ML13025A262

Potential for Breaches of Hesco Modular Flood Barriers and Earthen Embankments Affecting the Updated Hydrologic Analysis Results

ML13025A262
Person / Time
Site:	Watts Bar, Sequoyah
Issue date:	01/18/2013
From:	James Shea Tennessee Valley Authority
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
Download: ML13025A262 (47)

Text

Tennessee Valley Authority, 1101 Market Street, Chattanooga, Tennessee 37402 January 18, 2013 10 CFR 50.4 ATTN: Document Control Desk U.S. Nuclear Regulatory Commission Washington, D.C. 20555-0001 Sequoyah Nuclear Plant, Units 1 and 2 NRC Docket Nos. 50-327 and 50-328 Facility License Nos. DPR-77 and DPR-79 Watts Bar Nuclear Plant, Unit 1 NRC Docket No. 50-390 Facility Operating License No. NPF-90

Subject:

Potential for Breaches of HESCO Modular Flood Barriers and Earthen Embankments Affecting the Updated Hydrologic Analysis Results for Sequoyah Nuclear Plant, Units I and 2, and Watts Bar Nuclear Plant, Unit I

References:

1. Tennessee Valley Authority (TVA) Submittal to NRC Document Control Desk, "Impact of Potential Breaches of HESCO Modular Flood Barriers and Earthen Embankments on the Updated Hydrologic Analysis Results for Sequoyah Nuclear Plant, Units 1 and 2, and Watts Bar Nuclear Plant, Unit 1," dated October 30, 2012 (Accession No. ML12307A227)

2. NRC Meeting Summary, "Summary of December 13, 2012, Management Meeting with Tennessee Valley Authority on the Status of Hydrology,"

dated January 11, 2013 (Accession No. ML13010A137)

As committed to in the Reference 1 letter, Tennessee Valley Authority (TVA) has performed an evaluation of possible breach initiators for the HESCO modular flood barriers. These potential breach initiators include uncontrolled commercial river traffic and waterborne debris.

In addition, as discussed in a December 13, 2012 TVA-NRC meeting (Reference 2), the enclosed evaluation details the locations of the flood barriers in relation to the respective dam structures, provides description and results of the reasonable simulations performed for closing the public access (PA) gaps in the HESCO modular flood barriers, and provides Printed on recycled paper

U.S. Nuclear Regulatory Commission Page 2 January 18, 2012 technical details regarding the computational flow dynamics (CFD) analysis for Fort Loudoun and Tellico Reservoirs. A summary of an analysis of flooding levels for SQN Units 1 and 2 and WBN Unit 1 that assumes the HESCO modular flood barriers are not installed and includes failure of the earthen embankments if overtopped in the analysis at Fort Loudoun, Cherokee, Tellico, and Watts Bar Dams is also provided.

The information provided in the enclosure demonstrates that the potential for breaches of the HESCO modular flood barriers is not likely as evidenced by the evaluation of barrier locations, commercial river traffic, river traffic management, CFD model, and waterborne debris and overtopping. TVA has also demonstrated the ability to close PA gaps in the barriers ensuring the effectiveness of the barriers. In addition, the analysis of the flooding levels for SQN Units 1 and 2 and WBN Unit 1 that assumes the HESCO modular flood barriers are not installed demonstrates that the Probable Maximum Flood would increase above the current licensing basis and design basis requirements, reinforcing the decision to install the barriers in 2009. Because the barriers are necessary, permanent modifications to replace the temporary HESCO modular flood barriers will be implemented by October 2015 as previously committed to by TVA.

The enclosure contains no new regulatory commitments. Please address any questions regarding this request to Terry Cribbe at (423) 751-3850.

Resp ully, J1. . Sea Vic P sident, Nuclear Licensing

Enclosure:

Potential for Breaches of the HESCO Modular Flood Barriers and Earthen Embankments at Design Basis Flood Levels cc (Enclosure):

NRC Regional Administrator - Region II NRC Senior Resident Inspector - Watts Bar Nuclear Plant, Unit 1 NRC Senior Resident Inspector - Sequoyah Nuclear Plant

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS By letter dated October 30, 2012 (Reference 1), the Tennessee Valley Authority (TVA-)'

submitted a summary of an evaluation that identified changes to the Probable Maximum Flood (PMF) elevations at Sequoyah Nuclear Plant (SQN) Units 1 and 2 and Watts Bar Nuclear Plant (WBN) Unit 1 resulting from non-mechanistic postulated breaches of the HESCO modular flood barriers and a subsequent failure of the earthen embankments of the dams. This submittal was based upon discussions held between TVA and NRC staff on May 31, 2012, to discuss issues regarding the updated hydrologic analysis for SQN Units 1 and 2 and WBN Unit 1 (Reference 2). As part of the presentation, TVA described the circumstances that* in 2009, it had placed temporary barriers atop portions of four dams upstream of the SQN and WBN sites.

In previous correspondence regarding the barriers (References 3 through 7), TVA had described the barriers as temporary measures that were adequate to prevent overtopping of the specific upstream dams under certain hydrologic conditions. In a letter dated January 25, 2012'

.(Reference 8), the NRC discussed the review of this previous correspondence, and indicated the following:

"[TI*he'NRC staff finds that the sand baskets are not capable of resisting debris impact.

These documents neither discuss the ability of sand baskets to withstand debris impact, or mention whether the baskets are designed for impact of debris loads. The NRC staff is unable to conclude that these sand baskets were designed to withstand impacts from large debris during a -flood. If a design flood were to occur, there is a high likelihood that -

significant debris would accompany the flood waters which could impact the baskets. There is the potential for this debris to damage the baskets or push the individual baskets apart causing a breach. There would be no time to repair the baskets because the flood would already be in progress. Therefore, sand baskets that are'not designed and constructed to withstand impacts from large debris are-not acceptable as a long-term solution."

During the May 31, 2012 meeting, TVA confirmed that the modular flood barriers were intended to be temporary structures and that TVA intends to replace the modular flood barriers with permanent modifications by October 2015. To support the conclusion thatthe modular flood barriers represent an acceptable interim measure, TVA presented the results of the range of impact tests that were previously presented in the Reference 4 letter. In.addition, TVA

- presented the results of a qualitative assessment of the potential for barges on the Tennessee River to impact the modular flood barriers at each of the four affected dams. Finally, in an initiative to gain additional insight as to the potential impact on flood level at SQN Units 1 and 2 and WBN Unit 1 from a failure of a portion of the modular flood barriers at the upstream dam prior to the completion of the permanent dam modifications, TVA committed to perform a hydrologic analysis assuming such a failure. In a letter dated June 13, 2012 (Reference 9), TVA followed up the verbal commitment made at the May 31, 2012 meeting with the following written commitment (numbered as Commitment No. 6):

"By August 31, 2012, TVA will perform an analysis of the Design Basis Flood for SQN Units 1 and 2 and WBN Unit 1 that assumes a failure of a section of the HESCO flood barriers and earthen embankments at Fort Loudoun, Cherokee, Tellico, and Watts Bar dams."

By letter dated June 25, 2012 (Reference 10), the NRC confirmed the commitments made by TVA in the Reference 9 letter. In the Reference 10 letter,.the NRC further directed:

"With regard to Commitment No.6, please provide a summary of the results of TVA's analysis to the NRC within 60 days after its completion."

Page 1 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS TVA completed the analysis associated with Commitment No. 6 of the Reference 9 letter on August 31, 2012, which was provided by letter dated October 30, 2012 (Reference 1).

Following submittal of the analysis of the flooding levels for SQN Units 1 and 2 and WBN Unit 1 that assumes a failure of a section of the HESCO modular flood barriers-and earthen embankments at Fort Loudoun, Cherokee, Tellico, and Watts Bar Dams, TVA and NRC had a series of meetings to discuss status of hydrology issues including the potential for breaches of the HESCO modular flood barriers on December 3, 2012, and December 13, 2012,.

(References 11 and 12). This enclosure presents the results of an evaluation of possible breach initiators to determine their credibility. These potential breach initiators include uncontrolled commercial river traffic and waterborne debris. In addition, the enclosed evaluation details the exact locations of the flood barriers in relation to the respective dam structures, provides description and results of the reasonable simulations performed for closing the public access (PA) gaps in the HESCO modular flood barriers, provides additional technical details regarding the computational flow dynamics (CFD) analysis for Fort Loudoun and Tellico Reservoirs, and provides a summary of an analysis of the flooding levels for SQN Units 1 and 2 and WBN Unit 1 that assumes the HESCO modular flood barriers are not installed and includes failure of the earthen embankments if overtopped in the analysis at Fort Loudoun, Cherokee, Tellico, and Watts Bar Dams.

Purpose The simulations described in the Reference 1 letter postulate a non-mechanistic failure of the HESCO modular flood barriers resulting in a breach of the earthen embankment. However, the analysis does not take credit for the locations of the HESCO modular flood barriers that minimize the potential for impact failures during flooding, and does not assume any specific cause of the breach. 'Potential breach initiators include barge impacts.from commercial river traffic or waterborne debris impacts from material (such as logs) floating downstream carried by the flood waters. In addition, the potential for breaches are influenced by the flows through the reservoirs during PMF events, and the ability to close PA gaps left in the flood barriers prior to the PMF event reaching critical flood elevations. Finally, an analysis was performed to determine the impact on flooding levels at SQN Units 1 and 2 and WBN Unit 1 that assumes the HESCO modular flood barriers are not installed and includes failure of the earthen embankments ifovertopped in the analysis at Fort Loudoun, Cherokee, Tellico, and Watts Bar Dams. The results of this analysis reinforce the importance of ensuring that the earthen embankments at the affected dams are precluded from overtopping and reinforce the basis for installing the HESCO modular flood barriers until such time as permanent modifications to the dams can be implemented. The evaluations below address these areas of concern.

scope The factors considered in evaluating the potential failure mechanisms for the HESCO modular flood barriers include the following:

1. HESCO Modular Flood Barrier Locations - Locations of the HESCO modular flood barriers in relation to the edges of the reservoirs, and the topography of the reservoirs and earthen embankments underlying the modular flood barriers, reduces the potential for impact by commercial barges or other waterborne objects during a PMF event.

Page 2 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS

2. Commercial River Traffic - The small amount of commercial river traffic in each reservoir reduces the likelihood that commercial barges would be located in each reservoir during a PMF event, and then available to impact the modular flood barriers if they become uncontrolled.

3. River Traffic Management - The control of commercial transportation during high river levels and flows reduces the likelihood that commercial barges would become uncontrolled during a PMF event and then possibly impact the modular flood barriers.

4. Computational Flow Dynamics (CFD) Model for Fort Loudoun and Tellico Reservoirs - The actual flow profile through the reservoirs during high river levels and flows reduces the likelihood that commercial barges would be directed towards the modular flood barriers during a PMF event.

5. Waterborne Debris and Overtopping - The ability. of the HESCO modular flood barriers to withstand impacts from waterborne debris preventing overtopping of the underlying earthen embankments reduces the likelihood of a failure of the modular flood barriers during a PMF event.

6. Ability to Close Public Access (PA) Gaps in HESCO Modular Flood Barriers - The ability to implement procedures to close the gaps left for public access through the HESCO modular flood barriers increases the likelihood that the modular flood barriers will be effective during a PMF event.

In addition, an analysis was performed to determine the impact on flooding levels at SQN Units 1 and 2 and WBN Unit 1 that assumes the HESCO modular flood barriers are not installed and includes failure of the earthen embankments if overtopped in the analysis at Fort Loudoun, Cherokee, Tellico, and Watts Bar Dams. This analysis supports the TVA decision to install the HESCO modular flood barriers in 2009.

HESCO Modular Flood Barrier Locations contains drawings detailing the locations of the HESCO modular flood barriers.

From these drawings, it can be seen that the flood barriers are located on the land side (as opposed to the reservoir side) of roadways or earthen embankments upon which the flood barriers are installed, where possible. Locating the flood barriers in this manner provides additional protection from barge and waterborne debris impact.

For Watts Bar Dam and Fort Loudoun Dam, where commercial barge traffic currently exists, the following discussions describe the potential for impacts by uncontrolled barges on the HESCO modular flood barriers based on flood barrier location.

1. Watts Bar Dam During a PMF event, commercial barge traffic is restricted based on prediction of critical headwater elevations by TVA River Forecast Center. This restriction is enforced by lock closure for Watts Bar Dam at a spillway discharge flow rate of 100,000 cfs. At this point, the headwater elevation at Watts Bar Dam will be low enough that any commercial barge traffic Page 3 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS approaching the embankments at Watts Bar Dam would ground before reaching any portion of the barrier. As shown in Figure 1, a potential head-on barge strike of the HESCO modular flood barriers is averted by the barge grounding before impacting the flood barriers during a PMF event. This analysis uses a standard barge size of 10.5 ft by 35 ft by 195 ft.

The barge is assumed empty to maximize the amount out of the water and conservatively assumes the reservoir level at peak PMF elevation. A barge with more cargo would exhibit more draft and would also ground before reaching the flood barriers. It should be noted that the peak PMF elevation is approximately at the lowest elevation of the earthen embankments for Watts Bar Reservoir as shown in the figure. Therefore, the embankments would not be overtopped by a PMF event and the HESCO modular flood barriers are not needed to preserve the assumptions of the current hydrologic analysis. Based on this geometrical analysis, barge impact is not considered possible for the HESCO modular flood barriers installed at Watts Bar Dam, except for a section of approximately 75 ft near the South Embankment where a barge may, if precisely oriented, make contact with the barriers. However, since the peak PMF does not reach the bottom of these or any other barrier for Watts Bar Dam, a breach of the barrier by a barge at peak PMF would have no consequence.

Figure 1 - Barge Grounding Analysis for Watts Bar Dam

2. Fort Loudoun Dam During a PMF event, commercial barge traffic is restricted based on prediction of critical headwater elevations by TVA River Forecast Center. This restriction is enforced by lock closure for Fort Loudoun Dam at a spillway discharge flow rate of 60,000 cfs. At this point, the headwater elevation at Fort Loudoun Dam will be approximately elevation 812.0 ft. At this elevation, any commercial barge traffic approaching the South Embankment would ground before reaching any portion of the barrier. Thus, no impacts from commercial river traffic could damage the HESCO modular flood barriers installed on the South Embankment of Fort Loudoun Dam prior to river closure. At peak headwater elevation during a PMF event, impact from an uncontrolled commercial barge is possible but only if the barge is directed by flow towards the barriers. As shown in the CFD analysis performed for Fort Loudoun and Tellico Reservoirs presented later in this enclosure, flow through Fort Loudoun Page 4 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS Reservoir at PMF levels would direct uncontrolled commercial barges and other smaller objects and debris towards Fort Loudoun Dam instead of towards the HESCO modular flood barriers. Therefore, a breach of the barriers at Fort Loudoun Dam is not likely.

Commercial River Traffic Commercial river traffic on the Tennessee River primarily consists of commercial barges and towboats. Cherokee Dam is located above the navigable channel, and Cherokee Reservoir has no commercial river traffic. Tellico Reservoir has no commercial harbors or facilities, so no commercial river traffic is expected on the Little Tennessee River or through the Fort Loudoun/Tellico Channel between the two reservoirs. Therefore, there is no expectation of commercial barge impacts to the HESCO modular flood barriers located at Cherokee and Tellico Dams. However, there are commercial barges routinely in operation and moving through the Fort Loudoun Reservoir and Watts Bar Reservoir as shown in Table 1.

Table 1 - Commercial River Traffic Fort Loudoun Watts Bar Total Annual Lock Total Annual Lock Barge/Tow Traffic Barqe/TowTraffic 2008 244 263 2009 188 204 2010 220 251 2011 236 245 River Traffic Manaaement Commercial transportation on the Tennessee River is managed through regulations by the United States Coast Guard (USCG). Operation of the dams and locks is managed by TVA River Operations and the United States Army Corps of Engineers (USACE). The roles and responsibilities of these organizations are described in the Tennessee River Waterway Management Plan (https://www.tva.gov/river/naviqation/pdf/waterway plan.pdf) co-authored and implemented by TVA River Operations, USACE, and USCG.

Although the location of the HESCO modular flood barriers provides some protection of the flood barriers during a PMF event for the reservoirs that have routine commercial barge operations, the possibility of uncontrolled barges impacting the flood barriers is minimized by the following considerations as discussed in the Tennessee River Waterway Management Plan:

1. Commercial transportation on the Tennessee River is managed by the USCG as follows:

• The hydrological and meteorological factors that weigh into decisions to curtail commercial river traffic are described in the plan.

• Waterway management practices and procedures are included in the plan that describe the Marine Transportation Emergency Response Organization; system management Page 5 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS and controls as described by regulations including the issuance of safety advisories; the establishment of Safety Zones; and establishment of methods of communications.

" An outline of the issues associated with Waterway Management during high water, including impacts to navigating tows, impacts on moored, fleeted vessels and facilities, and impacts when navigation is halted, are described in the plan.

" The Marine Transportation Emergency Response Cycle is described in the plan, which includes the four phases of the response cycle - Watch, Action, Emergency, and Recovery.

2. During high river levels and flows, actions discussed in the plan are taken to prevent vessel casualties, pollution incidents, and barge breakaways, as follows:

• Upon the forecast of the large rain event, a teleconference is conducted to assemble the members of the emergency response team, including the USCG, TVA River Operations, and USACE, to discuss potential protective actions to be implemented including issuance of a safety advisory.

" Rainfall simulations are generated by TVA River Operations to forecast river levels and flows.

• The USCG establishes Safety Zones as further described below.

• Commercial river traffic operators are notified by the USCG of potential high flow conditions so the operators can begin taking protective actions, including banning of barge traffic from Safety Zones and relocating tows outside of the Safety Zones.

• Upon reaching spillway flows of 60,000 cfs at Fort Loudoun Dam and 100,000 cfs at Watts Bar Dam, the locks are closed respectively at those dams by the USACE, preventing further commercial barge traffic between reservoirs.

• As described in the safety advisory issued by the USCG in consultation with TVA River Operations and the USACE, commercial river traffic is to remain outside of the Safety Zones until the event is declared over.

3. Safety Zones as established by the USCG, in consultation with TVA River Operations, USACE, and industry-user groups, result in the protection of waterfront facilities including the dams and HESCO modular flood barriers, as follows:

• A Safety Zone as defined by regulations is the areas of land, water, or land and water, which are so designated by the USCG Captain of the Port for such time as he deems necessary to prevent damage or injury to any vessel or waterfront facility, to safeguard ports, harbors, territories, or waters of the United States.

• The USCG Captain of the Port has the authority granted to him by 33 CFR, Chapter 1, Section 1.01-30, to establish Safety Zones. As defined by 33 CFR, Chapter 1, Part 165, Subpart C, Section 165.20, a "Safety Zone is a water area, shore area, or water and Page 6 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS shore area to which, for safety or environmental purposes, access is limited to authorized persons, vehicles, or vessels. It may be stationary and described by fixed limits or it may be described as a zone around a vessel in motion."

Safety Zones are established to prevent damage or injury to any vessel or waterfront facility, which has the effect of protecting TVA dam assets including the-adjacent HESCO modular flood barriers by preventing commercial river traffic from approaching those assets.

For commercial river traffic which is not relocated downstream prior to lock closure, ample time is available to relocate tows outside of the Safety Zones before the PMF elevations are reached, as follows:

• As shown in Attachment 2, the following time intervals between reaching the action level threshold discharge flow rate and the peak PMF elevation at the respective dams are determined:

- For Fort Loudoun Dam, the peak PMF elevation is reached 7 days 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> after the 60,000 cfs discharge flow rate threshold is reachedfor the.21,400 square-mile storm, and 7 days 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after the 60,000 cfs discharge flow rate threshold is reached for the 7,980 square-mile storm (pages 1 and 2 of Attachment 2).

- For Watts Bar Dam, the peak PMF elevation is reached 7 days 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after the 100,000 cfs discharge flow rate threshold is reached for the 21,400 square-mile storm, and 8 days after the 100,000 cfs discharge flow rate threshold is reached for the 7,980 square-mile storm (pages 3 and 4 of Attachment 2).

• Reaching the threshold discharge flow rates triggers the USACE to close the Fort Loudoun Lock and Watts Bar Lock, and commercial operators are notified to relocate their tows outside of the Safety Zones.

No commercial river traffic is expected to be in the main river channel during the peak of the PMF event. This minimizes the possibility of a barge traveling downstream and impacting the HESCO modular flood barriers at Fort Loudoun Dam and Watts Bar Dam. Furthermore, with no commercial river traffic on either the Tellico or Cherokee Reservoirs, there is no possibility of barge impacts to the HESCO modular flood barriers located at those dams.

Computational Flow Dynamics (CFD) Model for Fort Loudoun and Tellico Reservoirs Fort Loudoun Dam is located at Tennessee River Mile (TRM) 602.30. Commercial harbors or docks are located upstream at approximately TRM 646.10, almost 44 miles upstream of the dam and lock. Assuming that a barge could break loose from this location and -navigate its way downstream, a CFD model for the Fort Loudoun and Tellico Reservoirs was developed to determine the likely movement of anuncontrolled barge during a PMF event.

The University of Tennessee at Chattanooga (UTC)' National Center for Computational Engineering (SimCenter) developed a CFD model to predict the movement of large objects such as barges and other waterborne debris during a PMF event. The CFD model includes sections*

Page 7 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS of the Tennessee and Little Tennessee Rivers near Fort Loudoun and Tellico Dams. The model includes the channel and downstream portion of the Tennessee River past the confluence of the Little Tennessee River.

The SimCenter utilized existing land area elevation data from United States Geological Survey (USGS), bathymetry data available from TVA (via US Army Corps of Engineers, USACE),

boundary conditions available from the SOCH (Simulated Open Channel Hydraulics) one-dimensional model from TVA, and solution algorithms developed at the SimCenter. The solution algorithms were used to predict the Tennessee River and Little Tennessee River flows at peak flooding conditions. Given the flowfield, the trajectory of large objects simulating barges and other waterborne objects under a zero-power condition was computed for a range of release locations.

The topography (Digital Elevation Model, DEM) was obtained from the National Map Viewer (USGS data). The land area used as the solution domain was chosen such that boundaries were placed far enough away from the complex flow near the dams such that the incoming and outgoing river flows would be reasonably uniform. The downloaded resolution was 1/3 arcsecond, which provides approximately a 10 m resolution at the earth's surface. The data were downloaded as a GeoTIFF, and code was developed to read the GeoTIFF's data into a useable form. At this point, the data consisted of a longitude/latitude/elevation triple. The raw GeoTIFF data are rendered in Figure 2.

I-igure z - -enaerea version OT ueo I I1-I- U:M Elevation uata The bathymetry data were obtained from TVA (via the USACE) as a set of sonar soundings that follow the path of the vessel that took the data. Data were obtained from TVA for the Tennessee River both upstream and downstream of Fort Loudoun Dam, but no data were available for the Little Tennessee River. Therefore, code was developed to read in the data and convert from the Tennessee State Plane (TNSP) coordinate system to UTM zone 16. This code Page 8 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS made use of the nad2nad utility, which is part of the PROJ.4 Cartographic Projections Library (http://trac.osgeo.oral/roi/). The available Tennessee River bathymetry data and results from this code for the Little Tennessee River were cross-checked with conversions made with ArcGIS for verification of correctness.

As a preliminary step before merging the bathymetry and topography data, the shoreline of the water system (as dictated by the pool level when the DEM data were collected) was identified to determine where the topographical data were valid. Along with the DEM data, satellite imagery was downloaded to confirm validity of the identified shoreline. The result of merging the DEM data and satellite imagery is shown in Figure 3, and shown as red areas in Figure 4.

Figure 3 - Satellite Imagery Tiles Downloaded from USGS after Merging Once the topographical and bathymetry data were both collected and converted to UTM coordinates, a process was developed to merge the two into one final geometry. This process first required that a domain be chosen that is at least marginally smaller than the land area covered by the topology, bathymetry, and the satellite imagery. Then, the UTM coordinate mesh was formed independently, and for every point that is in the mesh, the UTM coordinates were taken and converted to a longitude/latitude pair for indexing into the topography (DEM) dataset, which was taken in longitude/latitude also. The elevation for each UTM point in the mesh was interpolated, using area-weighted basis functions, from the DEM dataset in this manner. The Python UTM library was utilized to convert UTM zone 16 coordinates into longitude/latitude pairs. It should be noted that the UTM conversions performed in this way Page 9 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS were cross-checked against ArcGIS conversions, in order to verify that the calculations were correct.

A mechanism was then created to raise or lower features in the topology where data were not available (e.g., the Little Tennessee River where no bathymetry data were available, where the channel bottom was estimated to be 10 m below the pool level recorded in the DEM dataset).

There was also no bathymetry data for the Fort Loudoun/Tellico Channel between the two reservoirs, so it was also estimated to be 10 m. Other small side channels were estimated to be 8 m in depth. The specifics of these adjustments are shown as green areas in Figure 4.

The HESCO modular flood barriers and dams were also raised in a similar manner, since manmade structures are not available in the USGS datasets. Thus, since the elevations and locations of both the HESCO modular flood barriers and the dams themselves are known, the overlay shown as blue areas in Figure 4 was used to set the elevations of both to 839.9 ft.

While this height is approximately five feet too tall in most places, it does not affect the simulation, since the water levels are at 835 ft and below at steady state. The extra height serves to hold the water back during the solution transients.

The composite of these adjustments are shown in Figure 4.

Figure 4 - Composite Image Indicating Water Areas (Red), Areas Where Original Topology was Raised or Lowered (Green), and Areas Where the Elevation was Fixed (Blue)

After the bathymetry was applied to the water areas, bathymetry smoothing was necessary to address areas where depths were unknown. Therefore, a method was constructed to propagate known depths to the nearby areas where the depth is unknown. In order to determine a reasonable river bottom, a nine-point Laplacian smoothing was applied to the points in the mesh where the elevation is unknown. This set of points is the difference of the sonar track points and the points that are shaded red in Figure 4. The known topography was held fixed so that the river bottom, as it is formed by smoothing, merges smoothly with the Page 10 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS known land areas. The smoothing algorithm was executed until no further change in the elevations was detected. In this way, bathymetry at known locations is spread to the other points for which there are no bathymetry data or topographical data.

After the completion of the tasks above, the final topology is shown in Figure 5. The upper image is the actual topology in the proper scale, and the second image has the elevations exaggerated by a factor of 3 so that the relief is more visible.

Figure 5 - Completed Topography with Integrated Bathymetry A comparison of the USGS topography to the topography with integrated bathymetry is shown in Figure 6. Elevations have been exaggerated by a factor of 3 so that the relief is more visible.

Page 11 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS Figure 6 - Comparison of USGS Topography (Top) to Topography with Integrated Bathymetry (Bottom)

Boundary conditions were then established for the CFD flow solution algorithm. Data from the SOCH model for the limiting PMF event were used to determine the inlet heights and flow velocities at approximately LTRM 3.6 and TRM 605.75. Conditions used were:

• Little Tennessee Inlet: Water height = 833.0 ft, incoming velocity = 2.3 ft/s

• Tennessee Inlet: Water height = 835.6 ft, incoming velocity = 7.48 ft/s When necessary (depending on the domain extent used in the flow simulation and the type of simulation used), outflow targets at the Fort Loudoun and Tellico Dams were:

• Tellico Dam Outlet: Water height = 833.28 ft, discharge = 601,898 cfs

" Fort Loudoun Dam Outlet: Water height = 835.45 ft, discharge = 640,356 cfs Page 12 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASISFLOOD LEVELS Several solution algorithms and tools were available for predicting the flowfield once the topography was determined. These tools include the following:

• Tenasi Shallow Water Solver- Thisis a shallow water solver in alpha state, and is integrated into the larger Tenasi solver framework. Tenasi is the workhorse solver used to generate many flowfield results in the areas of aerodynamics and open-water hydrodynamics. By definition, a shallow-water solver is a two-dimensional solver that depth-averages each water column in the domain (Reference 13).

Tenasi-SWC'Shallow Water Solver- This is another shallow water solver in alpha state.

This solver is currently not parallel-capable, and therefore was not able to generate a solution in the time frame allotted.

• Tenasi Free-Surface Capturing- This is a validated solver used in the context of surface ship drag calculations and maneuvering (Reference 14).

Tenasi/lncompressible- This is a solver used for low-speed airflows and water flows, and can be used here for the river system, although the full surrounding land topography cannot be taken into account (Reference 15).

• OpenFOAM VoF - OpenFOAM is an open-source CFD solver, and has two solvers (InterFoam and LTSInterFoam) that showed some promise as a solution for this simulation.

This algorithm functions very similarly to Tenasi Free-Surface capturing, in that the combined water/air domain is solved, and a volume-of-fluid parameter is used to track the two distinct media and account for density and property differences between the two (Reference 16).1 Attempts were made at obtaining a solution with each of the above methods. However, it was determined that the Tenasi/lncompressiblesolution method offered the best tradeoff between time to solution and solution fidelity. For the Tenasi/lIncompressible solution, the solution domain was truncated at the Fort Loudoun and Tellico Dams, and a top boundary was defined for the domain at the PMF height of 835 ft.

The Tenasi flow solver is a configured subset of a modular suite of computational modeling and simulation software tools that have been developed at the UTC SimCenter beginning in 2002 and continuing to the present. This SimCenter technology and software tool/suite was originally designed as an in-house platform for developing and assessing cutting-edge computational modeling and simulation algorithms and methodology for complex real-world engineering and science problems. It now enables and supports the SimCenter integrated research and education programs, and also provides advanced modeling and simulation capabilities to support government and industry.

These capabilities encompass the following broad areas of computational simulation involving physical field phenomena that depend on space and time variables and that arise in complex engineering analysis and design problems:

1. Modeling - Physical/mathematical modeling and problem formulation for generic computational engineering problems.

Page 13 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS

2. Grid Generation - Computational algorithms and their implementation in grid-generation software for discretization of geometric surfaces, and volumes for computational simulation.

3. Simulation - Computational algorithms and their implementation in equation-solving software for problem models.

4. Verification and Validation - Testing of algorithms and software for correctness and validity of models.

5. Software Engineering - Development and management of computational engineering software for modern high-performance computers.

6. Commercialization Features - Development of features that enable application and use of technology for problem solving in commercially attractive engineering-problem categories or domains.

The Tenasi software tool/suite consists of modular software tools that are developed, managed, and maintained by an internal SimCenter Developer Group. It can be configured to perform complex field simulations in individual problem categories that are themselves highly specialized engineering applications of field phenomena. The software is also modularly constructed to facilitate use of some third-party freeware and commercially available software.

The Tenasi tool/suite has been developed in-house at the SimCenter, and has been used extensively to generate validated solutions across a wide range of application domains (References 15 and 17 through 24). The code has also been used extensively in hydrodynamic submarine and ship simulations (References 25 through 28), and in combined Eulerian/Lagrangian scenarios (Reference 29). The baseline flow solver in Tenasi Reference 17) employs a finite-volume, implicit scheme with high resolution fluxes and a Newton iteration procedure for time accuracy.

The Tenasi unstructured algorithm solves generic systems of field equations, and does so on meshes with mixed element types. In general, isotropic tetrahedral elements are employed in regions removed from surfaces giving rise to high gradients, and prisms or hexahedrals are extruded from the triangular or quadrilateral surface meshes for high gradient-inducing boundaries (i.e., in the boundary layer for the Navier-Stokes equations), giving rise to a highly anisotropic grid near these surfaces. The solver is node-centered; control volumes are built from a median dual surrounding each vertex of the mesh.

The Tenasi code base is automatically regression tested every three hours against a cumulatively assembled set of validation test cases. This testing procedure ensures that the code base is correct, and that changes to the code base authorized by the SimCenter development team do not change the validated solutions of the solver.

The specific version of the Tenasi codebase used in this simulation is SVN version 12609.

Figure 7 shows the extent of the domain that was simulated using Tenasi/Incompressible. The domain was truncated at the Fort Loudoun and Tellico Dams so that boundary conditions (see above) could be imposed such that desired flow rates were achieved through the respective Page 14 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS dams. A total pressure condition was imposed at the inflow boundaries (Tennessee River and Little Tennessee River in Figure 7) and outflow boundary conditions (based on a specified back pressure) were imposed at the Fort Loudoun and Tellico Dams, respectively. In each case, the flowrates at the dams were matched to the limiting PMF SOCH simulation provided by TVA; and the overall mass balance for the domain at solution terminus was 1%.

Figure 7 - Overview of the Computational Domain for the Flow Solution The top of the computational domain was created by assuming a flood state with a height of 835.6 ft, and intersecting this plane with the combined terrain/bathymetry profile. Since the flow simulation solves the three-dimensional Navier-Stokes equations, a volume mesh was generated using the bathymetry combined with the surface topography for the bottom of the domain, while the top of the domain was obtained as mentioned earlier. This volume mesh contained approximately one million nodes. A surface mesh showing the detail of the grid near the Tennessee River inflow is shown in Figure 8.

Page 15 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS The flow solution obtained using Tenasi/Incompressibleis shown in Figure 9, which shows velocities at the free surface shaded by velocity magnitude.

owing veiociny Magnituae A view of the velocity vectors in the channel connecting the Fort Loudoun and Tellico Reservoirs is shown in Figure 10. As can be seen from this figure, the higher flowrate in the Tennessee River forces flow from the Fort Loudoun Reservoir to the Tellico Reservoir through the Fort Loudoun/Tellico Channel. This is a reversal of the normal flow pattern, which is an expected effect under the limiting PMF condition.

Page 16 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS ire -iu - veiocity vectors in tne r-or LOUUUUrl/ I elllLU k,[nannf1eI Once the flow solution was determined, the barge trajectories were computed by inclusion of inertial particles into the flow system (Reference 30). The trajectory of each particle in the computational domain was computed by solving the following ordinary differential equations for location and velocity:

d

• Xp = Up d 1--+=

where Fp =FDrag + FMagnus + FSaffman + IFAddedMass and x. and up are the position and velocity vectors of the particles respectively. The fluid drag on the particle is modeled using an empirically derived drag equation given by:

FDrag = - CDp D Ur r where Dp is the particle diameter, p is the fluid density, Ur is the velocity of the particle relative to the surrounding fluid, Ur is the magnitude of the of relative velocity vector, and CD is the drag coefficient 0 given by:

CD=24 1+ Re,' forRep p 6 P <1000 CD = 0.424 for Rep, > 1000 Page 17 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS Rep is the particle Reynolds number given by:

Rep - pUrDp For computing the envelopes of large object trajectories in the flooded river, the force terms (except for drag term) are not required. Barges are modeled using a sphere whose wetted area equals that of a 195 ft barge with a width of 35 ft and a draft of 6 ft. The sphere density is chosen to match the laden weight of the barge, 1000 tons. The continuous phase velocities used to drive the time integration are computed at the free-surface or just below it in order to account for the draft of the barge. Since the particle is confined to the free surface, the gravity terms as well as displacements in the vertical direction were deactivated.

The release locations for the modeled barges included the following:

• TRM 608.3, which is located approximately at the northern inlet of the computational domain,

• First Class Harbor located on the western shore of the Tennessee River between TRM 608 and TRM 609,

" First Class Harbor located on the southern shore of the Tennessee River between TRM 606 and TRM 607,

• First Class Landing located on the opposite shore from Fort Loudoun Dam at approximately TRM 603,

• LTRM 3.5, which is located approximately at the southern inlet of the computational domain, and

• First Class Landing located on the western shore of the Little Tennessee River at approximately LTRM 3.0.

Figure 11 shows the results of the simulations of 1000-ton large objects representing barges released as 400,000 inertial particles initially at rest near the domain inlet at approximately TRM 608.3. The released inertial particles follow the primary Tennessee River channel very

-closely throughout its length. As the inertial particles approach the southern bend (TRM 606.5),

the flow constriction and the flow in the bend cause the trajectories to converge to one path as the inertial particles exit the bend at TRM 606. From here, the inertial particle momentum is responsible for the continuation of the course toward Fort Loudoun Dam. Therefore, it is concluded that uncontrolled barges on the Tennessee River will approach the dam spillways, rather than the HESCO modular flood barriers. More detail is shown in Figure 12, where it can be clearly seen that the inertial particle paths are drawn strongly towards the Fort Loudoun Dam instead of towards the HESCO modular flood barriers. The inertial particles also have enough momentum to overcome the Fort Loudoun/Tellico Channel suction, and approach the Fort Loudoun Dam spillways rather than being pulled down toward the South and approaching the HESCO modular flood barriers. Figure 12 shows a 3D detail view of the inertial particle trajectory, with the HESCO modular flood barriers highlighted.

Page 18 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS gure I i - i rajeciornes ior i -ion upjecis K'eieasea Tromn Figure 12 - Trajectories for 1000-ton Figures 13 and 14 show a similar computation performed with 100-ton and 10-ton inertial particles, which much more closely simulates other waterborne objects and debris. Like the 1000-ton inertial particles, the smaller inertial particles display the same behavior of coalescence at the river bend, and preferentially approaching the Fort Loudoun Dam spillways.

Page 19 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS Figure 13 - Trajectories for 100-ton Objects Released from Figure 14 - Trajectories for 10-ton Objects Released from TRM 608.3 1000-ton inertial particle releases from the two First Class Harbors between TRM 608 and TRM 609, and between TRM 606 and TRM 607, each behave very similarly to the releases from the Tennessee River northern inlet at TRM 608.3. These trajectories are shown in Figures 15 and 16. These trajectories again coalesce at the river bend, and approach the Fort Loudoun Dam spillways. One exception is a small number (3% probability) of inertial particle trajectories that beach and/or recirculate in the sloughs immediately adjacent to the First Class Harbor between TRM 606 and TRM 607. Releases from this First Class Harbor initially travel upriver until approximately TRM 607.5, and then migrate to the center of the channel, getting pulled downstream in a manner similar to inertial particles from both the northern inlet and the First Class Harbor between TRM 608 and TRM 609.

Page 20 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS ctories for 1000-ton Objects Released from First Class Harbor between TRM 608 and TRM 609 Figure 16 - Trajectories for 1000-ton Objects Released from between TRM 606 and TRM 607 1000-ton inertial particle trajectories when released from the First Class Landing near Fort Loudoun Dam at approximately TRM 603 are shown below in Figure 17. None of these inertial particles pose a threat to the HESCO modular flood barriers, with most (83% probability) approaching the Fort Loudoun Dam spillways (Figure 18) and the rest (17% probability) beaching on the shoreline adjacent to the landing (Figure 19). The inertial particles that approach the dam are entrained within the flow that naturally exits the dam, and therefore follows its streamlines quite closely.

Page 21 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS Figure 17 - I rajectones for lUUU-ton Objects Released from First I Near Fort Loudoun Dam at Approximately TRM 603 ire 18 - Trajectories for 1000-ton Objects Released from First Class Landing Near Fort Loudoun Dam at Approximately TRM 603 83% that Approach Fort Loudoun Dam Spillways Page 22 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS Figure 19 - Trajectories for 1000-ton Objects Released from First Class Landing Near Fort Loudoun Dam at Approximately TRM 603 17% that Beach As previously stated, there are no commercial interests requiring barge traffic on the Little Tennessee River, and no commercial river traffic exists in Tellico Reservoir. However, to gain additional insights, a simulation was performed for 1000-ton inertial particles released near the domain inlet at approximately LTRM 3.5. The simulation shows that released inertial particles approach the HESCO modular flood barriers with 68% probability. The remaining possibility is that the inertial particles become beached very near the First Class Landing on the northern shore of the Little Tennessee River at LTRM 3.

It should be noted that this particular result should be viewed in the context of the modifications that were required in order to perform the flow simulation. The flow outlet at the Tellico Dam was enlarged slightly to provide more flow area for the outlet boundary condition. Without this modification, it was extremely difficult to numerically "pull" the correct amount of flow over the dam; the required back pressure was too low. Thus, the area was enlarged, which has a significant effect on the computed inertial particle trajectories in the vicinity of Tellico Dam.

Therefore, although the trajectories approach the location of HESCO modular flood barriers, this small section of boundary is actually pulling water out of the domain by definition. Thus the projection that the inertial particles simulating barges would actually approach the HESCO modular flood barriers is suspect. Incidentally, each of the Little Tennessee River inertial particle trajectory computations is affected by this necessary geometry modification.

An additional simulation was performed for 1000-ton inertial particles trajectories released near the First Class Landing located on the western shore of the Little Tennessee River at approximately LTRM 3.0. Like the Little Tennessee River inlet releases, there is some probability (7%) of the inertial particles approaching the HESCO modular flood barriers, with the rest becoming grounded near the First Class Landing. Again, the inertial particle trajectories should not be considered accurate due to the boundary condition change discussed immediately above. Further, the inertial particles were released throughout the area, rather than Page 23 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS specifically against the shore where the First Class Landing is located. It is expected that objects against the shore would much more likely ground along the shoreline.

The combined flow/inertial particle computational solutions demonstrate that the trajectories of uncontrolled large objects including barges originating on the Tennessee River are not likely to approach the HESCO modular flood barriers placed between the Fort Loudoun and Tellico Dams. 1000-ton objects simulating barges were released from the northern inlet and from three different First Class Harbors or Landings on the Tennessee River. The simulated barges have enough momentum to overcome the channel suction and therefore approach the Fort Loudoun Dam spillways, rather than being pulled down toward the South and impacting the HESCO modular flood barriers.

Computation of simulated barge trajectories originating on the Little Tennessee River (from the southern inlet and from one First Class Landing) show a possibility that the HESCO modular flood barriers could be approached. However, these results should be considered indeterminate because of the enlargement of the outlet boundary condition in the vicinity of Tellico Dam. Also, there is no known commercial barge traffic on the Little Tennessee River, so this result only provides additional insight for predicting potential approach to the HESCO modular flood barriers.

Waterborne Debris and Overtopping.

An assumed failure of the HESCO modular flood barriers from waterborne debris other than commercial river traffic has been evaluated. Waterborne debris is primarily organic material such as timber (logs) and brush as well as non-organic material such as automobile tires, bottles, and other human waste. Most of this debris is small in size and mass. However, Waterborne logs traveling downstream are equivalent to low-speed missiles. The impacts of waterborne logs on the HESCO modular flood barriers were evaluated and failure of the flood barriers was determined to be unlikely due to these impacts. Failure of the HESCO modular flood barriers by overtopping has also been tested and found to be unlikely. The documentation supporting this evaluation was previously submitted to the NRC (Reference 4). It is presented here for completeness and context to further support the adequacy of the HESCO modular flood barriers as temporary measures.

1. Extensive laboratory and field testing of temporary flood, barriers has been conducted by the USACE Engineer Research and Development Center (ERDC). In July 2007, the ERDC issued a Final Report titled "Flood Fighting Structures Demonstration and Evaluation Program: Laboratory and Field Testing in Vicksburg, Mississippi." The purpose of this program was to develop real-world testing procedures for Rapid Deployment Flood Wall (RDFW) and other promising alternative flood-fighting technologies. To fulfill that purpose, ERDC developed a comprehensive laboratory and field testing program for the scientific evaluation of various products. Included in this program was the HESCO Bastion Concertainer, the same system used for the HESCO modular flood barriers by TVA.

2. Comprehensive laboratory testing of the HESCO Bastion Cohcertainer was conducted in a wave research basin at ERDC. The product was tested in a controlled laboratory setting, but under conditions that emulate real-world flood fighting. Stringent construction, testing, and removal protocols were developed for the laboratory. The protocol for the laboratory testing included both performance parameters (hydrostatic testing, hydrodynamic testing Page 24 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS with waves and overtopping, and structural debris impact testing with a floating log) and laboratory setting operational parameters (time, manpower, and equipment to construct and disassemble, suitability for construction and disassembly by unskilled labor, fill requirements, ability to construct around corners, disposal of fill.material, damage, repair, and reusability).

The laboratory testing included the construction of a skewed u-shaped structure. The length of the structure varied from approximately 69 ft to about 81 ft. Due to the restrictive height of the research basin walls,1the height of each-structure was limited to approximately 3 ft.

Laboratory testing of the structure wasinitiated in March 2004 and completed during August 2004.

The impact testing consisted of an apparatus designed, constructed, and installed to provide a log impact speed of 5 mph at an approximate angle of 70 degrees with the horizontal.

Two impact tests were performed. A nominal 12-inch diameter was used for one test and a nominal 16-inch diameter log was used for the other test. Both logs were 12 ft in length.

The logs were cut perpendicularly to their length with a chain saw and left rough with sharp edges. After testing, the levee was inspected (where possible) for weakness and/or failure before the second impact test was performed.

.3. Although impact by waterborne debris is not likely based on direction and velocity of flows past the areas where the HESCO modular flood barriers are located, results of the laboratory testing of the HESCO Bastion Con'certainer by ERDC demonstrates that the flood barriers would not be damaged by the limiting waterborne debris-expected at the peak PMF elevation.

The log impact tests were conducted at a water level of 24 inches. The 12-inch log impacted the structure and bounced back without causing noticeable damage. The structure displaced slightly and recovered to its original position. The 16-inch log impacted the structure and bounced back also without causing any noticeable damage. The structure displaced slightly and recovered to its original position, but vertical deformations of the sand fill ranging-from 4.02 to 0.72 inches were noted. Figure 20 shows the impact zone of the larger diameter log against the barrier wall.

Page 25 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS As shown in Figure 20, no significant damage to the barrier wall by the 16-inch log traveling at 5 mph (7.333 ft/sec) occurred. Results of computer modeling of the river system using HEC-RAS gives a velocity of 3.61 ft/sec at Mile 602.70 (Fort Loudoun Dam is at Mile 602.30) and a velocity of 7.36 ft/sec at Mile 606.43. This provides a velocity profile that indicates the potential log impact velocities that could impact the barrier wall are represented by the impact testing done by the USACE. Therefore, the barrier wall would survive potential impacts by waterborne debris.

4. During the above testing after a series of wave tests were performed on the structure, the reservoir level was raised from a height of 37.6 inches to a height of 38.8 inches. At this height, overtopping occurred. The structure withstood overtopping without failure.

Figure 21 shows the overtopped levee structure. Therefore, the barrier wall would survive any potential overtopping.

Page 26 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS Ability to Close Public Access (PA) Gaps in HESCO Modular Flood Barriers Reasonable simulations were performed to validate representative portions of the TVA River Operations Cherokee Dam, Fort Loudoun/Tellico Dams, and Watts Bar Dam Emergency Action Plans (EAPs) on December 11, 2012. During these simulations, required personnel were mobilized, reported, and closed selected representative PA gaps in the HESCO modular flood barriers prior to the calculated time that headwater elevations would reach the base of the HESCO modular flood barriers at each site during an actual PMF event. Additionally, EAP specified equipment was mobilized, delivered, and utilized in the required time. The details of these reasonable simulations are described below.

Following the guidance provided in the EAPs, a timed reasonable simulation was performed in order to verify that time dependent activities were achievable and required resources were available. This reasonable simulation included prediction of critical headwater elevations by TVA River Forecast Center, notification of required personnel, mobilization of required personnel and equipment at the affected dams, and construction of HESCO modular flood barriers for selected PA gap closures (Watts Bar Dam PA-3 and portion of Fort Loudoun Dam PA-2).

The following steps of the respective EAPs were simulated, including initial and followup notifications and communications between TVA and contractor organizations and actual construction of the HESCO modular flood barriers by the contractor:

1. During flooding events, the River Forecast Center monitors the headwater elevation behind the dams, and upon predicting a critical headwater elevation for each dam based on observed rainfall the River Forecast Center makes notification based on the River Operations Emergency Response Plan to the Asset Owner for River Operations non-power assets (the River Scheduling General Manager). This notification includes the time at which the critical headwater elevation is forecasted to occur. The Asset Owner directs the River Page 27 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS Emergency Operations Center (REOC) to activate in 'advisory' status and activates the on-duty non-power asset Incident Management Team (IMT). The IMT becomes the site Incident Command and contacts the contractor to mobilize and stage equipment and personnel at the location of the HESCO Concertainer units fill material on the dam site. The Incident Command contacts the REOC and requests that the REOC be placed on 'alert' status.

2. If the River Forecast Center observes a specified critical headwater elevation behind the dams, the Asset Owner is notified of the flooding condition. The Asset Owner declares a dam safety emergency - Condition YELLOW (based on procedure RO-SPP-35.1) and notifies the REOC and the Incident Command. The REOC goes to 'activation' status. The Incident Command implements the applicable PMF Barrier Closure Plan for the affected dam(s).

3. The REOC Incident Support Staff is activated to support site activities throughout activation of the Incident Command.

4. Incident Command verifies that equipment and operators are checked in and on site, notifies the REOC of the precise start time of HESCO modular flood barrier installation operations, and uses the respective dam PMF Barrier Closure Plan attached to the applicable EAP to complete HESCO modular flood barrier installation.

5. The contractor arrives on site and installs the HESCO modular flood barriers in accordance with the applicable PMF Barrier Closure Plan. These plans include the following:

" Listing of the equipment needed that is mobilized to the site by the contractor, such as heavy equipment like front loaders and dump trucks as well as hand tools,

" Names and phone numbers of the contractor primary and secondary contacts,

• Listing of the stockpiled material needs including locations, such as empty HESCO Concertainer units, miscellaneous parts needed to construct and connect the HESCO Concertainer units to each other, and fill material which meets Tennessee Department of Transportation Specifications 903.01 (e), 903.01 (f) or 903.22 Gradation #10,

• Manufacturer's instructions for installation of HESCO Concertainer units,

" Detailed instructions specific to the PA gap to be closed, including detailed location plans, and

• Expected installation time of the HESCO modular flood barriers for each PA gap.

The first step of the simulations resulted in notification of required personnel for Cherokee, Fort Loudoun/Tellico, and Watts Bar Dams within 10 minutes of simulating initial prediction of reaching the specified critical headwater elevation behind the dams, and declaration of a dam safety emergency - Condition YELLOW.

Required equipment and personnel, including Incident Command and contractor personnel, arrived at each site as shown in Table 2 below.

Page 28 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS Table 2 - Equipment and Personnel Arrival Times Site Equipment Arrival Time Personnel Arrival Time Cherokee Dam 4 hrs 19 min 2 hrs 1 min Fort Loudoun/Tellico Dams 6 hrs 32 min 5 hrs 7 min Watts Bar Dam 3 hrs 20 min 1 hr 20 min Estimates were determined for the total PA gap closure times for each site based upon actual measured times for selected PA gap closure simulations and extrapolations of these times to the other PA gaps. These representative times for the overall PA gap closure time estimates were determined by simulations of the following portions of the EAPs:

1. Travel times between Fort Loudoun PA-1 to the material staging area and return to Fort Loudoun PA-1 using a dump truck and simulating material loading and unloading were measured as representative of times required for material movement using dump trucks during construction.

2. Travel times between each of the three PA gaps at Watts Bar Dam to the material staging area and return to the three PA gaps using a skid steer and simulating material loading and unloading were measured as representative of times required for material movement using skid steers during construction.

3. Construction times for assembly of a 30 foot section at Fort Loudoun Dam PA-2, using the specified stacked HESCO modular flood barrier configuration, were measured as representative of times required for this type of HESCO modular flood barrier installation configuration.

4. Construction times for assembly of the complete section at Watts Bar Dam PA-3, using the specified single-layer HESCO modular flood barrier configuration, were measured as representative of times required for this type of HESCO -modular flood barrier installation configuration.

Table 3 shows the described PA gap closure times for each site as estimated above, the expected installation times provided in the applicable PMF Barrier"Closure Plans, the available time between reaching the EAP critical elevation headwater behind the dams and the headwaters reaching the base of the HESCO modular flood barriers as determined from the respective hydrographs, and the resulting available time margin between estimated PA gap closure times and available times from hydrographs.

Page 29 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS Table 3 - Estimated PA Gap Closure Times from Simulations Estimated Expected Available Available PA Gap PA Gap Time from Time Site Closure Time Closure Time Hydroqraphs Margin Cherokee Dam 2 hrs 5 min 6 hrs 33 hrs 30 hrs 55 min Fort Loudoun Dam 14 hrs 48 min 12 hrs 19 hrs 4 hrs 12 min Tellico Dam 2 hrs 13 min 6 hrs 27 hrs 24 hrs 47 min Watts Bar Dam 8 hrs 43 min 10 hrs 44 hrs 35 hrs 17 min Observations and conclusions from these reasonable simulations include the following:

" Required resources (personnel, equipment, and material) as described in each site's EAP are available, and were determined to be sufficient to close the PA gaps. These resources were verified through the reasonable simulations and previous inventory of materials.

Additionally, the ability of the EAPs to be effectively implemented by WVA and contractors was validated for the specific simulations performed.

• The PA gaps are accessible to varying degrees to perform the required actions, and accessibility was deemed acceptable. It should be noted that some gap locations provide the opportunity to use multiple pieces of equipment to close a single gap.

" High winds, heavy rains, and lightening could delay construction of HESCO modular flood barriers. However, as indicated in Table 3 above, sufficient margin is available to account for these delays.

• The TVA River Forecasting Center has the ability to predict weather for the Tennessee Valley 10 days in advance. There are various postulated 9-day rain events which result in the need for the HESCO modular flood barriers to prevent dam embankment overtopping.

The trigger point for notification of HESCO modular flood barrier construction crews and equipment is based on predicted headwater elevations at each dam. Therefore, it would be possible for the TVA River Forecasting Center to forecast the rain events and predict the trigger headwater elevations hours if not days in advance. Thus, prior to routes becoming impassible, notifications could be made to mobilize required equipment and personnel.

Additionally, there are multiple roadways between major metropolitan areas and respective sites to transport personnel and equipment. This provides redundant roadways to access each site if one roadway were to become impassible. Therefore, it is concluded that mobilization of personnel and equipment would not be adversely restricted.

Hydrologic Analysis Without HESCO Modular Flood Barriers A study to fulfill TVA's commitment to perform an analysis of the flooding levels for SQN Units 1 and 2 and WBN Unit 1 that assumes the HESCO modular flood barriers are not installed and includes failure of the earthen embankments if overtopped in the analysis at Fort Loudoun, Cherokee, Tellico, and Watts Bar Dams was completed. The results of this study are not included in the design basis at SQN Units 1 and 2 and WBN Unit 1.

Page 30 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND' EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS The simulation performed for this study used the USACE HEC-RAS software code. Previous hydrologic analyses were performed using the TVA Simulated Open Channel Hydraulics (SOCH) software code. However, TVA is evaluating replacing theSOCH model for flood routing calculations for the Tennessee River and selected tributaries to the use of USACE HEC-RAS for future hydrologic analyses. Comparisons between similar cases using the SOCH model and the currently developed USACE HEC-RAS model demonstrate that the resulting PMF elevations are similar, with results slightly lower using USACE HEC-RAS.

The configuration of the postulated breaches of the unprotected earthen embankments when overtopped is based upon a number of factors including the following:

a. The breach configurations are based on review of different methods discussed in the Predictionof Embankment Dam Breach Parameterspaper prepared by Tony Wahl with

-the U.S. Department of the Interior.

b. The breach configuration proposed for each dam is based on an approach used by the Federal Energy Regulatory Commission (FERC).

c. For each dam a single embankment was postulated to fail. The embankment selected to fail provides the greatest breach size allowing the largest volume of water downstream.

d. The size of the breach was a function of the breach depth which assumed failure down to bedrock.-

e. The location of the postulated breach is the main earthen embankment next to the known.

spillways-where bedrock elevations are

f. In addition to the failure of the main embankment at Watts Bar Dam, the West Saddle Dam and East Wall were postulated to fail consistent with the' current analysis approach.

g. For each dam downstream of SQN Units 1 and 2 and WBN Unit 1, embankments were assumed to not fail even if overtopped:.

Two basic storm situations have the potential to produce a maximum flood at SQN Units 1 and 2 and WBN Unit 1. These are (1) a sequence of March storms producing maximum rainfall on the 21,400-square-mile watershed above Chattanooga, hereafter called the 21,400-square-mile storm, and (2) a sequence of March storms centered and producing maximum rains in the basin to the west of the Appalachian Divide and above Chattanooga, hereafter called the 7,980-square-mile storm. This analysis used the 7,980 square-mile event, since this analysis is'expected to yield the highest flood levels at SQN Units 1 and 2 and WBN Unit 1. Additional details of the simulation are as follows:

a. Cherokee Dam South Embankment is postulated to partially fail and breach progressively over one hour at maximum headwater elevation.

b. Fort Loudoun Dam Section 1 of South Embankment is postulated to partially fail and breach progressively over one hour when overtopping reaches approximately two feet at elevation 835.3 ft.

Page 31 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS

c. Tellico Dam Main Dam Works Embankment is postulated to partially fail and breach progressively over one hour. The HESCO modular flood barriers installed on the Right Bank Saddle Dam, Main Dam Works Embankment, Saddle Dam 2, and Saddle Dam 3, were assumed to not be in place during the event setting the crest elevation of the embankments at elevation 830 ft. Tellico Dam was assumed to fail at the same time as Fort Loudoun Dam with 1.32 feet of overtopping at the Main Dam Works Embankment.

d. Watts Bar Dam East Embankment is breached progressively over one hour at elevation 768.0 ft.

e. Chickamauga Dam, Nickajack Dam, and Guntersville Dam are not assumed to fail.

The elevations are shown below in Table 4.

Table 4 - Results of Simulation Current Study Increase Site Calculated PMF (ft) Result (ft) In.PMF (ft)

SQN 722.0 726.8 4.8 WBN 739.2 746.3 7.1 As anticipated, the elevation produced by these breaches exceeds the current PMF elevation described in the License Amendment Request related to the updated hydrologic analysis submitted for SQN Units 1 and 2 and WBN Unit 1 (References 31 and 32). These results reinforce the importance of ensuring that the earthen embankments at the affected dams are precluded from overtopping and reinforce the basis for installing the HESCO modular flood barriers until such time as permanent modifications to the dams can be implemented consistent with the commitments made in the Reference 9 letter.

Conclusion TVA has completed a study analysis to determine the increase in PMF at SQN Units 1 and 2 and WBN Unit 1 ifthe HESCO modular flood barriers were not installed. The study demonstrates that PMF would increase above the current licensing basis and design basis requirements. Therefore, permanent modifications to replace the temporary HESCO modular flood barriers will be implemented by October, 2015 (Reference 9), eliminating the need for the temporary flood barriers and eliminating the possibility of breaches in the areas currently protected by the flood barriers.

In addition to the breach analysis, the likelihood of impacts and failure to the HESCO modular flood barriers from commercial river traffic and other waterborne debris during a PMF event has been evaluated. This evaluation included the following areas and conclusions:

1. The locations of the HESCO modular flood barriers in relation to the edges of the reservoirs, and the topography of the reservoirs and earthen embankments underlying the modular flood barriers, reduces the potential for impact by commercial barges or other waterborne objects during a PMF event. An evaluation of this area concludes that the locations and Page 32 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS configurations of the modular flood barriers minimize the possibility of impacts from commercial barges and waterborne debris.

2. The small amount of commercial river traffic in each reservoir reduces the likelihood that commercial barges would be located in each reservoir during a PMF event, and then available to impact the modular flood barriers if they become uncontrolled. An evaluation of this area concludes that there is no anticipated impact to the modular flood barriers at Cherokee and Tellico Dams because there is no commercial river.traffic on the respective reservoirs. In addition, the commercial river traffic on Fort Loudoun and Watts Bar Reservoirs is manageable minimizing the possibility of impacts from commercial barges.

3. The control of commercial transportation during high river levels and flows reduces the likelihood that commercial barges would become uncontrolled during a PMF event and then possibly impact the modular flood barriers. A review of the controls provided by the Tennessee River Waterway Management Plan and the regulations implemented by the USCG concludes that the possibility of commercial barges becoming uncontrolled during a PMF event is minimal.

4. The ability of the HESCO modular flood barriers to withstand impacts from waterborne debris and overtopping reduces the likelihood of a failure of the modular flood barriers during a PMF event. An evaluation of this area concludes that the modular flood barriers are robust and are not likely to be damaged by waterborne debris during a PMF event.

5. The actual flow profile through the reservoirs during high river levels and flows reduces the likelihood that commercial barges would be directed towards the modular flood barriers during a PMF event. An evaluation of the CFD model for Fort Loudoun Reservoir demonstrates that it is not likely that uncontrolled barges and other large waterborne objects would approach the HESCO modular flood barriers. A similar analysis for Tellico Reservoir, which does not currently have any commercial river traffic, demonstrates a small likelihood that if commercial river traffic were located on the reservoir during a PMF event there is some chance that an uncontrolled barge could approach the HESCO modular flood barriers.

6. The ability to implement procedures to close the gaps left for public access through the HESCO modular flood barriers (PA gaps) ensures the effectiveness of the modular flood barriers during a PMF event. A reasonable simulation verifies that the required resources (personnel, equipment, and material) as described in each site's EAP are available, and were determined to be sufficient to close the PA gaps. These resources were verified through the reasonable simulations and previous inventory of materials. Additionally, the ability of the EAPs to be effectively implemented by TVA and contractors was validated for the specific simulations performed.

References

1. TVA Submittal to NRC Document Control Desk, "Impact of Potential Breaches of HESCO Modular Flood Barriers and Earthen Embankments on the Updated Hydrologic Analysis Results for Sequoyah Nuclear Plant, Units 1 and 2, and Watts Bar Nuclear Plant, Unit 1,"

dated October 30, 2012 (Accession No. ML12307A227).

Page 33 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS

2. Carl F. Lyon, NRC, to NRC Document Control Desk, "Summary of May 31,2012, Senior Management Meeting With Tennessee Valley Authority on the Licensing Basis for Flooding/Hydrology," dated June 6, 2012 (ADAMS Accession No. ML12157A457).

3. TVA Submittal to NRC Document Control Desk, "Responses to Hydrology Action Items,"

dated January 14, 2011 (ADAMS Accession No. ML12136A439).

4. TVA Submittal to NRC Document Control Desk, "Re-submittal of Attachments for Responses to Hydrology Action Items," dated March 21, 2011 (ADAMS Accession No. ML110831044).

5. TVA Submittal to NRC Document Control Desk, "Response to Request for Additional Information Regarding Final Safety Analysis Report Section 2.4 (TAC No. ME3945)," dated April 20, 2011 (ADAMS Accession No. ML11112A137).

6. TVA Submittal to NRC Document Control Desk, "Revised Response to Request for Additional Information Question 3 Regarding Final Safety Analysis Report Section 2.4 (TAC No. ME3945)," dated May 20, 2011 (ADAMS Accession No. ML11145A163).

7. TVA Submittal to NRC Document Control Desk, "Information Presented in the May 11-12, 2011 TVA/NRC Meeting Regarding Final Safety Analysis Report Section 2.4,"

dated June 1, 2011 (ADAMS Accession No. ML11154A139).

8: Siva P. Lingam, NRC, to Joseph W. Shea, TVA, "Tennessee Valley Authority (TVA)

Long-Term Hydrology Issues for Operating Nuclear Plants -Browns Ferry Nuclear Plant, Units 1,2, and 3 (TAC Nos. ME5026, ME5027, and ME5028); Sequoyah Nuclear'Plant, Units I and 2 (TAC Nos. ME5029 and ME5030); and Watts Bar Nuclear Plant, Unit 1 (TAC No. ME5031)," dated January 25, 2012 (ADAMS Accession No. ML11241A166).

9. TVA Submittal to NRC Document Control Desk, "Commitments Related to Updated Hydrologic Analysis Results for Sequoyah Nuclear Plant, Units'l and 2, and Watts Bar Nuclear Plant, Unit 1,"dated June 13, 2012 (ADAMS Accession No. ML12171A053).

10. Eric J. Leeds, NRC, to Joseph W. Shea, TVA, "Confirmatory Action Letter - Watts Bar Nuclear Plant, Unit 1, and Sequoyah Nuclear Plant, Units 1 and 2, Commitments to Address External Flooding Concerns (TAC Nos. ME8805, ME8806, and ME8807)," dated June 25, 2012 (ADAMS Accession No. ML12165A527).

11. NRC Meeting Summary, "Public Meeting Summary - Flood Mode Operation Improvement Strategy for Watts Bar/Sequoyah - Docket Nos..50-390, 50-327, 50-328," dated December 17, 2012 (Accession No. ML12352A208).

12. NRC Meeting Summary, "Summary of December 13, 2012, Management Meeting with Tennessee Valley Authority on the Status of Hydrology," dated January 11, 2013 (Accession No. ML13010A137).

13. Sivakumar, P., Hyams, D.C., Taylor, L.K., and Briley, W.R., "A Primitive-Variable Riemann Method for Solution of the Shallow-Water Equations with Wetting and Drying," Journal of Computational Physics, Vol. 228; No.19, pp. 7452-7472, 2009.

Page 34 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS

14. Wilson, R., Nichols, III, S., Mitchell, B., Karman, S., Betro, V., Hyams, D., Sreenivas, K.,

Taylor, L., Briley, R., and Whitfield D., "Simulation of a Surface Combatant with Dynamic Ship Maneuvers," 9th Int. Conf. in Num. Ship Hydro., University of Michigan, 5-8 Aug.

2007.

15. Sreenivas, K., Hyams, D.G., Nichols, III, D. S., Mitchell, B., Taylor, L.K., Briley, W.R., and Whitfield, D.L., "Development of an Unstructured Parallel Flow Solver for Arbitrary Mach Numbers," AIAA Paper No. 2005-0325, 43rd Aerospace Sciences Meeting and Exhibit, Reno, NV, January 2005.

16. The OpenFOAM Foundation, http://www.openfoam.org.

17. Hyams, D. G., An Investigation of Parallel Implicit Solution Algorithms for Incompressible Flows on Unstructured Meshes, Ph.D. Dissertation, Mississippi State University, May 2000.

18. Sreenivas, K., Hyams, D.G., Nichols, Ill, D. S., Mitchell, B., Taylor, L.K., Briley, W.R., and Whitfield, D.L., "Development of an Unstructured Parallel Flow Solver for Arbitrary Mach Numbers," AIAA Paper No. 2005-0325, 43rd Aerospace Sciences Meeting and Exhibit, Reno, NV, January 2005.

19. Sreenivas, K., Pankajakshan, R., Nichols, III, D.S., Mitchell, B., Taylor, L., and Whitfield, D.L., "Aerodynamic Simulation of Heavy Trucks with Rotating Wheels," AIAA-2006-1394, 44th AIAA Aerospace Sciences Meeting and Exhibit, January 2006.

20. Nichols, Ill, D.S., Hyams, D., Sreenivas, K., Mitchell, B., Taylor, L., and Whitfield, D., "An Unstructured Incompressible Multi-Phase Solution Algorithm," AIAA-2006-1290, 44th AIAA Aerospace Sciences Meeting and Exhibit, January 2006.

21. Hyams, D., Sreenivas, K., and Whitfield, D.L., "Parallel FAS Multigrid for Arbitrary Mach Number, High Reynolds Number Unstructured Flow Solvers," AIAA-2006-2821, June 2006.

22. Sreenivas, K., Mitchell, B., Sawyer, S., Karman, Jr., S.L., Nichols, III, D.S., and Hyams, D.,

"Computational Prediction of Forces and Moments for Transport Aircraft," AIAA-2007-1088, 45th AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada, Jan. 8-11, 2007.

23. Karman, Jr., S.L., Mitchell, B., Sawyer, S., and Whitt, J., "Unstructured Grid Solutions of CAWAPI F-1 6XL by UT SimCenter," AIAA-2007-0681, 45th AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada, Jan. 8-11, 2007.

24. Hyams, D. G., Webster, R. W., and Sreenivas, K., "A Generalized Interpolative Interface for Parallel Unstructured Flow Solvers," 40th Fluid Dynamics Conference and Exhibit, Chicago, Illinois, June 28-1, 2010. Paper No. 2010-5097.

25. Wilson, R.V., Nichols, III, D.S., Mitchell, B., Karman, S.L., Hyams, D.G., Sreenivas, K.,

Taylor, L.K., Briley, W.R., and Whitfield, D.L., "Application of an Unstructured Free Surface Flow Solver for High Speed Transom Stern Ships," 26th Symposium on Naval Hydrodynamics, Rome Italy, September. 17-22, 2006.

Page 35 of 36

ENCLOSURE POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS

26. Almeida, T.G., Walker, D.T., Leighton, R.I., Alajbegovic, A., Pankajakshan, R., Taylor, L.K.,

Whiffield, D.L., and Ceccio, S.L., "A Reynolds-Averaged Model for the Prediction of Friction Drag Reduction by Polymer Additives," 26th Symposium on Naval Hydrodynamics, Rome Italy, September. 17-22, 2006.

27. Wilson, R., Nichols, III, S., Mitchell, B., Karman, S., Betro, V., Hyams, D;, Sreenivas, K.,

Taylor, L., Briley, R., and Whitfield D., "Simulation of a-Surface Combatant with Dynamic Ship Maneuvers," 9th Int. Conf. in Num. Ship Hydro., University of Michigan, 5-8 Aug.

2007.

28. Wilson, R., Lei, J., Karman, Jr., S.L., Hyams, D., Sreenivas, K., Taylor, L., and Whitfield D.,

.2008, "Simulation of Large Amplitude Ship Motions for Prediction of Fluid-Structure Interaction," Proceedings of the 27th ONR Symposium on Naval Hydrodynamics, Seoul, Korea, 5-10 Oct. 2008.

29. Pankajakshan, R., Mitchell, B.J., and Taylor, L.K., "Simulation of unsteady two-phase flows using a parallel Eulerian-Lagrangian approach," Computers & Fluids, Volume 41, Issue 1, February 2011, Pages 20-26.

30. Sabnis JS, Choi S-K, Buggeln RC, Gibeling HJ. Computation of two-phase shearlayer flow using an Eulerian-Lagrangiananalysis.AIAA Paper 88 3202; 1998.

31. TVA Submittal to NRC Document Control Desk, "Application to Revise Sequoyah Nuclear Plant Units 1 and 2 Updated Final Safety Analysis Report Regarding Changes to Hydrologic Analysis, (SQN-TS-12-02)," dated August 10, 2012 (ADAMS Accession No. ML12226A561).

32. TVA Submittal to NRC Document Control Desk, "Application to Revise Watts Bar Nuclear Plant Unit 1 Updated Final Safety Analysis Report Regarding Changes to Hydrologic Analysis, TAC No. ME8200 (WBN-UFSAR-12-01)," dated July 19, 2012 (ADAMS Accession No. ML12236A167).

Page 36 of 36

ENCLOSURE - ATTACHMENT I POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS L-nMolIo~LnI I . 3 1 4 1 5 1: a I 7 .I . 9 I 10 I 11 I 12-

-~ - - - .9. I*~

A A mmiv6-B B:

I C C D D GTAD.A WI GNDWf WiTM iTIWI E

F F tZA

.C DM MtIn"TWO P'IF MPORARY; BARRIER!

PLANANDSECTIONS HH.

WATlS8M PRO=C "U - w au Bumt ivammS.TS

-1. .1.

I I 2 ' .1 1, 6 1r 7 IL 4 .1 ý -

v-'age 1 of 5

ENCLOSURE -ATTACHMENT I POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS

~L~I]ZZLZLi Al-Al IMRl~ AWS.*n m-S ~uS5 At

!fTAD.. B

=7.SIrz cimal m d.cma tSI0 DRAINAGED DNETOTAZIL DRAINAGE TRAFFMCONTOLBARREER INLET DETAIL.

vage 2 of 5

ENCLOSURE - ATTACHMENT I POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS I ~-~M0LI~IOLI 2 3 I 4~ I .5 I 6 -,I 7 .1

,. 8 -,I 9 10

-- I 11 12

-- I-Zýrm ot[3100 9- 1 -3, 1

• 1 5 1 a 1 7, '1 a 9 1 10 1 11 1 12-rE - I A. A B

UTý

~c. 0.

D. D E' E F*

.G PMF TEMPORAF!Y BARRIER PLANAND-SECTION.S H*

.1 1 2 1 1 1 - C V 5, . 1; ýqý 1 7 , 1 lk 1:

vage 3 of 5

ENCLOSURE - ATTACHMENT I POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS I =Mali im"-, 1 V  %,-

i

// t d~L~~iiFiiio A 1*11Th baSifl IF

• m amiWm1*0. N lbbblUbMmIt fllt& ~

tbaatATla .lbflbAlbm amA) - Zfl I. NWIS X ASS an ) mIbi.a, a If mslWnaunVbmminfllflbMSAlmP CAMAI SAMOt DAWBARRIO IWSALLAT1I" 4 ~.m DblbAAb*1. SA ISA/LIw SI Atlfl

• smm ~lt - 055Mb am naMmlb am~l. moms

-v mba a ma mm B-B C-C v'age 4 of 5

ENCLOSURE - ATTACHMENT I POTENTIAL FOR BREACHES OF THE HESCO MODULAR FLOOD BARRIERS AND EARTHEN EMBANKMENTS AT DESIGN BASIS FLOOD LEVELS I-raolMlo I LI 2 I 31 I 4 I I B I ,7 I 8a I 9 1 10 I 11 I 12 .

-E A A

-. a~,A11I1I1111II1IIIItIlIfrrrn1TITyl~,,l.rt,1T¶1T1IIIIIIIIIIIIIIIILiJt~T]

'-1~ ATM?

MC OA SCWUfJM

,.r~4JJIUI~WIIIIIiJJJJJiIIThII~lIIIIIIiJJJUiIIIIJJJ1 SWAT- ATlAS Bt RAnTAMST B

.- C D D E W S "M3T CSM SaWS" E F F hi~-I--~iT izz~ * ~Aa a~~~~~~~. A*Af*

l ' t¢ na n an. A A a,,* I*lleu

,T., f

~

I~~1 II G-RAIICAICATI@T PMF TEMPORARY, BARRIER,

/ ~ 'PLAN

-i-i-i -i -- ) - --

'H.

eM92WTWWA WIý ARCIMAWAI DA TIWTW mAuman mcarrm WW.S~nm

-4 -

1 I 2 I 31 I 4 1 5 I 6 .I I I -

Iage 5 of 5

"D 0

z 214W0Sq. MI Oowfwtr..mC.nt~rod Marh Rlood Event b%4A140M M-n for WBd and 1Q1 PeekDaOiWE 641,318cefs INo tod~auqs owm or C1*kamnap OwnmIdurvap 64(1%), -0 Cutr~mt ock Configuration at Chckamasup Dame feon Loiadoun Dam MX R4()

I RM b02.3 lOCH -lawation a&WOkdirge Peic NW f1v 133 ilR

ýW, g1, 1 1 W 4

A I -A,a t S_10X1 AW 1-ZM

ýqý14W VwSIX) t~84~

I (S M"JIOW "

StS IIW0 Ve~y I I (.1 1120401 ý, Mi z

Wel 21400 Ný. 4 1 CS 201204M2144

\

0, PIak1Wp2O.5~ft I-810 -el N

NN 4320000 J >rn 4ZM 0

(A C

XCO)

M0

.19 f.-1~ -dou,r W-n lidw~teI I

- -4 touo,. Dam1 I. turep~

I ~ aC CA 0

?S0~J0 >--T x 0

1AP) (IX) 3200 M00 z 00 I50.000 140 rnx r"n 00 3/14 1/14 3/15 3/16 3/F 1/18 1/19 3/M0 )1 .121) 3/13 1/14 i//S 31)1. 1/13 3//8 3/29 I/ID 1/SI 4/1 4/) 413 cc w IJ -;F I' CD

-L 0

-.0

'a 0

-4 m

z

~M6

-n

?,NO0Sa.OM.gbid Wa Cenered March DeisW Stwtn for $a" en WON I

-0 Peek ONsd*W: 658W417 cfd (No Pikuajack Danom cwChkimcswam On Folume)

CurrantLock CoMfkmratkmoet Chkkarneug Daom Fort Loudown Darn mm T8PA602J SOCH-Ekvatlon a&WDcharge Pek MW pJev. "15t5t Zm

~.jI

~ A~3r*4.C1 M20403 o- mm m S-6, tto~ui 0p?..*- j ~IX$

(S33 ., FM> z alu

=.i Wi5037~0~r333337ko sot7S5~~W(n.5IC)030 0

P~kTW I1e~813I~ft F-0 S SOD

- fw odo Heawtlr, W"344.3070

\4

\

N N

\

N I too XG() m I jwrd3 Wm004.044 IAamte7 I 300,000S ic0 Co MO m

mw z CD

- ~

M f 44 CA

(

740 1Im000 z

0 3/14 3/33 3/1 3V16 /If / I 3/13 Il /211X3/21 3/27 1/ 3 /14 3/% A/26 I/" 3/28 )/)g V30~ 3/31 4/1 412 4/4 CD 0 76dIZh)

TU 0

z

~3j~ciii 2 fluO I WO1XK1 Z140D Sq. Ml Downwitotaim Centowed Manch flood Event -0 for WONend50M Peak isW ESw: 76;.48ft (No Nkkajak Dom or Clikkiseunp Don Faftrem)

Current Lock Configuration at Chickamneuge Denm It warn eec ZM TRM 529 SOCH Elevation ee

\ ~.ak D~dur.e I.248,S3cfx M 1W z (01,20401 it cs w~

5.61 14UO MI I. . 4 (A 01204M) n C8

-cs-xI2022m o 0 0

-- Wistt, Ows tIlisnslfrdwate

-wafts 8SwLa tinn 6c Ma

-a wBtai Weir 01 Rim Leak Iso 00 0 00,M 5.

M z

/ -1

1. 400,WOK z 0-

/

/

MOM.00 71.r 04 + .-

~.9~5~5d

+-.---~--~-~

~ I I -

3/13 3/14 )/IS 3/16 3117 Sil 3/19 3/20 3/21 1/22 3/13 3/24 3/12 312 31/27 3/,28 3/29 3/30 3/31 4/1 4/2 4/3

'Weir P7 was modIed in SegmentI asa sglve focal, bssliss Ownsaa possitivdisciztle 0fri tSfpi g~purjpwOts 1 CD wS 0

"D 0

z

~MIM 7.960 Sq. W Bulk Gap Cantered March Design Storm 0 for SQUandWBd Pok HW 8#ev 767.431.

(No NkMckaa Own or ChickamW arn Folem asrai r Watts Mckruapwe mM~~

t ~1 SZ/Mm X

SOCH -v thmon anmdOcharge

>z m S. Dam Z6W a." We a 7'Rin ktjrW00t 0v sql 5W4 I S 201Q~1.~>

sqi~~~~~~~ ~~ Z. ~e.i~~am C

-- Will L Rim gic 0

-Wit~ BiiRim D~l~i~r (z Mo z

/ ~0' ijK 00 1113 3/14 3/15 3/16 3/1 ;119 3/19 3120 3/21 12 3/23 1124 31 t) t IA 419 43D 11 4/1 4f 4/1C

,wt~f NI wasmndetecim segttsct I a, Ji NA2S DIU1 3(2 (2M

-u svwflis psltw ICO)~rhji~~

0