Toward machine-assisted tuning avoiding the underestimation of uncertainty in climate change projections

Hourdin, F.; Ferster, Brady; Deshayes, Julie; Mignot, Juliette; Musat, Ionela; Williamson, Daniel

doi:10.1126/sciadv.adf2758

Cited by 8 publications

(4 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent advances in model calibration (e.g., Hourdin et al, 2021Hourdin et al, , 2023 will be instrumental in better designing future PPE.…”

Section: 2mentioning

confidence: 99%

Bringing it all together: Science and modelling priorities to support international climate policy

Jones,

Adloff,

Booth

et al. 2024

Preprint

View full text Add to dashboard Cite

Abstract. We review how the international modelling community, encompassing Integrated Assessment models, global and regional Earth system and climate models, and impact models, have worked together over the past few decades, to advance understanding of Earth system change and its impacts on society and the environment, and support international climate policy. We then recommend a number of priority research areas for the coming ~6 years (i.e. until ~2030), a timescale that matches a number of newly starting international modelling activities and encompasses the IPCC 7th Assessment Report (AR7) and the 2nd UNFCCC Global Stocktake. Progress in these areas will significantly advance our understanding of Earth system change and its impacts and increase the quality and utility of science support to climate policy. We emphasize the need for continued improvement in our understanding of, and ability to simulate, the coupled Earth system and the impacts of Earth system change. There is an urgent need to investigate plausible pathways and emission scenarios that realize the Paris Climate Targets, including pathways that overshoot the 1.5 °C and 2 °C targets, before later returning to them. Earth System models (ESMs) need to be capable of thoroughly assessing such warming overshoots, in particular, the efficacy of negative CO2 emission actions in reducing atmospheric CO2 and driving global cooling. An improved assessment of the long-term consequences of stabilizing climate at 1.5 °C or 2 °C above pre-industrial temperatures is also required. We recommend ESMs run overshoot scenarios in CO2-emission mode, to more fully represent coupled climate - carbon cycle feedbacks. Regional downscaling and impact models should also use forcing data from these simulations, so impact and regional climate projections are as realistic as possible. An accurate simulation of the observed record remains a key requirement of models, as does accurate simulation of key metrics, such as the Effective Climate Sensitivity. For adaptation, improved guidance on potential changes in climate extremes and the modes of variability these extremes develop in, is a key demand. Such improvements will most likely be realized through a combination of increased model resolution and improvement of key parameterizations. We propose a deeper collaboration across modelling efforts targeting increased process realism and coupling, enhanced model resolution, parameterization improvement, and data-driven Machine Learning methods. With respect to sampling future uncertainty, increased collaboration between approaches that emphasize large model ensembles and those focussed on statistical emulation is required. We recommend increased attention is paid to High Impact Low Likelihood (HILL) outcomes. In particular, the risk and consequences of exceeding critical tipping points during a warming overshoot. For a comprehensive assessment of the impacts of Earth system change, including impacts arising directly from specific mitigation actions, it is important detailed, disaggregated information from the Integrated Assessment Models (IAMs) used to generate future scenarios is available to impact models. Conversely, methods need to be developed to incorporate potential future societal responses to the impacts of Earth system change into scenario development. Finally, the new models, simulations, data, and scientific advances, proposed in this article will not be possible without long-term development and maintenance of a robust, globally connected infrastructure ecosystem. This system must be easily accessible and useable across all modelling communities and across the world, allowing the global research community to be fully engaged in developing and delivering new scientific knowledge to support international climate policy.

show abstract

“…Recent advances in model calibration (e.g., Hourdin et al, 2021Hourdin et al, , 2023 will be instrumental in better designing future PPE.…”

Section: 2mentioning

confidence: 99%

Bringing it all together: Science and modelling priorities to support international climate policy

Jones,

Adloff,

Booth

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…As computers become more powerful, we can run land surface models as ensembles instead of a single realisation of the model, allowing us to obtain rigorously the uncertainty of the model prediction. Indeed, as a climate community, we should be moving towards using dataconstrained ensembles instead of a single realisation (Hourdin et al, 2023). HM would allow us to generate such ensembles to be used, for example, the Coupled Model Intercomparison Project.…”

Section: Future Avenuesmentioning

confidence: 99%

“…It has since been used in various domains of science and engineering, such as galaxy formation (Bower et al, 2010;Vernon et al, 2014), disease modelling (Andrianakis et al, 2015), systems biology models (Vernon et al, 2022), and traffic (Boukouvalas et al, 2014). In climate sciences, HM was also used to calibrate climate models of different complexities (Edwards et al, 2011;Williamson et al, 2013Williamson et al, , 2015Hourdin et al, 2023), ocean models (Williamson et al, 2017;Lguensat et al, 2023), atmospheric models (Couvreux et al, 2021;Hourdin et al, 2021;Villefranque et al, 2021) and ice sheet models (McNeall et al, 2013).…”

mentioning

confidence: 99%

Exploring the Potential of History Matching for Land Surface Model Calibration

Raoult,

Beylat,

Salter

et al. 2024

Preprint

View full text Add to dashboard Cite

Abstract. With the growing complexity of land surface models used to represent the terrestrial part of wider Earth system models, the need for sophisticated and robust parameter optimisation techniques is paramount. Quantifying parameter uncertainty is essential for both model development and more accurate projections. In this study, we assess the power of history matching by comparing results to variational data assimilation, commonly used in land surface models for parameter estimation. Although both approaches have different setups and goals, we can extract posterior parameter distributions from both methods and test the model-data fit of ensembles sampled from these distributions. Using a twin experiment, we test whether we can recover known parameter values. Through variational data assimilation, we closely match the observations. However, the known parameter values are not always contained in the posterior parameter distribution, highlighting the equifinality of the parameter space. In contrast, while more conservative, history matching still gives a reasonably good fit and provides more information about the model structure by allowing for non-Gaussian parameter distributions. Furthermore, the true parameters are contained in the posterior distributions. We then consider history matching's ability to ingest different metrics targeting different physical parts of the model, helping to reduce parameter space further and improve model-data fit. We find the best results when history matching is used with multiple metrics; not only is the model-data fit improved, but we also gain a deeper understanding of the model and how the different parameters constrain different parts of the seasonal cycle. We conclude by discussing the potential of history matching in future studies.

show abstract

“…computing-intensive, way to calibrate a climate model is to generate a perturbed parameter ensemble (PPE), in which model parameters are varied within expert-defined ranges (Bellprat et al, 2012;Hauser et al, 2012). PPEs have been shown to allow a thorough exploration of model parametric uncertainty regarding global climate sensitivity (Piani et al, 2005;Sanderson et al, 2008;Hourdin et al, 2023). Here our main focus is to explore and quantify atmospheric model parametric uncertainty in the estimation of contrail radiative forcing, the model being ARPEGE-climat, the atmospheric component of the CNRM-CM6-1 model (Voldoire et al, 2019).…”

mentioning

confidence: 99%

Parametric Sensitivity and Constraint of Contrail Cirrus Radiative Forcing in the Atmospheric Component of CNRM-CM6-1

Perini,

Terray,

Cariolle

et al. 2023

Preprint

View full text Add to dashboard Cite

Abstract. The impact of aviation on climate change due to CO2 emissions no longer needs to be demonstrated. However, the impact of non-CO2 effects such as those from contrails is still subject to large uncertainties. An often neglected source of uncertainty comes from climate model sensitivity to numerical parameters representing subgrid-scale processes. Here we investigate the sensitivity of contrail radiative forcing due parametric uncertainty based on the atmospheric component of the CNRM-CM6-1 coupled model. A perturbed parameter ensemble is generated from the sampling of twenty-two adjustable parameters involved in convection, cloud microphysics and radiative transfer processes. A surrogate model based on multi-linear regression is used to explore the full range of contrail radiative forcing due to parametric uncertainty. Based on an optimization algorithm and a climatological skill score, we find a constrained range of contrail radiative forcing from equally skillful model versions with different sets of parameters. We find a contrail radiative forcing best-estimate of 56 mW.m-2 with a 5–95 % confidence interval of 38–70 mW.m-2. Finally, a sensitivity analysis shows that model parameters controlling contrail's lifetime play a major role in the estimation of contrail radiative forcing.

show abstract

Toward machine-assisted tuning avoiding the underestimation of uncertainty in climate change projections

Cited by 8 publications

References 59 publications

Bringing it all together: Science and modelling priorities to support international climate policy

Bringing it all together: Science and modelling priorities to support international climate policy

Exploring the Potential of History Matching for Land Surface Model Calibration

Parametric Sensitivity and Constraint of Contrail Cirrus Radiative Forcing in the Atmospheric Component of CNRM-CM6-1

Contact Info

Product

Resources

About