| Summary: | My PhD research focuses on the role of extratropical processes on the variability and prediction of tropical cyclone (TC) activity. In my first Ph.D. project, I examined the relative importance of local and remote forcing on TC frequency using the WRF model. I showed that the remote processes outside the North Atlantic, particularly extratropical Rossby wave breaking, play an important role in modulating the interannual variability of Atlantic TC frequency and that the remote impacts may exceed the direct impacts of local SST in some years. In my second project, I developed a skillful hybrid prediction scheme for multi-year variability of Atlantic TC activity. I employed a Poisson regression model and took SST indices averaged over the Atlantic main development region (MDR) and the Atlantic subpolar gyre region (SPG) from the CESM large-ensemble hindcasts as predictors. The model skillfully predicts various Atlantic TC indices (including landfalling TC frequency) during the past ~60 years. I found that the Atlantic SPG SST is a more important source of predictability than the Atlantic MDR SST on the multi-year time scale. I also explored the optimal ensemble size for skillful hybrid prediction and compared initialized CESM hindcasts with uninitialized ones to investigate the roles of internal variability and external forcing in TC prediction. My third research project focused on summertime stationary waves, which provide a unified dynamic framework integrating tropical and extratropical impacts on TC activity. I investigated the interannual variability of summertime subtropical stationary waves, their response to anthropogenic forcing, and the implications for regional TC projection. Through observational analysis, I found that the variability of the subtropical stationary waves can be largely explained by the longitudinal displacement of the zonal wavenumber-1 component and the intensity change of the zonal wavenumber-2 component. Both aspects are significantly correlated to the variability of TC activity over the North Pacific and North Atlantic. The connection of subtropical stationary waves to regional TC activity can be explained by variations in environmental vertical wind shear, tropospheric humidity, and frequency of Rossby wave breaking. It is also found that the stationary waves are strongly related to precipitation anomalies over various oceanic regions, indicating the impacts of low-frequency climate modes. To disentangle the relative impacts of different climate modes, semi-idealized CESM2 experiments are carried out to investigate the response of stationary waves to different SST forcing, and results show that stationary waves are strongly modulated by the ENSO, Pacific meridional mode, and variations in the tropical North Atlantic SST. Finally, a simple stationary wave model is used to identify the underlying physical mechanisms, and diabatic heating is demonstrated to play a central role in driving stationary wave variability on the interannual timescale. CMIP6 projections show that the Pacific tropical upper-tropospheric trough (TUTT) will likely weaken and the Atlantic TUTT will likely strengthen in a warming climate, which suggests the increasing TC activity over the Central Pacific and decreasing TC activity over the North Atlantic. Further examination indicates that the inter-model spread in Pacific TUTT projection is closely related to the warming contrast of inter-basin SST over the tropical North Pacific and tropical North Atlantic and that the inter-model spread in Atlantic TUTT is strongly regulated by the relative SST warming over the Atlantic main development region, with respect to the global tropical mean. Impacts of SST changes on TUTT projections can be understood through their modulation of tropical precipitation and latent heating. Overall, stationary waves provide a useful framework to understand the impacts of various forcing agents on TC activity over the North Atlantic and North Pacific. My analysis helps better understand the underlying mechanisms for the variability of TC activity and can increase our confidence in TC projection.
|
|---|