ML18012A568

Technical Evaluation Report: Storm Surge_Seabrook_January2018

ML18012A568
Person / Time
Site:	Seabrook
Issue date:	01/12/2018
From:	Taylor Engineering
To:	Office of New Reactors
Thompson J
References
Download: ML18012A568 (43)

Text

Storm Surge Technical Evaluation Report Seabrook Station Hazard Reevaluation Review Prepared for By March 2017

TECHNICAL EVALUATION REPORT: SEABROOK STATION STORM SURGE NextEra reported in the Flood Hazard Reevaluation Report (FHRR; October 2016 Revision) that the reevaluated hazard, including associated effects, for site flooding due to storm surge is 23.35 ft-NAVD88 (North American Vertical Datum of 1988) at the seawall away from site structures with maximum flow depths that exceed 20.23 ft-NAVD88 near site structures. This flood-causing mechanism is described in NextEras current design basis. The current design basis hazard for site flooding due to the Combined Effect scenario of a probable maximum hurricane (PMH) induced probable maximum storm surge (PMSS) at the site is 21.03 ft-NAVD88 for peak water surface elevation due to runup with maximum flow depths of 20.23 ft-NAVD88 at site structure walls.

NextEra submitted a FHRR on September 25, 2015. A revised FHRR was submitted by NextEra dated October 26, 2016. The revised FHRR contains introductory sections that describe the changes to the FHRR. Specific to storm surge, the changes to the FHRR relate to an additional flooding case for the PMSS event that featured higher storm-generated stillwater levels, a larger wave height near the site, and a lower wave period. The revised FHRR and additional PMSS analysis stems from NextEra and audits conducted between the original and revised FHRR submittals. Specific to storm surge, the audits discussed the Delft3D grid setup, the Taylor Engineering staff independent storm surge simulations, and comparisons of the water levels developed by different model setups and storms simulations.

The Taylor Engineering staff describes its evaluation of site flooding from storm surge, including associated effects, against the relevant regulatory criteria based on present-day methodologies and regulatory guidance below.

1.0 Historical Storm Surge Data Information Submitted by NextEra NextEra reviewed the historical hurricane tracks within a 100 mile radius (see Figure 1) of the Seabrook station and found that from 1854 to present, most historical hurricanes move in a northerly direction, and there are no recorded landfalling hurricanes at the latitude of Seabrook station that come from the east or southeast.

NextEra also performed a site-specific climatological study to expand the search for storm data.

To support a site-specific climatological study, NextEra conducted online searches, reviews of numerous research papers, queries of National Weather Service (NWS) reports and databases, and examination of non-tropical storm related storm surges along the Atlantic coastline to develop the list of storms used in the climatological study. NextEra searched sources that included the NWS forecast offices in the region, National Center for Environmental Prediction (NCEP) (2014), National Oceanic and Atmospheric Administrations (NOAA) Earth System Research Laboratory (ESRL) 3/6 Hourly 20th century reanalysis data composites website (NOAA, 2014l), the NOAA ESRL 6-Hourly NCEP/National Center for Atmospheric Research (NCAR) reanalysis data composites (NOAA, 2014q), NOAA ESRL 3-Hourly NCEP North American Regional Reanalysis (NARR) (NOAA, 2014n), National Climatic Data Center (NCDC) storm archives (NCDC, 2014), and the Storm Prediction Center (SPC) severe storm reports (NOAA, 2014o). Additional storm cases are recorded by numerous refereed journal articles and research papers (e.g., Armstrong, 2013; Blake et al., 2012; Burt, 2012; Cooperman and Rosendal, 1963; Halverson and Rabenhorst, 2013; Hayden, 1889; McQueen et al., 1956). Table 1 lists squall line 1

storms identified in the storm search analysis and Table 2 lists storms that produced the highest wind speeds and lowest pressures at one or more analysis periods for a given direction (FHRR, Sec. 4.4.9.1).

Figure 1 Historical Hurricane Tracks within 100 mile Radius of Seabrook (Source: FHRR, Figure 4-24) 2

Table 1 List of Squall Line Storms Identified in the Storm Search Analysis (Source: FHRR, Table 4-30) 3

Table 2 List of Synoptic Storms Used for Delft3D Model Input (Source: FHRR, Table 4-31) 4

Taylor Engineering Staff Technical Evaluation NextEra performed a wide and exhaustive search of historical storm and surge data. NextEra identified and adequately described all relevant storms and surges.

2.0 Probable Maximum Hurricane Information Submitted by NextEra NextEra performed a hurricane climatology study for the area near Seabrook station; applied WindRiskTech (2014) technique to generate large numbers of synthetic hurricane tracks; and applied deterministic, coupled numerical modeling to simulate storm intensity along each track.

Figure 2 shows the hurricane climatology study methodology flow diagram. Using the SLOSH model, NextEra generated 20,400 tropical cyclone events to represent the tropical cyclone events that may take place at Seabrook over an extended time period and applied the Delft3D-FLOW and Delft3D-WAVE models on select 27 storm tracks (see Table 3, Figure 3, and Figure 4) to determine how the detailed nearshore bathymetry, more accurate representation of the hurricane and detailed hydrodynamics would affect the predicted storm surge. NextEra used the below criteria to select the 27 storm tracks (FHRR, Sec. 4.4.9.2.5).

1) 10 tracks with maximum wind velocity near the site,

2) 10 tracks with maximum storm surge from the SLOSH simulations, and

3) 10 tracks to map the entire range of the relationship between the SLOSH simulated surge and the Delft3D simulations.

NextEra stated that the above study provides high water level values with a probability of exceedance of the order of 10-5 per year.

Taylor Engineering Staff Technical Evaluation NextEra provided the details of the site and region-specific climatological study and estimations of the ranges of PMH parameters. Taylor Engineering staff concludes NextEra has identified and described acceptable PMH parameters. However, NextEra did not provide the specific parameters values of the PMH or how it estimated the hypothetical PMH.

5

Figure 2 Hurricane Climatology Methodology Flow Diagram (Source: FHRR, Figure 4-36) 6

Table 3 Surge and Wind Speed for Identified Track Subset (Source: FHRR, Table 4-40) 7

Figure 3 Track Set with Maximum Wind Speeds Near Seabrook (Source: FHRR, Figure 4-42)

Figure 4 Tracks Set Producing Maximum Storm Surge with SLOSH model at Seabrook (Source: FHRR, Figure 4-43) 8

3.0 Probable Maximum Wind Storm.

Information Submitted by NextEra NextEra cited Nuclear Regulatory Commission (NRC) guidelines ANSI/ANS (1992) Section 7.2.2.1 and NRC (2013a) Section 3.2.2 for the area in the vicinity of the site that states the PMSS and seiche are calculated from the PMWS [probable maximum wind storm]. NextEra also cited that ANSI/ANS (1992) Section 7.2.2.3.1 further indicates that parameters of the PMWS should be determined by a meteorological study. Following these guidelines, NextEra conducted a site-specific northeaster climatological study for Seabrook including an extensive data search.

The northeaster climatological study results provided Seabrook sites PMWS wind speed, wind direction, and pressure values. NextEra described the data search as follows (FHRR, Sec.

4.4.9.1.1):

A comprehensive search was conducted to identify significant synoptic storms, generally referred to as Nor'easters in the region, some of which combine with remnant tropical systems that impacted the region around Seabrook. Among the sources used in this search are the NWS forecast offices in the region, National Center for Environmental Prediction (NCEP) (2014), NOAAs Earth System Research Laboratory (ESRL) 3/6 Hourly 20th century reanalysis data composites website (NOAA, 2014l), the NOAA ESRL 6-Hourly NCEP/ National Center for Atmospheric Research (NCAR) reanalysis data composites (NOAA, 2014q),

NOAA ESRL 3-Hourly NCEP North American Regional Reanalysis (NARR)

(NOAA, 2014n), National Climatic Data Center (NCDC) storm archives (NCDC, 2014), and the Storm Prediction Center (SPC) severe storm reports (NOAA, 2014o). Additional storm cases are recorded by numerous refereed journal articles and research papers (e.g., Armstrong, 2013; Blake et al., 2012; Burt, 2012; Cooperman and Rosendal, 1963; Halverson and Rabenhorst, 2013; Hayden, 1889; McQueen et al., 1956). These storms are selected because of their historic impacts in the region, as well as their meteorological significance, including record-breaking high winds and/or low pressure readings.

Table 4 shows the 35 potential northeaster storms that the above-described storm search produced. After gridded data analysis, Table 5 shows a subset of the storms listed in Table 4 that produced the highest wind speeds and lowest pressures at one or more analysis periods for a given direction and thus were used for Delft3D model input. The latitude and longitude listed in the tables represent the location of the grid with the highest wind speed from a given direction (FHRR, Sec. 4.4.9.1.1).

Table 6 to Table 9 show the highest wind speeds and the pressures generated in the northeaster climatology study. NextEra evaluated extreme wind events which produced the highest wind speeds and greatest pressure gradients over the entire period of record of a large region that were considered transpositionable to the Seabrook site. So all wind and pressure data associated with each event were shifted so that the theoretical storm occurred directly over the Seabrook site or in a location which would produce the worst case-yet physically possible scenario and therefore maximizes potential storm surge. NextEra assumed the combination of extremely rare wind and pressure events over a large region of transpositionability to the site produced representative PMWS data.

9

Table 4 Summary of Squall Line Storms Identified in the Storm Search Analysis (Source: FHRR, Table 4-30) 10

Table 5 Summary of Synoptic Storms Used for Delft3D Model Input (Source: FHRR, Table 4-31) 11

Table 6 Highest 3-Hour Average Wind Speeds from 0° to 180° Compared to the Return Frequency 3-Hour Average Wind Speed Climatology at Portland (Source: FHRR, Table 4-34)

Table 7 Highest 6-Hour Average Wind Speeds from 0° to 180° Compared to the Return Frequency 6-Hour Average Wind Speed Climatology at Portland (Source: FHRR, Table 4-35)

Table 8 Highest 3-Hour Average Wind Speeds from 180° to 360° Compared to the Return Frequency 3-Hour Average Wind Speed Climatology at Portland (Source: FHRR, Table 4-36) 12

Table 9 Highest 6-Hour Average Wind Speeds from 180° to 360° Compared to the Return Frequency 6-Hour Average Wind Speed Climatology at Portland (Source: FHRR, Table 4-37)

Taylor Engineering Staff Technical Evaluation NextEra provided sufficient detail about the development of the PMWS. NextEra has identified and described an acceptable PMWS.

4.0 Antecedent Water Levels.

Information Submitted by NextEra NextEra used recorded maximum monthly tide elevations from NOAAs Portland, Maine (NOAA 8418150) tidal station to calculate the 10% exceedance high tide for the antecedent water levels.

NextEra applied the Weibull plotting position to estimate a 10% exceedance high tide of 7.38 ft-NAVD88 (7.61 ft-MSL) and a normal distribution to estimate a 95 percent confidence interval of 7.26 to 7.49 ft-NAVD88 for the 10% exceedance high tide (see Table 10) (FHRR, Sec. 4.4.6.1).

Table 10 Select Stations Results for 10% Exceedance High and Low Confidence Tide Values (Source: FHRR, Table 4-23)

NextEra presented estimates of the sea level rise from many nearby stations between the end of 2013 and years 2030, 2050, and 2113 (see Table 11). NextEra selected the sea level rise rate at Boston, MA for the antecedent water level (AWL) estimation. Based on the estimated sea level rise rate at Boston station, NextEra estimates the sea level rise between the end of 2013 and 2030 at 0.16 ft, the end of 2013 and March of 2050 at 0.35 ft, and between the end of 2013 and 2113 at 0.96 ft (see Table 11) (FHRR, Sec. 4.4.6.2).

13

Table 11 Estimated Sea Level Rise (Source: FHRR, Table 4-24)

Taylor Engineering Staff Technical Evaluation Taylor Engineering staff added NextEras three sea level rise rates to the 10% exceedance high tide of 7.38 ft-NAVD88, to calculate a total AWL of 7.54 ft-NAVD88 for the short-term period between the end of 2013 and years 2030; 7.73 ft-NAVD88 in the event an extension of the current license is granted for the period between the end of 2013 and years 2030; and 8.34 ft-NAVD88 for the long-term period between the end of 2013 and 2113.

To independently confirm NextEras AWL calculation, Taylor Engineering staff used the Portland, Maine (NOAA 8418150) tidal station data to calculate a 10% exceedance high tide of 6.72 ft-NAVD88, an expected 50-year sea level rise of 0.34 ft, and a total AWL of 7.06 ft-NAVD88. Thus, the Taylor Engineering staff verified that NextEra provided a more conservative calculation of antecedent water levels.

NextEra used an acceptable antecedent water level for its surge analysis.

Table 12 provides the Taylor Engineering staff-estimated tidal datums from NOAA tidal gage Station 8418150 (Portland, ME) data.

Table 122 NOAA Station 8418150, Portland, ME Tidal Data and Seabrook Plant Datum Station Name NOAA 8418150 - Portland, ME Latitude 43° 39.4' N Longitude 70° 14.8' W Tide Datum Elevation (ft-NAVD88)

Mean Higher High Water (MHHW) 4.65 Mean High Water (MHW) 4.22 Mean Sea Level (MSL) -0.31 Mean Tide Level (MTL) -0.35 Mean Low Water (MLW) -4.91 Mean Lower Low Water (MLLW) -5.25 NGVD29 -0.72 Plant Datum NGVD Plant Datum Elevation -0.72 14

5.0 Storm Surge Models 5.1 Surge Propagation Models Information Submitted by NextEra NextEra applied the two-dimensional depth-averaged Delft3D Version 4.00.01 model software package to simulate the storm surge. In Delft3D-FLOW, the hydrodynamics of storm surge conditions are simulated by solving the Navier-Stokes equations for incompressible free surface flow. The Navier-Stokes equations are reduced to two-dimensional, depth-averaged flow with Delft3D-FLOW (Deltares, 2011c). The Navier-Stokes equations for incompressible flow are solved under the shallow water and Boussinesq assumptions (FHRR, Sec. 4.4.3.1).

NextEra developed five Delft3D rectangular model grids for storm surge modeling (a) one sufficiently large coarse domain to ensure all potentially significant regions and features that could affect the storm surge results were captured and appropriate boundary conditions analyzed; and (b) four refined grids located close to the Seabrook site. Model simulations used domain decomposition so the model conveys the information from the coarse grid to provide boundary conditions for the finer grids, which will vary with time during the evolution of the simulation (FHRR, Sec. 4.4.2).

NextEra applied a roughness value of 0.02 and 0.04 for deep ocean and nearshore, respectively.

NextEra also applied 538 ft2/s (50 m2/s) in the overall model domain for horizontal eddy viscosity and horizontal eddy diffusivity (FHRR, Sec. 4.4.4.1).

NextEra stated that the following guidelines were followed during model grid development (FHRR, Sec. 4.4.2):

• Size between adjacent cells should vary less than 20%,

• Orthogonality should be less than 0.2 for offshore areas, but as close to zero as possible,

• Smoothness between adjacent grid cell lengths is generally preferred to be less than 1.2 in the area of interest, and

• aspect ratio should be less than 2.

• Courant number should be less than 42 = ~5.66.

NextEra applied astronomical tidal components to describe a water level forcing at the model boundary during the storm surge model calibration phase.

Taylor Engineering Staff Technical Evaluation After submittal of the original FHRR (September 2015), Taylor Engineering staff reviewed NextEra's Delft3D_FLOW model, and the parameters, data, and conditions used in the model.

Review of the initial Delft3D_FLOW nested grid system indicated inconsistency in the most refined grid (site grid) location and bathymetry and topography features. Through the audit process and discussions with NextEra, the final Delft3D_FLOW simulations, within the revised FHRR (October 2016) applied a nested grid system with correction features for the most refined grid (site grid).

15

5.2 Wave Models Information Submitted by NextEra NextEra applied the Delft3D-WAVE Simulating WAVEs Nearshore (SWAN) model software package to simulate wave transformation. NextEra describes the SWAN model as a spectral wave model that evaluates the refracted wave height and wave angle based on a spectrum of waves using linear wave theory (Booij et al., 1999; Deltares, 2011c). The SWAN model accounts for (refractive) wave propagation due to current and water depth and represents the physical processes of wave generation by wind, dissipation due to whitecapping, bottom friction, depth-induced wave breaking, and nonlinear wave-wave interactions (both quadruplets and triads) explicitly with state-of-the-art formulations. Wave blocking by currents is also explicitly represented in the model. The SWAN model is based on the discrete spectral action balance equation and is fully spectral (across all directions and frequencies). The latter implies that short-crested random wave fields propagating simultaneously from widely different directions can be accommodated (e.g., a wind sea with superimposed swell). SWAN computes the evolution of random, short-crested waves in coastal regions with deep, intermediate, and shallow water depths and ambient currents (FHRR, Sec. 4.4.1).

There are three generations of wave models available to compute the sea surface state in Delft3D-WAVE (i.e., SWAN) (Deltares, 2011e). First generation wave models do not consider nonlinear wave interactions. Second generation models parameterized these interactions and include the coupled hybrid and coupled discrete formulations. Third generation models explicitly represent all the physics relevant for the development of the sea state in two dimensions, without assumptions regarding the spectral space. Further, energy terms are described explicitly with the addition of bottom dissipation and reflection, diffraction, and refraction terms. For Seabrook, the model computes the sea state from the hurricane using the third-generation model.

The Delft3D-WAVE computations accounted for the following processes (Deltares, 2011e):

depth-induced breaking, nonlinear triad interactions, bottom friction, wind growth, whitecapping, refraction and frequency shift (FHRR, Sec. 4.4.3.2).

NextEra applied the Delft3D-FLOW and Delft3D-WAVE modules to simulate the coupled effects of flow movement (i.e., storm surge) and wave propagation (i.e., wave spectra, height, period, and setup) through a water body when acted upon by external forcing functions (i.e., wind and atmospheric pressure fields). The coupling allowed for the accounting of the effect of flow on the waves (via setup, current refraction, and enhanced bottom friction) and the effect of waves on current (via forcing, enhanced turbulence, and enhanced bed shear stress). The coupling of the Delft3D-FLOW and Delft3D-WAVE models occurred every 30 min throughout the simulation (FHRR, Sec. 4.4.3.3).

In the SWAN model, NextEra opted to apply (1) wave growth from wind field, white capping process of energy dissipation, and wave nonlinear interaction; (2) depth-induced shoaling and refraction in the model and current-induced shoaling; (3) a vegetation based roughness of 0.0628 m for use in wind-wave modeling to most accurately represent wave growth within Hampton harbor; and (4) a constant ratio of breaking wave height to breaking wave depth of 0.73 (FHRR, Sec. 4.4.4.2).

Taylor Engineering Staff Technical Evaluation The Delft3D-WAVE model applied the same nested grids as the Delft3D-FLOW model so the comments in the Taylor Engineering staff technical evaluation comments in Section 5.1 also pertain to the Delft3D-WAVE model grids. Through the audit process and discussions with 16

NextEra, the final Delft3D_FLOW simulations applied a nested grid system with correction features for the most refined grid (site grid).

5.3 Topography and Bathymetry Information Submitted by NextEra NextEra obtained (1) the coastal bathymetry data from the Generalized Bathymetric Charts of the Ocean (GEBCO) project; (2) the 1/9 Arc Second (approximately 3 m) National Elevation Dataset (NED) from the U.S. Geological Survey (USGS) National Map Viewer; and (3) the local site bathymetry and topography from NOAA (1947, 1954, 1983) and USGS (2011b) (FHRR, Sec.

4.4.2).

Taylor Engineering Staff Technical Evaluation As mentioned in the Taylor Engineering staff technical evaluation of Sections 5.1 and 5.2, the original FHRR submittal (September 2015) included bathymetry and topography features within the most refined nested grid (site grid) that did not align with existing bathymetry and topography features near the site. The revised FHRR (October 2016) contains Delft3D nested grids with bathymetry and topography features that align with existing features near the site.

6.0 Numerical Model Validation Information Submitted by NextEra NextEra selected Hurricane Bob (August 16 to 21, 1991) and Hurricane Donna (September 3 to 13, 1960) (see Figure 5) to provide data for calibration and verification of various Delft3D model parameters because (a) these hurricanes were both Category 2 on the Saffir-Simpson scale near Seabrook, (b) calibration data are available, and (c) the tracks of the storms are relevant to Seabrook (FHRR, Sec. 4.4.7.1.2).

NextEra reported the procedure for the Seabrook Delft3D-FLOW and Delft3D-WAVE model calibration and verification (FHRR, Sec. 4.4.7.1.3) as follows:

• Selected a historical calibration event (Hurricane Bob) based on availability of observed storm surge data, the available resolution of the wind and pressure forcing data, and the magnitude of the event;

• Performed a series of sensitivity simulations of the wind drag coefficient for the calibration event;

• Performed a series of sensitivity simulations of the Mannings roughness coefficient for the calibration event.

• Ran the Hurricane Donna as validation to determine if the final calibration parameter values are acceptable for storm surge modeling.

17

Figure 5 Tracks of Hurricane Bob and Hurricane Donna (Source: FHRR, Figure 4-25)

Based on the above procedure, NextEra judged model performance by comparing model simulated and observed values and computing for the Nash-Sutcliffe model quotient efficiency (NSE) and Root Mean Square Error (RMSE).

= = Equation 1

=

= Equation 2 where t = time step; n = total number of time steps; y0 = observed value; ym = simulated value; and ybar0 = average of all observed values.

NextEra considered NSE values closer to 1 as indicative of better model performance and values closer to - as indicative of poor model performance. An NSE value of 0 suggests the models predictive power is equal to a model that simply reproduces the average of the observed time series. Numerical models producing an NSE value of 0 or less add no additional value. An NSE 18

value of 1 suggests an ideal model that has no error in reproduction of observed data. Values between 0 and 1 suggest that use of the model adds value to the prediction; however, the transformation is not ideal. The RMSE represents the sample standard deviation between simulated and observed values.

NextEra divided the model calibration into two parts tide calibration (Figure 6) and surge (Figure

7) calibration. NextEra provided a tide calibration RMSE of 0.214 m and an NSE of 0.957 based on comparison with measured tides at the Portland Station. Comparisons of modeled and observed peak storm surge elevations show an R2 (coefficient of determination) of 0.9081 (for Hurricane Bob) and 0.4694 (for Hurricane Donna). Notably, an R2 = 0 indicates the modeled peak storm surge elevations do not fit the data and an R2 = 1 indicates the modeled peak storm surge elevations perfectly fit the data. Tables 13, 14, and 15 provide summaries of the Delft3D-FLOW model tide and model surge calibration and Delft3D-Wave model surge calibration. Tables 16 and 17 show the comparison of model simulated and observed peak surge for model calibration (Hurricane Bob) and model verification (Hurricane Donna).

Figure 6 Resynthesized and Simulated Water Level Comparison, Portland, ME (Source: FHRR, Figure 4-27) 19

Figure 7 Delft3D Simulated and Observed Time Series Storm Surge Comparison at Newport (RI) during Hurricane Bob (Source: FHRR, Figure 4-27) 20

Table 13 Summary of Parameters for Delft3D-FLOW Model Tide Calibration (Source: FHRR, Table 4-25) 21

Table 14 Summary of Parameters for Delft3D-FLOW Model Storm Surge Calibration (Source: FHRR, Table 4-26) 22

Table 15 Summary of Parameters for Delft3D-WAVE Model Surge Calibration (Source: FHRR, Table 4-27) 23

Table 16 Comparison of Observed and Simulated Peak Storm Surge Elevations for Hurricane Bob (Source: FHRR, Table 4-28)

Table 17 Comparison of Observed and Simulated Peak Storm Surge Elevations for Hurricane Donna (Source: FHRR, Table 4-29)

Taylor Engineering Staff Technical Evaluation NextEras model-predicted high water levels reasonably approximated observed levels and present an acceptable validation of the models at select locations. NextEra did not provide indications on model performance near the Seabrook site; due to a lack of measured high storm surge levels near the site during historical tropical or extra-tropical storms.

7.0 Numerical Model Error and Uncertainty Information Submitted by NextEra For numerical model error, NextEra provided a tidal calibration RMSE of 0.214 m and an NSE of 0.957 based on comparison with measured tides at the Portland Station. Comparisons of modeled and observed peak storm surge elevations show an R2 (coefficient of determination) of 0.9081 (for Hurricane Bob) and 0.4694 (for Hurricane Donna). Notably, an R2 = 0 indicates the modeled peak storm surge elevations do not fit the data and an R2 = 1 indicates the modeled peak storm surge elevations perfectly fit the data.

For model uncertainty, NextEra performed a series of sensitivity simulations for the model wind drag coefficient and Mannings roughness coefficient during the model calibration. However, the FHRR did not present the results of the sensitivity simulations and did not indicate how these model parameters change water elevations at the Seabrook site.

Taylor Engineering Staff Technical Evaluation NextEra provided information on the error associated with the tide analysis and provided comparisons of measured water levels for locations relatively near to the site during strong 24

storms. NextEra did not present detailed and specific information on how modeling error and uncertainty was incorporated in the modeling processes.

8.0 Storm Surge Water Levels.

8.1 Deterministic Storm Surge Water Levels.

Information Submitted by NextEra Table 5-1 in the revised FHRR (October 2016) contains the summary water levels for the PMSS analysis that applied the Delft3D-FLOW and 3D-WAVE model. The table contains maximum stillwater level for the combined events analysis (17.75 ft-NAVD88) and the maximum total water level including wave runup (23.35 ft-NAVD88). Revised FHRR Section 4.4.9.4 discusses the 17.75 ft-NAVD88 maximum stillwater level and documents the inclusion of a 4.0 ft sensitivity margin within the 17.75 ft-NAVD88 level. The FHRR states the 23.35 ft-NAVD88 total water level including wave runup occurs at the site seawall away from site structures. NextEra also provided the Seabrook site elevation of the top of the revetment (flood barrier) at 19.23 ft-NAVD88 (20.00 ft-Plant Datum) (FHRR, Sec. 4.4.9.4). Based on the simulated water levels and wave conditions, and knowing the site elevation, NextEra developed wave overtopping estimates summarized in revised FHRR Table 4-41 and Figure 4-46 that estimates an overtopping duration of approximately 4.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> under PMSS conditions.

Taylor Engineering Staff Technical Evaluation The discussion below shows the results of Taylor Engineering staff independent storm surge simulations conducted to provide additional estimates of the deterministic PMH storm surge water levels near Seabrook. The independent simulations for Seabrook applied the SWAN+ADCIRC model to develop the water level and waves conditions near the Seabrook site. The simulations applied the SWAN+ADCIRC mesh developed for the New England region with detailed mesh areas for the Seabrook and Pilgrim nuclear sites. Taylor Engineering staff applied the mesh in a test simulation on Taylor Engineerings high performance computer, Merlin, with Hurricane Sandy (2012) forcing that demonstrated the ability to reasonably reproduce measured values in the New England region.

The independent simulations applied time-varying, two-dimensional wind and pressure fields for tropical storms with varying parameters. Taylor Engineering staff developed the tropical storm parameter combinations to produce the Probable Maximum Intensity (PMI) storm conditions at PNPS. Table 18 presents the range of parameters tested in the initial sensitivity simulations. The site location, north of Cape Cod and interior of an inlet to the east of the site created a complex storm surge response at the site, which required examination of many storm track angles and landfall locations.

25

Table 18 Tropical Storm Parameters in ADCIRC Sensitivity Simulations Central Pressure (mb) 920, 925 Radius to Maximum Winds (RMW) (nm) 30, 45 Forward Velocity (kt) 25, 40 Track Angle at Landfall (degrees counter-30, 10, 0, -10, -20, -30 clockwise from north)

Landfall Location (RMW from reference site; to 0, 0.25, 0.5, 1 west of point along Cape Cod south shoreline)

The initial test simulations applied only the ADCIRC (hydrodynamic) model to expedite run times which allowed examination of storm surge sensitivity to the tropical storm parameters. Based on these results, the final simulations applied the SWAN+ADCIRC model, which couples the hydrodynamic model to a spectral wave model (SWAN) to include wave-induced water level effects near the site. Review of the refined simulation results showed angles at landfall of 0, 10, and 30 degrees counter-clockwise from north produced the highest water levels at the site. Due to the site location, several landfall location and storm track angle combinations created elevated storm surge levels at the site. The final simulations applied the parameter values and ranges listed in Table 19.

Table 19 Tropical Storm Parameters in SWAN+ADCIRC Final Simulations Central Pressure (mb) 925 Radius to Maximum Winds (RMW) (nm) 45 Forward Velocity (kt) 25 Track Angle at Landfall (degrees counter-30, 10, 0 clockwise from north)

Distance from Site as Storm Track Passes (nm) 1, 7, 24 Review of FHRR Storm Surge Simulations NextEra applied SLOSH and Delft3D models to estimate the PMSS at the site. Approximately 20,400 SLOSH storm surge simulations identified the combination of storm tracks and meteorological parameters that NextEra expects to cause the largest storm surge at Seabrook.

NextEra completed more detailed simulations using 27 Delft3D-FLOW and Delft3D-WAVE models simulations to further evaluate the storms identified by the SLOSH model screening analysis that resulted in large surges at Seabrook. NextEra simulated PMWS and PMH storm conditions to calculate the PMSS. Table 5-1 in the revised FHRR (October 2016) contains the summary water levels for the PMSS analysis that applied the Delft3D-FLOW and 3D-WAVE model. The table contains maximum stillwater level for the combined events analysis (17.75 ft-NAVD88) and the maximum total water level including wave runup (23.35 ft-NAVD88). Revised FHRR Section 4.4.9.4 discusses the 17.75 ft-NAVD88 maximum stillwater level and documents 26

the inclusion of a 4.0 ft sensitivity margin within the 17.75 ft-NAVD88 level. The FHRR states the 23.35 ft-NAVD88 total water level including wave runup occurs at the site seawall away from site structures. Revised FHRR Section 2.2.9.5 states the runup at the seawall occurs approximately 20 m (66 ft) from any structures.

Comparison of FHRR and Independent Simulation Water Levels The results from the independent simulations with the PMI forcing allow comparison of the FHRR PMSS water levels. Table 20 presents details of the final three independent simulations with comparisons to the FHRR PMSS water levels for the Seabrook site. Figure 8 shows the tracks for the final three independent simulations.

Table 20 contains a summary of the revised FHRR PMH water levels with information listed for antecedent water level, stillwater level, wave conditions, and total water level including wave runup. The FHRR PMH water levels show a stillwater level of 17.8 NAVD88-ft including the 4 ft sensitivity margin and a total water level of 23.4 ft-NAVD88 at the seawall away from the structures. The three right hand columns in Table 20 show results from the final three Taylor Engineering staff independent SWAN+ADCIRC simulations that applied PMH-level wind forcing at the site. 1 The Taylor Engineering independent simulations show stillwater levels near the site seawall that range from 17.8 to 20.6 ft-NAVD88; near the revised FHRR stillwater levels. The Taylor Engineering independent simulation results show total water levels, including wave runup, near the site that range from 19.6 to 21.3 ft-NAVD88. Notably, the revised FHRR Table 5-1 provides a total water level of 23.4 ft-NAVD88, but states this value occurs near the seawall and should not be deemed equal to the total water level near site structures.

With consideration of all information provided by NextEra in the original and revised FHRR documents, NextEra provided an acceptable PMSS within the revised FHRR.

1 Note: Mentions of staff in this table refer to Taylor Engineering staff.

27

Table 20 Summary of Water Levels at Site based on FHRR and Taylor Engineering Independent Simulations Se abrook FHRR PMH Staff SWAN+ADCIRC Staff SWAN+ADCIRC Staff SWAN+ADCIRC Parame te r (re v. FHRR, 10/2016) Inde pe nde nt Run#1 Inde pe nde nt Run#2 Inde pe nde nt Run#3 Peripheral Pressure (mb) 1013 1013 1013 Central Pressure (mb) 925 925 925 Radius of Maximum Winds (NM) Probabilistic Evaluation 45 45 45 Forward Speed (kt) of Water Level, no 25 25 25 Angle at Landfall (deg CCW from specific PMH hurricane 30 10 0 N)

Landfall Location (first land) 70.51 W, 41.35 N 70.73 W, 41.34 N 70.83 W, 41.33 N Distance as T rack Passes Site 24.3 NM West of Site 6.6 NM West of Site 0.9 NM East of Site 10% Astronomical High T ide 7.38 6.71 6.71 6.71 (ft-NAVD88)

Sea Level Rise (ft)1 0.2 0.34 0.34 0.34 Location N/A Marsh NE of Site Marsh NE of Site Marsh NE of Site Revaluated Stillwater (ft-NAVD88) 17.8 20.6 18.9 17.8

[FHRR T able 5-1]

FHRR T able 5-1 CDB Stillwater 14.8 14.8 14.8 14.8 (ft-NAVD88)

T op of Revetment (Flood 19.23 19.23 19.23 19.23 Protection) (ft-NAVD88) [FHRR top of revetment top of revetment top of revetment top of revetment Section 4.4.9.4] (flood barrier) (flood barrier) (flood barrier) (flood barrier)

Location Seawall Marsh NE of Site Marsh NE of Site Marsh NE of Site Revaluated PMH SWAN Hs/T m 4.0/4.0 5.9/4.7 5.1/4.3 4.6/4.3 (ft/sec) [FHRR Section 4.4.9.4]

Location Seawall/Site Seawall/Site Seawall/Site Seawall/Site Reevaluated T otal Water Level 23.4 (at seawall) 21.3 19.9 19.6 (ft-NAVD88) [FHRR T able 5-1]2,3 FHRR T able 5-1 CDB T otal Water 21.03 at site structure 21.03 at site structure 21.03 at site structure 21.03 at site structure Level (ft-NAVD88) walls walls walls walls 1)

Apply projection for 50 years (remaining operating life of plant) for staff assessment of SLR 2)

Staff total water level applies an average slope method (USACE CEM; small berm width structure) for Indep. Run #1 (negative freeboard) 3)

Staff total water level applies an overtopping method (Van der Meer & Bruce (2014); volumetric based) for Indep. Runs #2/#3 (positive freeboard) 28

Figure 8 Storm Tracks for Top Three Storms in the Independent Analysis Water Levels 8.2 Probabilistic Storm Surge Water Levels.

Information Submitted by NextEra NextEras storm surge evaluation for the Seabrook site involved components that were probabilistically informed. The probabilistically-informed components of the approach focused on development and statistical analysis of synthetically derived data. Specifically, NextEra generated 20,400 synthetic storm tracks and used SLOSH to estimate surge heights for each storm track. Next, NextEra used Delft3D to generate refined surge heights for 27 of the 20,400 synthetic storm tracks. A statistical analysis (linear regression model) was used to define a relationship between SLOSH and Delft3D surge heights using the 27 synthetic results. The linear model was used to estimate refined surge heights for the remaining synthetic data points.

Finally, NextEra performed a statistical analysis to fit a distribution to the synthetically derived surge data and to select the PMSS. Figure 2 provides an overview of NextEras approach.

29

Taylor Engineering Staff Technical Evaluation The Taylor Engineering staff performed an initial review of NextEras probabilistically-informed approach, including performance of several initial sensitivity studies. As part of its initial review, Taylor Engineering staff identified several topics for which additional review or information was required. These topics related to NextEras methodological approach for development of the aleatory model as well as consideration of epistemic uncertainties (including sensitivity of results to modeling decisions and assumptions). For example, topics included treatment of tides and sea level rise, storm parameter ranges and discretization used in generation of synthetic storm parameter sets, the approach used to assign return periods to wind speeds, the sufficiency of sample size for large surge events, the assumed linear relationship between SLOSH and Delft3D results, and the non-parametric approach used to fit a distribution to synthetic surge data.

In recognition of the uncertainty associated with the probabilistically-informed assessment as well as the sensitivity of results to various modeling decisions and assumptions, NextEra adjusted their probabilistically-informed PMSS elevation by adding a 4.00 ft sensitivity factor.

The resultant stillwater elevation was 17.8 ft-NAVD88. This stillwater elevation is generally consistent with the results of independent, deterministic assessments performed by the Taylor Engineering staff. As a result, the Taylor Engineering staff was able to conclude that the results of the assessment were reasonable and the Taylor Engineering staff did not perform detailed, further review of the probabilistically-informed assessment methodology.

9.0 Wave Runup, Inundation, and Drawdown.

Information Submitted by NextEra As the PMH event produced a water surface elevation (WSEL) of 17.75 ft-NAVD88 and as the top of the revetment (flood barrier) elevation is at 19.23 ft-NAVD88, NextEra estimated the water depth, wave height, and wave period and applied the USACE Coastal Engineering Manual (USACE, 2011) to estimate a wave runup height of 5.6 ft (Figure 9). Adding the PMH-generated WSEL and wave runup height, NextEra provided the peak water surface elevation at 23.35 ft-NAVD88. NextEra stated that this runup occurs at the vertical seawall section, approximately 20 m from any structures (FHRR, Sec. 4.4.9.4).

Table 21 provides NextEras estimated maximum flow depths and related velocities at select points of interest (POI) (see POIs in Figure 10 and Figure 11) resulting from overtopping waves.

(FHRR, Sec. 4.4.9.6).

NextEra used the PMSS model and selected an observation point located 1 km past the Seabrook inlet, in the general area of the intake structures to ensure an accurate low water elevation to estimate the drawdown from a PMSS. NextEra used the 10% exceedance low tide (see Table 10) and a constant 100 miles per hour wind at a constant direction of 270 degrees (nautical convention) to estimate a low water elevation of -21.42 ft-NAVD88 (-20.65 ft-Plant Datum) (FHRR, Sec. 4.13.2).

30

Figure 9 Wave Runup on Impermeable Vertical Wall (Source: FHRR, Figure 4-45, from USACE CEM) 31

Table 21 Wave Overtopping Maximum Flow Depths and Velocities (Source: FHRR, Table 4-41) 32

33 34 Figure 10 Seabrook Points of Interest (POI), East (Source: FHRR, Figure 4-9) 35

Figure 11 Seabrook Points of Interest (POI), West (Source: FHRR, Figure 4-10) 36

Taylor Engineering Staff Technical Evaluation NextEras estimation of surge-induced flood wave runup, inundation, and drawdown are acceptable.

10.0 Flood Event Duration Information Submitted by NextEra NextEra estimated the duration of surge-induced flood duration from overtopping waves at approximately 4.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> (Figure 12).

Figure 12 Time Series of Wave Overtopping (Source: FHRR, Figure 4-46)

Taylor Engineering Staff Technical Evaluation Based on review of NextEras Delft3d and overtopping results and the Taylor Engineerings independent simulations, NextEra's estimate of the duration of the flood event is acceptable.

11.0 Hydrostatic and Hydrodynamic Forces.

Information Submitted by NextEra NextEra applied methods presented in the USACE Coastal Engineering Manual (CEM) (USACE, 2011) and Federal Emergency Management Agency (FEMA) P-259 (FEMA, 2012) to calculate hydrostatic and hydrodynamic forces generated by flood water waves during the PMSS. Table 22 provides NextEras estimated hydrostatic and hydrodynamic forces from a PMSS at the points of interest (see Figure 10 and Figure 11) (FHRR, Sec 4.11.2).

37

Table 22 Calculated Hydrostatic and Hydrodynamic Forces at Points of Interest (Source: FHRR, Table 4-52) 38

39 Taylor Engineering Staff Technical Evaluation Taylor Engineering staff independently verified hydrostatic and hydrodynamic force calculations based on water surface and ground elevations and depth and flow velocities found in FPL-081-CALC-021_Hydrostatic and Hydrodynamic Loading Calculation Rev 1.pdf document. Taylor Engineering staff calculations substantially agreed with force values listed in FHRR Table 4-52.

12.0 Debris and Water-Borne Projectiles.

Information Submitted by NextEra NextEra did not provide an analysis of debris and water-borne loadings because of NextEras opinion that maximum water depths on site due to local intense precipitation (LIP) or PMSS flooding are on the order of 1 - 2 feet and are transient in nature. The low levels and short durations will not support transport of debris of significant size, so no further evaluation of water-borne projectiles and debris loading is warranted (FHRR, Sec. 4.12).

Taylor Engineering Staff Technical Evaluation NextEras estimated depths do not exceed 1.5 ft at any FHRR Point of Interest. Taylor Engineering staff agrees that this depth would not support transportation of any debris of significant size.

13.0 Effects of Sediment Erosion and Deposition.

Information Submitted by NextEra NextEra did not provide an analysis of sediment erosion and deposition because of NextEras opinion that the drainage area for Seabrook consists mostly of concrete and paved surfaces which contain none or very few unconsolidated particles. Therefore, Seabrook cannot provide the amount of sediment necessary to lead to any accumulation at the points of interest (FHRR, Sec. 4.11).

Taylor Engineering Staff Technical Evaluation Taylor Engineering staff agrees with NextEras opinion that as Seabrook consists mostly of concrete and paved surfaces, significant sediment erosion is not likely and thus Seabrook cannot provide enough sediment to result in deposition at the points of interest.

14.0 Consideration of Other Site-Related Evaluation Criteria.

Information Submitted by NextEra 40

NextEra did not provide discussion of how seismic and non-seismic information was used in the postulation of worst-case storm surge scenarios.

Taylor Engineering Staff Technical Evaluation NextEra did not discuss how seismic and non-seismic information was used in the postulation of worst-case storm surge scenarios.

15.0 Conclusion The information on flooding from storm surge that is specific to the data needs of the Integrated Assessment is described in Section 5 of the FHRR.

The Taylor Engineering staff confirmed NextEras conclusion that the reevaluated hazard for flooding from storm surge is not bounded by the current design basis flood hazard; therefore, NextEra should include flooding from storm surge within the scope of the Integrated Assessment.

41

16.0 References ANSI/ANS, 1992, American National Standards Institute/American Nuclear Society (ANSI/ANS),

Determining Design Basis Flooding at Power Reactor Sites, ANSI/ANS-2.8-1992, La Grange Park, Illinois, July 28, 1992.

Armstrong, 2013, Armstrong, T., Superstorm of 1993 Storm of the Century, National Weather Service (NWS), http://www.erh.noaa.gov/ilm/archive/Superstorm93/.

Blake et al., 2012, Blake, E.S., T.B. Kimberlain, R.J. Berg, J.P. Cangialosi, and J.L. Beven, Tropical Cyclone Report, National Hurricane Center, 157 pp.

Booij et al., 1999, Booij, N., L.H. Holthuijsen, and R. Ris, A third-generation wave model for coastal regions, Part I, Model Description and Validation, Journal of Geophysical Research, Volume 104 (C4), pp. 7649-7666, 1999.

Burt, 2012, Burt, C., The Great Blizzard of 1888; Americas Greatest Snow Disaster, Weather Underground Blog, http://www.wunderground.com/blog/weatherhistorian/the-great-blizzard-of-1888-americas-greatest-snow-disaster.

Cooperman and Rosendal, 1963, Cooperman, A.I. and H.E. Rosendal, Mean Five-Day Pressure Pattern of the Great Atlantic Coast Storm, March 1962, Monthly Weather Review.

Deltares, 2011c, Delft3D-FLOW, Simulation of multi-dimensional hydrodynamic flows and transport phenomena, including sediments, User Manual Version 3.15, Revision 18392, 2600 MH Delft, The Netherlands, September 7, 2011.

Deltares, 2011e, DELFT3D-WAVE, Simulation of short-crested waves with SWAN, User Manual Version 3.04, Revision 15779, 2600 MH Delft, The Netherlands, May 18, 2011.

FEMA, 2012, Federal Emergency Management Agency (FEMA), Engineering Principles and Practices for Retrofitting Flood-Prone Residential Structures (Fourth Edition), FEMA P-259, January 2012.

Halverson and Rabenhorst, 2013, Halverson, J.B. and T. Rabenhorst, Hurricane Sandy: The Science and Impacts of a Superstorm, Weatherwise, 18 pp.

Hayden, 1889, Hayden, E., Great Storm off the Atlantic Coast of the United States, March 11 to 14th, 1888 Part 1, Part 2 National Geographic Magazine, Volume 1, No. 1.

McQueen et al., 1956, McQueen, H.R. and H.C. Keith, The Ice Storm of January 7-10, 1956 over the Northeastern United States, Monthly Weather Review.

NCDC, 2014, National Climatic Data Center (NCDC), Hourly/Sub-Hourly Observational Data, http://www.ncdc.noaa.gov/cdo-web/.

NCEP, 2014, National Center for Environmental Prediction (NCEP), North American Regional Reanalysis, http://www.emc.ncep.noaa.gov/mmb/rreanl/.

42

NOAA, 1947, National Oceanic and Atmospheric Administration (NOAA), National Ocean Service (NOS) Hydrographic Survey: H07140, released 1947, National Geophysical Data Center (NGDC): Bathymetry & Digital Elevation Models Website, http://maps.ngdc.noaa.gov/viewers/bathymetry/, accessed January 2014.

NOAA, 1954, National Oceanic and Atmospheric Administration (NOAA), National Ocean Service (NOS) Hydrographic Survey: H08096 & H08097, released 1983, National Geophysical Data Center (NGDC): Bathymetry & Digital Elevation Models Website, http://maps.ngdc.noaa.gov/viewers/bathymetry/, Accessed January 2014.

NOAA, 1983, National Oceanic and Atmospheric Administration (NOAA), National Ocean Service (NOS) Hydrographic Survey: H10129, released 1983, National Geophysical Data Center (NGDC): Bathymetry & Digital Elevation Models Website, http://maps.ngdc.noaa.gov/viewers/bathymetry/, Accessed January 2014.

NOAA, 2014l, National Oceanic and Atmospheric Administration (NOAA), Earth System Research Laboratory - Physical Science Division, 20th Century Reanalysis (V2) Data Composites, http://www.esrl.noaa.gov/psd/data/composites/subdaily_20thc/.

NOAA, 2014n, National Oceanic and Atmospheric Administration (NOAA), Earth System Research Laboratory - Physical Science Division, 3-Hourly NCEP North American Regional Reanalysis (NARR) Composites, http://www.esrl.noaa.gov/psd/cgi-bin/data/narr/plothour.pl.

NOAA, 2014o, National Oceanic and Atmospheric Administration (NOAA), Storm Prediction Center, Surface and Upper Air Maps, http://www.spc.noaa.gov/obswx/maps/, November 1998 to present.

NOAA, 2014q, National Oceanic and Atmospheric Administration (NOAA), Earth System Research Laboratory - Physical Science Division, 6-Hourly NCEP/NCAR Reanalysis Data Composites, http://www.esrl.noaa.gov/psd/data/composites/hour/.

NRC, 2013a, U.S. Nuclear Regulatory Commission (NRC), Guidance for Performing a Tsunami, Surge and Seiche Flooding Safety Analysis Revision 0, Japan Lessons-Learned Project Directorate Interim Staff Guidance, JLD-ISG-2012-06, January 4, 2013, ADAMS Accession No. ML12314A412.

USACE, 2011, U.S. Army Corps of Engineers (USACE), Coastal Engineering Manual - Parts I-VI, EM 1110-2-1100, Change 3, September 2011.

USGS, 2011b, United States Geological Survey (USGS), New Hampshire LIDAR: 001001 to 001005, 001034 to 001038, 001064 to 001069, 001096 to 001100, 001126, and 001127, Center for LIDAR Information Coordination and Knowledge Website, https://lta.cr.usgs.gov/LIDAR, accessed January 2014.

WindRiskTech, 2014, Hurricane risk at Seabrook, New Hampshire, Final Report.

43