Search for: All records

Abstract We show that explainable neural networks can identify regions of oceanic variability that contribute predictability on decadal timescales in a fully coupled Earth‐system model. The neural networks learn to use sea‐surface temperature anomalies to predict future continental surface temperature anomalies. We then use a neural‐network explainability method called layerwise relevance propagation to infer which oceanic patterns lead to accurate predictions made by the neural networks. In particular, regions within the North Atlantic Ocean and North Pacific Ocean lend the most predictability for surface temperature across continental North America. We apply the proposed methodology to decadal variability, although the concept is generalizable to other timescales of predictability. Furthermore, while our approach focuses on predictable patterns of internal variability within climate models, it should be generalizable to observational data as well. Our study contributes to the growing evidence that explainable neural networks are important tools for advancing geoscientific knowledge. 

Abstract The intensity of deep convective storms is driven in part by the strength of their updrafts and cold pools. In spite of the importance of these storm features, they can be poorly represented within numerical models. This has been attributed to model parameterizations, grid resolution, and the lack of appropriate observations with which to evaluate such simulations. The overarching goal of the Colorado State University Convective CLoud Outflows and UpDrafts Experiment (C 3 LOUD-Ex) was to enhance our understanding of deep convective storm processes and their representation within numerical models. To address this goal, a field campaign was conducted during July 2016 and May–June 2017 over northeastern Colorado, southeastern Wyoming, and southwestern Nebraska. Pivotal to the experiment was a novel “Flying Curtain” strategy designed around simultaneously employing a fleet of uncrewed aerial systems (UAS; or drones), high-frequency radiosonde launches, and surface observations to obtain detailed measurements of the spatial and temporal heterogeneities of cold pools. Updraft velocities were observed using targeted radiosondes and radars. Extensive datasets were successfully collected for 16 cold pool–focused and seven updraft-focused case studies. The updraft characteristics for all seven supercell updraft cases are compared and provide a useful database for model evaluation. An overview of the 16 cold pools’ characteristics is presented, and an in-depth analysis of one of the cold pool cases suggests that spatial variations in cold pool properties occur on spatial scales from O (100) m through to O (1) km. Processes responsible for the cold pool observations are explored and support recent high-resolution modeling results. 

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.