Bolivar et al. (2025) How Do Tropical Cyclones Directly Simulated in High-Resolution Climate Models Differ from Statistically Dynamically Generated Storms?

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Journal of Climate
• Year: 2025
• Date: 2025-12-02
• Authors: Ana J. Bolivar, Colin M. Zarzycki
• DOI: 10.1175/jcli-d-24-0454.1

Research Groups

[Information not available in the provided abstract.]

Short Summary

This study compares high-resolution climate models and statistical–dynamical downscaling (SDD) models for simulating landfalling tropical cyclones (TCs). It finds that while SDD offers computational advantages and improves some metrics, it also introduces unphysical behaviors and overrepresents landfalling storms, highlighting distinct biases in both techniques that can be better understood through their comparison.

Objective

• To compare the simulation of landfalling tropical cyclones (TCs) using high-resolution climate models and statistical–dynamical downscaling (SDD) models to assess their respective strengths, weaknesses, and inherent biases for studying TC climatology.

Study Configuration

• Spatial Scale: Global (HighResMIP, GCMs) with a focus on landfalling TCs in the North Atlantic Ocean.
• Temporal Scale: Climatological, utilizing subdaily kinematic and monthly thermodynamic fields for forcing.

Methodology and Data

• Models used: High-resolution climate models (from HighResMIP), Statistical–Dynamical Downscaling (SDD) models.
• Data sources: Objectively tracked TCs from High-Resolution Model Intercomparison Project (HighResMIP) output, observed landfalls from the International Best Track Archive for Climate Stewardship (IBTrACS), reanalysis storm tracks, HighResMIP subdaily kinematic and monthly thermodynamic fields.

Main Results

• SDD models are computationally less expensive and improve several metrics compared to global climate models (GCMs).
• SDD introduces unphysical behaviors in TCs that are not observed in GCMs.
• Downscaling with SDD leads to more uniform behavior across models but overrepresents the proportion of landfalling storms, even after correcting for genesis location.
• SDD TC climatologies are more sensitive to the choice of model forcing than to the grid spacing of the model forcing.
• Comparing both techniques provides insights into the biases inherent in HighResMIP TC climatology.
• A simple metric for TC seed survival reveals that mechanisms contributing to biases in TC climatology vary among HighResMIP models.

Contributions

• Provides a direct comparison of high-resolution global climate models and statistical–dynamical downscaling models for simulating landfalling tropical cyclones, a critical aspect for risk assessment.
• Identifies specific advantages of SDD (computational cost, some metric improvements) and significant disadvantages (unphysical behaviors, overrepresentation of landfalls).
• Highlights that using both techniques together can reveal and help target key areas for improvement in TC climatology simulations.
• Offers insights into the varying mechanisms contributing to biases across different HighResMIP models, enhancing understanding of model limitations.
• Crucial for improving confidence in future studies on tropical cyclone changes and their implications for community preparedness.

Funding

[Information not available in the provided abstract.]

Citation

@article{Bolivar2025How,
  author = {Bolivar, Ana J. and Zarzycki, Colin M.},
  title = {How Do Tropical Cyclones Directly Simulated in High-Resolution Climate Models Differ from Statistically Dynamically Generated Storms?},
  journal = {Journal of Climate},
  year = {2025},
  doi = {10.1175/jcli-d-24-0454.1},
  url = {https://doi.org/10.1175/jcli-d-24-0454.1}
}