Diagnosing Earth System Model Biases

Last updated: February 20, 2026

Earth System Models (ESMs) are essential---though imperfect---tools for understanding mechanisms of present and future climate variability. Across generations of model development, conspicuous biases in their simulations of circulation and precipitation have persisted. I am interested in the relative degree to which parameterized processes (e.g., cloud microphysics, convection) versus resolved elements (e.g., topography) contribute to these pervasive model biases, though with particular emphasis on the latter.

Mountains and high terrain strongly affect winds, vertical motion, and moisture transport, helping to organize major climate features such as storm tracks and monsoon systems. As ESMs operate at finite spatial resolution, they rely on simplified representations of Earth's topography that are generally overly-smooth with reduced peak heights. The strength of topography-atmosphere interactions in ESMs thus varies depending on how complex terrain is represented.

Through a combination of multi-model assessments of ESMs with varying spatial resolutions, idealized experiments implementing elevated surface height boundary conditions in coupled and atmosphere-only ESMs, and perturbed parameter ensembles that simultaneously adjust microphysical parameterizations around modified topography, my research has taken a multi-pronged approach to investigating the influence of topography on model biases. The ultimate goal of this work is to better understand the mechanical and thermodynamical processes that link orography to regional circulation. By generating such an understanding of the physical sources of model biases, we can hopefully identify pathways for improving the fidelity and reliability of Earth System Models, particularly in regions where climate is strongly influenced by terrain.

RELATED PUBLICATIONS:

• Meegan-Kumar, D., Elsaesser, G.S., Battisti, D.S., Colose, C., Wu, J., Sexton, J., Baldwin, J.W. (2025). Optimizing Topographic Boundary Conditions for East Pacific Climate Simulation. Journal of Climate, 38 (11), 2497-2524. DOI: 10.1175/JCLI-D-24-0316.1

• Meegan-Kumar, D., Baldwin, J. W. (in prep) Topography Shapes Distinct Patterns of North American Monsoon Rainfall Biases with Increasing Earth System Model Resolution. For submission to AGU Advances

• Olivas Ordoñez, K., Meegan-Kumar, D., Elsaesser, G.S., Wu, J., Baldwin, J.W. (in prep) Influence of Elevated Andes Topography on South American Precipitation Biases. For submission to Geophysical Research Letters

• Kramer, S.M., Karnauskas, K.B., Elling, M.T., Zhang, L., Liu, H., Chen, Y., Amaya, D.J., Nazarenko, L., Yang, W., Vecchi, G.A., Meegan-Kumar, D., Baldwin, J.W., Samanta, D. ColdBlobMIP: A Multi-Model Assessment of the Atmospheric Response to the North Atlantic Warming Hole. Geophysical Research Letters, 52, e2025GL117784. DOI: 10.1029/2025GL117784