ClimateBench: A benchmark dataset for data-driven climate projections

Watson-Parris, Duncan; Rao, Yuhan; Olivié, Dirk; Seland, Øyvind; Nowack, Peer; Camps‐Valls, Gustau; Stier, Philip; Bouabid, Shahine; Dewey, M. C.; Fons, Emilie; Gonzalez, Jessenia; Harder, Paula; Jeggle, Kai; Lenhardt, Julien; Manshausen, Peter; Novitasari, Maria; Ricard, Lucile; Roesch, Carla

doi:10.1002/essoar.10509765.1

Cited by 1 publication

(8 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The pattern scaling model is obtained by fitting a linear regression model using the same training data as the other models. The plain GP model is analogous to the baseline GP emulator in Watson-Parris et al (2021), but differs in two aspects: (i ) we adopt a simpler construction for the covariance with a Matérn-3/2 kernel with automatic relevance determination, and (ii ) our model takes as input global aerosols emissions whereas Watson-Parris et al ( 2021) use spatially-resolved aerosols emission maps. Scores are computed over the 2080-2100 period since the start of all SSPs is quite similar.…”

Section: Shared Socio-economic Pathways Emulationmentioning

confidence: 99%

“…This is because computing their posterior distribution involves a matrix inversion, which has a cubic computational cost in the number of training samples. Fortunately, unlike neural networks which require large amounts of data (Watson-Parris et al, 2021), GPs excel in scenarios with limited data (Rasmussen & Williams, 2005). Consequently, it is possible to develop skilful GP emulators with limited training data.…”

Section: Computational Efficiency and Scalabilitymentioning

confidence: 99%

“…Going beyond surface temperature emulation, we can explore how FaIRGP can be used to emulate precipitations. One approach is to combine the Gaussian process GP emulator for precipitation described in Watson-Parris et al (2021) with FaIRGP. By leveraging Gaussian conjugacy relationships, a natural cross-covariance between the two emulators is induced.…”

Section: Opportunities For Precipitation Emulationmentioning

confidence: 99%

“…Conducting detection and attribution studies with emulators is not exclusive to FaIRGP, and could in principle be conducted with any emulator as discussed in Watson-Parris et al (2021). However, the strength of FaIRGP lies in its ability to input temperature ranges directly into a known probability density function, providing a precise probability between 0 and 1 of such temperatures to occur under a given emission scenario.…”

Section: Application To Detection and Attributionmentioning

confidence: 99%

“…Statistical modelling enjoys greater flexibility and has produced powerful emulators, capable of successfully approximating ESMs' spatially resolved climate projections with visually indistinguishable outputs. More recently, statistically driven emulators drawing from advances in statistical machine learning have demonstrated a remarkable capacity at regressing global emission profiles onto spatially resolved temperature and precipitation maps (Watson-Parris et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Probabilistic climate emulation with physics-constrained Gaussian processes

Bouabid¹,

Sejdinović²,

Watson-Parris³

2023

Preprint

View full text Add to dashboard Cite

Emulators, or reduced complexity climate models, are surrogate Earth system models that produce projections of key climate quantities with minimal computational resources. Using time-series modeling or more advanced machine learning techniques, statistically-driven emulators have emerged as a promising venue, producing spatially resolved climate responses that are visually indistinguishable from state-of-the-art Earth system models. Yet, their lack of physical interpretability limits their wider adoption. In fact, the exploration of future emission scenarios still relies on one-dimensional energy balance models which can lack the flexibility to capture certain climate responses. Here, we propose a statistically-driven emulator that hinges on an energy balance model. Using Gaussian processes, we formulate an emulator of temperature response that exactly satisfies the physical thermal response equations of an energy balance model. The result is an emulator that (1) enjoys the flexibility of statistical machine learning models and can be fitted against observations to produce spatially-resolved predictions (2) has physically-interpretable parameters that can be used to make inference about the climate system. This model shows skilfull prediction of annual mean global distribution of temperature, even over scenarios outside the range of observed emissions of greenhouse gases and aerosols during training &#8212; improvement in RMSE of 27% against the energy balance model and of 60% against a physics-free machine learning model. In addition, the Bayesian nature of our formulation provides a principled way to perform uncertainty quantification on the predictions. We outline how the probabilistic nature of this emulator makes it a natural candidate for detection and attribution studies and discuss extension to a precipitation emulator also incorporating physical constraints. We hope that by combining the ability of machine learning techniques to capture complex patterns with the interpretability of energy balance models, our work will produce a reliable tool for efficiently and accurately exploring future climates.

show abstract

Section: Shared Socio-economic Pathways Emulationmentioning

confidence: 99%

Section: Computational Efficiency and Scalabilitymentioning

confidence: 99%

Section: Opportunities For Precipitation Emulationmentioning

confidence: 99%

Section: Application To Detection and Attributionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations