, although differences exist with the simulated responses.more » « less
-
Previous work identified an anthropogenic fingerprint pattern in 𝑇AC (𝑥, 𝑡), the amplitude of the seasonal cycle of mid- to upper tropospheric temperature (TMT), but did not explicitly consider whether fingerprint identification in satellite 𝑇AC(𝑥,𝑡) data could have been influenced by real-world multidecadal internal variability (MIV). We address this question here using large ensembles (LEs) performed with five climate models. LEs provide many different sequences of internal variability noise superimposed on an underlying forced signal. Despite differences in historical external forcings, climate sensitivity, and MIV properties of the five models, their 𝑇AC (𝑥, 𝑡) fingerprints are similar and statistically identifiable in 239 of the 240 LE realizations of historical climate change. Comparing simulated and observed variability spectra reveals that consistent fingerprint identification is unlikely to be biased by model underestimates of observed MIV. Even in the presence of large (factor of 3-4) inter-model and inter-realization differences in the amplitude of MIV, the anthropogenic fingerprints of seasonal cycle changes are robustly identifiable in models and satellite data. This is primarily due to the fact that the distinctive, global-scale fingerprint patterns are spatially dissimilar to the smaller-scale patterns of internal 𝑇AC(𝑥,𝑡) variability associated with the Atlantic Multidecadal Oscillation and the El Niño~Southern Oscillation. The robustness of the seasonal cycle D&A results shown here, taken together with the evidence fromidealized aquaplanet simulations, suggest that basic physical processes are dictating a common pattern of forced𝑇AC(𝑥,𝑡) changes in observations and in the five LEs. The key processes involved include GHG-induced expansion of the tropics, lapse-rate changes, land surface drying, and sea ice decrease.more » « less
