A residential storm surge damage model was developed for a coastal Texas county. The model was used to investigate uncertainty in damage estimates under varying initial assumptions of structure elevations. Results indicated that the model was highly-sensitive to such assumptions, and discrepancies between damage estimates approached $2 billion for some scenarios. A survey of home elevations within the county was conducted to correct the initial assumptions. After correction, variability between the estimates decreased significantly, and the results suggested commonly employed elevation assumptions may be prone to damage overestimation. Additional aspects of uncertainty were then incorporated into the model and an uncertainty analysis (UA) was performed with respect to assumptions of the spatial independence between model parameters. The UA results showed a large degree of variability between independence assumptions. Finally, a global sensitivity analysis of the model was performed with inconclusive results, as they varied by independence assumption.