Using machine learning to build temperature-based ozone parameterizations for climate sensitivity simulations

Anumber of studies have demonstrated the importance of ozone in climate change simulations, for example concerning global warming projections and atmospheric dynamics. However, fully interactive atmospheric chemistry schemes needed for calculating changes in ozone are computationally expensive. Climatemodelers therefore often use climatological ozonefields, which are typically neither consistent with the actual climate state simulated by eachmodel norwith the specific climate change scenario. This limitation applies in particular to standardmodeling experiments such as preindustrial control or abrupt 4xCO2 climate sensitivity simulations.Herewe suggest a novelmethod using a simple linearmachine learning regression algorithm to predict ozone distributions for preindustrial and abrupt 4xCO2 simulations. Using the atmospheric temperature field as the only input, the regression reliably predicts three-dimensional ozone distributions atmonthly to daily time intervals. In particular, the representation of stratospheric ozone variability ismuch improved comparedwith a fixed climatology, which is important for interactions with dynamical phenomena such as the polar vortices and theQuasi-Biennial Oscillation. Ourmethod requires training data covering only a fraction of the usual length of simulations and thus promises to be an important stepping stone towards a range of new computationally efficientmethods to consider ozone changes in long climate simulations.We highlight key development steps to further improve and extend the scope ofmachine learning-based ozone parameterizations.

[1]  Arthur L. Samuel,et al.  Some Studies in Machine Learning Using the Game of Checkers , 1967, IBM J. Res. Dev..

[2]  B. Hunt The Need for a Modified Photochemical Theory of the Ozonosphere , 1966 .

[3]  P. Crutzen The influence of nitrogen oxides on the atmospheric ozone content , 1970 .

[4]  M. Molina,et al.  Stratospheric sink for chlorofluoromethanes: chlorine atomc-atalysed destruction of ozone , 1974, Nature.

[5]  J. Pyle A calculation of the possible depletion of ozone by chlorofluorocarbons using a two-dimensional model , 1980 .

[6]  J. D. Mahlman,et al.  Stratospheric Sensitivity to Perturbations in Ozone and Carbon Dioxide: Radiative and Dynamical Response. , 1980 .

[7]  J. Haigh,et al.  Ozone perturbation experiments in a two‐dimensional circulation model , 1982 .

[8]  Daniel Cariolle,et al.  Southern hemisphere medium-scale waves and total ozone disturbances in a spectral general circulation model , 1986 .

[9]  Donald J. Wuebbles,et al.  Radiative Forcing of Climate Changes in the Vertical Distribution of Ozone , 1990 .

[10]  D. Rind,et al.  A simple lightning parameterization for calculating global lightning distributions , 1992 .

[11]  D. Rind,et al.  Modeling Global Lightning Distributions in a General Circulation Model , 1994 .

[12]  R. Tibshirani Regression Shrinkage and Selection via the Lasso , 1996 .

[13]  Joanna D. Haigh The Impact of Solar Variability on Climate , 1996, Science.

[14]  Arthur E. Hoerl,et al.  Ridge Regression: Biased Estimation for Nonorthogonal Problems , 2000, Technometrics.

[15]  B. Hannegan,et al.  Stratospheric ozone in 3-D models : A simple chemistry and the cross-tropopause flux , 2000 .

[16]  Oliver Wild,et al.  Fast-J: Accurate Simulation of In- and Below-Cloud Photolysis in Tropospheric Chemical Models , 2000 .

[17]  M. Prather,et al.  Fast-J2: Accurate Simulation of Stratospheric Photolysis in Global Chemical Models , 2002 .

[18]  David W. J. Thompson,et al.  Interpretation of Recent Southern Hemisphere Climate Change , 2002, Science.

[19]  P. Braesicke,et al.  Changing ozone and changing circulation in northern mid‐latitudes: Possible feedbacks? , 2003 .

[20]  J. C. McConnell,et al.  Doubled CO2‐induced cooling in the middle atmosphere: Photochemical analysis of the ozone radiative feedback , 2004 .

[21]  B. Soden,et al.  Robust Responses of the Hydrological Cycle to Global Warming , 2006 .

[22]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[23]  I. Kirchner,et al.  Effect of zonally asymmetric ozone on stratospheric temperature and planetary wave propagation , 2007 .

[24]  J. Penner,et al.  Global atmospheric chemistry: Integrating over fractional cloud cover , 2007 .

[25]  Daniel Cariolle,et al.  A revised linear ozone photochemistry parameterization for use in transport and general circulation models: multi-annual simulations , 2007 .

[26]  T. Shepherd,et al.  The Impact of Stratospheric Ozone Recovery on the Southern Hemisphere Westerly Jet , 2008, Science.

[27]  N. Gillett,et al.  Sensitivity of Southern Hemisphere climate to zonal asymmetry in ozone , 2008 .

[28]  Warm and sensitive Paleocene-Eocene climate , 2009 .

[29]  N. Gillett,et al.  Sensitivity of climate to dynamically‐consistent zonal asymmetries in ozone , 2009 .

[30]  J. A. Pyle,et al.  Geoscientific Model Development Evaluation of the new UKCA climate-composition model – Part 1 : The stratosphere , 2009 .

[31]  Veronika Eyring,et al.  Impact of stratospheric ozone on Southern Hemisphere circulation change: A multimodel assessment , 2010 .

[32]  Chris Harris,et al.  Design and implementation of the infrastructure of HadGEM3: the next-generation Met Office climate modelling system , 2010 .

[33]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[34]  The effect of zonally asymmetric ozone heating on the Northern Hemisphere winter polar stratosphere , 2011 .

[35]  Veronika Eyring,et al.  Ozone database in support of CMIP5 simulations: results and corresponding radiative forcing , 2011 .

[36]  T. Shepherd,et al.  A Robust Mechanism for Strengthening of the Brewer–Dobson Circulation in Response to Climate Change: Critical-Layer Control of Subtropical Wave Breaking , 2011 .

[37]  J. Lamarque,et al.  Pre-industrial to end 21st century projections of tropospheric ozone from the Atmospheric Chemistry and Climate Model Intercomparison Project (ACCMIP) , 2012 .

[38]  P. Braesicke,et al.  Implementation of the Fast-JX Photolysis scheme (v6.4) into the UKCA component of the MetUM chemistry-climate model (v7.3) , 2012 .

[39]  T. Nathan,et al.  Pathways for Communicating the Effects of Stratospheric Ozone to the Polar Vortex: Role of Zonally Asymmetric Ozone , 2012 .

[40]  Chao Li,et al.  Deep-ocean heat uptake and equilibrium climate response , 2013, Climate Dynamics.

[41]  Karl E. Taylor,et al.  An overview of CMIP5 and the experiment design , 2012 .

[42]  Veronika Eyring,et al.  Analysis of Present Day and Future OH and Methane Lifetime in the ACCMIP Simulations , 2012 .

[43]  K. Taylor,et al.  Forcing, feedbacks and climate sensitivity in CMIP5 coupled atmosphere‐ocean climate models , 2012 .

[44]  E. Guilyardi,et al.  A first look at ENSO in CMIP5 , 2012 .

[45]  Trevor Hastie,et al.  An Introduction to Statistical Learning , 2013, Springer Texts in Statistics.

[46]  P. Jöckel,et al.  Chemical contribution to future tropical ozone change in the lower stratosphere , 2013 .

[47]  Mark Lawrence,et al.  An overview of the Geoengineering Model Intercomparison Project (GeoMIP) , 2013 .

[48]  E. Guilyardi,et al.  ENSO representation in climate models: from CMIP3 to CMIP5 , 2013, Climate Dynamics.

[49]  P. J. Young,et al.  Long‐term ozone changes and associated climate impacts in CMIP5 simulations , 2013 .

[50]  Nitesh V. Chawla,et al.  Computational Intelligent Data Analysis for Sustainable Development , 2013 .

[51]  Bryan Lawrence,et al.  Storing and manipulating environmental big data with JASMIN , 2013, 2013 IEEE International Conference on Big Data.

[52]  R. Sausen,et al.  Interactive ozone induces a negative feedback in CO2‐driven climate change simulations , 2014 .

[53]  G. Schmidt,et al.  The QBO in two GISS global climate models: 1. Generation of the QBO , 2014 .

[54]  S. Madronich,et al.  Solar Ultraviolet Radiation in a Changing Climate , 2014 .

[55]  N Marwan,et al.  Prediction of extreme floods in the eastern Central Andes based on a complex networks approach , 2014, Nature Communications.

[56]  S. Brönnimann,et al.  The coupled atmosphere–chemistry–ocean model SOCOL-MPIOM , 2014 .

[57]  Jürgen Kurths,et al.  Identifying causal gateways and mediators in complex spatio-temporal systems , 2015, Nature Communications.

[58]  Annette Osprey,et al.  A large ozone-circulation feedback and its implications for global warming assessments , 2014, Nature climate change.

[59]  Veronika Eyring,et al.  Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization , 2015 .

[60]  S. Marsland,et al.  An atmospheric mechanism for ENSO amplitude changes under an abrupt quadrupling of CO2 concentration in CMIP5 models , 2016 .

[61]  Ball F eb 2 01 6 Ozone observations reveal lower solar cycle spectral vari - , 2016 .

[62]  T. Stocker,et al.  Response of the AMOC to reduced solar radiation – the modulating role of atmospheric chemistry , 2016 .

[63]  L. Polvani,et al.  Reduction of Climate Sensitivity to Solar Forcing due to Stratospheric Ozone Feedback , 2016 .

[64]  P. Nowack,et al.  Stratospheric ozone changes under solar geoengineering: implications for UV exposure and air quality , 2016 .

[65]  Anja Schmidt,et al.  Emergence of healing in the Antarctic ozone layer , 2016, Science.

[66]  F. Marone,et al.  Rapid Mapping of Lithiation Dynamics in Transition Metal Oxide Particles with Operando X-ray Absorption Spectroscopy , 2016, Scientific Reports.

[67]  G. Hegerl,et al.  Beyond equilibrium climate sensitivity , 2017 .

[68]  N. Harnik,et al.  Radiative effects of ozone waves on the Northern Hemisphere polar vortex and its modulation by the QBO , 2017 .

[69]  L. Polvani,et al.  Reduced Southern Hemispheric circulation response to quadrupled CO2 due to stratospheric ozone feedback , 2017 .

[70]  Andrew Stuart,et al.  Earth System Modeling 2.0: A Blueprint for Models That Learn From Observations and Targeted High‐Resolution Simulations , 2017, 1709.00037.

[71]  Jakob Runge,et al.  Early prediction of extreme stratospheric polar vortex states based on causal precursors , 2017 .

[72]  Peter Braesicke,et al.  On the role of ozone feedback in the ENSO amplitude response under global warming , 2017, Geophysical research letters.

[73]  A. Kitoh,et al.  Impact of interactive chemistry of stratospheric ozone on Southern Hemisphere paleoclimate simulation , 2017 .

[74]  J. Gregory,et al.  Relationship of tropospheric stability to climate sensitivity and Earth’s observed radiation budget , 2017, Proceedings of the National Academy of Sciences.

[75]  Donald D. Lucas,et al.  Machine Learning Predictions of a Multiresolution Climate Model Ensemble , 2018 .

[76]  Pierre Gentine,et al.  Could Machine Learning Break the Convection Parameterization Deadlock? , 2018, Geophysical Research Letters.

[77]  N. Abraham,et al.  Quasi-Newton methods for atmospheric chemistry simulations: implementation in UKCA UM vn10.8 , 2018, Geoscientific Model Development.

[78]  Jessica L. Neu,et al.  Tropospheric Ozone Assessment Report:Assessment of global-scale model performance for global and regional ozone distributions, variability, and trends , 2018 .

[79]  Peter Braesicke,et al.  The Impact of Stratospheric Ozone Feedbacks on Climate Sensitivity Estimates , 2018 .

[80]  L. Polvani,et al.  The response of the ozone layer to quadrupled CO2 concentrations. , 2018, Journal of climate.

[81]  Thomas Vandal,et al.  Intercomparison of machine learning methods for statistical downscaling: the case of daily and extreme precipitation , 2017, Theoretical and Applied Climatology.

[82]  The response of the ozone layer to quadrupled CO2 concentrations. , 2019, Journal of climate.

[83]  1.2 The impact of solar variability on climate , 2020, Earth's climate response to a changing Sun.