The Effects of High Horizontal Resolution on Tropical Cyclones’ Potential Intensity and Upper Ocean Heat Content in the Community Earth System Model

The potential intensity (PI) of tropical cyclones (TCs) defines the theoretical maximum strength of a tropical storm within a given large-scale environmental setting. The upper ocean, as the primary interface for TC interaction, plays a crucial role in this dynamic: its warmth acts as fuel for the development of TCs, while storm-induced mixing brings colder water from the thermocline to the surface, consequently reducing sea surface temperature. The vertical temperature profile of the upper ocean is thus a pivotal factor in both the development of TCs, including their rapid intensification, and the feedback they give to the ocean itself. Previous research suggests that increasing the resolution in Global Climate Models (GCMs) enhances the simulation of TCs. With advancements in high-performance computing, there is a transition to higher-resolution models due to their improved representation of climate mean states and extreme weather events. Here we report on the influence of varying horizontal model resolutions on coupled ocean-atmosphere variables, such as PI, and on upper ocean thermal properties.

In our study, we analyze simulations data from the Community Earth System Model version 1.3 (CESM1.3), conducting experiments in both a standard low-resolution (LR) configuration and a high-resolution (HR) counterpart. The simulation period covers 250 years, extending from 1850 to 2100. The LR simulation employs a 1° atmospheric resolution and a nominal 1° ocean and sea-ice resolution. The HR setup, by contrast, features a 0.25° atmospheric resolution and a 0.1° ocean and sea-ice resolution. Additionally, data from six models within the European PRIMAVERA (PRocess-based climate sIMulation: AdVances in high-resolution modeling and European climate Risk Assessments) project is examined to validate the consistency of our findings with those from CESM. Our research addresses the following questions: 1. How do the magnitudes and spatial distributions of PI compare between the LR and HR configurations? 2. What variations exist in the mixing layer depth, ocean heat potential, and variable mixing length among these resolutions? 3. What factors contribute to these simulation differences? 4. What implications do these results hold for projections of future TC activity? We find that increasing horizontal resolution of the ocean model, which allows for some effects of ocean eddies to be explicitly resolved, has a larger impact on potential intensity calculations than changing atmospheric model resolution alone. Documenting these dependencies is an important step to understanding uncertainties in future projections of TC climatologies using GCMs with different spatial resolutions.