Improved modeling of the role of mangroves in storm surge attenuation

Mangroves have been proven to be effective in storm surge attenuation but it remains an important challenge to accurately quantify such bio-shielding effects using numerical simulations, as it is very difficult to comprehensively represent the ecological characteristics of mangroves at both large and small scales. In this study, a numerical method is developed and implemented in the Coastal and Estuarine Storm Tide (CEST) model in order to investigate the attenuation effect of mangroves on storm surge. This numerical method employs an improved drag force formula, which involves the development of new abstract tree models and use of a landscape scale data map of mean mangrove tree height for the study area. The storm surge observed in the South Florida mangrove zone caused by Hurricane Wilma (2005) is used to verify the numerical model. The numerical results indicate a maximum surge of approximately 4.3 m, and a decay rate of peak storm surge height of approximately 18 cm/km across the areas with a mixture of mangrove islands and open water, and nearly 24 cm/km through areas with dense mangrove forest. Results also show that short mangroves (< 4 m) can outperform tall mangroves on surge attenuation when the water depth is low (< 4 m). Extensive comparisons are also made with the conventional Manning coefficient based method that incorporates the mangrove drag force into bed friction; it is found that the current method predicts better inundation extents for Wilma (2005), hence a more accurate quantification of the attenuation of storm surge due to mangroves. •Mangrove effects on storm surge modeled by a porosity plus drag force (PDB) method.•Both small scale and landscape scale mangrove tree characteristics are important.•The PDB method outperforms the conventional Manning coefficient method (MBF).•Mangroves cause 18 to 24 cm/km decay rate of storm surge peaks.•Short mangroves (height less than 4 m) may attenuate storm surges better than tall ones.