Stochastic parameterizations account for uncertainty in the representation of unresolved subgrid processes by sampling from the distribution of possible subgrid forcings. Some existing stochastic parameterizations utilize data‐driven approaches to characterize uncertainty, but these approaches require significant structural assumptions that can limit their scalability. Machine learning models, including neural networks, are able to represent a wide range of distributions and build optimized mappings between a large number of inputs and subgrid forcings. Recent research on machine learning parameterizations has focused only on deterministic parameterizations. In this study, we develop a stochastic parameterization using the generative adversarial network (GAN) machine learning framework. The GAN stochastic parameterization is trained and evaluated on output from the Lorenz '96 model, which is a common baseline model for evaluating both parameterization and data assimilation techniques. We evaluate different ways of characterizing the input noise for the model and perform model runs with the GAN parameterization at weather and climate time scales. Some of the GAN configurations perform better than a baseline bespoke parameterization at both time scales, and the networks closely reproduce the spatiotemporal correlations and regimes of the Lorenz '96 system. We also find that, in general, those models which produce skillful forecasts are also associated with the best climate simulations.

Deep learning models, such as convolutional neural networks, utilize multiple specialized layers to encode spatial patterns at different scales. In this study, deep learning models are compared with standard machine learning approaches on the task of predicting the probability of severe hail based on upper-air dynamic and thermodynamic fields from a convection-allowing numerical weather prediction model. The data for this study come from patches surrounding storms identified in NCAR convection-allowing ensemble runs from 3 May to 3 June 2016. The machine learning models are trained to predict whether the simulated surface hail size from the Thompson hail size diagnostic exceeds 25 mm over the hour following storm detection. A convolutional neural network is compared with logistic regressions using input variables derived from either the spatial means of each field or principal component analysis. The convolutional neural network statistically significantly outperforms all other methods in terms of Brier skill score and area under the receiver operator characteristic curve. Interpretation of the convolutional neural network through feature importance and feature optimization reveals that the network synthesized information about the environment and storm morphology that is consistent with our understanding of hail growth, including large lapse rates and a wind shear profile that favors wide updrafts. Different neurons in the network also record different storm modes, and the magnitude of the output of those neurons is used to analyze the spatiotemporal distributions of different storm modes in the NCAR ensemble.

Forecasting severe hail accurately requires predicting how well atmospheric conditions support the development of thunderstorms, the growth of large hail, and the minimal loss of hail mass to melting before reaching the surface. Existing hail forecasting techniques incorporate information about these processes from proximity soundings and numerical weather prediction models, but they make many simplifying assumptions, are sensitive to differences in numerical model configuration, and are often not calibrated to observations. In this paper a storm-based probabilistic machine learning hail forecasting method is developed to overcome the deficiencies of existing methods. An object identification and tracking algorithm locates potential hailstorms in convection-allowing model output and gridded radar data. Forecast storms are matched with observed storms to determine hail occurrence and the parameters of the radar-estimated hail size distribution. The database of forecast storms contains information about storm properties and the conditions of the prestorm environment. Machine learning models are used to synthesize that information to predict the probability of a storm producing hail and the radar-estimated hail size distribution parameters for each forecast storm. Forecasts from the machine learning models are produced using two convection-allowing ensemble systems and the results are compared to other hail forecasting methods. The machine learning forecasts have a higher critical success index (CSI) at most probability thresholds and greater reliability for predicting both severe and significant hail.

Clouds are critical for weather and climate prediction. The multiple scales of cloud processes make simulation difficult. Often models and measurements are used to develop empirical relationships for large‐scale models to be computationally efficient. Machine learning provides another potential tool to improve our empirical parameterizations of clouds. To explore these opportunities, we replace the warm rain formation process in a General Circulation Model (GCM) with a detailed treatment from a bin microphysical model that causes a 400% slowdown in the GCM. We analyze the changes in climate that result from the use of the bin microphysical calculation and find improvements in the rain onset and frequency of light rain compared to high resolution process models and observations. We also find a resulting change in the cloud feedback response of the model to warming, which will significantly impact the climate sensitivity. We then replace the bin microphysical model with several neural networks designed to emulate the autoconversion and accretion rates produced by the bin microphysical model. The neural networks are organized into two stages: the first stage identifies where tendencies will be nonzero (and the sign of the tendency), and the second stage predicts the magnitude of the autoconversion and accretion rates. We describe the risks of overfitting, extrapolation, and linearization by using perfect model experiments with and without the emulator. We can recover the solutions with the emulators in almost all respects, and get simulations that perform as the detailed model, but with the computational cost of the control simulation.

Demands to manage the risks of artificial intelligence (AI) are growing. These demands and the government standards arising from them both call for trustworthy AI. In response, we adopt a convergent approach to review, evaluate, and synthesize research on the trust and trustworthiness of AI in the environmental sciences and propose a research agenda. Evidential and conceptual histories of research on trust and trustworthiness reveal persisting ambiguities and measurement shortcomings related to inconsistent attention to the contextual and social dependencies and dynamics of trust. Potentially underappreciated in the development of trustworthy AI for environmental sciences is the importance of engaging AI users and other stakeholders, which human–AI teaming perspectives on AI development similarly underscore. Co‐development strategies may also help reconcile efforts to develop performance‐based trustworthiness standards with dynamic and contextual notions of trust. We illustrate the importance of these themes with applied examples and show how insights from research on trust and the communication of risk and uncertainty can help advance the understanding of trust and trustworthiness of AI in the environmental sciences.