Sanderson, D.R.; Gravens, M.B., and Permenter, R.L., 2019. Methodology for identifying a subset of representative storm surge hydrographs from a coastal storm modeling database. Journal of Coastal Research, 35(5), 1095–1105. Coconut Creek (Florida), ISSN 0749-0208.

Freely accessible databases that contain the results of large coastal storm modeling efforts are available online. Some of these databases explore coastal risks broadly (e.g., overall vulnerability to sea-level rise and storm surge), whereas other databases contain more specific modeling results (e.g., time series of storm surge hydrographs and associated significant wave heights). In the case of the latter, these databases can be composed of both synthetically defined, probabilistic storms as well as historically based storms. Some coastal storm modeling efforts have resulted in upwards of 1000+ simulated storm events. In the context of numerical models that use as input storm surge and significant wave height time series (e.g., those that rely on precomputed relational databases), use of a full probabilistic storm suite can prove to be infeasible. A methodology for identifying a subset of representative storm events from a probabilistic database is outlined in this paper. By utilizing storm surge and significant wave height time series resulting from high-fidelity numerical modeling of the U.S. Army Corps of Engineers North Atlantic Coast Comprehensive Study, a representative storm suite was developed for Crisfield, Maryland. The results of using a representative storm suite and a full storm suite were compared by simulating 75 iterations (life cycles) of a 216-year period using the Monte Carlo life cycle model G2CRM. By performing a nonparametric extreme value analysis, the stage-frequency curves resulting from the representative and the full storm suites compared favorably.