loading page

Influence of Forcing Conditions on Total Water Level Prediction and Spatiotemporal Patterns in Delaware Bay, USA
  • +3
  • David Fernando Muñoz Pauta,
  • Dongxiao Yin,
  • Jiannan Tian,
  • Roham Bakhtyar,
  • Kyle Mandli,
  • celso ferreira
David Fernando Muñoz Pauta
University of Alabama

Corresponding Author:dfmunoz1@crimson.ua.edu

Author Profile
Dongxiao Yin
Louisiana State University
Author Profile
Jiannan Tian
University of Alabama
Author Profile
Roham Bakhtyar
U.S. Army Corps of Engineers
Author Profile
Kyle Mandli
Columbia University in the City of New York
Author Profile
celso ferreira
George Mason University
Author Profile

Abstract

Accurate forecasts of total water level (i.e., a combination of tides, surge, wave and freshwater components) is imperative for stakeholders and federal agencies to adopt strategies for potential flooding hazards in a timely-manner. In that regard, the National Water Center in partnership with several federal agencies have been providing forecast services to the United States since 2017. However, the complex interaction of dynamical forcing conditions among other factors (e.g., anthropogenic activities, land cover change, etc.) reduce the National Water Model’s (NWM) ability to provide accurate Total Water Level (TWL) prediction in Coastal Transition Zones (CTZs). In this study, we use an existing inland to coastal model coupling framework (i.e., NWM, HWRF, Delft3D-FM and ADCIRC) to analyze the influence of dynamical forcing conditions (e.g., local wind, surge and river discharge) on TWL prediction in Delaware Bay, USA. In addition, we quantify the contribution of each component in TWL for Hurricanes Isabel and Sandy based on a systematic set of scenarios generated in Delft3D-FM. It is revealed that in both hurricanes, storm surge-induced water level is the main contributor to TWL followed by astronomical tides. River discharge induced-water level is rather small compared to the other components. Analyses of spatial variation of TWL as well as temporal variation of error in prediction suggest that wind forcing plays a key role in TWL prediction followed by river discharge. Moreover, our results suggest that the wind module of Delft3D-FM greatly improves the model performance at TWL peak when compared to the other forcing.