- 1Institute for Atmospheric and Climate Science, ETH Zurich, Zurich, Switzerland (nora.zilibotti@env.ethz.ch)
- 2Department of Applied Mathematics and Theoretical Physics, University of Cambridge, Cambridge, UK
Earth system models are widely used to make projections not only about the mean atmospheric state under climate warming, but also about the circulation on synoptic to seasonal timescales and their related weather extremes. However, the statistics and characteristics of synoptic weather systems, such as extratropical cyclones, exhibit substantial biases relative to observations and reanalysis data. Although model resolution and the representation of moist processes have been pinpointed as important contributors to these biases, the exact pathway by which they affect the cyclone evolution and the coupling between the surface and upper-level flow needs further investigation.
Here, we present a spectral analysis that reveals pronounced biases in the extratropical upper-level kinetic energy, especially at the upper end of synoptic scales, when comparing Community Earth System Model version 2 large ensemble simulations (CESM2-LENS) to ERA5. Focusing on the North Pacific storm track, we show that upper-level eddy kinetic energy (EKE) is underestimated by up to 30% and upper-level forcing as measured by QG omega forcing is reduced in CESM2. In addition, we observe differences in the vertical structure of diabatic heating between CESM2 and ERA5. CESM2 exhibits weak and permanent heating in the planetary boundary layer, whereas ERA5 shows more intermittent, localised heating that extends further into the free troposphere. We discuss possible relationships between these biases and cyclone properties in the North Pacific storm track. This provides a pathway by which model biases in both the upper and lower levels can influence the structure and evolution of extratropical cyclones, potentially amplifying upper-level errors.
How to cite: Zilibotti, N., Wernli, H., and Schemm, S.: An in-depth analysis of the North Pacific storm track bias in CESM2-LENS, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13451, https://doi.org/10.5194/egusphere-egu26-13451, 2026.