Authors: Jorge Bravo Mendez, Marouane Temimi – Stevens Institute of Technology
Title: Assessing streamflow forecast over the Hackensack River Watershed using physics- and AI-driven weather prediction models
Presentation Type: Poster
Abstract: Accurate streamflow forecasting is crucial for effective water resource management, flood prediction, and disaster preparedness. The Hackensack River Watershed, spanning northern New Jersey and southern New York, lies within the densely populated New York City metropolitan area. As one of the most developed floodplains on the U.S. East Coast, it is highly flood-prone, with hurricanes like Irene (2011) and Sandy (2012) causing significant impacts.
Regional numerical weather prediction (NWP) models, such as the Weather Research and Forecasting (WRF) model and the Model for Prediction Across Scales (MPAS), play a key role in providing meteorological inputs for hydrological models like the National Water Model (NWM). However, differences in model formulations, spatial resolution capabilities, and computational efficiency may significantly impact streamflow predictions. We evaluate the performance of MPAS and WRF in meteorological forecasting and their subsequent effects on streamflow simulations using the NWM. A comparative analysis is conducted to assess how differences in precipitation and other forecast variables translate into hydrological outputs, particularly in terms of runoff generation, peak discharge, and forecast accuracy. In addition, the use of AI-based weather.