Could Machine Learning Break the Convection Parameterization Deadlock?

By Pierre Gentine, Michael S. Pritchard, Stephan Rasp, G. Reinaudi, G. Yacalis. Geophysical Research Letters, 2018

417 citations recorded by the source corpus.

Representing unresolved moist convection in coarse‐scale climate models remains one of the main bottlenecks of current climate simulations. Many of the biases present with parameterized convection are strongly reduced when convection is explicitly resolved (i.e., in cloud resolving models at high spatial resolution approximately a kilometer or so). We here present a novel approach to convective parameterization based on machine learning, using an aquaplanet with prescribed sea surface temperatures as a proof of concept. A deep neural network is trained with a superparameterized version of a climate model in which convection is resolved by thousands of embedded 2‐D cloud resolving models. The machine learning representation of convection, which we call the Cloud Brain (CBRAIN), can skillfully predict many of the convective heating, moistening, and radiative features of superparameterization that are most important to climate simulation, although an unintended side effect is to reduce some of the superparameterization's inherent variance. Since as few as three months' high‐frequency global training data prove sufficient to provide this skill, the approach presented here opens up a new possibility for a future class of convection parameterizations in climate models that are built “top‐down,” that is, by learning salient features of convection from unusually explicit simulations.

Subfields: Atmospheric Science, Global and Planetary Change

View Paper: https://www.semanticscholar.org/paper/6117239f75255bc281051b0b5876102f820b337f

How researchers see this paper

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

In Atmospheric Science

In Global and Planetary Change

Citations in the corpus

Transformation-based Adversarial Video Prediction on Large-Scale Data (2020, background): “For example, high resolution cloud simulation [23] and prediction of short-term changes in the sea ice extent using satellite images have both been identified as a "high leverage" machine learning applications for tackling climate change [56].”
Tackling Climate Change with Machine Learning (2019, background): “trained a deep neural network to emulate the behavior of a high-resolution cloud simulation, and found that the network gave similar results for a fraction of the cost [483] and was stable in a simplified global model [484].”
Machine learning and artificial intelligence to aid climate change research and preparedness (2019, background): “Recent application of neural nets to climate sciences include dryland disturbance (Buckland et al 2019), inverse problems for remote sensing (Krasnopolsky and Schiller 2003) and replacing costly components of climate models (Gentine et al 2018).”
Integrating Scientific Knowledge with Machine Learning for Engineering and Environmental Systems (2020, methodology): “Due to the substantial time cost of running such complex models, there is an increasing interest in using ML models as surrogate models (a model approximating simulation-driven input–output data) to predict target variables at a higher resolution [103, 254].”
Using Machine Learning to Parameterize Moist Convection: Potential for Modeling of Climate, Climate Change, and Extreme Events (2018, background): “, 1998) and for convective and boundary-layer processes (Brenowitz & Bretherton, 2018; Gentine et al., 2018; Krasnopolsky et al., 2010, 2013; Rasp et al., 2018).”
A Transdisciplinary Review of Deep Learning Research and Its Relevance for Water Resources Scientists (2017, methodology): “In a recent self-archived paper, authors have employed an MLP to learn fine-resolution dynamics such as convective heating, moistening and cloud-radiative transfer feedbacks from high-resolution simulations, to replace existing multiple parameterization schemes (Gentine et al., 2018).”
Quantifying Progress Across Different CMIP Phases With the ESMValTool (2020, methodology): “Stevens, 2019) or completely novel approaches to parameterize clouds and convection in climate models such as, for instance, machine learning based cloud parametrizations (Gentine et al., 2018; Rasp et al., 2018)”
A global spatially contiguous solar-induced fluorescence (CSIF) dataset using neural networks (2018, neutral): “, 2018) or to bypass parameterization (Gentine et al., 2018).”
Confronting the Challenge of Modeling Cloud and Precipitation Microphysics (2020, methodology): “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).”
Perspectives on the Future of Land Surface Models and the Challenges of Representing Complex Terrestrial Systems (2020, background): “Given the increasing emphasis on machine learning approaches and the successes of machine learning in solving problems in ESM (Gentine et al., 2018) or offline hydrologic model (Fang et al., 2017; Shen, 2018) behavior, designing models with an emphasis on modular complexity to allow for such hybrid…”