Recent advances have allowed coupling of the ADCIRC hydrodynamic model and the SWAN wave model for improved predictions of surge and waves during hurricane conditions. In hindcasts of recent major storms, including Gustav and Ike of the 2008 hurricane season, the SL16 mesh in generally capable of predicting peak surge values to within one-half meter. The first body of work in this thesis focuses on improving model accuracy through adjustment of spatially variable roughness coefficients. We examine many land coverage datasets, which are used to assign nodal bottom friction and bathymetric values, and develop a best possible coverage for modeling the 2008 hurricane season. A new land coverage dataset for coastal Louisiana is developed by merging the NOAA C-CAP post-Katrina 2006 land coverage, and a 2007 vegetation type survey from USGS. This new merged land coverage is applied to the SL16 ADCIRC+SWAN mesh, developing an updated version of the mesh with different bathymetric and bottom friction parameters based on the new land coverage. Hindcast using the new version of SL16 are compared to the original. Very little difference in overall model validation is observed. A peripheral focus of this thesis is to improve the efficiency of modeling storm surge by incorporating a nested mesh approach. With this approach, the user extracts a small mesh from a large scale mesh for a particular area of interest, and then applies boundary conditions from a large scale simulation. This set-up allows the nested mesh to reproduce nearly equivalent results as the large scale simulation, for the particular area of interest. The nested mesh approach improves efficiency, allowing more simulations to be evaluated with limited computational resources. In this paper, we demonstrate the use of the nested mesh for a real design project. Using the nested mesh approach, we also explore the possibility of parameter optimization for a particular area of interest.The second body of work focuses on surge and wave attenuation over wetlands in Southern Louisiana, with a sensitivity study of timing of wind-forcing, canopy coefficients, and bottom friction coefficients. We focus on hindcasts of hurricanes Gustav and Ike to illustrate the processes which govern surge and wave attenuation over wetlands. We are particularly interested in the timing and duration of wind stress, and whether or not surge levels reach steady state conditions, where surface wind stress forcing is in balance with the surge levels, and bottom stress has limited influence on the governing equations. Finally, a sensitivity analysis is provided to determine the impact on surge levels with increased canopy coverage in the marshes, which effectively removes windstress from the governing equations. A second sensitivity evaluates the influence of bottom friction on peak surge levels for various storm forward speeds. Increased bottom friction, which can be associated with wetland restoration, plays a very minor roll in the slower moving storms, where surge reaches stready state conditions. Canopy, which removes windstress in the model, is found to drastically reduce surge for all storm forward speeds. To illustrate these processes, we have selected the Caernarvon marsh as a study area.