Search for: All records

Extreme atmospheric wind and precipitation events have created extensive multiscale coastal, inland, and upland flooding in United States (U.S.) coastal states over recent decades, some of which takes days to hours to develop, while others can take only several tens of minutes and inundate a large area within a short period of time, thus being laterally explosive. However, their existence has not yet been fully recognized, and the fluid dynamics and the wide spectrum of spatial and temporal scales of these types of events are not yet well understood nor have they been mathematically modeled. If present-day outlooks of more frequent and intense precipitation events in the future are accurate, these coastal, inland and upland flood events, such as those due to Hurricanes Joaquin (2015), Matthew (2016), Harvey (2017) and Irma (2017), will continue to increase in the future. However, the question arises as to whether there has been a well-documented example of this kind of coastal, inland and upland flooding in the past? In addition, if so, are any lessons learned for the future? The short answer is “no”. Fortunately, there are data from a pair of events, several decades ago—Hurricanes Dennis and Floyd in 1999—that we can turn to for guidance in how the nonlinear, multiscale fluid physics of these types of compound hazard events manifested in the past and what they portend for the future. It is of note that fifty-six lives were lost in coastal North Carolina alone from this pair of storms. In this study, the 1999 rapid coastal and inland flooding event attributed to those two consecutive hurricanes is documented and the series of physical processes and their mechanisms are analyzed. A diagnostic assessment using data and numerical models reveals the physical mechanisms of downstream blocking that occurred. 

Abstract The Lagrangian tracking approach is often used in numerical modeling (NM) to simulate and predict the movement of marine particles such as plastic, oil spills, and floating wreckage. The uncertainties in NM reduce prediction accuracy as a result of the coarse temporal and spatial resolution, along with waves, winds, and currents. From 29 August to 22 December 2020, the United States Defense Advanced Research Projects Agency's Ocean of Things program deployed 422 floating drifters in the Gulf of Mexico, providing an opportunity of using the observed trajectories as ground truth to train Artificial Intelligence (AI) models to correct NM float trajectory predictions. A Regional Ocean Model System (ROMS) and AI Hybrid model was developed to implement AI to correct the ROMS‐predicted 1‐day float trajectories. The AI model is built on a convolutional neural network and Gated Recurrent Unit. The results of the ROMS‐AI Hybrid model show that 82.0% of the trajectory predictions were improved at the 24 hr, with the corresponding overall mean separation error decreasing by 11.82 km, from 20.56 to 8.74 km, which is a 57% improvement, and the error growth rate decreasing from 5.06 km per 6 hr to 1.95 km per 6 hr. The evident improvement indicates that the ROMS‐AI Hybrid model can correct the ROMS simulation to improve the 1‐day prediction of the float trajectories and shows great potential to predict the cluster of the floats.