The Effect of Grid Resolution on Hydrodynamic Modeling of Biscayne Bay

Abstract

Environmental Fluid Dynamics Code Plus (EFDC+) is a model that has been proven to be effective in hydrodynamic simulation. Understanding water surface elevation (WSE) and salinity dynamics in estuaries is crucial for comprehending the complex interplay of these vital hydrodynamic parameters, as they significantly influence ecological processes, water quality, and habitat suitability in these critical coastal ecosystems. The study aims to develop and assess five EFDC+ models with varying resolutions for Biscayne Bay (BB), Florida. The impact of grid resolution on model performance in WSE and salinity simulations was explored by comparing model outputs with observational data. The study employed a Medium-resolution model grid as a baseline, featuring a finer resolution (~130 m × 130 m) in North BB and a coarser resolution (~300 m × 300 m) in Central and South BB, seamlessly integrated. The Fine-resolution model was created by halving the "i-direction" and "j-direction" of the Medium-resolution grid; in contrast, the Coarse2-, Coarse3-, and Coarse4-resolution models are developed by multiplying these directions by factors of 2, 3, and 4, respectively. All models adopted the WGS 84 UTM zone 17N coordinate system, with meters as the unit of measurement. A geo-processed bathymetry at a 20 m x 20 m horizontal spatial resolution and 0.01 m vertical accuracy, along with other input data from Alarcon et al. (2022), was utilized for simulations spanning January 2012 to December 2018, with a warm-up phase from January 2017 to June 2017. The assessment of WSE simulation at three measurement stations consistently showed performance trends across models. The accuracy of model simulations exhibited a gradual reduction, ranging from -0.1 cm to 1.7 cm, with an increasing trend as resolution decreases. Salinity simulations, categorized into four regions, exhibit varying model performance. While models generally performed poorly and irregularly in salinity simulation and consistently fare better in simulating salinity in Open and South Bay sections, challenges arise in accurately replicating low salinity levels below 15 ppt in Nearshore locations. Notably, the North Bay division shows greater variability across resolutions, with Coarse3 and Coarse4 models diverging greatly from others. The results suggest a need for the establishment of salinity measurement stations at the tidal open mouths of the Bay or using HYbrid Coordinate Ocean Model (HYCOM) salinity data for the tidal BCs. Also, installing salinity measurement stations in the North Bay is essential. This initiative is crucial for conducting studies that can yield improved results and management recommendations, particularly given the intricate geometry and potential complexities in dynamics.