Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

A Spatiotemporal-Aware Weighting Scheme for Improving Climate Model Ensemble Predictions

Journal Article · · Journal of Machine Learning for Modeling and Computing
Multimodel ensembling has been widely used to improve climate model predictions, and the improvement strongly depends on the ensembling scheme. In this work, we propose a Bayesian neural network (BNN) ensembling method, which combines climate models within a Bayesian model averaging framework, to improve the predictive capability of model ensembles. Our proposed BNN approach calculates spatiotemporally varying model weights and biases by leveraging individual models' simulation skill, calibrates the ensemble prediction against observations by considering observation data uncertainty, and quantifies epistemic uncertainty when extrapolating to new conditions. More importantly, the BNN method provides interpretability about which climate model contributes more to the ensemble prediction at which locations and times. Thus, beyond its predictive capability, the method also brings insights and understanding of the models to guide further model and data development. In this study, we design experiments using an ensemble of CMIP6 climate model simulations to illustrate the BNN ensembling method's capability with respect to prediction accuracy, interpretability, and uncertainty quantification (UQ). We demonstrate that BNN can correctly assign larger weights to the regions and seasons where the individual model fits the observation better. Moreover, its offered interpretability is consistent with our understanding of localized climate model performance. Additionally, BNN shows an increasing uncertainty when the prediction is farther away from the period with constrained data, which appropriately reflects our trustworthiness of the models in the changing climate.
Research Organization:
Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
Sponsoring Organization:
USDOE Laboratory Directed Research and Development (LDRD) Program; USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR); USDOE Office of Science (SC), Biological and Environmental Research (BER)
Grant/Contract Number:
AC05-00OR22725
OSTI ID:
1906596
Journal Information:
Journal of Machine Learning for Modeling and Computing, Journal Name: Journal of Machine Learning for Modeling and Computing Journal Issue: 4 Vol. 3; ISSN 2689-3967
Publisher:
Begell HouseCopyright Statement
Country of Publication:
United States
Language:
English

References (37)

Systematic land climate and evapotranspiration biases in CMIP5 simulations: CMIP5 BIASES journal January 2014
A climate model projection weighting scheme accounting for performance and interdependence: Model Projection Weighting Scheme journal January 2017
Weighting of model results for improving best estimates of climate change journal September 2009
Climate model dependence and the replicate Earth paradigm journal December 2012
Regional and seasonal intercomparison of CMIP3 and CMIP5 climate model ensembles for temperature and precipitation journal February 2014
Multi-model ensemble predictions of precipitation and temperature using machine learning algorithms journal May 2020
Performance metrics for climate models journal January 2008
Evaluating the present-day simulation of clouds, precipitation, and radiation in climate models journal January 2008
Prospects and Caveats of Weighting Climate Models for Summer Maximum Temperature Projections Over North America journal May 2018
Future Projections of Global Pluvial and Drought Event Characteristics journal November 2018
Robust Future Changes in Meteorological Drought in CMIP6 Projections Despite Uncertainty in Precipitation journal June 2020
Evaluation of CMIP6 GCMs Over the CONUS for Downscaling Studies journal November 2022
Global assessment of trends in wetting and drying over land journal September 2014
Climate change will affect global water availability through compounding changes in seasonal precipitation and evaporation journal June 2020
Taking climate model evaluation to the next level journal January 2019
Improving multiple model ensemble predictions of daily precipitation and temperature through machine learning techniques journal March 2022
Quantifying uncertainty in European climate projections using combined performance-independence weighting journal November 2019
Geographical and Temporal Weighted Regression (GTWR) journal March 2015
Challenges in Combining Projections from Multiple Climate Models journal May 2010
An Overview of CMIP5 and the Experiment Design journal April 2012
Simulation of Daily Extreme Precipitation over the United States in the CMIP5 30-Yr Decadal Prediction Experiment journal April 2019
Improving Antarctic Total Ozone Projections by a Process-Oriented Multiple Diagnostic Ensemble Regression journal December 2013
A Representative Democracy to Reduce Interdependency in a Multimodel Ensemble journal July 2015
Climate Model Dependence and the Ensemble Dependence Transformation of CMIP Projections journal March 2015
Constraining Future Summer Austral Jet Stream Positions in the CMIP5 Ensemble by Process-Oriented Multiple Diagnostic Regression journal January 2016
Is Institutional Democracy a Good Proxy for Model Independence? journal November 2016
Projecting ozone hole recovery using an ensemble of chemistry–climate models weighted by model performance and independence journal August 2020
ERA5-Land: a state-of-the-art global reanalysis dataset for land applications journal January 2021
Connecting spatial and temporal scales of tropical precipitation in observations and the MetUM-GA6 journal January 2017
Skill and independence weighting for multi-model assessments journal January 2017
Climate pattern-scaling set for an ensemble of 22 GCMs – adding uncertainty to the IMOGEN version 2.0 impact system journal January 2018
European daily precipitation according to EURO-CORDEX regional climate models (RCMs) and high-resolution global climate models (GCMs) from the High-Resolution Model Intercomparison Project (HighResMIP) journal January 2020
A spatiotemporal weighted regression model (STWR v1.0) for analyzing local nonstationarity in space and time journal December 2020
MSDM v1.0: A machine learning model for precipitation nowcasting over eastern China using multisource data journal June 2021
Latent Linear Adjustment Autoencoder v1.0: a novel method for estimating and emulating dynamic precipitation at high resolution journal August 2021
The software architecture of climate models: a graphical comparison of CMIP5 and EMICAR5 configurations journal January 2015
Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization journal January 2016

Similar Records

Multimodel Ensemble Predictions of Precipitation using Bayesian Neural Networks
Conference · Fri Apr 01 00:00:00 EDT 2022 · OSTI ID:1871085

Multimodel ensemble predictions of precipitation using bayesian neural networks
Conference · Fri Apr 01 00:00:00 EDT 2022 · OSTI ID:1869077

Uncertainty quantification of a deep learning fuel property prediction model
Journal Article · Sat Sep 23 20:00:00 EDT 2023 · Applications in Energy and Combustion Science · OSTI ID:2536669