Factors that contributed to the Hurricane Gaston flooding event in Richmond, Virginia

Tropical cyclone precipitation forecasting remains a difficult challenge to operational meteorologists. While precipitation forecasting skill has not appreciably improved, inland freshwater flooding has become the leading cause of death in tropical cyclones over the past 50 years. One of the major forecast challenges is predicting tropical cyclone interaction with coastlines, terrain, and mid-latitude synoptic weather systems. Weak tropical cyclones are particularly problematic for forecasters as numerical models often do not initialize the circulations or moist processes well, failing to capture mesoscale features relevant to significant precipitation events. This research focuses on Hurricane Gaston of 2004 to study the dynamics of an extreme rainfall event in Richmond, Virginia and to test numerical model improvements for their effect on precipitation forecasting.

Hurricane Gaston made landfall in South Carolina on 29 August as a minimal Category 1 hurricane. On 30 August, over 12 inches of rain fell in the Richmond, Virginia, metro area causing extensive flooding, eight deaths, and an estimated $18 million in damage. The event was largely unforecasted as the Hydrometeorological Prediction Center predicted only an inch of rain and the operational NCEP North American Mesoscale (NAM) model predicted less than one inch of rain. A full analysis of Gaston's lifecycle and evolution post-landfall is combined with an ingredients based methodology to diagnose the factors contributing to the heavy rainfall. An initial case analysis indicates that shallow supercells were present in the convective rainband that trained over southeastern Virginia through the afternoon and evening of 30 August. Daytime heating combined with the advection of high θe air combined to destabilize the atmosphere in the right front quadrant of Gaston, an area with a favorable shear profile for supercells. Weak baroclinicity at Gaston's cloud boundary provided the forcing necessary to initiate convection.

A modeling study will investigate how accurately the ingredients necessary for the heavy rainfall can be represented in simulations. Assimilation of asynoptic satellite data is evaluated as a method to improve the initial moisture state in the model and reduce convective spin-up time, which has been found to be a major impediment to accurate modeling of Gaston's evolution. Preliminary simulations show high sensitivity to initial condition analyses and suggest significant forecast improvement is possible.