This will depend on the vulnerability assessment you are to perform. If your interest is to analyze vulnerability as a function of the sensitivity of the population, you could use the wave height relating to mortality functions. If you are assessing buildings's vulnerability you could use wave heights relating to infrastructures's damage functions.
I agree with Ignacio Aguirre, first of all depends of the main focus of the risk analysis processes (buildings, population, breakwaters, port infrastructure, etc.).
If what you want is the curve of vulnerability of residential houses to wave height
for coastal inundation studies you may want to have a look at the following report: "Multi-hazard analysis for the city of Cairns (QLD)", see Section 7 (Inundation Simulation). The report can be found in: https://www.researchgate.net/publication/261798246_Multi-hazard_analysis_for_the_city_of_Cairns_(QLD).
Hope this helps,
Augusto
Article Multi-hazard analysis for the city of Cairns (QLD)
Do not forgot that you have to consider wave height close and along to the shoreline Consequently wave height at the breaking point is probably the best indicator. You may use a numerical model or empirical rules from Goda and others (see en SPM, 1984)