The El Niño-Southern Oscillation Modulation of Tropical Cyclones in CMIP6 Models

The El Niño-Southern Oscillation (ENSO) is known to modulate global tropical cyclone (TC) activity, which is also known as the ENSO-TC relationship. In response to anthropogenetic induced global warming, the intensity, frequency, and nature of ENSO may evolve, partly due to changes in the mean state of the tropical Pacific upon which ENSO operates. Moreover, TC’s characteristics change with climate change. Consequently, these changes may also impact the ENSO-TC relationship.

The present work studies the ENSO-TC relationship within 12 CMIP6 models. First, we explore the sensitivity of the ENSO-TC relationship to the index used to describe ENSO activity in models, given their biases in the spatial pattern of SST anomalies during ENSO events. The CMIP6 TCs are downscaled using the Columbia tropical cyclone HAZard model (CHAZ), a statistical-dynamical downscaling model, using CMIP6 models’ environmental fields. A thorough comparison will then be made between the CMIP6 models’ ENSO-TC relationship and the observed one, followed by an assessment of the anticipated changes in this relationship based on future projections. Preliminary results suggest while CMIP6 models capture the ENSO variability of tropical Pacific SST well, the observed ENSO-TC relationship is not well-captured in the 12 CMIP6 models downscaled here. Among the models, GISS-E2-1-G exhibits an ENSO-TC relationship that mostly aligns with observations, while CNRM-CN6-1's relationship deviates the most. As the climate continues to warm up, the majority (9) of the 12 CMIP6 models project an increase in anomalous TC activity during El Niño years in the South Pacific regions and during La Niña years in the South Indian Ocean basin. In the North Pacific, models project an increase in anomalous TC activity during La Niña years. However, no clear consensus is observed in other TC basins in the Northern Hemisphere. These changes will be discussed by investigating TC-relevant large-scale environmental fields from CMIP6 models.