Modeling Tropical Cyclone Induced Inland Flooding at Tar Pamlico River Basin of North Carolina

Landfalling tropical cyclones often produce heavy precipitation and result in river and flash floods. Such floods can not only cause loss of human lives and properties, but also lead to ecological disasters in the affected watershed areas, estuaries and coastal waters. In order to better understand and simulate large coastal watershed hydrology and hydro-meteorological processes associated with tropical cyclones (TC) - induced inland flooding, the Weather Research and Forecasting (WRF) model and the Annualized Agricultural Nonpoint Source Pollution Model (AnnAGNPS) have been employed in this study. The study focuses on four major hydro-meteorological identities and their interactions: 1) previous rainfall events, 2) synoptic atmospheric environment, 3) landfalling hurricane, and 4) surface and ground water hydrology. The research is divided into two parts. Part one focuses on the investigation of the impacts of previous rainfall events on watershed surface runoff while part two studies the impacts of the synoptic atmospheric environment on landfalling hurricanes and the resulting effect on surface runoff. Hurricane Floyd was chosen in this study as a special case because it produced massive flooding as a result of the combined effects of previous rainfall events from Hurricane Dennis and the synoptic atmospheric environment.;The modeling results indicate that the AnnAGNPS model performs well in predicting the total amount of watershed runoff. However Muskingum channel routing is needed for AnnAGNPS to improve the hydrographs of flow discharge during hurricane events. Sensitivity analysis of soil saturated hydrological conductivity (Ks) indicates that both base flow and event total runoff are sensitive to Ks. Base flow increases as Ks increases when K s ≥15 m/day, but slightly decreases when K s > 15 m/day which is out of assumption of linear relationship from Darcy's law. Peak runoff exponentially decreases as Ks increases. The results show that without the preceding Hurricane Dennis, the outlet discharge as a result of Hurricane Floyd would have been as much as 37% lower than that caused by the combined Dennis-Floyd effect.;Precipitation during the landfall of Hurricane Floyd (1999) was simulated by using the WRF model with two-way nested domains. The horizontal grid spacing of the inner-most domain is 2 km. The WRF model could reproduce the northeast- southwest oriented narrow and intense band of precipitation that developed just inland of the coast over North Carolina. The distribution pattern of accumulated precipitation is consistent with the observed precipitation pattern. The magnitudes of the simulated precipitation are generally within 30% of the observed values at the weather stations in the study area.;The effects of large-scale atmospheric environment on Hurricane Floyd - induced precipitation and flooding have also been investigated. Through a vortex removal technique, Hurricane Floyd vortex was mostly removed to obtain the large-scale atmospheric environmental field at the model initial time. It is shown that the environment-induced precipitation can be as high as 21.5% of total precipitation in domain 3 which covers most of the North Carolina and 7.3% in the focused hydrological study area. Idealized hydrological simulation results demonstrate that without large scale environmental impacts the discharge would have been as much as 10.4% lower at the Tarboro gauge station if assuming TC and synoptic environment interaction is linear.;Using simulated precipitation from the WRF, AnnAGNPS model along with Muskingum routing was able to reproduce the hydrograph and total volume of surface runoff at the watershed outlet. This implies that with improved precipitation forecasts from numerical weather prediction models such as WRF, it is possible to make skillful river runoff forecasts using distributed hydrological models.