A document by Ariel L Morrison. Click on the document to view its contents.

Seasonal-to-decadal climate prediction is crucial for decision-making in a number of industries, but forecasts on these timescales have limited skill. Here, we develop a data-driven method for selecting optimal analogs for seasonal-to-decadal analog forecasting. Using an interpretable neural network, we learn a spatially-weighted mask that quantifies how important each grid point is for determining whether two climate states will evolve similarly. We show that analogs selected using this weighted mask provide more skillful forecasts than analogs that are selected using traditional spatially-uniform methods. This method is tested on two prediction problems within a perfect model framework using the Max Planck Institute for Meteorology Grand Ensemble: multi-year prediction of North Atlantic sea surface temperatures, and seasonal prediction of El Niño Southern Oscillation. This work demonstrates a methodical approach to selecting analogs that may be useful for improving seasonal-to-decadal forecasts and understanding their sources of skill.

We use neural networks and large climate model ensembles to explore predictability of internal variability in sea surface temperature anomalies on interannual (1-3 year) and decadal (1-5 and 3-7 year) timescales. We find that neural networks can skillfully predict SST anomalies at these lead times, especially in the North Atlantic, North Pacific, Tropical Pacific, Tropical Atlantic and Southern Ocean. The spatial patterns of SST predictability vary across the nine climate models studied. The neural networks identify “windows of opportunity” where future SST anomalies can be predicted with more certainty. Neural networks trained on climate models also make skillful SST predictions in historical observations, although the skill varies depending on which climate model the network was trained. Our results highlight that neural networks can identify predictable internal variability within existing climate datasets and show important differences in how well patterns of SST predictability in climate models translate to the real world.

Climate and Earth system models are important tools to assess the benefits and risks of stratospheric aerosol injection (SAI) relative to those associated with anthropogenic climate change. A “controller” algorithm has been used to specify injection amounts of sulfur dioxide in SAI experiments performed with the Community Earth System Model (CESM). The experiments are designed to maintain specific temperature targets, such as limiting global mean temperature to 1.5ºC above the pre-industrial level. However, the influence of natural climate variability on the injection amount has not been extensively documented. Our study reveals that more than 70% of the year-to-year variation in the total injection amount (excluding the long-term trend) in CESM SAI experiments is attributed to the El Niño-Southern Oscillation (ENSO). A simplified statistical model further suggests that the intrinsic, lagged response of the controller to the climate can increase the variance of global mean temperature in the model simulations.

Stratospheric aerosol injection (SAI) has been proposed as a possible complementary solution to limit global warming and its societal consequences. However, the climate impacts of such intervention remain unclear. Here, we introduce an explainable artificial intelligence (XAI) framework to quantify how distinguishable an SAI climate might be from a pre-deployment climate. A suite of neural networks is trained on Earth system model data to learn to distinguish between pre- and post-deployment periods across a variety of climate variables. The network accuracy is analogous to the “climate distinguishability” between the periods, and the corresponding distinctive patterns are identified using XAI methods to gain insights into the emerging signals from SAI. For many variables, the two periods are less distinguishable under SAI than under a no-SAI scenario, suggesting that the specific intervention modeled decelerates future climatic changes. Other climate variables for which the intervention has negligible effect are also highlighted.

Earth system models are a powerful tool to simulate the response to hypothetical climate intervention strategies, such as stratospheric aerosol injection (SAI). Recent simulations of SAI implement tools from control theory, called “controllers”, to determine the quantity of aerosol to inject into the stratosphere to reach or maintain specified global temperature targets, such as limiting global warming to 1.5\textdegree C above pre-industrial temperatures. This work explores how internal (unforced) climate variability can impact controller-determined injection amounts using the Assessing Responses and Impacts of Solar climate intervention on the Earth system with Stratospheric Aerosol Injection (ARISE-SAI) simulations. Since the ARISE-SAI controller determines injection amounts by comparing global annual-mean surface temperature to predetermined temperature targets, internal variability that impacts temperature can impact the total injection amount as well. Using an offline version of the ARISE-SAI controller and data from CESM2 earth system model simulations, we quantify how internal climate variability and volcanic eruptions impact injection amounts. While idealized, this approach allows for the investigation of a large variety of climate states without additional simulations and can be used to attribute controller sensitivities to specific modes of internal variability.

As oceanic moisture evaporates, it leaves a signature on sea surface salinity. Roughly 10% of the moisture that evaporates over the ocean is transported over land, allowing the salinity fields to be a predictor of terrestrial precipitation. This research is among the first in published literature to assess the role of sea surface salinity for improved predictions on low-skill summertime subseasonal timescales for terrestrial precipitation predictions. Neural networks are trained with the CESM2 Large Ensemble using North Atlantic salinity anomalies to quantify predictability of U.S. Midwest summertime heavy rainfall events at 0 to 56-day leads. Using explainable artificial intelligence, salinity anomalies in the Caribbean Sea and Gulf of Mexico are found to provide skill for subseasonal forecasts of opportunity, e.g. confident and correct predictions. Further, a moisture-tracking algorithm applied to reanalysis data demonstrates that the regions of evaporation identified by neural networks directly provide moisture that precipitates in the Midwest.

Stratospheric aerosol injection (SAI has been proposed as a potential method for mitigating risks and impacts associated with anthropogenic climate change. One such risk is widespread permafrost thaw and associated carbon release. While permafrost has been shown to stabilize under different SAI scenarios, natural variability may lead to a wide range of projected climate futures under SAI. Here we use the 10-member ensemble from the ARISE-SAI-1.5 simulations to assess the spread in projected active layer depth and permafrost temperature across boreal permafrost soils and specifically in four peatland and Yedoma regions. The forced response in active layer depth and permafrost temperature quickly diverge between an SAI and non-SAI world, but individual ensemble members overlap for several years following SAI deployment. Projected permafrost variability may mask the forced response to SAI and make it difficult to detect if and when SAI is stabilizing permafrost in any single realization. We find that it may take more than a decade of SAI deployment to detect the effects of SAI on permafrost temperature and almost 30 years to detect its effects on active layer depth. Not only does natural variability make it more difficult to detect SAI’s influence, it could also affect the likelihood of reaching a permafrost tipping point. In some realizations, SAI fails to prevent a tipping point that is also reached in a non-SAI world. Our results underscore the importance of accounting for natural variability in assessments of SAI’s potential influence on the climate system.

Current global actions to reduce greenhouse gas emissions are very likely to be insufficient to meet the climate targets outlined under the Paris Agreement. This motivates research on possible methods for intervening in the Earth system to minimize climate risk while decarbonization efforts continue. One such hypothetical climate intervention is stratospheric aerosol injection (SAI), where reflective particles would be released into the stratosphere to cool the planet by reducing solar insolation. The climate response to SAI is not well understood, particularly on short-term time horizons frequently used by decision makers and planning practitioners to assess climate information. This knowledge gap limits informed discussion of SAI outside the scientific community. We demonstrate two framings to explore the climate response in the decade after SAI deployment in modeling experiments with parallel SAI and no-SAI simulations. The first framing, which we call a snapshot around deployment, displays change over time within the SAI scenarios and applies to the question “What happens before and after SAI is deployed in the model?” The second framing, the intervention impact, displays the difference between the SAI and no-SAI simulations, corresponding to the question “What is the impact of a given intervention relative to climate change with no intervention?” We apply these framings to annual mean 2-meter temperature, precipitation, and a precipitation extreme in the first two experiments to use large ensembles of Earth system models that comprehensively represent both the SAI injection process and climate response, and connect these results to implications for other climate variables.

Soil moisture influences near-surface air temperature by partitioning downwelling radiation into latent and sensible heat fluxes, through which dry soils generally lead to higher temperatures. The strength of this coupled soil moisture-temperature (SM-T) relationship is not spatially uniform, and numerous methods have been developed to assess SM-T coupling strength across the globe. These methods tend to involve either idealized climate-model experiments or linear statistical methods which cannot fully capture nonlinear SM-T coupling. In this study, we propose a nonlinear machine learning-based approach for analyzing SM-T coupling and apply this method to various mid-latitude regions using historical reanalysis datasets. We first train convolutional neural networks (CNNs) to predict daily maximum near-surface air temperature (TMAX) given daily SM and geopotential height fields. We then use partial dependence analysis to isolate the average sensitivity of each CNN’s TMAX prediction to the SM input under daily atmospheric conditions. The resulting SM-T relationships broadly agree with previous assessments of SM-T coupling strength. Over many regions, we find nonlinear relationships between the CNN’s TMAX prediction and the SM input map. These nonlinearities suggest that the coupled interactions governing SM-T relationships vary under different SM conditions, but these variations are regionally dependent. We also apply this method to test the influence of SM memory on SM-T coupling and find that our results are consistent with previous studies. Although our study focuses specifically on local SM-T coupling, our machine learning-based method can be extended to investigate other coupled interactions within the climate system using observed or model-derived datasets.

We develop and demonstrate a new interpretable deep learning model specifically designed for image analysis in earth system science applications. The neural network is designed to be inherently interpretable, rather than explained via post hoc methods. This is achieved by training the network to identify parts of training images that act as prototypes for correctly classifying unseen images. The new network architecture extends the interpretable prototype architecture of a previous study in computer science to incorporate absolute location. This is useful for earth system science where images are typically the result of physics-based processes, and the information is often geo-located. Although the network is constrained to only learn via similarities to a small number of learned prototypes, it can be trained to exhibit only a minimal reduction in accuracy compared to non-interpretable architectures. We apply the new model to two earth science use cases: a synthetic data set that loosely represents atmospheric high- and low-pressure systems, and atmospheric reanalysis fields to identify the state of tropical convective activity associated with the Madden-Julian oscillation. In both cases, we demonstrate that considering absolute location greatly improves testing accuracies. Furthermore, the network architecture identifies specific historical dates that capture multivariate, prototypical behaviour of tropical climate variability.

Predicting Pacific Decadal Oscillation (PDO) transitions and understanding the associated mechanisms has proven a critical but challenging task in climate science. As a form of decadal variability, the PDO is associated with both large-scale climate shifts and regional climate predictability. We show that artificial neural networks (ANNs) predict PDO persistence and transitions from 12 months onward. Using layer-wise relevance propagation to investigate the ANN predictions, we demonstrate that the ANNs utilize oceanic patterns that have been previously linked to predictable PDO behavior. For PDO transitions, ANNs recognize a build-up of ocean heat content in the off-equatorial western Pacific 12-27 months before a transition occurs. The results support the continued use of ANNs in climate studies where explainability tools can assist in mechanistic understanding of the climate system.