🎉 Our paper on parametric and predictive uncertainty quantification in the E3SM Land Model has been published in JAMES!

Welcome 👋

Summary

Our paper, “A Framework for Parametric and Predictive Uncertainty Quantification in the E3SM Land Model: Assessing Site and Observable Generalizability,” has been published in the Journal of Advances in Modeling Earth Systems (JAMES)! In this work, we develop an integrated uncertainty quantification framework that combines parameter perturbation experiments (PPE), statistical emulation, and Bayesian inference to efficiently constrain model parameters and improve predictive capability. Using FLUXNET data from multiple evergreen forest sites, we show that posterior parameter distributions vary substantially across sites and observables, even within the same plant functional type, highlighting the importance of context-dependent observational constraints. At the core of this study is a PPE–emulator–MCMC framework that enables scalable uncertainty propagation and Bayesian calibration, transforming traditionally computationally expensive workflows into tractable inference problems. By leveraging a Gaussian process emulator, the framework accelerates simulations by orders of magnitude and supports efficient sensitivity analysis, calibration, and probabilistic prediction. This framework further supports model–observation system co-design by enabling OSSE-informed optimal experimental design, helping identify the most informative observations for reducing uncertainty. It also demonstrates the potential of surrogate-based (AI/ML-enabled) approaches for accelerating uncertainty quantification in Earth system models, ultimately strengthening predictive capability for hydrologic and carbon-cycle processes relevant to climate–energy system planning. If you are interested, feel free to check out the paper or reach out to discuss!

Jiang, Z., Isenberg, N. M., Subba, T., Woo, H.‐M., Serbin, S. P., Urban, N. M., & Kuang, C. (2026). A framework for parametric and predictive uncertainty quantification in the E3SM Land Model: Assessing site and observable generalizability. Journal of Advances in Modeling Earth Systems, 18, e2025MS005562. https://doi.org/10.1029/2025MS005562