To adequately evaluate risk associated hurricane flooding, numerous surge events must be considered, and the cost associated with high resolution numerical modeling for several storms is excessive. The Joint Probability Method with Optimal Sampling (JPM-OS) has been recently shown to be a reliable method in estimating extreme value probabilities of hurricane flooding – it relies heavily on a hurricane surge matrix comprised of surge values from several hurricane scenarios (with varying meteorological and climate change characteristics). Surge Response Functions (SRFs) are physics-based equations developed using scaling laws to adequately scale surge response in dimensionless space; they serve as surrogates to high resolution numerical models in estimating hurricane peak surge to populate the JPM-OS surge matrix. Research presented in this dissertation is primarily focused on the development of dimensionless formulations using physics-based scaling laws to account for the contribution of forward speed (v_f), approach angle (theta) and Sea Level Rise (SLR). These parameters are incorporated into pre-existing SRFs for open coast locations and bays. For the bays, in addition to accounting for the effects of v_f and theta in the SRFs, a new dimensionless formulation for the influence of storm size (R_p) is included in the SRFs.   To account for the influence of v_f in the SRFs, the dimensionless formulations primarily consist of the time it takes for surge to build up (over the shelf, for open coast SRFs and within the bays, for bay SRFs). The formulation for the influence of theta primarily accounts for the rotation of the hurricane wind field as the storm makes landfall. For the influence of R_p in the bays, the new formulation scales R_p with the farthest distance through which water mass will move inside the bay, from its center of gravity. A simple correction based on a linear model is derived to account for the influence of SLR on surge response at open coast locations and in bays. The developed dimensionless formulations for v_f and theta (and R_p for bay SRFs) are incorporated into the SRFs to obtain revised versions of the response functions. For open coast locations, the revised SRFs estimate peak surge with an increased accuracy (based on root-mean-square errors of modeled versus SRF-estimated peak surge) of up to 12.5% reduction in root-mean-square errors. In addition, the new formulations improve the predictions of 65% of surge events of 2 m or greater. For the bays, the revised SRFs reduce the root-mean-square errors (by up to 54% in Matagorda Bay), when compared to the previous formulation. These results indicate that the new formulations, which include v_f and tehta (and R_p for bay SRFs), significantly improve the accuracy of the SRFs. Application of the revised open coast SRFs to the JPM-OS framework shows only minor impacts of v_f and theta variation on surge versus return period curves (about 5.2% maximum increase in surge for theta varying from -80 degrees to +80 degrees, and a maximum of 6.7% for fvvarying from 1.54 m/s to 10.8 m/s). Climate change parameters however show a much more significant impact on the surge versus return period curves. SLR variation from 0.5 m to 2.0 m yields a maximum of 42.4% increase in surge, while hurricane intensification from 0.5 degrees C to 1.5 degrees C yields an increase of up to 11.3% in surge.

Performance-based engineering (PBE) requires advances in hazard characterization, structural modeling, and nonlinear analysis techniques to fully and efficiently develop the fragility expressions and other tools forming the basis for risk-based design procedures. This research examined and extended the state-of-the-art in hazard characterization (wind and surge) and risk-based design procedures (seismic). State-of-the-art hurricane models (including wind field, tracking and decay models) and event-based simulation techniques were used to characterize the hurricane wind hazard along the Texas coast. A total of 10,000 years of synthetic hurricane wind speed records were generated for each zip-code in Texas and were used to statistically characterize the N-year maximum hurricane wind speed distribution for each zip-code location and develop design non-exceedance probability contours for both coastal and inland areas. Actual recorded wind and surge data, the hurricane wind field model, hurricane size parameters, and a measure of storm kinetic energy were used to develop wind-surge and wind-surge-energy models, which can be used to characterize the wind-surge hazard at a level of accuracy suitable for PBE applications. These models provide a powerful tool to quickly and inexpensively estimate surge depths at coastal locations in advance of a hurricane landfall. They also were used to create surge hazard maps that provide storm surge height non-exceedance probability contours for the Texas coast. The simulation tools, wind field models, and statistical analyses, make it possible to characterize the risk-consistent hurricane events considering both hurricane intensity and size. The proposed methodology for event-based hurricane hazard characterization, when coupled with a hurricane damage model, can also be used for regional loss estimation and other spatial impact analyses. In considering seismic hazard, a risk-consistent framework for displacement-based seismic design of engineered multistory woodframe structures was developed. Specifically, a database of probability-based scale factors which can be used in a direct displacement design (DDD) procedure for woodframe buildings was created using nonlinear time-history analyses with suitably scaled ground motions records. The resulting DDD procedure results in more risk-consistent designs and therefore advances the state-of-the-art in displacement-based seismic design of woodframe structures.