The Influence of Tropopause Temperature Biases on Climate Model Simulations of Tropical Cyclones

Abstract Potential intensity (PI) is a key indicator for tropical cyclone (TC) activity, yet it exhibits considerable variability across global climate model (GCM) simulations, even with identical sea surface temperatures (SSTs). We show that the spread in PI across GCMs is primarily driven by differences in outflow temperature, a consequence of different upper atmospheric temperatures. To explore the impacts of these biases on TC activity, we conduct several idealized experiments with altered temperature profiles. In these experiments, global TC frequency and accumulated cyclone energy change by ∼35% and hurricane frequency by ∼80%. There are smaller but still significant impacts on lifetime maximum intensity. These findings highlight an underappreciated role of upper atmospheric model biases in modulating TC activity in GCMs, how future changes in TC activity may be influenced by responses of upper atmospheric temperature to anthropogenic emissions, and that TCs are more directly influenced by PI than SST alone.