Prognostic Validation of a Neural Network Unified Physics Parameterization

By Noah D. Brenowitz, Chris Bretherton. 2018

Weather and climate models approximate diabatic and sub‐grid‐scale processes in terms of grid‐scale variables using parameterizations. Current parameterizations are designed by humans based on physical understanding, observations, and process modeling. As a result, they are numerically efficient and interpretable, but potentially oversimplified. However, the advent of global high‐resolution simulations and observations enables a more robust approach based on machine learning. In this letter, a neural network‐based parameterization is trained using a near‐global aqua‐planet simulation with a 4‐km resolution (NG‐Aqua). The neural network predicts the apparent sources of heat and moisture averaged onto (160 km)2 grid boxes. A numerically stable scheme is obtained by minimizing the prediction error over multiple time steps rather than single one. In prognostic single‐column model tests, this scheme matches both the fluctuations and equilibrium of NG‐Aqua simulation better than the Community Atmosphere Model does.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Stratospheric Aerosol Injection as a Deep Reinforcement Learning Problem (2019, background): “Recently, the advent of deep learning has led to a number of successful emulation attempts of full GCMs used for weather prediction (Düben, 2018), as well as for sub-grid scale processes (Brenowitz, 2018; Rasp, 2018), including precipitation (O’Gorman, 2018).”
WeatherBench: A Benchmark Data Set for Data‐Driven Weather Forecasting (2020, background): “In recent years, several studies have explored the potential of hybrid models, in which a physical model is combined with data-driven components, for example replacing heuristic cloud or radiation parameterizations (Chevallier et al., 1998; Krasnopolsky et al., 2005; Rasp et al., 2018; Brenowitz and”
Confronting the Challenge of Modeling Cloud and Precipitation Microphysics (2020, background): “Another approach that has gained recent traction within the atmospheric science community has been to develop parameterizations wholly via machine learning (e.g., Brenowitz & Bretherton, 2018; Gentine et al., 2018; O’Gorman & Dwyer, 2018; Rasp et al., 2018).”
Deep learning to represent subgrid processes in climate models (2018, unknown): “Unstable modes have been the main issue in previous studies that used shallow architectures (30, 31).”
Can Machines Learn to Predict Weather? Using Deep Learning to Predict Gridded 500‐hPa Geopotential Height From Historical Weather Data (2019, background): “…in observed and modeled atmospheric states (Kurth et al., 2018; Lagerquist, McGovern, & Gagne, 2019; Liu et al., 2016), improve parameteriza-46 tions in GCMs (e.g., Brenowitz & Bretherton, 2018; Rasp, Pritchard, & Gentine, 2018), and predict uncertainty in weather forecasts (Scher & Messori, 2018).”
Improving Data‐Driven Global Weather Prediction Using Deep Convolutional Neural Networks on a Cubed Sphere (2020, methodology): “In order to nudge the DLWP model towards learning to predict longer-term weather and improve its long-term stability, we train the model to minimize error on multiple iterated predictive steps using a multi-time-step loss function similar to the strategies in [4] and [25].”
Integrating Scientific Knowledge with Machine Learning for Engineering and Environmental Systems (2020, background): “Already, ML-based parameterizations have shown success in geology [52, 109], atmospheric science [40, 103], and hydrology [29].”
Using Machine Learning to Parameterize Moist Convection: Potential for Modeling of Climate, Climate Change, and Extreme Events (2018, background): “We do not discuss these results further given recent advances in using different training and architecture for ANN parameterizations [Brenowitz and Bretherton, 2018; Rasp et al., 2018], and we instead focus on our promising results for the RF parameterization.”
Integrating Physics-Based Modeling with Machine Learning: A Survey (2020, background): “Already, ML-based parameterizations have shown success in geology [43, 95] and atmospheric science [32, 33, 90, 186].”