Machine Learning Emulation of 3D Cloud Radiative Effects

The treatment of cloud structure in numerical weather and climate models is often greatly simplified to make them computationally affordable. Here we propose to correct the European Centre for Medium‐Range Weather Forecasts 1D radiation scheme ecRad for 3D cloud effects using computationally cheap neural networks. 3D cloud effects are learned as the difference between ecRad's fast 1D Tripleclouds solver that neglects them and its 3D SPARTACUS (SPeedy Algorithm for Radiative TrAnsfer through CloUd Sides) solver that includes them but is about five times more computationally expensive. With typical errors between 20% and 30% of the 3D signal, neural networks improve Tripleclouds' accuracy for about 1% increase in runtime. Thus, rather than emulating the whole of SPARTACUS, we keep Tripleclouds unchanged for cloud‐free parts of the atmosphere and 3D‐correct it elsewhere. The focus on the comparably small 3D correction instead of the entire signal allows us to improve predictions significantly if we assume a similar signal‐to‐noise ratio for both.

[1]  R. Hogan,et al.  Machine Learning Emulation of Urban Land Surface Processes , 2021, Journal of Advances in Modeling Earth Systems.

[2]  Peter Ukkonen.  Exploring Pathways to More Accurate Machine Learning Emulation of Atmospheric Radiative Transfer , 2021, Journal of Advances in Modeling Earth Systems.

[3]  Thomas Nagler,et al.  Synthia: multidimensional synthetic data generation in Python , 2021, J. Open Source Softw..

[4]  Torsten Hoefler,et al.  The digital revolution of Earth-system science , 2021, Nature Computational Science.

[5]  Matthew Chantry,et al.  Machine Learning Emulation of Gravity Wave Drag in Numerical Weather Forecasting , 2021, Journal of advances in modeling earth systems.

[6]  Robin J. Hogan,et al.  Copula-based synthetic data augmentation for machine-learning emulators , 2020, Geoscientific Model Development.

[7]  Damian Podareanu,et al.  Predicting atmospheric optical properties for radiative transfer computations using neural networks , 2020, Philosophical Transactions of the Royal Society A.

[8]  R. Hogan,et al.  Copula-based synthetic data generation for machine learning emulators in weather and climate: application to a simple radiation model , 2020, ArXiv.

[9]  Robert Pincus,et al.  Accelerating Radiation Computations for Dynamical Models With Targeted Machine Learning and Code Optimization , 2020, Journal of Advances in Modeling Earth Systems.

[10]  Hwan‐Jin Song,et al.  Evaluation of Neural Network Emulations for Radiation Parameterization in Cloud Resolving Model , 2020, Geophysical Research Letters.

[11]  J. Thepaut,et al.  The ERA5 global reanalysis , 2020, Quarterly Journal of the Royal Meteorological Society.

[12]  Robin J. Hogan,et al.  Evaluating and improving the treatment of gases in radiation schemes: the Correlated K-Distribution Model Intercomparison Project (CKDMIP) , 2020, Geoscientific Model Development.

[13]  J. Dudhia,et al.  WRF‐TEB: Implementation and Evaluation of the Coupled Weather Research and Forecasting (WRF) and Town Energy Balance (TEB) Model , 2020, Journal of Advances in Modeling Earth Systems.

[14]  R. Hogan,et al.  Parametrizing cloud geometry and its application in a subgrid cloud‐edge erosion scheme , 2020, Quarterly Journal of the Royal Meteorological Society.

[15]  A. Benedetti,et al.  An aerosol climatology for global models based on the tropospheric aerosol scheme in the Integrated Forecasting System of ECMWF , 2019, Geoscientific Model Development.

[16]  R. Hogan,et al.  Entrapment: An Important Mechanism to Explain the Shortwave 3D Radiative Effect of Clouds , 2019, Journal of the Atmospheric Sciences.

[17]  Matthew R. Norman,et al.  Using Deep Neural Networks as Cost‐Effective Surrogate Models for Super‐Parameterized E3SM Radiative Transfer , 2019, Geophysical Research Letters.

[18]  Robert Pincus,et al.  Balancing Accuracy, Efficiency, and Flexibility in Radiation Calculations for Dynamical Models , 2019, Journal of advances in modeling earth systems.

[19]  Peter A. G. Watson,et al.  Applying Machine Learning to Improve Simulations of a Chaotic Dynamical System Using Empirical Error Correction , 2019, Journal of advances in modeling earth systems.

[20]  Noah D. Brenowitz,et al.  Spatially Extended Tests of a Neural Network Parametrization Trained by Coarse‐Graining , 2019, Journal of Advances in Modeling Earth Systems.

[21]  Robin J. Hogan,et al.  A Flexible and Efficient Radiation Scheme for the ECMWF Model , 2018, Journal of Advances in Modeling Earth Systems.

[22]  W. Collins,et al.  Reducing uncertainties in climate models , 2018, Science.

[23]  Noah D. Brenowitz,et al.  Prognostic Validation of a Neural Network Unified Physics Parameterization , 2018, Geophysical Research Letters.

[24]  Pierre Gentine,et al.  Deep learning to represent subgrid processes in climate models , 2018, Proceedings of the National Academy of Sciences.

[25]  Vanessa Sochat,et al.  Singularity: Scientific containers for mobility of compute , 2017, PloS one.

[26]  Inna,et al.  Radiation in numerical weather prediction , 2017 .

[27]  B. Mayer,et al.  Representing 3‐D cloud radiation effects in two‐stream schemes: 1. Longwave considerations and effective cloud edge length , 2016 .

[28]  B. Mayer,et al.  Representing 3‐D cloud radiation effects in two‐stream schemes: 2. Matrix formulation and broadband evaluation , 2016 .

[29]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[30]  M. Zelinka,et al.  An observational radiative constraint on hydrologic cycle intensification , 2015, Nature.

[31]  B. Mayer,et al.  A three-dimensional parallel radiative transfer model for atmospheric heating rates for use in cloud resolving models—The TenStream solver , 2015 .

[32]  R. Hogan,et al.  Incorporating the Effects of 3D Radiative Transfer in the Presence of Clouds into Two-Stream Multilayer Radiation Schemes , 2013 .

[33]  Thomas Reichler,et al.  Analysis and Reduction of Systematic Errors through a Seamless Approach to Modeling Weather and Climate , 2010 .

[34]  George Mozdzynski,et al.  A Reduced Radiation Grid for the ECMWF Integrated Forecasting System , 2008 .

[35]  Vladimir M. Krasnopolsky,et al.  Decadal Climate Simulations Using Accurate and Fast Neural Network Emulation of Full, Longwave and Shortwave, Radiation* , 2008 .

[36]  Robin J. Hogan,et al.  Tripleclouds: An Efficient Method for Representing Horizontal Cloud Inhomogeneity in 1D Radiation Schemes by Using Three Regions at Each Height , 2008 .

[37]  R. Betts,et al.  Changes in Atmospheric Constituents and in Radiative Forcing. Chapter 2 , 2007 .

[38]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[39]  D. Stensrud.  Parameterization Schemes: Keys to Understanding Numerical Weather Prediction Models , 2007 .

[40]  Robert Pincus,et al.  The Accuracy of Determining Three-Dimensional Radiative Transfer Effects in Cumulus Clouds Using Ground-Based Profiling Instruments , 2005 .

[41]  D. Chalikov,et al.  New Approach to Calculation of Atmospheric Model Physics: Accurate and Fast Neural Network Emulation of Longwave Radiation in a Climate Model , 2005 .

[42]  J. Morcrette,et al.  A fast, flexible, approximate technique for computing radiative transfer in inhomogeneous cloud fields , 2003 .

[43]  Frédéric Chevallier,et al.  Use of a neural‐network‐based long‐wave radiative‐transfer scheme in the ECMWF atmospheric model , 2000 .

[44]  Roger Davies,et al.  Effects of Cloud Heterogeneities on Shortwave Radiation: Comparison of Cloud-Top Variability and Internal Heterogeneity , 1999 .

[45]  Alain Chedin,et al.  A Neural Network Approach for a Fast and Accurate Computation of a Longwave Radiative Budget , 1998 .

[46]  A. Heidinger,et al.  Finite-Cloud Effects in Longwave Radiative Transfer , 1996 .

[47]  A. Chédin,et al.  Une méthode utilisant les techniques neuronales pour le calcul rapide de la distribution verticale du bilan radiatif thermique terrestre , 1996 .

[48]  Graeme L. Stephens,et al.  The Parameterization of Radiation for Numerical Weather Prediction and Climate Models , 1984 .