Search NASA⌕ Search

SEARCH · Search NASA

Results for “Bayesian model averaging”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2

Improved Nested Sampling and Surrogate-Enabled Comparison With Other Marginal Likelihood Estimators

Estimating marginal likelihood is of central importance to Bayesian model selection and/or model averaging. The nested sampling method has been recently used together with the Metropolis-Hasting (M-H) sampling algorithm for estimating marginal likelihood. This study develops a new implementation of nested sampling by using the DiffeRential Evolution Adaptive Metropolis (DREAMzs) sampling algorithm. The two implementations of nested sampling are evaluated for four models of a synthetic groundwater flow modeling. The DREAMzs-based nested sampling outperforms the M-H-based nested sampling in terms of their accuracy, convergence, efficiency, and stability, which is attributed to the fact that DREAMzs is more robust than M-H for parameter sampling. The nested sampling method is also compared with four other methods (arithmetic mean, harmonic mean, stabilized harmonic mean, and thermodynamic integration) commonly used for estimating marginal likelihood and posterior probability of the four groundwater models. The comparative study requires hundreds of millions of model executions, which would not be possible without using surrogate models to replace the original models. Using the arithmetic mean estimator as the reference, the comparison reveals that thermodynamic integration outperforms nested sampling in terms of accuracy and stability, whereas nested sampling is more computationally efficient to reach to convergence. Here, the harmonic mean and stabilized harmonic mean methods give the worst marginal likelihood estimation. Accurate marginal likelihood estimation is important for accurate estimation of posterior model probability and better predictive performance of Bayesian model averaging.

54 ENVIRONMENTAL SCIENCES↗

Bayesian weighting of climate models based on climate sensitivity

Using climate model ensembles containing members that exhibit very high climate sensitivities to increasing CO 2 concentrations can result in biased projections. Various methods have been proposed to ameliorate this ‘hot model’ problem, such as model emulators or model culling. Here, we utilize Bayesian Model Averaging as a framework to address this problem without resorting to outright rejection of models from the ensemble. Taking advantage of multiple lines of evidence used to construct the best estimate of the earth’s climate sensitivity, the Bayesian Model Averaging framework produces an unbiased posterior probability distribution of model weights. The updated multi-model ensemble projects end-of-century global mean surface temperature increases of 2 °C for a low emissions scenario (SSP1-2.6) and 5 °C for a high emissions scenario (SSP5-8.5). These estimates are lower than those produced using a simple multi-model mean for the CMIP6 ensemble. The results are also similar to results from a model culling approach, but retain some weight on low-probability models, allowing for consideration of the possibility that the true value could lie at the extremes of the assessed distribution. Our results showcase Bayesian Model Averaging as a path forward to project future climate change that is commensurate with the available scientific evidence.

59 BASIC BIOLOGICAL SCIENCES↗

Attributing runoff changes to climate variability and human activities: Uncertainty analysis using four monthly water balance models

Hydrological simulations to delineate the impacts of climate variability and human activities are subjected to uncertainties related to both parameter and structure of the hydrological models. To analyze the impact of these uncertainties on the model performance and to yield more reliable simulation results, a global calibration and multimodel combination method that integrates the Shuffled Complex Evolution Metropolis (SCEM) and Bayesian Model Averaging (BMA) of four monthly water balance models was proposed. The method was applied to the Weihe River Basin (WRB), the largest tributary of the Yellow River, to determine the contribution of climate variability and human activities to runoff changes. The change point, which was used to determine the baseline period (1956-1990) and human-impacted period (1991-2009), was derived using both cumulative curve and Pettitt’s test. Results show that the combination method from SCEM provides more skillful deterministic predictions than the best calibrated individual model, resulting in the smallest uncertainty interval of runoff changes attributed to climate variability and human activities. This combination methodology provides a practical and flexible tool for attribution of runoff changes to climate variability and human activities by hydrological models.

runoff change, uncertainty assessment, monthly wat↗

Quantifying uncertainty in physics-based predictions of rare-isotope production cross sections via Bayesian-inspired model averaging across nuclear mass tables

Accurate prediction of fragmentation cross sections is essential for rare-isotope beam production, planning new-isotope searches, and designing experiments to study the most exotic regions of the nuclear chart. However, existing reaction models and phenomenological cross-section parametrizations often exhibit significant deviations over broad regions of mass and charge. In this work, a Bayesian-inspired model-averaging framework is developed to combine abrasion-ablation (AA) calculations based on multiple nuclear mass tables into a single statistically weighted estimate. For the calibrated systems, the model weights are assigned empirically according to the relative quality of fit to measured cross sections, thereby reducing systematic model bias while preserving the underlying physics content of the AA description. The weights are constrained using proton-rich fragmentation data for the 78 Kr and 124 Xe projectiles. The resulting parameter trends are then propagated to the 92 Mo and 144 Sm systems through a controlled scaling procedure. In the present implementation, the excitation-energy prescription is fixed, while the averaging is performed across nuclear-mass inputs; the framework provides both weighted cross sections and associated uncertainty estimates. Applied to proton-rich fragmentation, the present approach provides a practical basis for interpolation and limited extrapolation in regions relevant to rare-isotope production. The resulting predictions are used to assess the production of very proton-rich nuclei, and candidate new isotopes are discussed.

Bayesian methods↗

Treatment of Uncertainty in Groundwater Modeling: A (Limited) Research Perspective

Groundwater systems are complex, exhibiting significant heterogeneity across multiple scales, multiple interacting processes, and nonlinear behaviors. Groundwater systems are also open, often with poorly defined boundaries and time-dependent, uncertain boundary forcings (e.g., recharge). As a consequence, a groundwater model cannot be expected to precisely reflect the system it represents and the model results are uncertain quantities. The dominant language of uncertainty is probability and the majority of research in the treatment of uncertainty in groundwater modeling has been directed at developing and improving probabilistic methods for these models. One area of recent research has been methods for assessing the impact of model uncertainty in groundwater analyses. This line of research is part of a long-term trend of increasing appreciation for the importance of providing estimates of uncertainty in groundwater model outputs. The research was motivated by observations of nonuniqueness in model parameterizations, the sensitivity of groundwater model output to model structural choices, and a history of practice that established the importance of iterative improvement in the conceptual basis of groundwater models. Among the methods proposed for assessing model uncertainty, we discuss Bayesian model averaging, it’s theoretical basis and research development. We outline critical arguments and discuss alternative approaches. Finally, we briefly describe a regulatory application of groundwater modeling that used Bayesian model averaging concepts to resolve a key conceptual model uncertainty.

54 ENVIRONMENTAL SCIENCES↗

Evaluating marginal likelihood with thermodynamic integration method and comparison with several other numerical methods

Evaluating marginal likelihood is the most critical and computationally expensive task, when conducting Bayesian model averaging to quantify parametric and model uncertainties. The evaluation is commonly done by using Laplace approximations to evaluate semianalytical expressions of the marginal likelihood or by using Monte Carlo (MC) methods to evaluate arithmetic or harmonic mean of a joint likelihood function. This study introduces a new MC method, i.e., thermodynamic integration, which has not been attempted in environmental modeling. Instead of using samples only from prior parameter space (as in arithmetic mean evaluation) or posterior parameter space (as in harmonic mean evaluation), the thermodynamic integration method uses samples generated gradually from the prior to posterior parameter space. This is done through a path sampling that conducts Markov chain Monte Carlo simulation with different power coefficient values applied to the joint likelihood function. The thermodynamic integration method is evaluated using three analytical functions by comparing the method with two variants of the Laplace approximation method and three MC methods, including the nested sampling method that is recently introduced into environmental modeling. The thermodynamic integration method outperforms the other methods in terms of their accuracy, convergence, and consistency. The thermodynamic integration method is also applied to a synthetic case of groundwater modeling with four alternative models. The application shows that model probabilities obtained using the thermodynamic integration method improves predictive performance of Bayesian model averaging. As a result, the thermodynamic integration method is mathematically rigorous, and its MC implementation is computationally general for a wide range of environmental problems.

97 MATHEMATICS AND COMPUTING↗

Quantified limits of the nuclear landscape

The chart of the nuclides is limited by particle drip lines beyond which nuclear stability to proton or neutron emission is lost. Predicting the range of particle-bound isotopes poses an appreciable challenge for nuclear theory as it involves extreme extrapolations of nuclear masses well beyond the regions where experimental information is available. Still, quantified extrapolations are crucial for a wide variety of applications, including the modeling of stellar nucleosynthesis. We use microscopic nuclear global mass models, current mass data, and Bayesian methodology to provide quantified predictions of proton and neutron separation energies as well as Bayesian probabilities of existence throughout the nuclear landscape all the way to the particle drip lines. Here, we apply nuclear density functional theory with several energy density functionals. We also consider two global mass models often used in astrophysical nucleosynthesis simulations. To account for uncertainties, Bayesian Gaussian processes are trained on the separation-energy residuals for each individual model, and the resulting predictions are combined via Bayesian model averaging. This framework allows to account for systematic and statistical uncertainties and propagate them to extrapolative predictions. We establish and characterize the drip-line regions where the probability that the nucleus is particle- bound decreases from 1 to 0. In these regions, we provide quantified predictions for one- and two-nucleon separation energies. According to our Bayesian model averaging analysis, 7759 nuclei with Z ≤ 119 have a probability of existence ≥ 0.5. The extrapolation results obtained in this study will be put through stringent tests when new experimental information on existence and masses of exotic nuclei becomes available. In this respect, the quantified landscape of nuclear existence obtained in this study should be viewed as a dynamical prediction that will be fine-tuned when new experimental information and improved global mass models become available.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Local Bayesian Dirichlet mixing of imperfect models

Abstract To improve the predictability of complex computational models in the experimentally-unknown domains, we propose a Bayesian statistical machine learning framework utilizing the Dirichlet distribution that combines results of several imperfect models. This framework can be viewed as an extension of Bayesian stacking. To illustrate the method, we study the ability of Bayesian model averaging and mixing techniques to mine nuclear masses. We show that the global and local mixtures of models reach excellent performance on both prediction accuracy and uncertainty quantification and are preferable to classical Bayesian model averaging. Additionally, our statistical analysis indicates that improving model predictions through mixing rather than mixing of corrected models leads to more robust extrapolations.

97 MATHEMATICS AND COMPUTING↗

Extracting the Pion Distribution Amplitude from Lattice QCD through Pseudo-Distributions

The Light-Cone Distribution Amplitude (LCDA) encodes the non-perturbative information of the leading Fock component of the hadron wave function, therefore required for processes including exclusive hadron production. As the Pseudo-Nambu-Goldstone boson of QCD, the nonperturbative structure of the pion is of particular interest. Progress on the Lattice QCD calculation of the pion LCDA on O(a) -improved Wilson fermion ensembles at several lattice spacings is presented. Excited-state systematics are taken into account within a Bayesian Model Averaging framework. A Renormalization-Group-Invariant (RGI) ratio of matrix elements is formed for further extraction of the pion LCDA.

Kovner, Daniel↗

Selection of polynomial chaos bases via Bayesian model uncertainty methods with applications to sparse approximation of PDEs with stochastic inputs

Generalized polynomial chaos (gPC) expansions allow us to represent the solution of a stochastic system using a series of polynomial chaos basis functions. The number of gPC terms increases dramatically as the dimension of the random input variables increases. When the number of the gPC terms is larger than that of the available samples, a scenario that often occurs when the corresponding deterministic solver is computationally expensive, evaluation of the gPC expansion can be inaccurate due to over-fitting. We propose a fully Bayesian approach that allows for global recovery of the stochastic solutions, in both spatial and random domains, by coupling Bayesian model uncertainty and regularization regression methods. It allows the evaluation of the PC coefficients on a grid of spatial points, via (1) the Bayesian model average (BMA) or (2) the median probability model, and their construction as spatial functions on the spatial domain via spline interpolation. The former accounts for the model uncertainty and provides Bayes-optimal predictions; while the latter provides a sparse representation of the stochastic solutions by evaluating the expansion on a subset of dominating gPC bases. Moreover, the proposed methods quantify the importance of the gPC bases in the probabilistic sense through inclusion probabilities. We design a Markov chain Monte Carlo (MCMC) sampler that evaluates all the unknown quantities without the need of ad-hoc techniques. The proposed methods are suitable for, but not restricted to, problems whose stochastic solutions are sparse in the stochastic space with respect to the gPC bases while the deterministic solver involved is expensive. We demonstrate the accuracy and performance of the proposed methods and make comparisons with other approaches on solving elliptic SPDEs with 1-, 14- and 40-random dimensions.

97 MATHEMATICS AND COMPUTING↗

A Bayesian framework for adsorption energy prediction on bimetallic alloy catalysts

Abstract For high-throughput screening of materials for heterogeneous catalysis, scaling relations provides an efficient scheme to estimate the chemisorption energies of hydrogenated species. However, conditioning on a single descriptor ignores the model uncertainty and leads to suboptimal prediction of the chemisorption energy. In this article, we extend the single descriptor linear scaling relation to a multi-descriptor linear regression models to leverage the correlation between adsorption energy of any two pair of adsorbates. With a large dataset, we use Bayesian Information Criteria (BIC) as the model evidence to select the best linear regression model. Furthermore, Gaussian Process Regression (GPR) based on the meaningful convolution of physical properties of the metal-adsorbate complex can be used to predict the baseline residual of the selected model. This integrated Bayesian model selection and Gaussian process regression, dubbed as residual learning, can achieve performance comparable to standard DFT error (0.1 eV) for most adsorbate system. For sparse and small datasets, we propose an ad hoc Bayesian Model Averaging (BMA) approach to make a robust prediction. With this Bayesian framework, we significantly reduce the model uncertainty and improve the prediction accuracy. The possibilities of the framework for high-throughput catalytic materials exploration in a realistic setting is illustrated using large and small sets of both dense and sparse simulated dataset generated from a public database of bimetallic alloys available in Catalysis-Hub.org.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Accounting for uncertainty in complex alluvial aquifer modeling by Bayesian multi-model approach

Alluvial aquifers by nature are complex caused by varied depositional environments. Developing a reliable groundwater model to represent an alluvial aquifer is non-trivial. Also, relying on a single best calibrated model may not be sufficient because of an inadequate choice of model parameter values. To better understand groundwater dynamics and improve model prediction reliability, this study presents a Bayesian multi-model uncertainty quantification (BMMUQ) framework to account for model parameter uncertainty in complex alluvial groundwater modeling. The methodology was applied to the agriculturally intensive Mississippi River alluvial aquifer (MRAA), Northeast Louisiana. An aquifer architecture was first constructed using 7,259 well logs in the MRAA area which covers three fluvial deposits (alluvium, braided-stream terrace, and braided-stream terrace-loess). A 12-layer MODFLOW model was then developed to address the alluvial aquifer complexity and well calibrated through a genetic algorithm. This study quantified model parameter uncertainty in hydraulic conductivity and specific storage of sand facies. Bayesian model averaging (BMA) with the Expectation Maximization (EM) algorithm was adopted to derive posterior model weights and head variances of 50 alternative conceptual groundwater flow models, and thereby obtains BMA ensemble model predictions instead of only relying on the best calibrated conceptual model. Overall, results show that an estimated around 950 million m3 of groundwater storage loss occurs in 2015 with respect to the beginning of 2004, due to high groundwater demand for irrigation in the MRAA area. Explicitly quantifying model uncertainty can produce more reliable groundwater level predictions from BMA ensemble model. The presented groundwater modeling framework improves our understanding of the MRAA and provides a valuable tool to assist agricultural water management.

54 ENVIRONMENTAL SCIENCES↗

The Global LAnd Surface Satellite (GLASS) evapotranspiration product Version 5.0: Algorithm development and preliminary validation

An accurate estimation of spatially and temporally continuous global terrestrial evapotranspiration (ET) is essential in the assessment of surface energy, water and carbon cycles. The Global LAnd Surface Satellite (GLASS) ET product Version 4.0 (v4.0) based on the Bayesian model averaging (BMA) method was generated to estimate global terrestrial ET. However, certain uncertainty for the GLASS ET product v4.0 limits its application. In this study, we introduced the deep neural networks (DNN) merging framework to improve terrestrial ET estimation for GLASS ET product Version 5.0 (v5.0) generation by integrating five satellite-derived ET products [Moderate Resolution Imaging Spectroradiometer (MODIS) ET product (MOD16), Shuttleworth–Wallace dual-source ET product (SW), Priestley–Taylor-based ET product (PT-JPL), modified satellite-based Priestley–Taylor ET product (MS-PT) and simple hybrid ET product (SIM)]. We compared the performance of DNN method against other merging methods, including GLASS ET algorithm v4.0 (BMA), the gradient boosting regression tree (GBRT) method and the random forest (RF) method, based on 195 global eddy covariance (EC) flux towers covering observations from 2000 through 2015. Validations indicated that the DNN had the highest accuracy among four merging methods across different land cover types, yielding the highest average determination coefficients (R 2 , 0.62), root-mean-squared-error (RMSE, 24.1 W/m 2 ) and Kling–Gupta efficiency (KGE, 0.77) with a of 99% confidence interval. Compared with GLASS ET algorithm v4.0, the DNN improved on the R 2 by approximately 7% (p < 0.01) and the KGE by 10%. Based on the DNN, we then generated 8-day GLASS ET product v5.0 globally with a 1 km spatial resolution from 2001 to 2015 driven by GLASS vegetation and surface net radiation (R n ) datasets and Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA2) datasets. Finally, this global terrestrial ET product provides a valuable dataset for monitoring regional and global water resources and environmental changes.

54 ENVIRONMENTAL SCIENCES↗

Bayesian Hierarchical Model Uncertainty Quantification for Future Hydroclimate Projections in Southern Hills-Gulf Region, USA

The study investigates the hierarchical uncertainty of multi-ensemble hydroclimate projections for the Southern Hills-Gulf region, USA, considering emission pathways and a global climate model (GCM) as two main sources of uncertainty. Forty projections of downscaled daily air temperature and precipitation from 2010 to 2099 under four emission pathways and ten CMIP5 GCMs are adopted for hydroclimate modeling via the HELP3 hydrologic model. This study focuses on evapotranspiration (ET), surface runoff, and groundwater recharge projections in this century. Climate projection uncertainty is characterized by the hierarchical Bayesian model averaging (HBMA) method, which segregates emission pathway uncertainty and climate model uncertainty. HBMA is able to derive ensemble means and standard deviations, arising from individual uncertainty sources, for ET, runoff, and recharge. The model results show that future recharge in the Southern Hills-Gulf region is more sensitive to different climate projections and exhibits higher variability than ET and runoff. Overall, ET is likely to increase and runoff is likely to decrease in this century given the current emission path scenarios. Runoff are predicted to have an 18% to 20% decrease and ET is predicted to have around a 3% increase throughout the century. Groundwater recharge is likely to increase in this century with a decreasing trend. Recharge would increase about 13% in the early century and will have only a 3% increase in the late century. All hydrological projections have increasing uncertainty towards the end of the century. The HBMA result suggests that the GCM uncertainty dominates the overall hydrological projection uncertainty in the early century and the mid-century. The emission pathway uncertainty becomes important in the late century.

54 ENVIRONMENTAL SCIENCES↗

Future projections and uncertainty assessment of precipitation extremes in the Korean peninsula from the CMIP5 ensemble

Abstract Projections of changes in extreme climate are sometimes predicted by multimodel ensemble methods that combine forecasts from individual simulation models using weighted averaging. One method to assign weight to each model is the Bayesian model averaging (BMA) in which posterior probability is used. For the cases of extreme climate, the generalized extreme value distribution (GEVD) is typically used. We applied the approach of GEV‐embedded BMA to a series of 35 years of the annual maximum daily precipitation data (both historical data and data gathered from simulation experiments for future periods) over the Korean peninsula as simulated by the models in the Coupled Model Intercomparison Project Phase 5 (CMIP5). Simulation data under two Representative Concentration Pathway (RCP) scenarios, namely RCP4.5 and RCP8.5, were used. Observed data and 17 CMIP5 models for 12 gird cells in Korea have been examined to predict future changes in precipitation extremes. A simple regional frequency analysis of pooling observations from three stations in each cell was employed to reduce the estimation variance and local fluctuations. A bias correction technique using the regression‐type transfer function was applied to these simulation data. Return levels spanning over 20 and 50 years, as well as the return periods relative to the reference years (1971–2005), were estimated for two future periods, namely Period 1 (2021–2050) and Period 2 (2066–2095). From these analyses, relative increase observed in the spatially averaged 20‐year (50‐year) return level was approximately 23% (16%) and 45% (36%) in the RCP4.5 and RCP8.5 experiments, respectively, by the end of the 21st century. We concluded that extreme rainfalls will likely occur two times and four times more frequently in the RCP4.5 and RCP8.5 scenarios, respectively, as compared to in the reference years by the end of the 21st century.

Lee, Youngsaeng↗

Bayesian and LASSO Regressions for Comparative Permeability Modeling of Sandstone Reservoirs

In this research, Bayesian model averaging (BMA) and least absolute shrinkage and selection operator regression (LASSO) algorithms were adopted for the core permeability modeling as a function of well log and core measurements. More specifically, the core permeability (dependent factor) was modeled given the well log and core data (independent variables) and then predicted in non-cored intervals of a well in a sandstone formation. The BMA is a stochastic linear modeling and a Bayesian parameter selection. Among 50 linear models generated as a function of the independent variables and depicted in Occam’s window, the best model of the highest posterior probability is determined. The best model has the optimal subset parameters that most influences the response factor. The core permeability modeling was again conducted using the LASSO algorithm, which adopts a completely different way of modeling and subset selection by using the penalized least-squared equation. Both BMA and LASSO resulted in a very accurate prediction of the core permeability by achieving perfect measured and calculated permeability matching. Matches for the two algorithms were quantified by computing the adjusted R-squared and root-mean-squared prediction error. In addition, results of the BMA and LASSO were compared to the conventional multiple linear regression (LM). The LASSO algorithm led to an accurate matching similar to LM, but slightly better than BMA. Therefore, both BMA and LASSO represent integrated procedures for accurate permeability estimation similar to conventional regression analysis.

58 GEOSCIENCES↗

Beyond the proton drip line: Bayesian analysis of proton-emitting nuclei

The limits of the nuclear landscape are determined by nuclear binding energies. Beyond the proton drip lines, where the separation energy becomes negative, there is not enough binding energy to prevent protons from escaping the nucleus. Predicting properties of unstable nuclear states in the vast territory of proton emitters poses an appreciable challenge for nuclear theory as it often involves far extrapolations. In addition, significant discrepancies between nuclear models in the proton-rich territory call for quantified predictions. With the help of Bayesian methodology, we mix a family of nuclear mass models corrected with statistical emulators trained on the experimental mass measurements, in the proton-rich region of the nuclear chart. Separation energies were computed within nuclear density functional theory using several Skyrme and Gogny energy density functionals. We also considered mass predictions based on two models used in astrophysical studies. Quantified predictions were obtained for each model using Bayesian Gaussian processes trained on separation-energy residuals and combined via Bayesian model averaging. Here, we obtained a good agreement between averaged predictions of statistically corrected models and experiment. In particular, we quantified model results for one- and two-proton separation energies and derived probabilities of proton emission. This information enabled us to produce a quantified landscape of proton-rich nuclei. The most promising candidates for two-proton decay studies have been identified. Finally, the methodology used in this work has broad applications to model-based extrapolations of various nuclear observables. It also provides a reliable uncertainty quantification of theoretical predictions.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

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

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.

54 ENVIRONMENTAL SCIENCES↗