Performance Assessments of Hurricane Wave Hindcasts
2021
Landfalling tropical cyclones (TC) generate extreme waves, introducing significant property, personal, and financial risks and damage. Accurate simulations of the sea state during these storms are used to support risk and damage assessments and the design of coastal structures. However, the TCs generate a complex surface gravity wave field as a result of the inherently strong temporal and spatial gradients of the wind forcing. This complexity is a significant challenge to model. To advance our understanding of the performance of these models on the eastern seaboard of the United States, we conduct an assessment of four hindcast products, three based on WAVEWATCH-III and the other using the Wave Modeling project, for six major landfall TCs between 2011–2019. Unique to our assessment was a comprehensive analysis of these hindcast products against an array of fixed wave buoys that generate high quality data. The analysis reveals a general tendency for the wave models to underestimate significant wave height (Hs) around the peak of the TC. However, when viewed on an individual TC basis, distinct Hs error patterns are evident. Case studies of hurricanes Sandy and Florence illustrate complex Hs bias patterns, likely resulting from various mechanisms including insufficient resolution, improper wind input and source term parameterization (e.g., drag coefficient), and omission of wave–current interactions. Despite the added challenges of simulating complex wave fields in shallow coastal waters, the higher resolution Wave Information Study and National Centers for Environmental Prediction (ST4 parameterization only) hindcasts perform relatively well. Results from this study illustrate the challenge of simulating the spatial and temporal variability of TC generated wave fields and demonstrate the value of in-situ validation data such as the north Atlantic buoy array.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
60
References
2
Citations
NaN
KQI