Click on figures for larger images.

Figure 1-1. Example of a scenario H3_D8 with wave height of 1.25 meters and wave direction of 169 degrees: A, Wave height; B, Wave direction; C, Wave period; D, Wind speed; and, E, Wind direction. Red bars show the normalized distribution of other relevant variables, constrained by height and direction. Blue dots show the data for the observation times that fall within this scenario.

Figure 1-2. Comparison of the best-match condition (yellow dot) to the target values for the scenario (magenta asterisk) against all observations (blue dots) for scenario H3_D8. A, wave height and direction, B, wind speed and wave direction, and C, wave height and period.

In this appendix, we define scenarios for reducing the full-time series to representative conditions and identify time periods that represent these scenarios for running simulations with spatially varying boundary conditions. As described in the Methods section of this report, scenarios were defined by drawing observations from 5 wave height ranges (i.e., bins) and 16 wave direction ranges, yielding 80 unique scenarios. The specifications for each scenario are defined in table 1–1. In addition to being defined by wave height and direction, the scenarios are further described by the conditional mean values of wave period, wind speed, and wind direction based on the observed data (from NDBC buoy 42040) for each scenario (fig. 1–1).

The period chosen to represent each scenario was identified by matching the multivariate wave height and direction of that scenario to the values defining each bin. The match was made by evaluating the likelihood of each observation given the bin-defining values. Observations were given an uncertainty equal to 0.1 times the width of the bin to allow matching observations that fall on bin boundaries. It is expected that the best-match scenario would come from a short time period (6 hours) where the conditions consistently met the scenario conditions in order to avoid choosing transitional conditions that may produce unrepresentative conditions. The comparison of the best match to the H3_D8 scenario shows that this approach effectively selected time periods with representative wave period and offshore wind conditions (fig. 1–2).