Text

NRC-2017-000688 (Formerly FOIA/PA-2017-0690) - Resp 5 - Final, Agency Records Subject to the Request Are Enclosed, Part 12 of 15
	ML20366A020
Person / Time
Issue date:	12/29/2020
From:	; NRC/OCIO
To:	;
Shared Package
ML20366A007	List: ML20366A008; ML20366A009; ML20366A010; ML20366A011; ML20366A012; ML20366A013; ML20366A014; ML20366A015; ML20366A016; ML20366A017; ML20366A018; ML20366A019; ML20366A020; ML20366A021; ML20366A022; ML20366A023;
References
	FOIA, FOIA/PA-2017-0690, NRC-2017-000688
	Download: ML20366A020 (66)
	v • d • e

FLOOD HAZARD REEVALUATION REPORT, REV.1 IN RESPONSE TO THE 50.54(f) INFORMATION REQUEST REGARDING NEAR-TERM TASK FORCE RECOMMENDATION 2.1: FLOODING for the THREE MILE ISLAND NUCLEAR POWER STATION ROUTE 441 SOUTH, MIDDLETOWN, PENSYLVANIA 17057 Renewed Facility Operating License No. DPR-50 NRC Docket No. 50-289 Exelon.

Exelon Generation Company, LLC P.O Box 80S387 Chicago, llllnols 60680-5387 Prepared by:

AMEC Environment & Infrastructure, Inc.

502 W. Germantown Pike, Suite 850, Plymouth Meeting, Pennsylvania 19462 Rev. 1 Submittal Date: March 11, 2013 Printed Name Affiliation Originator:~ l-J/,..1,- AMEC Ve,lffer: ~$' AMEC App,,..r: J."I{.-<¥ A~ . nt1 -

AMEC Lead Responsible Engineer: 'i3 ,' f..C. M<= .!b, /e Branch Manager Senior Manager Design Engineering: ~ - ll- ,,.

Corporate Acceptance: 3/11/13 Page 1of64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl Contents

1. UST OF ACRONYMS .................................................................................................................................... 5

2. PURPOSE .................................................................................................................................................... 6

a. Background ............................................................................................................................................ 6

b. Requested Actions ................................................................................................................................. 7

c. Requested Information .......................................................................................................................... 7

3. SITE INFORMATION .................................................................................................................................... 8

a. Detailed Site Information ....................................................................................................................... 8

b. Current Design Basis Flood Elevations for All Flood Causing Mechanisms............................................ 9

c. Flood Related Changes to the Licensing Basis and Any Flood Protection Changes (including mitigation) Since License lssuance ............................................................................................................... 11

d. Changes to Watershed and Local Area since License Issuance ........................................................... 12

e. Current Licensing Basis Flood Protection and Pertinent Flood M itigation Features........................... 13

4.

SUMMARY

OF FLOOD HAZARD REEVALUATION ...................................................................................... 13

a. Local Intense Precipitation ................................................................................................................... 13

b. Flooding in Streams and Rivers ............................................................................................................ 17 Calibrated Hydrologic Model ................................................................................................................... 19 Scenario 1................................................................................................................................................. 24 Scenario 2 ................................................................................................................................................. 27 Scenario 3 ................................................................................................................................................. 33 Governing PMF Peak Discharge ............................................................................................................... 37 Assumed Dam Failures During Precipitation Event ................................................................................. 37

c. Dam Breaches and Failures...................................................................................................... ............ 39 Combined Storage Volume of Small Dams within the Watershed ...................... .................................... 39 Seismically-Induced Dam Failure ............................................................................................................. 41

d. Storm Surge.......................................................................................................................................... 44

e. Seiche .................................................................................................................................................... 44

f. Tsunami ................................................................................................................................................ 44

g. Ice Induced Flooding ............................................................................................................................ 44

h. Channel Migration or Diversion ........................................................................................................... 47

i. Combined Effect Flood ......................................................................................................................... 50 Governing PMF Hydraulic Analysis .......................................................................................................... SO Wind-Generated Waves ........................................................................................................................... 53 Hydrodynamic Load ................................................................................................................................. 56 Debris Load .............................................................................................................................................. 57 THREE MILE ISLAND NUCLEAR POWER STATION Page 2 of 64 RCN: TMl-140

NTTF Recommendatio n 2.1 (Hazard Reevaluatio ns): Flooding M arch 11, 2013 Exelon Corporation Rev 1

5. COM PARISON WITH CURRENT BASIS FLOOD HAZARD ............................................................................ 59

6. REFERENCES ............................................................................................................................................. 60 List of Tables Table 1: Summary of Land Use Changes in Susquehanna River Wat ershed ................................................... 13 Table 2: l -hr/ 1-sq-mi PMP Distribution .......................................................................................................... 15 Table 3 : LIP Predicted Flooding Results at Pathways to TMI Unit 1 Safety Related Struct ures (Maximum Water Surface Elevation, Flooding Depth, and Velocity).. ............................................................................... 17 Table 4 Observed Rainfall of Tropical St orm Lee and Tropical Storm Agnes ................................................... 20 Table 5

Hydrologic Soils Groups and Loss Rates ............................................................................................ 22 Table 6: Mean Monthly Base flow per Unit Area............................................................................................ 25 Table 7: Rainfall Distribution Sensitivity Results.............................................................................................. 26 Table 8: M ean Monthly Base flow per Unit Area............................................................................................ 28 Table 9: Comparison of Flows from 1996 Rain-on-Snow to those using t he Tropical St orm Lee Loss Rat es .. 29 Table 10: Alternatives for Scenario 2 .............................................................................................................. 32 Table 11: Mean Monthly Base flow per Unit Area .......................................................................................... 33 Table 12: Scenario 3 Results ........................................................................................................................... 36 Table 13: Precipitation Driven PMF Scenario Results ...................................................................................... 37 Table 14: Maximum peak flow from the precipitation scenarios.................................................................... 38 Table 15: Dams Modeled in Calibrated HEC* HMS model ................................................................................ 39 Table 16: Weighted height, length and st orage volumes for the individual and composit e dams added t o estimate t he hydrologic dam break peak fl ow rate......................................................................................... 41 Table 17: Results of the seismically-induced dam failure scenarios ... ............................................................. 42 Table 18: Three most Severe Ice Jam events................................................................................................... 45 Table 19: Peak Flows and gage heights for t he January, 1996 Ice Jam .......................................................... 46 Table 20: Peak Flow Comparison ..................................................................................................................... 47 Table 21 : Results summary from the NU REG/ CR 7046 Still Water Analysis ....................... ............................. 52 Table 22 : Wind Generated Wave Predicted Flooding Elevations and Depth Result s at Pathways t o TMI Unit 1 Safety Related Structures.................................................................................................................... .......... 56 Table 23 : Predicted Hydrodynamic Results at TMI Unit 1 Safety Relat ed Structures..................................... 57 Table 24: Summary Comparison with Current Licensing Basis Flood Hazard .................................................. 59 List of Figures Figure 1: 1-HR PMP Distribution ...................................................................................................................... 16 Figure 2: Watershed and Subbasin Delineated ................................................................................................ 21 Figure 3 : Preliminary Constant Loss Rat e (mm/ hr) .......................................................................................... 22 Figure 4 : Susquehanna River HEC-HMS Model Calibration Results at the Harrisburg USGS Gage ................. 23 Figure 5 : Susquehanna River HEC-HMS Model Calibration Results at the Marietta USGS Gage .................... 24 Figure 6 : Scenario 1.1 Input Volume ................................................................................................................ 26 Figure 7: Scenario 1 Results Hydrographs at Three Mlle Island ..................................................................... 27 Figure 8: Input Volume Scenario 2.2 ............................................................................................................... 31 Figure 9: Scenario 2 Results - Hydrographs at Three M ile Island .................................................................... 32 Figure 10: Scenario 3.4 Input Volume .............. ............................................................................................... 35 THREE MILE ISLAND NUCLEAR POWER STATION Page 3 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Figure 11: Scenario 3 Results - Hydrographs at Three M ile Island .................................................................. 36 Figure 12: Probable Maximum Flood Hydrograph at Three Mile Island ......................................................... 38 Figure 13: Composite and Individual Dam Locations....................................................................................... 40 Figure 14: Peak flow results at Three Mile Island from the dam breach analysis ........................................... 43 Figure 15: 60-Day Discharge, January 1996 Ice Jam ........................................................................................ 46 Figure 16: Comparison of 1908 and 1943 Middletown, PA USGS Quadrangles .............................................. 49 Figure 17: Comparison of 1963 and 1999 Middletown, PA USGS Quadrangles.............................................. 49 Figure 18: Flood Stage versus Discharge "Curve at Intake Screen and Pump House........................................ 52 Figure 19: NUREG/CR 7046 Analysis PMF Water Surface Elevation vs. Time Hydrograph at the Intake Screen and Pump House ............................................................................................................... ............................... 53 THREE MILE ISLAND NUCLEAR POWER STATION Page 4 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1

1. LIST OF ACRONYMS QC degree(s) Celsius (or Centigrade)

!!F degree(s) Fahrenheit ac acre(s)

ANS American Nuclear Society ANSI American National Standards Instit ute CFR Code of Federal Regulations cfs cubic (foot)feet per second D-A-D depth-area-duration (curves)

DEM Digital Elevation Model EM Engineer Manual ESRI Environmental Systems Research Institute FEMA Federal Emergency Management Agency FTP file transfer protocol GIS Geographic Information System GHCN Global Historical Climat ology Network GHCND Global Historical Climatology Network Data HEC-HMS Hydrologic Engineering Center Hydrologic Modeling System HEC-RAS Hydrologic Engineering Center River Analysis System HHA hierarchical hazard assessment HMR Hydrometeorological Report hr hour(s)

HUC Hydrologic Unit Code In Inch km kilometer(s) 2 km square kilometer(s)

Landsat Land-Use Satellite m meter(s) 2 mi square mlle(s) mi mlle(s) min minute(s) mm millimeter(s)

MSL mean sea level NAVD-88 North American Vertical Datum of 1988 NCDC National Climatic Data Center NED National Elevation Dataset NGVD-29 National Geodetic Vertical Datum of 1929 NHD National Hydrography Dataset NLCD National Land Cover Database NOAA National Oceanic and Atmospheric Administration NRC United States Nuclear Regulatory Commission NRCC Northeast Regional Climate Center THREE M ILE ISLAND NUCLEAR POWER STATION Page 5 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 NRCS Natural Resources Conservation Service NWS National Weather Service OBE operating basis earthquake PCSWMM PC Storm Water Management Model PMF probable maximum flood PMP probable maximum precipitation PPT Precipitation depth PMSA Probable Maximum Snowpack Accumulation PSAR Preliminary Safety Analysis Report scs Soil Conservation Service SSCs structures, systems, and components SSE safe shutdown earthquake TMI Three Mile Island Nuclear Power Station UFSAR Updated Final Safety Analysis Report USGS United States Geological Survey WSE Water Surface Elevation

2. PURPOSE

a. Background In response to the nuclear fuel damage at the Fukushima Dai-lchl power plant due to the March 11, 2011 earthquake and subsequent tsunami, the United States Nuclear Regulatory Commission (NRC) established the Near Term Task Force (NTTF) to conduct a systematic review of NRC processes and regulations, and to make recommendations to the Commission for its policy direction. The NTTF reported a set of recommendations that were Intended to clarify and strengthen the regulatory framework for protection against natural phenomena.

On March 12, 2012, the NRC Issued an Information request pursuant to Title 10 of the Code of Federal Regulations, Section 50.54 (f) (10 CFR 50 .54(f) or 50.54(f)) (Reference 4) which included six (6) enclosures:

1. [NTTF) Recommendation 2.1: Seismic

2. [NTTF) Recommendation 2.1: Flooding

3. [NTTF) Recommendation 2.3: Seismic

4. [NTTF] Recommendation 2.3: 'F looding

5. [NTTF] Recommendation 9.3: IEP

6. Licensees and Holders of Construction Permits In Enclosure 2 of Reference 4, the NRC requested that licensees 'reevaluate the flooding hazards at their sites against present-day regulatory guidance and methodologies being used for early site permits and combined license reviews.'

On behalf of Exelon Generation Company, LLC (Exelon), this report provides the information requested in the March 12, 50.54(f) letter; specifically, the information listed under the 'Requested Information' section of Enclosure 2, paragraph 1 ('a' through 'e'). The 'Requested Information' section of Enclosure 2, paragraph THREE MILE ISLAND NUCLEAR POWER STATION Page 6 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl 2 ('a' through 'd'), Integrated Assessment Report, will be addressed separately if the current design basis floods do not bound the reevaluated hazard for all flood causing mechanisms.

b. Requested Actions Per Enclosure 2 of Reference 4, Addressees are requested to perform a reevaluation of atl appropriate external flooding sources, including the effects from local intense precipitation (LIP) on the site, probable maximum flood (PMF) on streams and rivers, storm surges, seiches, tsunami, and dam failures. It is requested that t he reevaluation apply present-day regulatory guidance and methodologies being used for ESP and calculation reviews including current techniques, software, and methods used in present-day standard engineering practice to develop the flood hazard. The requested information will be gathered in Phase 1 of the NRC staff's two phase process to implement Recommendation 2.1, and will be used to identify potential 'vulnerabilities' (see definition below).

For the sites where the reevaluated flood exceeds the design basis, addressees are requested to submit an interim action plan that documents actions planned or taken to address the reevaluated hazard with the hazard evaluat ion.

Subsequently, addressees should perform an integrated assessment of the plant to identify vulnerabilities and actions to address t hem. The scope of the integrated assessment report will include full power operations and other plant configurations that could be susceptible due to the status of the flood protection features. The scope also includes those features of the ultimate heat sinks (UHS) that could be adversely affected by the flood conditions and lead to degradation of the flood protection (the loss of UHS from non-flood associated causes are not included). It is also requested that the integrated assessment address the entire duration of the flood conditions.

A definition of vulnerability in the context of {Enclosure 2) is as follows: Plant-specific vulnerabilities are those features important to safety that when subject to an increased demand due to the newly calculated hazard evaluation have not been shown to be capable of performing their intended functions.

c. Requested Information Per Enclosure 2 of Reference 4, the final report should be provided documenting results, as wetl as pertinent site information and detailed analysis, and include the following:

a. Site information related to the flood hazard. Relevant structures, systems, and components (SSCs) important to safety and the UHS are included in the scope of this reevaluation, and pertinent data concerning these SSCs should be included. Other relevant sit e data includes the following:

i. Detailed site information (both designed and as-built), including present-day site layout, elevation of pertinent SSCs important to safety, site topography, as well as pertinent spatial and temporal data sets; ii. Current design basis flood elevations for all flood causing mechanisms; iii. Flood-related changes to the licensing basis and any flood protection changes (including mitigat ion) since license issuance; iv. Changes to the watershed and local area since license issuance; THREE MILE ISLAND NUCLEAR POWER STATION Page 7 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1

v. Current licensing basis flood protection and pertinent flood mitigation features at the site; vi. Additional site details, as necessary, to assess the flood hazard (i.e., bathymetry, walkdown results, et c.)

b. Evaluation of the flood hazard for each flood causing mechanism, based on present-day methodologies and regulatory guidance. Provide an analysis of each flood causing mechanism that may impact the site including local intense precipitation and site drainage, flooding in streams and rivers, dam breaches and failures, storm surge and seiche, tsunami, channel migration or diversion, and combined effects. Mechanisms that are not applicable at the site may be screened-out; however, a justification should be provided. Provide a basis for inputs and assumptions, methodologies and models used including input and output files, and other pertinent data.

c. Comparison of current and reevaluated flood causing mechanisms at the site. Provide an assessment of the current design basis flood elevation to the reevaluated flood elevation for each flood causing mechanism. Include how the findings from Enclosure 2 of the 50.54(f) letter (I.e.,

Recommendation 2.1 flood hazard reevaluations) support this determination. If the current design basis flood bounds the reevaluated hazard for all flood causing mechanisms, include how this finding was determined.

d. Interim evaluation and actions taken or planned to address any higher flooding hazards relative to the design basis, prior to completion of the integrated assessment described below, If necessary.

e. Additional actions beyond Requested Information item 1.d taken or planned to address flooding hazards, if any.

3. SITE INFORMATION

a. Detailed Site Information Three Mile Island Nuclear Power Station (TMI) is located on Three Mile Island, approximately 2.5 miles south of Middletown, Pennsylvania. The station midpoint longitude is 76.43'30" west and latitude 40°09'15 north (Reference 13). Three Mile Island Is one of the largest of a group of several islands in the Susquehanna River and Is situated about 900 ft from its east bank (Reference 13). It Is elongated parallel to the flow of the river, with its longer axis oriented approximately due north and south. The island is about 11,000 ft In length and 1,700 ft in width (Reference 13).

The southeasterly flowing Susquehanna River makes a sharp change in direction, to nearly due south, in t he vicinity of Middletown (Reference 13). After this directional change just north ofThree Mile Island, the river widens to approximately 1.5 miles (Reference 13). York Haven Dam Is located just downstream of the station. The Conowingo Dam is located approximately 50 miles downstream of TMI Unit 1 (Reference 13).

The Susquehanna River Watershed to TMI Unit 1 is approximately 24,901 square miles and spans Pennsylvania and parts of New York State.

The following SSC are important to safety and are located below site grade elevation (304' elevation):

o LPI/Decay Heat Removal pumps and valves o Emergency electrical power distribution center, lC ESV 480V MCC o HPl/makeup pumps and valves o Emergency feedwater pumps and valves THREE MILE ISLAND NUCLEAR POWER STATION Page 8 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl The following SSC are important to safety and are located on 305' elevation:

o Emergency electrical power diesel generators o Emergency electrical power distribution center lA & 1B ESV 480V MCC The following SSC are important to safety and are located at 322' elevation or above:

o Stat ion batteries o Vital power inverters o Vital instrument power distribution centers o All emergency electrical power distribution centers (except as noted above) o Control room

b. Current Design Basis Flood Elevations for All Flood Causing Mechanisms The design basis was reviewed to determine which flood-causing mechanisms are considered in the current design basis flood. Below is a summary of flood-causing mechanisms considered and not considered in the design basis.

1. Local Intense Precipitation The design basis does not address LIP.

1. Flooding in Streams and Rivers The design basis flood elevation at the Intake Screen and Pump House {ISPH) is 313.3 ft NGVD-29 for a river flooding event (Reference 13).

The Susquehanna River is the principal source of flooding for TMI Unit 1. The TMI Unit 1 design flood peak discharge of 1,100,000 cfs was established based on the PMF as defined by the US Army Corps of Engineers (USACE) in 1967 (time of the Preliminary Safety Analysis Report (PSAR)) (Reference 13). The design of the dike was based upon the peak flow rate of 1,100,000 cfs (Reference 13). In 1969, the USACE issued a revised PMF peak river flow of 1,625,000 cfs at TMI Unit 1 (Reference 13). The original license (1973) for TMI Unit 1 included a commitment that "the plant would be provided with component protection to the degree which will assure a safe and orderly shutdown for the level of flooding postulated by the official value of the new Probable Maximum Flood, as modified by existing upstream flood cont rol projects (Q =

1,625,000 cfs)" (Reference 13). In a 1970 study, the predicted water level for the PMF peak discharge of 1,625,000 cfs was determined to be elevation 309 ft National Geodetic Vertical Datum of 1929 (NGVD-29) at the ISPH (Reference 13). In 2011, the st age discharge relationship at TMI Unit 1 was re-evaluated and t he current predicted water level for the PMF peak discharge of 1,625,000 cfs is 313.3 ft NGVD-29 at the ISPH (Reference 13).

The 1969 USACE PMF study was performed using the USACE HEC-1 hydrologic modeling software. The input parameters included the unit hydrograph or coefficients for its computation, rainfall mass curve ratio, sub-basin mean rainfall, infiltration indices, ratio of exponential recession curve, base flow, unit graph duration and drainage area (Reference 26). The PMF study was performed using the Probable Maximum Storm (PMS) rainfall from HMR 40 (Reference 35) Routing coefficients were optimized for each reach by reconstituting the observed hydrographs of several floods. Routing of the PMF through the reaches was accomplished by the Muskingum method, using two constant coefficient s for each reach. The routing reaches and coefficient values used were obtained from a November 1968 SRBC Study "Unit Hydrographs and Flood Routing" (Reference 26).

THREE MILE ISLAND NUCLEAR POWER STATION Page 9 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 The 1968 USACE PMF study indicates that previous studies "show an initial toss of one inch is acceptable for large flood producing storms in the Potomac, Susquehanna and Delaware River Basins. Infiltration loss rotes during runoff should show an increment of reduced loss for the winter and spring condition of snow cover and soif conditions; for this reason, 0.05 inch per hour was used rather than o higher figure. No effort was mode to determine regional loss rotes." (Reference 26) The use of a single conservative infiltration loss rate applied across the ent ire wat ershed did not consider variation due to soils, nor was it adjusted for calibration to extreme storm events. The 1968 USACE PMF study did not consider combined effects precipitation events or rain-on-snow events.

3. Dam Breaches and Failures Upstream Dam Breaches and Failures Per USFAR Section 2.6.2.1, the peak predicted water elevation is less than 301.6 ft NGVD-29 (Reference 13).

The potential for an upstream dam failure to use floodin at the site was investigated during the original licensin rocess and review Per t he TMI Unit 1 UFSAR (section 2.6.3), a 2005 study confirmed that, wit h river flow of 1,700 cfs (minimum daily flow of record at Harrisburg) and postulated failure and complete removal of both the York Haven Dam and the East Channel Dam, approximately 882 cfs of water will be available at the service water screen and pump house (Reference 13). This corresponds to a river water surface elevation of 271.93 ft above mean sea level (Ref erence 13). The low-flow intake canal assures that the intake structure has continuous access to river waters at water surface elevations of 270 ft or above (Reference 13). The 2005 study confirmed the original license basis. TMI Unit 1 can safely shutdown after an extreme event resulting in a complete failure and removal of both downstream impoundment dams.

4. Storm Surge The design basis does not address storm surge.

5. Seiche The design basis does not address seiche.

6. Tsunami The design basis does not address Tsunami.

7. Ice Induced Flooding The design basis does not address ice induced flooding.

8. Channel Migration or Diversion The design basis does not address channel migration or diversion.

THREE MILE ISLAND NUCLEAR POWER STATION Page 10of64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1

9. Combined Effects The design basis does not address wind generated waves during the PMF.

c. Flood Related Changes to the Licensing Basis and Any Flood Protection Changes (including mitigation) Since License Issuance Flood-Related Changes to the Ucenslng Basis Since License Issuance The following is a summary of flood-related changes to the licensing basis since license issuance:

1969 - USACE issued a revised PMF peak river flow of 1,625,000 cfs at TMI Unit 1 (Reference 13).

1970 - Peak flood elevation was determined to be 309 ft NGVD-29 at the ISPH during the PMF peak river flow of 1,625,000 cf s (Reference 13).

September 2011 - A re-analysis of the river stage discharge relationship at TMI Unit 1 concluded that the water level at the peak licensing basis event flow of 1,625,000 cfs was 313.3 ft NGVD-29 as compared to 309 ft NGVD-29 as originally (1970) analyzed. TMI Unit 1 flood protection barriers was re-evaluated and modified to establish a minimum level of protection of 313.5 ft NGVD-29 elevat ion.

The original analysis was superseded by an updated river stage discharge analysis. A steady-state, one-dimensional hydraulic analysis was completed using a HEC-RAS hydraulic model of the Susquehanna River. The hydraulic model extends approximately 13.7 miles downstream of TMI Unit 1 to the Marietta USGS Gage Station, and 5.4 miles upstream of TMI Unit 1. Topographic information for t he model was obtained from the PA DCNR LiDAR data for Dauphin, York, and Lancaster Counties. Bathymetric data (from measurements obtained and evaluated by Ocean Surveys, Inc. in April 2005) along the Susquehanna River around TMI Unit 1 upstream of the York Haven Dams was incorporated into the HEC-RAS model. Additional survey data for the top of dams and the river bottoms, representing the East and Main Channels below each dam (the "Conewago Rapids" ), were incorporated into t he HEC-RAS model from the 1999 York Haven Hydrostation Fish Passage Project. For all other cross sections, bathymetric data was obtained from FEMA's effective (1978) HEC-2 hydraulic model.

Additional parameters (such as Manning's 'n' values, ineffective flow areas, and blocked obstructions) were added based on review of PAMAP orthophotographs for Dauphin, Lancaster, and York Counties as well as local observations. Structures, including the East Channel York-Haven Dam, West Channel York-Haven Dam, Shocks M ill Railway Bridge, upstream TMI Unit 1 access bridge, and downstream TMI Unit 1 access bridge, were included in the HEC-RAS model. A USGS stage-discharge rating curve for the Marietta Gage Station was used to establish the downstream boundary conditions of the HEC-RAS model. The HEC-RAS model was run at the measured peak flow rate from the 1972 Tropical Storm Agnes event to calibrate by adjustment of the Manning ' n' values. A range of discharge values were used in the hydraulic analysis to develop a stage-discharge rating curve at each location of interest along TMI Unit 1.

THREE MILE ISLAND NUCLEAR POWER STATION Page 11 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Flow pattern complexities along the Susquehanna River in the vicinity of TMI Unit 1, caused by islands, dams, and bridges, raised concerns that the 10 HEC-RAS model would produce unacceptable errors in predicting flood elevations. To address these concerns, a Two Dimensional (2D) Finite-Element Computer Model, RiverFLO-2D, was used to simulate the PMF hydraulics (flow depths and velocity vectors). The 20 model reach included Three Mile Island, Shelly Island, Hill Island, the York Haven Dams, and the two (2) access bridges. The 20 model was calibrated to observed peak WSEs from Tropical Storm Agnes. WSEs from the appropriate models were used to develop stage-discharge rating curves.

Flood Protection Changes Since License Issuance The following is a summary of flood protection changes since license issuance:

1994 - As a result of the GL 88-20 and the IPEEE risk assessment process, equipment was procured and a plan based on portable equipment was implemented to provide core cooling for events with a water level up to 320 ft elevation.

1995 - Inadequate flood protection for conduits leading into the Auxiliary Building from the BWST Tunnel was identified and corrected.

1998 - A flood protection deficiency with ISPH floor drains was identified and corrected (LER 98-07).

1999 - Flood protection deficiencies with Turbine Bldg and RB Personnel hatch area floor drains were identified and corrected (LER 99-010).

August 2010 - Flood protection deficiency with floor drains in the air intake tunnel was identified and corrected. Investigation into adequacy of control building sewage line flood protection was initiated.

January 2011 - Flood protection design basis document, flood barrier system drawings and several minor flood protection modifications issued.

April 2011 - Major flood emergency procedure overhaul and validation exercise completed, training completed and revision issued.

In September 2011, component evaluations and modificat ions of gates, seals and other barriers were completed to raise the level of flood protection to 313.5 ft NGVD-29.

In June 2012, a design change to provide a qualified flood protection seal for the Reactor Building seismic gap was completed.

In August 2012, flood protection walkdowns in response to the 50.54f request identified missing seals inside AIT conduits (LER 2012-02). Flood protection for the AIT conduits was restored prior to November 15, 2012.

d. Changes to Watershed and Local Area since License Issua nce Available studies of the watershed prior to 1974 were reviewed to determine the changes to the Susquehanna River watershed. The most significant change since the 1969 PMF was the construction of three flood control dams in t he upper portions of the watershed in 1979 (Tioga, Hammond, and Cowanesque Dams). Land use has also changed in the Susquehanna River watershed due to development, changes in land use and planning practices, such as efforts to reforest agricultural areas. In addition, stormwater management practices such as Pennsylvania Stormwater Management Act 167 (enacted since 1978) plans and changes in agricultural practices have been implemented throughout the watershed t o THREE MILE ISLAND NUCLEAR POWER STATION Page 12 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 achieve peak flow reduction. Table 1 below summarizes the land use changes in the watershed based on an evaluation of available land use data obtained from a 1970 study and 2011 study from the Susquehanna River Basin Commission.

Table 1: Summary of Land Use Changes in Susquehanna River Watershed Land Use (percent of land area)

Referenced Study Forest Grass Cultivated Urban Water/Wetland Other 1970 Susquehanna River Basin Study 1 51% 20% 25% 4% NA NA (Reference 12) 2011 Nutrients and Suspended Sediment in the Susquehanna River 69% NA 21% 7% 2% 1%

Basin (Reference 11) 1 Estimated based on land use data provided In Reference 12.

NA - Land use category not reported in referenced study and therefore not available for comparison.

e. Current Licensing Basis Flood Protection and Pertinent Flood Mitigation Features An earthen dike surrounds the site. The dike elevation Is 310 ft NGVD-29 at the north end and slopes gradually to 304 ft NGVD-29 elevation at the south end. The dike provides protection for river flooding events with flows up to 1,100,000 cfs (UFSAR Section 2.6.S, Reference 13).

A system of barriers and seals prevent water intrusion into safety related structures for events where the water elevation is up to 313.5 ft NGVD-29. The flood gates and other temporary elements are Installed based on forecasted river elevation or rising river levels (UFSAR Section 2.6.5, Reference 13).

Additional mitigation capability to prevent core damage Is provided by a system of portable equipment for events with river levels up to 320 ft NGVD-29 elevation. This capability Is not a license basis requirement.

4.

SUMMARY

OF FLOOD HAZARD REEVALUATION The following is a summary evaluation of the flood hazard at TMI Unit 1 for each flood causing mechanism described in NUREG/CR-7046. These evaluations are based on acceptable industry standard methodologies and regulatory guidance. An analysis to identify each flood causing mechanism that may impact the site was performed including local intense precipitation and site drainage, flooding in streams and rivers, dam breaches and failures, channel migration or diversion, and combined effects. Mechanisms that are not applicable at the site (I.e., storm surge, seiche, and tsunami) have been screened-out as described below.

a. Local Inte nse Precipitation The LIP is a measure of the extreme precipitation (high intensity/short duration) at a given location. The duration of the event and the support area are needed to qualify an extreme precipitation event fully.

Generally, the amount of extreme precipitation decreases with Increasing duration and increasing area.

NUREG/CR-7046 (Reference 6) specifies that the LIP should be equivalent to the 1-hr, 2.56-km2 (1-mi2) probable maximum precipitation (PMP) at the location of the site.

The LIP event was evaluated for TMI Unit 1 to determine the associated flooding elevation and velocities assuming the active and passive drainage features are non-functioning. The analysis applies to the safety related buildings of TMI Unit 1 and areas where water flows through TMI Unit 2 into TMI Unit 1. Severe blockage of the 60-in culvert pipe that drains to the river across the south east dike was included in the analysis. The LIP evaluation was performed in accordance with the NRC's "Design-Basis Flood Estimation for THREE MILE ISLAND NUCLEAR POWER STATION Page 13 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Site Characterization at Nuclear Power Plants in the United States of America," dated November 2011 (NUREG/CR-7046). The LIP evaluation was developed in Calculation Package 1101-122-E410-013 (Reference 18).

The runoff caused by the UP event was estimated using the FL0-20 software. The software uses shallow water equations to route stormwater throughout the site. FL0-2D depicts site topography using a digital elevation model (DEM) to characterize grading, slopes, drainage divides, and low areas of the site. The DEM is a grid model developed from the composite ground surface (LiDAR and survey data). The methodology used within the FL0-2D software included the rainfall function, and the levee function to inc.orporate site security features which could impact the natural drainage characteristics of the site.

A composite ground surface of t he TMI site was developed using LiDAR data obtained in 2012, LiDAR data from validated portions of the publically available Pennsylvania Department of Conservation and Natural Resources (PA DCNR) 2008 Pennsylvania Spatial Data Access (PASDA) LiDAR data set, and supplemental ground survey obtained by AMEC in 2011 and 2012. The supplemental survey captured additional site features including the top elevations of concrete jersey barriers and concrete blocks, which were modeled in FL0-2D as levee features.

The model was created with boundaries along the riverside of the flood protection dike surrounding the east, north, western, and southern sides of the site. The model boundary along the southern end of the site is approximately 700 ft south of the flood protection dike.

Manning's Roughness Coefficients (n-values) were used to characterize the site's surface roughness and calculate effects on f low depths and velocities. Land cover for the site was evaluated using interpretation of the 2008 PASDA Orthoimagery. Interpretation of then-values was field verified on site.

Per NUREG/CR-7046 recommendations, runoff losses were ignored during the LIP event in order to maximize the water elevation on site from the event (Reference 18). Only overland flow and open channel systems were modeled and considered in the LIP f looding analysis.

Active and passive drainage system components (e.g., pumps, gravity storm drain systems, small culverts, inlets, etc.) were considered non-functional or clogged during the LIP event, per Case 3 in NUREG/CR-7046 (Reference 18); except that the 60-in culvert pipe located in the south east dike and isolation valve SO-V-120 were modeled. Due to its large diameter, the culvert was considered analogous to a bridge crossing, which would make it unlikely to be completely blocked during the LIP event. The area that could potentially provide a source for debris is within the TMI owner controlled area and is either paved or covered with gravel or paved surfaces with little vegetation or loose materials available. In addition, the water levels and velocities during the LIP event over most of the local drainage area to the GO-inch pipe are relatively low (generally water levels were less than 3 ft and velocities were less than 2 fps) and would not be expected to damage structures or create large water borne debris. Due to the culvert opening size, configuration of the inlet, and the lack of debris supply on site, it was assumed this culvert would not be fully blocked during an LIP event. However, the culvert was assumed to be 50% blocked to provide a conservative analysis. The remaining storm sewers and culverts were assumed blocked per Case 3 in Appendix B of NUREG/CR-7046.

PC Storm Water Management Model (PCSWMM) software was used to estimate the entrance losses from the 60-ln culvert pipe that drains to the river from the south east dike. As-built information (Reference 62) showing the culvert pipe location, culvert pipe inverts, pipe diameter, pipe length, pipe cross section and pipe material type was input into the PCSWMM model to estimate the rating curve. The rating curve estimated by the PCSWMM model was used as a direct input into the FL0-2D model. The inflow and outflow grid elements corresponding to the inlet/outlet of the 60-in culvert were set as a hydraulic THREE MILE ISLAND NUCLEAR POWER STATION Page 14 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 structure component in FLO-20. The stage-area curve calculated for the culvert was used to estimate the flow losses from the surface due to the operation of the 60-in culvert pipe assuming 50% blockage. The specified hydraulic structure outflow elements in the FLO-20 model used the outflow stage-discharge curve to remove flow at each water surface stage during the model run. The FLO-2D model calculated the water surface elevations based on t he remaining flow in the model at each time step.

The outlet of the 60-inch culvert is above the normal river level. Therefore, the influence of external high water events on the culvert outflow was not considered. NUREG/CR-7046 does not require a coincident river flooding condition be considered for LIP flood hazard evaluations. The LIP event is considered a high intensity, short duration event that occurs at the site. If the UP event were to cause external high water levels from the Susquehanna River, it would be expected that due to the drainage basin size compared to t he site area, that the LIP peak flood levels would not coincide with external-high water events.

Per NUREG/CR-7046, the 1-hr, 1 square mile PMP event was developed using HMR 52 (Reference 36). The total PMP depth per square mile for the 1-hr event was interpolated from the PMP depth cont our map provided in Figure 24 of HMR 52 (Reference 36). The distribution of the 1-hr PM P was developed for the 5-,

15-, and 30-minute t ime intervals, with the GO-minute interval being the 1-hr PMP depth. The dept h for each time interval was calculated using the ratios obtained from Figures 36, 37, and 38 of HMR 52. The 1-hr PMP distribution is provided in Table 2 and Figure 1 below. The 1-hr PMP was modeled in FLO-2D to calculate the subsequent site flooding.

Table 2: 1-hr/1-sq-mi PMP Distribution Time Percent Total PMP Cumulative Depth Reference (minutes) (%) (inches) 0 0% 0.00 N/A 5 33.48% 5.96 Figure 36 of HMR 52 15 52.68% 9.38 Figure 37 of HMR 52 30 75.59% 13.46 Figure 38 of HMR 52 60 100% 17.81 Figure 24 of HMR 52 THREE MILE ISLAND NUCLEAR POWER STATION Page 15 of 64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 20 18 - J

-;; 16

,~ :.-i.-

"~ 14

.r:.

- ,_....:,..,,!.-'Y" --

£ 12 0.

I,

~-

Q" 10 '.D 3 ~

J,, ,.

~

~ 8

]

C 6 ,.

"ii J a:: 4 I

J 2

0 I

0 10 20 30 40 so 60 Time (minutes)

Figure 1: 1-HR PM P Distribution To determine the flooding elevation associated with the LIP, the 1-sq-mile/ 1-hr storm was applied evenly across the site, and the model was allowed to run for 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months to ensure that only the areas of static ponding would remain.

A summary of the maximum predicted flooding results for pathways where flood water could enter TMI Unit 1- safety related structures is provided in Table 3.

THREE MILE ISLAND NUCLEAR POWER STATION Page 16 of 64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl Table 3 : LIP Predicted Flooding Results at Pathways to TMI Unit 1 Safety Related Structures (Maximum Water Surface Elevation, Flooding Depth, and Velocity)

Mall. Flooding Flooding Minimum Depth Above Duration Mall.

Water Max. M inimum AbOlle Pathway Safety Related MH. Water Surface Flooding Pathway Elevation Velocity Water Elevation Minimum Building Depth (NGVD29) to Enter Water to Enter Pathwav Elevation Pathway ft ft ft ft/sec ft hrs (NAVD88) (NGVD29)

Air Intake Pagoda Air Intake 313.5 304.4 305.2 4.2 1.5 0 0

& Tunnel BWSTTunnel 305.0 304.5 305.3 0.6 1.4 0.3 0.80 Auxiliary Sumps (WDL*V-612)

Building, IJnit 1 Entrance on south 305.0 304.4 305.2 3.8 0.7 0.2 0.49 wall (A-116)

Entrance on north wall (TMI-FG-81)'

Entrance on Control Building, northeast wail 306.0 304.4 305.2 4.7 0 0 0 Unit I (TMl*FG-82)'

Turbine Building Drains (#1)2 Entrance on north Diesel 305.0 304.3 305.1 0.8 0.6 0.1 0.63 wall (TMI-FG*Dl)

Generating Entrance on east Building 305.0 304.4 305.2 0.7 0 0.1 0.76 wall (TM I-FG*D3) u Entrance on south 305.0 304.4 305.2 4.7 0 0.2 0.59 wall (TMI-FG-Al)'*'

Entrance on west Fuel Handling 301.5 304.4 305.2 4.7 2.0 3.7 23.96 wall (FH-208)

Building, Unit 1 Turbine Building Drains (#6, 7, 9, & 305.0 304.4 305.2 4.7 0 0.2 0.59

10) '*'

Intake Scrten & Entrance on south Pump House, side (TM I-FG-El & 308.0 304.3 305.1 1.0 0.7 0 0 1

Unit l E2A/B/C)

Reactor Building, None NA 304.6 305.4 0.7 1.4 NA NA Unit I

'Data Extracted from Exterior Grid Adj acent to Outside Entrance Leading to that Door.

' Velocity assumed zero for Interior doors.

Technical Evaluat ion 1467688-04 (Reference 64}, which was performed by Exelon to evaluate impacts caused by the LIP event, shows that without any active mitigation, some water would enter the Fuel Handling Building, Auxiliary Building, and Diesel Generator Building. However the limited volume would not adversely affect any safe shutdown functions.

b. Flooding in Streams and Rivers The PMF in rivers and streams adjoining the site should be determined by applying the PMP to the drainage basin in which the site is located. The PMF is based on a transformation of PMP rainfall on a watershed to flood flow. The PMP is a deterministic estimate of the theoretical maximum depth of precipitation that can occur at a time of year of a specified area. A rainfall-to-runoff transformation function, as well as runoff characteristics, based on t he topographic and drainage system network characteristics and watershed properties are needed to appropriately develop the PMF hydrograph. The PMF hydrograph is a time history THREE MILE ISLAND NUCLEAR POWER STATION Page 17 of 64 RCN:TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 of the discharge and serves as the input parameter for other hydraulic models which develop the flow characteristics including f lood flow and elevation.

The precipitation driven PMF discharge was determined from the evaluation of three combined-effect flood scenarios defined by NUREG/CR-7046, Appendix H.1 Floods Caused by Precipitation Events (Reference 6). A deterministic HEC-HMS model was developed and used to evaluate the combined-effect floods. The model was based on the best available geospatial data and was calibrated to a recent severe storm event (Tropical Storm Lee, September 2011).

The three precipitation driven combined-effect flood scenarios evaluated in this report are consistent with NUREG/CR-7046, Appendix H.1 Floods Caused by Precipitation Events (Reference 6):

Scenario 1 - Combination of:

Mean monthly base flow;

Median soil moisture;

Antecedent or subsequent rain: the lesser of (1) rainfall equal to 40 percent of PMP and (2) a 500-year rainfall; and

PMP.

Scenario 2 - Combination of:

*Mean monthly base flow;

Probable maximum snowpack; and

A 100-year, snow-season rainfall.

Scenario 3 - Combination of:

Mean monthly base flow;

A 100-year snowpack; and

Snow-season PMP.

This evaluation was performed with the hydrologic parameters, inputs, and assumptions recommended in NUREG/CR-7046. Hydrologic parameters, inputs, and assumptions based on Federal regulatory guidance from other agencies (i.e., NRCS, USGS, USACE), previous studies, and engineering judgment were used to develop parameters where NUREG/CR-7046 did not provide guidance. The methodology in this evaluation is consistent with the following guidance documents:

NRC Office of Standards Development, Regulatory Guide: RG 1.59 - Design Basis Floods for Nuclear Power Plants, Revision 2, dated August 1977.

American National Standard for Determining Design Basis Flooding at Power Reactor Sites (ANSI/ANS 2.8-1992).

NUREG/CR-7046 "Design-Basis Flood Estimation for Site Characterization at Nuclear Power Plants in the United States of America," publication date November 2011.

The purpose of this evaluation was to determine the governing PMF peak discharge at TMI Unit 1. The PMF scenario producing the greater calculated peak discharge of the three scenarios evaluated (listed above) Is identified as t he governing PMF. The governing PMF is combined with a conservative assumption that all dams will fail during a precipitation driven event as to determine the PMF hydrograph described in this THREE MILE ISLAND NUCLEAR POWER STATION Page 18 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 section. The evaluation of waves induced by 2-year wind speed applied along the critical direction on top of the PMF peak water surface elevation is discussed in Section 4.i "Combined Effect Flood".

The following sub sections describe the inputs, assumptions, methodology, and results for the calibrated hydrologic model and each of the three PMF scenarios.

Calibrated Hydrologic Model A hydrologic rainfall-runoff model calibrated to an extreme event was developed to estimate the PMF peak f low rates during a PMP event, and meet the standards established in NUREG/CR-7046 (Reference 6). The USACE Hydrologic Engineering Center - Hydrologic Modeling System (HEC-HMS) software, version 3.5 was used to simulate the hydrologic processes of the watershed. The HEC-HMS Model was calibrated within acceptable tolerance to recorded stream gage data from Tropical Storm Lee, which occurred between September 4th and 11th, 2011. The calibrated hydrologic model was developed in Calculation Package 1101-122-E410-010 (Reference 15). Since recorded stream gage data does not exist at TMI Unit 1, calibration was performed to the closest upstream and downstream bounding USGS stream gages in Harrisburg and Marietta, respectively. Tropical Storm Lee was chosen for the calibration since it is a recent extreme event that reflects a well-defined and severe storm for the basin. This event produced stream gage readings reflecting the basin's response to an extreme precipitation event with saturated soils caused by antecedent storms during current watershed conditions. Current land use and the addition of three USACE flood control dams in the watershed, which were not reflected in previous studies, were included in the analysis. Heavy rainfall from Hurricane Irene in late August 2011, as well as severe summer storms occurring prior to Hurricane Irene, resulted in saturated soil conditions in the Susquehanna River Watershed prior to the arrival of Tropical Storm Lee. These conditions are similar to Scenario 1 of NUREG/CR-7046, Appendix H.1 Floods Caused by Precipitation Events. NUREG/CR-7046 recommends a severe storm event is used for calibration (Reference 6). Rain gages were available for this storm as well as corresponding streamflow gages. Tropical Storm Lee was isolated from snowmelt effects. Calibration parameters obtained from the Hurricane Agnes HEC-1 model (Reference 28) inputs were used for initial values for the calibration.

The weighted average rainfall throughout the Susquehanna River Basin for Tropical Storm Lee is generally greater than Tropical Storm Agnes in the upper basins and lesser than Tropical Storm Agnes in the lower basins, as shown in Table 4. However, Tropical Storm Lee was considered a more appropriate event for calibration given it occurred during current watershed conditions with an antecedent storm. In addition, the duration and temporal distribution of Tropical Storm Lee is comparable to the 72-hour PMP event. PMP rainfall throughout the watershed based on HMR 40 is also provided in Table 4.

THREE MILE ISLAND NUCLEAR POWER STATION Page 19 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Table 4 Observed Rainfall of Tropical Storm lee and Tropical Storm Agnes Tropical Storm Lee Basin Tropical Storm Agnes Basin HMR 40 Basin Average 1 2 3 Discharge Gage Average Rainfall (in) Average Rainfall (in) Rainfall (in)

Waverly, NY 7.68 4.47 15.5 Chemung, NY 5.64 9.38 18.4 Wilkes-Barre, PA 6.99 6.63 14.2 Lewisburg, PA 6.3 9.92 16.4 Newport, PA 6.6 10.17 18.8 Harrisburg, PA 6.63 9.06 12.7 l

Basin weighted average estimated from Reference 65 2

Reference 28 3

Reference 3S Precipitation gages were evaluated t hroughout the watershed to identify gages that produced hourly rainfall data for Tropical Storm Lee while it was passing over the Susquehanna River Watershed. Hourly precipitation data from fourteen (14) rain gaging stations locat ed throughout the wat ershed in New York and Pennsylvania was obtained from the National Oceanic and Atmospheric Administration (NOAA)

National Climate Data Center. The Thiessen Polygon Method (Reference 25) was used to estimate aerially averaged values from point rainfall data obtained from the fourteen (14) rain gaging stations.

Stream flow data from 33 gage stations located in New York and Pennsylvania was obtained from the Unit ed States Geological Survey (USGS). Subbasins were delineated based on the locations of High Hazard dams and USGS stream flow gages within the watershed. The locations of the USGS stream flow gages provided 24 potential calibration points within the HEC-HMS model upstream of the final calibration point, and one USGS stream flow gage just downstream of Three M ile Island, at Marietta, PA. The tot al calibrated model watershed to Marietta, PA is 25,960 square miles. The subbasins were delineated based on preliminary USGS HUC-12 boundaries, refined using USGS t errain data and processed with ArcGIS computer software to produce updated boundaries to the watersheds to be used in the HEC-HMS model. The drainage boundary map is provided in Figure 2 below.

THREE MILE ISLAND NUCLEAR POWER STATION Page20 of64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 NEW YORK Legtnd

[ j ><EC-HMS lullllolln PENNS

{

l Figure 2: Watershed and Subbasln Delineated Soils data was obtained from the National Resource Conservation Service (NRCS). Hydrologic Soils Groups

{HSGs) for each soil type were identified for each subbasin using ArcGIS. Identification of HSGs was used to obtain hydraulic conductivity to establish the preliminary constant loss rate for the watershed in the model prior to calibration. The preliminary constant loss rate was calculated based on Table 11 in the HEC-HMS Technical Reference Manual (Reference 22). The constant loss rates were calculated based on the HSG using the ArcGIS Zonal Statist ics function In order to estimate the weighted basin average for the losses.

Table 5 provides the acceptable range in loss rates used to bound the calibration of the loss rate parameters. Figure 3 shows the distribution of preliminary loss rates throughout the watershed. The average percent impervious area for each subbasin was calculated from the National Land Cover Database (NLCD) 2006 Percent Developed Impervious data published by the USGS. The weighted average was calculated using the ArcGIS Zonal Statistics function.

THREE MILE ISLAND NUCLEAR POWER STATION Page 21 of 64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Table 5 - Hydrologic Solls Groups and Loss Rates Hydro logic Range of Loss Rates Description Soll Group In/hr mm/hr A Deep sands; Deep loess; Aggregated silts 0.30 - 0.45 7.62 - 11.43 B Shallow loess; Sandy loams 0.15 - 0.30 3.81 - 7.62 C Clay loams; Shallow sandy loams; Soils low In organic content; Soils usually high In clay 0.05-0.15 1.27 - 3.81 D Soils that swell significantly when wet; Heavy plastic clays; Certain saline soils 0.0-0.0S 0.0-1.27 NEW YORK Legend Pt*un1Ntr; Consbtnt Lon Figure 3: Preliminary Constant Loss Rat.e (mm/hr)

Calibration of the HEC-HMS model required an initial set of parameters to begin the iterative process of calibration. Initial model parameters were chosen based on available previous studies and soils data. The hydrologic model used the initial and constant loss method, Clark Unit Hydrograph, mean monthly base flow method, and Muskingham routing method, which were used In the 1975 Hurricane Agnes hydrologic model (Reference 28). During calibration, Muskingum parameters K and X, initial loss rate, and time of concentration were Initially set equal to the values from the HEC-1 Tropical Storm Agnes Model (Reference 28), since the model reflects a high flow condition in the watershed. These values were used as the init ial conditions and were further refined during the calibration process.

THREE M ILE ISLAND NUCLEAR POWER STATION Page 22 of 64 RCN:TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 The measured stream gage data during Tropical Storm Lee was used as a basis for calibrating these preliminary parameters through an iterative process. Calibration was performed for each individual subbasin within the 25,960-square-mile Susquehanna River watershed to Marietta working from the upstream to downstream subbasins. Characteristics used to measure the accuracy of flt included peak discharge, volume, time to peak, and shape of the resulting hydrograph. Peak discharge and volume were prioritized in terms of best fit when calibrating the model. Calibration is considered complete when peak discharge is within 25% of observed data at a gage (Reference 29).

The calibrated HEC-HMS model produced peak flows within approximately 10% of the peak discharge for Tropical Storm Lee, as recorded at the Harrisburg USGS gage and within approximately 5% of the peak discharge for Tropical Storm Lee, as recorded at the Marietta USGS gage. The calibration results are provided in Figure 4 and Figure 5 below.

- ~--------------..:,_ ______________________

t -

I...

l'dWI _M,llr,IDl'.MII.._

Susquehanna River at Harrisburg, PA. USGS Gage 01570500 Model Peak Obs peak Model volume Obs volume Model peak time Obs peak time flow(cms) flow (ems} (mm) (mm) 14,983.81 16,735.25 08sep201119:00 09sept201101:00 91.74 106.83 Figure 4: Susquehanna River HEC* HMS Model Calibration Results at the Harrisburg USGS Gage THREE MILE ISLAND NUCLEAR POWER STATION Page 23 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporat ion Rev 1

- ~------------------------------------,

1-Susquehanna River at Marietta, PA. USGS gage 01576000 Model Peak Obs peak Model volume Obs volume Model peak time Obs peak time flowlcms) flowlcms) Imm) Imm) 17.946.4 18.830.70 09sep2011 01:00 09sept20l 1 05:00 98.58 113.24 Figure 5: Susquehanna River HEC-HMS Model Calibration Results at the Marietta USGS Gage Scenario 1 Per NUREG/CR-7046, Appendix H.1 Floods Caused by Precipitation Events (Reference 6), the Scenario 1 combined-effect precipitat ion was evaluated to include:

Mean monthly base flow;

Median soil moisture;

Antecedent or subsequent rain: the lesser of (1) rainfall equal to 40 percent of PMP and (2) a 500-year rainfall; and

PMP.

Scenario 1 was evaluated using the calibrated hydrologlc model developed in Calculation Package 1101-122-E410-010 (Reference 15) and derived precipitation inputs to reflect the combination of mean monthly base flow, antecedent rain, and the PMP event, as well as median soil moisture conditions. Scenario 1 was evaluated in Calculation Package 1101-122-E410-011 (Reference 16).

Mean monthly base flow was previously calculated in Calculation Package C-1101-122-E410-010 (Reference 15) by taking the yearly average of the individual average daily stream flow data for each gage obtained from the USGS. Mean monthly base flow was calculated by averaging the mean base flow values annually and dividing them by area to obtain a unit area mean monthly base flow.

THREE MILE ISLAND NUCLEAR POWER STATION Page 24 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Table 6: Mean Monthly Base flow per Unit Area cfs/Square Mile ems/Square Kilometer Annual Annual 1.4584 0.0159 Subbasin loss parameters from the calibrated HEC-HMS model were used. These values were derived from the hydrologic model calibration to Tropical Storm lee. As previously discussed, Tropical Storm lee was a recent extreme event that reflects a well-defined and severe storm for the basin. This event produced stream gage readings reflecting the basin's response t o an extreme precipitation event with saturated soils caused by antecedent storms during current watershed conditions (land use and current configuration of USACE flood control dams). Heavy rainfall from Hurricane Irene in late August 2011, as well as severe summer storms occurring prior to Hurricane Irene, caused saturated soil conditions in the Susquehanna River Watershed prior to the arrival of Tropical Storm Lee. This reflects a similar condition to Scenario 1 and, therefore, the calibrated loss rates were determined appropriate for this evaluation.

HMR 40, prepared by the National Weather Bureau, contains the site-specific PMP for the Susquehanna River Basin to Harrisburg. HMR 40 utilizes a 72-hr event to account for runoff from the headwaters in the basin to reach Harrisburg before the event has terminated. The duration of the PMP should be long enough to enable a drop of water from the furthest point in the drainage area to reach the location where the flood flow is being calculated. This is called the time of concentration. Shorter durations do not allow all of the drainage area to contribute to the flood. HMR 40, which is specific to the Susquehanna River Basin, uses a 72-hr time of concentration for Harrisburg. HMR 51 indicates that site-specific studies should be used when available. Previous comparison with the application of HMR 51/52 rainfall depth and spatial distributions versus HMR 40, resulted with HMR 40 producing more conservative cumulative rainfall across the basin.

This is due to the multi-centered spatial distribution in HMR 40, which covers the entire Susquehanna River Basin. Precipitation depths from HMR 40 were used with the critical spatial and temporal distribution, also determined in HMR 40, throughout the watershed. An average value for the PMP for each subbasin was determined using ArcGIS Zonal Statistics, providing an aerially weighted average for each subbasin. The basin-wide average, 72-hr PMP for the Susquehanna River Basin at Harrisburg is 12.7 inches, obtained from Table 1 in HMR 40 (Reference 35).

The 500-year all-season precipitation was compared to the 40% PMP to determine the lesser of these events, to be used as the antecedent storm event in the Scenario 1 evaluation. To determine the 40% PMP, 40% of each subbasin value was used as the 40% PMP rainfall value to compare to the 500-year rainfall.

Existing data available for the 500-year all season precipitation from Northeast Regional Climate Center (NRCC) were unreliable at the border of Pennsylvania and New York. Therefore, the average all-season 500-year rainfall for the Susquehanna River Basin was developed using a Log Pearson Type Ill distribution analysis of precipitation gages throughout the watershed. Precipitation values were obtained from NOAA climate gage data. The data was extracted for a limited number of sites in and around the Susquehanna River Basin for the period of 1 January 1948 to the present. The average 500-year precipitation of the Susquehanna River Basin to TMI Unit 1 was distributed throughout the basin using the spatial distribution available in HMR 40. The spatial distribution of HMR 40 was developed using extreme events within the Susquehanna River watershed and, therefore, was considered appropriate for estimating the spatial distribution of the 500-year event for this evaluation. A grid of the spatially distributed values was used to calculate an aerially weighted average 500-year rainfall for each subbasin, using ArcGIS Zonal Statistics. The 40"/4 PMP as determined the lesser event when compared to the 500-year rainfall and, therefore, was used as the antecedent storm for Scenario 1.

THREE MILE ISLAND NUCLEAR POWER STATION Page 25 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl Two precipitation runs for Scenario 1 determined the conservative antecedent rainfall event timing for the combined storm events. ANS 2.8 and NUREG/CR-7046 recommend a configuration of the Scenario 1 precipitation event consisting an antecedent storm conditions followed by dry conditions prior to the PMP event (Reference 9 and 6). Due to the size of the watershed, the duration of the precipitation and dry conditions were set t o 72 hours8.333333e-4 days 0.02 hours 1.190476e-4 weeks 2.7396e-5 months . This Is consistent with the example provided in Sect ion C.4 of NUREG/CR-7046 and the duration of the PMP estimated in HMR 40. NUREG/CR-7046 does not provide specific recommendations for temporal distribution of PMP rainfall events; therefore, front-weighted and center-weighted precipitation distributions were considered. HMR 40 previously determined that the front-weighted precipitation distribution produces the most conservative result. Results from this analysis, comparing center-weighted to front-weighted precipitation distribution scenarios, are provided below In Table 7 for Scenario 1.1 (front-weighted precipitation distribution). This configuration yielded the governing discharge within Scenario 1. The subbasin precipitation hyetograph for Scenario 1.1 is provided in Figure 6.

Subbasln C1_W1010 Scenario 1.1 : 40% All Season PMP + 100% All Season PMP (Front-Weighted Temporal Distribution) 14 ee 12 S 10 Q.

~ 8 C

j 6

-i

g 4

it 2 0

11112010 11312010 1,s,2010 1n12010 11912010 111112010 111312010 Slmulatlon Time

- - Preclpltauon Only Figure 6: Scenario 1.1 Input Volume Table 7: Rainfall Distribution Sensitivity Results Peak Temporal Discharge Peak Discharge Scenario Conditions Distribution (ems) {cfs)

Scenario 1,1 0.4 All Season PMP + All Season PMP Front-Weighted 32,705 1,154,810 Scenario 1.2 0.4 All Season PMP + All Season PMP Center-Wel11hted 31,872 1,125 396 THREE MILE ISLAND NUCLEAR POWER STATION Page 26 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 The Scenario 1 peak discharge is 1,154,810 ds. This peak discharge is derived from the front-weighted distribution calculated with a 40% PMP antecedent event combined with a 100% PMP event in succession, in addition to a mean monthly base flow. The hydrographs for Scenario 1 are presented in Figure 7.

1200000 ,. *

,, 1-

' ' Scenario I.I f\ - - Scenulo 1.2 1100000 1000000 ' ...

' < T L

I I 900000 / I

, \ \ f

800000 I I \ \ '

I

\ '\

I'

[ 700000 GI r.i 600000 a:

3 l! 500000 r ..

,f l

'\

\

.,\

\

I

~+--

400000 300000 rr II I I \ '

'*1 ...

200000 ,, r r. I

\

I

' ' ~

~

100000 V

/

- J

~' ' ... - ... j

.J ,.,. ~

f

~

0 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 Time (hrs)

Figure 7: Scenario 1 Results Hydrographs at Three Mlle Island Scenario 2 Per NUREG/CR-7046, Appendix H.1 Floods Caused by Precipitation Events (Reference 6), the Scenario 2 combined-effect precipitation was evaluated to Include:

Mean monthly base flow;

Probable maximum snowpack; and

A 100-year, snow-season rainfall.

Scenario 2 was evaluated using the calibrated hydrologic model developed in Calculation Package 1101-122-E410-010 (Reference 15) and derived precipitation inputs to reflect the combination of mean monthly base flow, probable maximum snowpack, and a 100-year snow season rainfall. Scenario 2 was evaluated in Calculation Package 1101-122-E410-011 (Reference 16).

Mean monthly base flow was previously calculated In Calculation Package C-1101-122-E410-010 (Reference 15) by taking the yearly average of the individual average daily stream flow data for each gage obtained from the USGS. Mean monthly base flow was calculated by averaging the mean base flow values annually and dividing them by area to ,obtain a unit area mean monthly base flow.

THREE MILE ISLAND NUCLEAR POWER STATION Page 27 of64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Table 8: Mean Monthly Base flow per Unit Area ds/Square Mile ems/Square Kilometer Annual Annual 1.4584 0.0159 The 100-year cold season rainfall was evaluated with daily precipitation, snowfall and snow depth data obtained from the NCDC online archive from the GHCN. Precipitation values for the 100-year snow season rainfall and probable maximum snow pack were derived through statistical analysis of climate gage data obtained from NOAA. A cold/snow "season" is defined to be the period between 1 October and 30 April, which is consistent with HMR 42 (Reference 34). For the 100-year rainfall estimates, all precipitation that fell was assumed to be liquid (i.e., rainfall) regardless of the actual phase of precipitation. In order to calculate the worst cold season storm possible, variation in precipitation type was ignored and assuming all liquid precipitation produces the most conservative runoff conditions.

The data was extracted for a limited number of sites in and around the Susquehanna River Basin for the period of 1 January 1948 to the present. Data from valid sites were collated such that a Gumbel, Generalized Extreme Value (GEV), and Log Pearson Type Ill distributions were fitted to seasonal maximums.

These distribution types were chosen to meet NUREG/CR-7046 guidance to compute multiple types of distributions (Reference 6). The 100-year return periods were determined by inputting the distribution parameters into the cumulative density function (CDF) for the particular distribution. The 0.99 CDF value, which corresponds to the 1 in 100 year probability was used in the analysis. This analysis yielded a point 100-year probability rainfall at each gage analyzed. The values used in this analysis were calculated from the Log Pearson Type Ill distribution, which was determined to be the most conservative distribution. The average 100-year precipitation of the entire Susquehanna River Basin to TMI Unit 1 was distributed throughout the basin using the spatial distribution available in HMR 40 and the SCS Type II temporal distribution. A grid of the spatially distributed values was used to calculate an aerially weighted average 100-year rainfall for each subbasin, using ArcGIS Zonal Statistics.

The infiltration loss rates based on calibration to Tropical Storm Lee were used for this analysis. NUREG/CR-7046 Section 5.3 states that "the soils may be frozen during a rain-on-snow event and, therefore, should be assumed to allow no infiltration." The use of the Tropical Storm Lee calibrated loss rates was confirmed to be appropriate for a rain-on-snow event in the Susquehanna River Basin by comparison to the 1996 Susquehanna rain-on-snow event. The January 1996 rain-on-snow event produced record discharges and melt rates across the Susquehanna River Basin. Two meteorological studies of the 1996 event were available; one for the entire basin (References 44) and the other concentrated on the Loyalsock Creek and Lycoming Creek watersheds near Williamsport, Pennsylvania (References 45). Both meteorological studies identified these watersheds to have been hardest hit from the rainfall and coinciding melt. The Loyalsock/Lycoming subbasin study provided comprehensive meteorological inputs (Reference 45), which were used to model the 1996 event for these subbasins in the calibrated HEC-HMS model. The Loyalsock/Lycoming subbasin study utilized the Energy Budget method of snowmelt runoff calculation, which is referenced and simplified in EM 1110-2-1406, in order to calibrate parameters in their models. The analysis includes a model of Lycoming Creek; however, ice jams (Reference 45) in Lycoming Creek during the 1996 event produced artificially high peak flow rates at the Trout Run USGS stream flow gage.

Therefore, the Lycoming Creek gage information is discounted in final consideration of applicable loss rates for t his event.

THREE MILE ISLAND NUCLEAR POWER STATION Page 28 of 64 RCN: TMl-140

NTTF Recommendation 2.1 {Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 The model input parameters obtained from this study include hourly change in snowpack depth, t emperatures, and rainfall. The study indicates that approximately 1 meter of snowpack was depleted throughout the event, and resulted in 10 cm t o 15 cm of available water for runoff. This indicates that the density of water was greater than 10% of the total snow depth. Typically, a standing snowpack experiences densification and can produce snow water equivalent percentage as high as 40% for ripe snow (Reference 66). Based on this study, an assumed 10% snow-water-equivalent would provide a conservative estimate. However, based on the snow-water-equivalent map of the 1996 event, approximately 5 inches (12.7 cm) of snow water equivalent was available before the storm in the Loyalsock Creek Basin. This suggests an approximate 12.4% snow-water-equivalent was available prior to the 1996 event in t he Loyalsock Creek Basin. For this evaluation, a 12% snow-water-equivalent was assumed, which provides an overall contribution from snowmelt of 12.36 cm.

A sensitivity analysis was performed using 10% and 12% snow-water-equivalents to evaluate the performance of loss rates during a rain on snow event using the observed and conservat ive estimate of snow-water-equivalents during the 1996 event. The snow-water-equivalent assumed for each case (10%

and 12%) were added to hourly precipitation values for the event, obtained from the Millheim, Pennsylvania gage. The HEC-HMS model calibrated to Tropical Storm Lee was used to compare the response to a historical extreme rain-on-snow event for a sample set of the watershed. While this event occurred during January 1996, it was more typical of a spring event with cold temperatures and snow pack initially, followed by temperature increase with high winds causing rapid melt, and then a spring-like precipit ation event (Reference 45). The results of the sensitivity analysis are provided in Table 9.

Table 9: Comparison of Flows from 1996 Rain-on-Snow to those using the Tropical Storm Lee Loss Rates Observed Tropical Storm Model Flow Area Flow Lee Calibrated Using Tropical model loss rate Storm Lee % Difference Calibrated Loss Rates Sq miles cfs in/hr cfs %

Loyalsock Creek- 494 55,800 0 .16 53,658 -3.8 10%SWE Loyalsock Creek- 494 55,800 0.16 58,704 5.0 12%SWE The modeled flow based on the Tropical Storm Lee calibrated loss rates and a conservative 10% snow-water-equivalent provided a very good approximation of the observed flow in the Loyalsock Creek Basin.

Based on previous studies a standing snowpack can experiences densification and can produce snow water equivalent percentage as high as 40% for ripe snow (Reference 66). The modeled flow based on t he Tropical Storm Lee calibrated loss rates and 12% snow-water-equivalent provided a conservative estimate of the observed flow in the Loyalsock Creek. This sensitivity analysis indicates that the watershed infiltration loss rates during Tropical Storm Lee are similar to the infiltration loss rates during a historic extreme rain-on-snow event (in 1996). Therefore, loss rat es calibrated to Tropical St orm Lee provide an appropriate basis for the rain-on-snow analysis in the Susquehanna River Watershed.

THREE MILE ISLAND NUCLEAR POWER STATION Page29 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Probable maximum snowpack accumulation (PMSA) was calculated based on the techniques applied in HMR 42 and Buckler (References 34 and 30, respectively). At each valid site within the Susquehanna River Basin, the average seasonal snowfall and the maximum historical snowfall were calculated. The ratio of maximum to average seasonal snowfall was calculated and plotted against the average seasonal snowfall.

An enveloping (straight line) relationship was developed from the data so every data point lay beneath the line. Using the average seasonal snowfall for the basin as a whole, the corresponding ratio was found from the enveloping relationship. Using this ratio and the average seasonal snowfall for the basin, the PMSA was calculated. Since the PMSA value is not spatially distributed throughout the basin, the 100-year snowpack (evaluated under Scenario 3) was used to spatially distribute the PMSA throughout the basin. This technique involved multiplying the PMSA by the ratio of the 100-year snowpack in each subbasin to the average 100-year snowpack of the entire Susquehanna River Watershed to TMI Unit 1. Since snowpack depth is not the limiting factor for a rain on snow scenario, the PMSA was not further refined. The value used in this study represents a conservative estimate of the PMSA in the Susquehanna River Basin.

Snowmelt contribution was evaluated using equations from Tables 5-2 and 5-3 of EM 1110-2-1406 (Reference 39), to calculate the melt rate for rain-on-snow and rain-free scenarios. USACE EM 1110-2-1406 has been referenced in NUREG/CR-7046 as one of many most relevant publications for calculating a design basis flood at a nuclear site. EM 1110-2-1406 also states that the Energy Budget method referenced in it is suitable for PMP events over large watersheds.

The melt rate was used throughout the duration of storm events. Both sources of melt were considered during the model run for rain-on-snow and rain free time periods. The Susquehanna River Basin is mostly forested with 60% or more forest cover; therefore, appropriate equations and associated assumptions from EM 1110-2-1406 were used. Basin forest cover was calculated using land-Use Satellite (Landsat) imagery of land use and calculated to be approximately 61%. Snowmelt contribution is calculated with adequate snowpack to melt based on the assumption of ripe snow as described in EM 1110-2-1406 being mostly saturated and ready to melt.

Temperatures across the basin were assumed to be 17 degrees Celsius {62.6°F). Per EM 1110-2-1406, air temperature during the rain event is a critical variable based on the relationship between air temperature and snowmelt (Reference 39). The temperatures during the month of March from 1948 to present for days with measurable precipitation of a minimum of 1/lOOth of an inch, which is the minimum recorded depth by NOAA were obtained. Average maximum daily temperature values were obtained by calculating the mean of the daily maximum temperatures that occurred during days with accumulated precipitation during the month of March for the period of record available. The results of this analysis yielded an average daily maximum temperature of 7.55 *c (45.6 °F). During the 1996 Susquehanna River snowmelt event, temperatures ranged from 10 to 17 *c (Reference 45). When compared to the average daily maximum temperature, the temperature of 17 *c used in this analysis provides a conservative estimate of temperature for the basin. Diurnal fluctuations in temperature during the precipitation event were not considered since assuming zero reduction in temperatures at night produces a more conservative result.

Wind speed (v), during this event, is considered to be equal to the average wind speed for the month of March, 10.5 miles per hour (mph). EM-1110-2-1406 rain-on-snow equations require average wind speed at SO ft (15.24m). Hourly wind speed data was obtained from the NOAA National Climatic Data Center {NCDC) online archive from the Global Surface Hourly database. The data was extracted for five Weather Bureau Army Navy (WBAN) stations that had available wind speed data within the Susquehanna River Basin for the period of 1 January 1948 to the present. The analysis included dates from Feb 15 to April 14 for each available year of record. The recorded anemometer height changes for each WBAN station was obtained over the years of record. This allowed the wind speed to be adjusted to 50 ft (15.24 m) as required in EM THREE MILE ISLAND NUCLEAR POWER STATION Page30of64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl 1110-2-1406 by applying equation 11-2-9 from Chapter 2 of Coastal Engineering Manual 1110-2-1100. EM 1110-2-1406 and EM 1110-2-1100 are referenced in NUREG/CR-7046 as being the most relevant publications by the USACE for estimating design-basis floods.

Snow water equivalent was used to determine the available water to contribute to the precipitation hyetograph as a limiting factor. The Snow Water Equivalent equation, was obtained from Equation 10.1 In Andy Ward's Environmental Hydrology (Reference 41), with the conservat ive assumption that ripe snowpack is at a density of 0.5 g/cm3* Snowmelt calculations considered that the total melt contribution from the available snowpack does not exceed the snow water equivalent of the snowpack. The sum of snow melt was computed and compared with the snow water equivalent for each subbasin for each time step in the simulation. If the sum of the snow melt {Including the calculated melt for the current time step) is less than the total snow water equivalent, the full calculated melt rate from USACE equations was used for the current time step. If the sum of the snow melt (including t he calculated melt for the current time step) Is greater than the snow water equivalent, and the sum of t he snow melt alone jwlthout the calculated melt for the current time s,tep) is less than the snow water equivalent, the melt for the current time st ep was assumed to be the difference between the snow wat er equivalent and the total melt. All subsequent time steps were assumed to have no melt, since the snow water equivalent was assumed to have been exhausted.

The Snow Water Equivalent equation from Ward (Reference 41) yielded an available snow water equivalent depth of 2.45 m over the entire basin. This value was used as a limiting factor when snowmelt rate is applied in mm/ hr in addition to rain or rain free time periods of the hyetograph.

Subbasin C1_W1010 Scenario 2.2: 100-yr Cool Season Precipitation (SCS Distribution) + Probable Maximum Snowpack 50 -r------ - - - - - - - - - - - - - - - - - -

45 +--------+-------------

40 - + - - - -

gE 35 130 0

..c 25 o

-~ 20 ._ _ _ _ _ _ __

0,.

-~ 15 CL.

10 > - - - -

5 +-------__, - - ,'~

0

~-----' .-.-,.,..,., '"-...-.-..~----------

1/1/2010 1/3/2010 1/5/2010 117/2010 1/9/2010 1/11/2010 1113/2010 Slmulatlon Time

- - - - - Precipitation Only - - Snowmell + Precipitation Figure 8: Input Volume Scenario 2.2 THREE MILE ISLAND NUCLEAR POWER STATION Page 31 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 In order to maximize the effects of runoff from snowmelt, the model run time was extended to 13 days.

The effect of timing of the precipitation event on the snow-melt was evaluated by adjusting the beginning of the storm event in the model. Scenario 2.1 was run with the SCS Type II 100-year Snow Season Rainfall st th beginning on the 1 day. Scenario 2.2 was run with the SCS Type 11100-year Snow Season Rainfall on the 4 th day. Scenario 2.2, which had the precipitation event occurring on the 4 day, was found to produce the critical modeling configuration for Scenario 2. The results are shown in Table 10.

Table 10: Alternatives for Scenario 2 Peak Peak Temporal Start oflOO yr Scenario Conditions Dlschar1e Discharge Distribution Cool Season PPT (ems) (cfs)

Probable M aximum Snowpack + 100 yr Cool Precipitation Scenario 2.1 SCS Type II 23,621 834,047 Season PPT starts on 1st Day Probable Maximum Snowpack + 100 yr Cool Precipitation Scenario 2.2 SCS Type II 23,732 837,977 Season PPT starts on 4th Day Hydrographs from the results of Scenario 2 are presented in Figure 9. The Scenario 2 peak discharge is th 837,977 ds. This scenario is a PMSA with a 100-year cool season precipitation event that begins on the 4 day of melt.

900000

- Scenario 2.1 I 800000 A. , ,

J

- - Scenario 2.2 700000 I '

I \

\

\ '

\

600000 I

I I

't 500000 \ I I '\

\

'~ I CII a:

I \

~ 400000 u::: '

300000 '

\

I 200000 J r * ' \

~ ......

I I

~

100000

- _)_

I

~-- _r- ..

~ - ~

0 - -

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 Time (hrs)

Figure 9: Scenario 2 Results

Hydrographs at Three Mile Island THREE MILE ISLAND NUCLEAR POWER STATION Page 32 of 64 RCN : TMl-140

NTTF Recommendation 2.1 {Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Scenario 3 Per NUREG/CR-7046, Appendix H.1 Floods Coused by Precipitotion Events (Reference 6), the Scenario 3 combined-effect precipitation was evaluated to include:

Mean monthly base flow;

A 100-year snowpack; and

Snow-season PMP.

Scenario 3 was evaluated using the calibrated hydrologic model developed In Calculation Package 1101-122-E410*010 (Reference 15) and derived precipitation inputs to reflect the combination of mean monthly base flow, 100-year snowpack, and a snow-season PMP. Scenario 3 was evaluated In Calculation Package 1101-122-E410-011 (Reference 16).

Mean monthly base flow was previously calculated in Calculation Package C-1101-122-E410-010 (Reference 15) by taking the yearly average of the individual average daily stream flow data for each gage obtained from the USGS. Mean monthly base flow was calculated by averaging the mean base flow values annually and dividing them by area to obtain a unit area mean monthly base flow.

Table 11: Mean Monthly Base flow per Unit Area cfs/Square Mile ems/Square KIiometer Annual Annual 1.4584 0.0159 The precipitation depth value for the snow season PMP was determined by obtaining values from the month of March in HMR 33 (Reference 36). HMR 33 is applicable to watersheds up to 1,000 square miles but lt provides Depth-Area-Duration (DAD) curves, which allow for an extrapolation to larger watershed areas available for specific months of the year. HMR 53 Is applicable to 10-square-mfle watersheds and does not provide DAD curves or a similar mechanism to extrapolate for larger watersheds. HMR 33 precipitation values for the 72-hour storm duration were extrapolated to the 25, 960-square-mlle watershed (total watershed area of the hydrologic model) using Depth-Area-Duration curves. The month of March was evaluated since a snowpack could potentially exist when a seasonal probable maximum precipitation could occur. March was chosen for the Cold/Snow Season PMP since It has the warmest temperatures, highest average wind speed, and could have a significant available snowpack for snowmelt available. April was also considered; however, upon review of available meteorological data, it was concluded that adequate snowpack would not be available in April for a significant rain on snow event.

Graphical functions were created depicting Duration versus Percent Precipitation to extrapolate HMR 33 for a 72-hour duration and 25,960-square-mile watershed. The snow season PMP was calculated to be 9.23 Inches. The snow season PMP was distributed spatially using the spatial distribution of HMR 40 for extreme events. Aerially-weighted average precipitation values for each subbasin were used In the HEC-HMS Model to reflect the snow season probable maximum rainfall.

The infiltration loss rates based on calibration to Tropical Storm lee were used for this analysis. As discussed under Scenario 2, the loss rates calibrated to Tropical Storm Lee are appropriate for the snowmelt evaluation. This evaluation was based on a sensitivity analysis of 10% and 12% snow-water-equlvalent, and indicates that the watershed infiltration loss rates during Tropical Storm Lee are similar to the infiltration loss rates during a historic extreme rain-on-snow event (in 1996). Therefore, loss rates THREE MILE ISLAND NUCLEAR POWER STATION Page 33 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 calibrated to Tropical Storm Lee provide an appropriate basis for the rain-on-snow analysis in the Susquehanna River Watershed.

For the 100-year snowpack, data from valid sites were collated such that a Gumbel, GEV, and Log Pearson Type Ill distributions were fitted to seasonal maximums. Dis.tribution types were chosen to meet NUREG/OR-7046 guidance to compute multiple types of distributions (Reference 6). The 100-year return periods were determined by inputting the distribution parameters into the CDF for the particular distribution. The 0.99 CDF value was also used which corresponds to the 1 in 100 year probability. Runoff from snowmelt for the 100-year snowpack reflects the methodology used to melt the PMSA, with the available snowpack varying over the basin for the 100-year snowpack. Snow depths for the 100-year snowpack range from 31.2 in (792.5 mm) to 80.55 in (2046 mm) across the basin.

Snowmelt contribution was evaluated using equations from Tables 5-2 and 5-3 of EM 1110-2-1406 (Reference 39), to calculate the melt rate for rain-on-snow and rain-free scenarios. The melt rate was used throughout the duration of storm events. Both sources of melt were considered during the model run for rain-on-snow and rain free time periods. The Susquehanna River Basin is mostly forested with 60% or more forest cover; therefore, appropriate equations and associated assumptions from EM 1110-2-1406 were used. Basin forest cover was calculated using Landsat imagery of land use and calculated to be approximately 61%. Snowmelt contribution is calculated with adequate snowpack to melt based on the assumption of ripe snow as described in EM 1110-2-1406 being mostly saturated and ready to melt.

Temperatures across the basin were assumed to be 17 *c (62.6 °F). Per EM 1110-2-1406, air temperature during the rain event is a critical variable based on the relationship between Air Temperature and Snowmelt (Reference 39). An analysis of temperatures during the month of March from 1948 to present for days with measurable precipitation of a minimum of 1/lOOth of an inch, which is the minimum recorded depth by NOAA. Average maximum daily temperature values were obtained by calculating the mean of the daily maximum temperatures that occurred during days with accumulated precipitation during the month of March for the period of record available. The results of this analysis yielded an average daily maximum temperature of 7.55 *c (45.6 °F). During the 1996 Susquehanna River snowmelt event (Reference 45) temperatures ranged from 10 to 17 *c. When compared to the average daily maximum temperature, the temperature of 17 *c used in this analysis provides a conservative estimate of temperature for the basin.

Diurnal fluctuations in temperature during the precipitation event were not considered since assuming zero reduction in temperatures at night produces a more conservative result.

Wind speed (v), during this event, is considered to be equal to the average wind speed for the month of March, 10.5 mph. As discussed in Scenario 2, EM-1110-2-1406 rain-on-snow equations require average wind speed at 50 ft (15.24 m). Hourly wind speed data across the basin from 5 sites were used to develop an average wind speed for the month of March, with speeds adjusted to 50 ft (15.24 m). Wind data was used for sites with documented anemomet er heights for the years of record so recorded values could be adjusted per Coastal Engineering Manual 1110-2-1100, Chapter 2, Equation 11-2-9.

Snow water equivalent was used to determine the available water to contribute to the precipitation hyetograph as a limiting factor. The Snow Water Equivalent equation, was obtained from Equation 10.1 in Andy Ward's Environmental Hydrology (Reference 41), with the conservative assumption that ripe 3

snowpack is at a density of 0.5 g/cm

Snowmelt calculations considered that the total melt contribut ion from the available snowpack does not exceed the snow water equivalent of the snowpack. The sum of snow melt was computed and compared with the snow water equivalent for each subbasin for each time step in the simulation. If the sum of the snow melt (including the calculated melt for the current time step) is less than the total snow water equivalent, the full calculated melt rate from USACE equations shown was used for the current time step. If the sum of the snow melt (including the calculated melt for the current THREE MILE ISLAND NUCLEAR POWER STATION Page 34 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl time step) is greater than the snow water equivalent, and the sum of the snow melt alone (without the calculated melt for the current time step) is less than the snow water equivalent, the melt for the current time step was assumed to be the difference between the snow water equivalent and the total melt. All subsequent time steps were assumed t o have no melt, since the snow water equivalent was assumed to have been exhausted.

Scenario 3 combined the 100-year Snowmelt with the snow season PMP extrapolated from HMR 33.

Scenario 3 included the combination of various precipitation distributions such as front-weighted versus center-weighted hyetographs, and precipitation starting on the 1st day or the 4 th day. Scenario 3.5 is analyzed to determine if a tail-weighted precipitation distribution is more critical. The precipitation hyetograph for scenario 3.4 is shown In Figure 10, which was determined to be the critical model scenario.

Subbasin C1_W1010 Scenario 3.4: Cool Season Probable Maximum Precipitation (Center-Weighted Distribution)+ 100-yr Snowpack 14 12 + - - - - - - - - - -- -- - - - - - - - - - - - - - - - -

E° g 10 fj o

ia + - - - - - - - - - -- ~- - - -

C:

0

'w

,.. 6

t

Q.

o

£ 4 2

0 + - - - - - ~ - --L-- ~ - - - - ~ -

1/112010 1/3/2010 1/5/2010 117/2010 1/9/2010 1/11/2010 1/13/2010 Simulatlon Time

Precipitation Only - - Snowmen+ Precipitation Figure 10: Scenario 3.4 Input Volume Results from the Scenario 3 combinations are presented in Table 12. Hydrographs of the alternative Scenario 3 runs are presented in Figure 11. The Scenario 3 peak discharge is 1,481,745 cfs. This scenario is a 100-year snowpack with a snow season PMP event that begins on the 4th day of melt.

THREE MI LE ISLAND NUCLEAR POWER STATION Page 35 of 64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Table 12: Scenario 3 Results Temporal Beginning of March Peak Flow Peak Flow Scenario Conditions Distribution PMP (ems) (cfs)

Scenario 3.1 100 yr Snowpack + March PMP Front-Weighted Precipitation starts on 37,646 1,329,277 1st Day Scenario 3.2 100 yr Snowpack + March PMP Front-Weighted Precipitation starts on 41,714 1,472,911 4th Day Scenario 3.3 100 yr Snowpack + March PMP Center-Weighted Precipitation starts on 41,752 1,474,246 1st Day Scenario 3.4 100 yr Snowpack + March PMP Center-Weighted Precipitation starts on 41,964 1,481,745 4th Day Scenario 3.S 100 yr Snowpack + March PMP Tall-End-Weighted Precipitation starts on 38,694 1,366,282 4th Day 1600000 1500000 , .

,, ',..-*v.. -

- scenario 3.1

- scenario 3.2 1400000 /"\ I, ;

- - Scenario 3.3 I

1300000 ,-I \ - - - Scenario 3.4

......... Scenario 3.5 1200000 I \ I , ' f \

II \ \ ,

I 1100000 1000000 I I \ ', I f\ \

~

Qj 900000 800000

,I j

\ . I

\ I

\

I a:

3 700000

{ \ I \

' \ ,' \

0 600000 I l *., \ \

500000 400000 I I I I I

\ ' \'

\ / \ ,

-' \

\

I I t X \ * '*

300000 I I .. \ *.*.

200000 .I I

' ... \

~

'- ~ .....

~-...-*-

I ***

J J * * ****

I' 100000 0 -

_J :,_...r I "' ""'

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 Time (hrs)

Figure 11: Scenario 3 Results - Hydrographs at Three Mlle Island THREE MILE ISLAND NUCLEAR POWER STATION Page 36 of 64 RCN:TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl Governing PMF Peak Dischar ge The results of the precipitation driven scenarios performed in this calculation package are provided in Table 13 below.

Table 13: Precipitation Driven PMF Scenario Results Scenario Conditions Peak Flow (ds)

Scenario 1 0.4*AII Season PMP + All Season PMP 1,154,810 Scenario 2 Probable Maximum Snowpack + 100-year Snow Season Rainfall 837,977 Scenario 3 100-yr Snowpack + Snow Season PMP 1,481,745 Scenario 3, the 100-yr Snowpack plus Snow Season PMP, Is the governing PMF peak discharge scenario with a peak discharge of 1,481,745 cfs.

Assumed Dam Failures During Precipitati on Event The effect of upstream dam breaches and failures due to a precipitation event was evaluated in accordance with NUREG/CR-7046, Section 3.9 performed in Calculation C-1101-122-E410-011 (Reference 16). The analysis was performed using the calibrated HEC-HMS model developed in Calculation C-1101-122-E410-010 (Reference 15).

(b)(3) 16 U SC § 8240-1 (d) (b)(4), (b)(7)(F)

The dam breach parameters used as input to the HEC-HMS model were based on an empirical formula published in the Journal of Geotechnical and Geoenvironmental Engineering (Reference 38) that followed a US Bureau of Reclamation publication (Reference 33) expressing the need for a new breach model specifically for high hazard dams. The equations are based on widely accepted equations found in Froehlich (Reference 32) with empirical data to close the gap between idealized parameters used in the breach analysis wlth actual recorded breach events. Xu and Zhang (Reference 38) used the 75 failure cases that had sufficient information to develop regression equations, and subdivided breaching parameters into two groups, geometric and hydrographic.

The piping elevation was set to 2/3 the height of the dam with a piping coefficient of 0.7 for all dams (Reference 38)

._ _____________________________ (b)(3) 16 U SC § 8240-1 (d) (b)(4) (b;(7)(F)

T e t mes ep at which each dam will breach will vary since the timing is dependent on when the peak impoundment water surface will occur.

I"" '6 U SC § 8240*1(d) (b)(4,. (b)(7)(~

The dam break evaluation Investigated two different scenarios. The first scenario determined the peak flow at TMI Unit 1 with all upstream dams breached. The second evaluated the Impact that the modeled non-critical dams have on the critical PMF. Table 14 shows the peak flows associated with the critical dam THREE MILE ISLAND NUCLEAR POWER STATION Page 37 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporat ion Rev 1 breach scenario. The peak discharge during the governing PMF event when all dams were assumed to fail is 1,491,618 cfs.

Table 14: Maximum peak flow from the precipitation scenarios Precipitation Scenario Conditions Peak Flow jcms) Peak Flow Ids)

Scenario 3

Composite Dams+ No Scenario 3 41,981 1,482,342 Breach Scenario 3 + Composite Dams+ Alf Scenario 3 42,244 1,491,618 Dams Breach Scenario 3 + Composite Dams+ Alf r )(3) 1ti u s c § tl<'40 I (d)

(b}(3) 16 USC ..

Scenario 3

. . . P~*m t**

IBreach (b)(4) (b)(7)(F) 1

§ 8240 Hd~ (ll) *** **

(4) (b)(7)(F)

The PMF hvdro~raoh at Three Mlle Island Is present ed in Figure 121

(:>)(3) 16 U SC § 8240.1 (d) (b)(4) (tl)(7)(F) r I

Based on the stage-discharge relationship developed in Calculation 1101-122-E410-003 and provided in Section (Fig~ e 18), the estimate peak water surface elevation of the governing PMF scenario with dam (b)(3) 1G USC .Jallure is .......... NGVD-29 at the ISPH.

§ 8240 rrar

,., ",m,r, (h)'

(b)(3) 16 ll SC § 8240 1(d), (b)(4), (b)(7)(F)

Figure 12: Probable Ma>1imum Flood Hydrograph at Three Mile Island THREE MILE ISLAND NUCLEAR POWER STATION Page 38 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl

c. Dam Breaches and Failures Breach or failure of an artificial barrier used to impound water for multiple possible functions, including flood control (attenuation), recreation, water supply, hydroelectric, sediment storage, aquatic habitat, stormwater (quantity/quality) management, or a combination thereof, located within the watershed of an adjacent stream/river or upslope of SSCs important to safety. Onsite cooling or auxiliary water reservoirs and onsite levees should also be Included in the definition of dams. Flood waves resulting from the breach of upstream dams should consider domino-type or cascading dam failures.

Upstream dam breaches and failures were evaluated for a seismic event in accordance with NUREG/CR-7046, Section 3.9 and Appendix H.2 performed in Calculation C-1101-122-E410-012 (Reference 17). The dam failure analysis was performed using the calibrated HEC-HMS model developed in Calculation C-1101-122-E410-010 (Reference 15).

The calibrated HEC-HMS model 1~n;)~,5~~~i) 1 82 Basin. Table 15 shows the top of am inputs or t e model.

0 1

~ - lusACE flood control dams in the Susquehanna River Table 15: Dams Modeled In Calibrated HEC-HMS model ib)(3) 16 USC § 8240 1(d) (b)(4) (b)(/)(F)

Combined Storage Volume of Smal I Dams within the Watershed The USACE National Inventory of Dams Database (NID) reports that there are 126 dams in the Susquehanna River Basin upstream of TMI Unit 1. A screening evaluation (Reference 16) was conducted to identify dams having a more substantial impact at the site if they were to fall. IT~l~~f1,~)~;F~ § 8240 1 I

- 1dl in Tab~

~ ent dams having enough impact of failure at the site to be mo~ele[ in73,vT<3ualfy. fhe remaining .... ~~~~,~~)s(~i (b)(3) 16 U SC -L:::.._J with storage volume ranging from 6.5 ac-ft to 345,320 ac-ft, were considered small and re 4). (b)(7)(F)

~}2i~.}i~:(b) enough to be modeled collectively. The height and storage volume of the (b)(3) 16 were obtained from the NID Database. To account for the combined storage volume associated wit sma I upstream dams in the hydrologic dam breach flow rate, composite reservoirs representative of the combined storage volume were inserted into the HEC-HMS model (Reference 3). This methodology provides a conservative estimate of the hydrologlc dam breach flow rate because the attenuation of each individual dam breach hydrograph between reservoirs is not modeled. The composite dams were developed using the following assumptions:

THREE MILE ISLAND NUCLEAR POWER STATION Page 39 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1

The dam height of each composite dam was calculated based on a weighted average of individual dam heights and storage volumes.

The stage-storage relationship for each composite dam is a simplified linear relationship.

Breach parameter and time is based on of the methodology currently discussed in the sections below.

Starting water surface elevation for composite dams is assumed to be top of dam.

Composite dams are assumed to have no auxiliary spillway.

There Is no specified release from the composite dams prior to breaching.

The composite dams are located at the downstream end of the subbasin of the associated grouping.

(6)(3} l6 0 s C § 824o-1(d\ I 1(t,)( t) (h)(7)(FJ iwere developed based on he assumptions described above and inserted Into the (b)(3) 16 USC .J:IEC:.HMSmodeL.Anaddltionali.ndividualdam ***** Dam, was also added to the HEC-HMS model.

f;RgA11d~k~)

K 2

Jo'l\ffFttil . . .~iSllEE!J~ ~hqw~.Jhe.Jo.cation of..the _ _ _ _ ____, Individually modeled dams In the HEC-HMS model, (l) (tJ){?)(I l the location of the small upstream dams that were combined to form the composite dams, and the locatlon of the composite dams. This configuration of dams in the watershed was used in the evaluation of dam failure due to precipitation driven and seismic events.

b){3) 16 US L § t!<'40-1(d) (b)(4) (D)(l)( Fl Figure 13: Composite and Individual Dam Locations THREE MILE ISLAND NUCLEAR POWER STATION Page 40 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations}: Flooding March 11, 2013 Exelon Corporation Rev 1 Table 16: Weighted height, length and storage volumes for the Individual and composite dams added to estimate the hydrologlc dam break peak flow rate.

HEC-HMS Number of Weighted Weighted Length Length Total Storage Total Storage Dam ID Individual Dams Height (ft) Helght (m) (ft) (ml (ac-ft) (1000 m*)

OTSL 1 11 3,35 70 21 34 46475 57326 07 C1-1 4 34 04 10.38 324 98 78 8098 9988 74 C1-2 29 15.04 4.58 17140 52 26 9286.50 11454.73 C2 22 84. 12 19.55 82200 250 61 33271 .53 41039.83 C3 17 17 78 5 42 43000 131 .10 4286 00 5286.70 C4 24 14 74 4 49 22900 69 82 7084 40 8738,48 cs 13 9.58 2.92 2842.26 866.54 45879,30 56591.28 C6 3 55.82 17.02 1258,00 383.54 25499.00 31452.55 Seismically-Induced Dam Failure NUREG/CR-7046 (Appendix H.2) and ANS-2.8 (Section 9.2.1.2} provide the following two (2) alternative combinat,ions for seismically-induced dam failure:

1. 25-year flood, dam failure caused by the safe shutdown earthquake (SSE} coincident with the peak flood, and 2-year wind speed applied in the critical direction.

2. 1/2 PMF or 500-year flood, whichever is less, dam failure caused by the operating basis earthquake (QBE) coincident with the peak flood, and 2-year wind speed applied in the critical direction.

The best available interpretation of the above requirements is that seismically-induced dam failure occurs coincident with the peak pool levels for the 25-year flood, 500-year flood, or 1/2 PMF entering the dam; implying that runoff downstream of the dam to the site is not included.

This approach is reasonable for a single upstream dam but seems Inappropriate for multiple dams. It would not be reasonable to assume that peak flood pool levels for all upstream dams occur coincident with an earthquake. Developing floods for multiple upstream dams seemed best accomplished by using watershed-wide precipitation events, not upstream flood-flow, to represent the above combinations. Therefore, TMI Unit l's seismically-induced dam failure analysis is based on the following precipitation-based combinations:

1. 25-year precipitation throughout the site's watershed, dam failure caused by SSE, and 2-year wind speed applied in the crit ical direction.

2. 1/2 PMP or 500-year precipitation, whichever is less, throughout the site's watershed, dam failure caused by the QBE, and 2-year wind speed applied in the critical direction.

This would be consistent with NRC expectations that, as stated in NUREG/CR-7046 and ANS-2.8, combinations are thought to have a probability-of-exceedance of less than 1 x 10'6* Also, using watershed-wide precipitation events would be conservative because it would include runoff downstream of the dams to the site. Since all dams are assumed to fail during the lower-magnitude (QBE} earthquake, only combination #2 was included in the analysis.

Precipitation depths for each basin for the 1/2 PMP events were determined using the spatial distribution of HMR 40. To determine the 1/2 PMP, the rainfall values derived in Calculation C-1101-122-E410-011 (Reference 16) for the all season PMP were divided in half. The 500-year precipitation was derived as the antecedent rainfall in Alternative 1 from Calculation C-1101-122-E410-011 (Reference 16). When the total precipitation depths are compared the1/2 PMP is the lesser event.

THREE MILE ISLAND NUCLEAR POWER STATION Page 41 of 64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Note that all dams are assumed to fail when the earthquake occurs (the timing of which is established based on optimal impact to the site), which may not result in a particular dam failing at its peak water level.

Dam breach parameters including breach time formation, side slope factor, top width, bottom width, average width, and average breach depth were estimated using dam dimensions provided by the USACE and the approach described in Xu and Zhang (Reference 38). The piping elevation was set to 2/3 the height of the dam with a piping coefficient of 0.7 for all dams (Reference 38). Dam breach inputs were entered in the HEC-HMS model for each dam and the reservoirs were set to breach simultaneously to reflect a potential seismic event scenario. Breach times, beginning with the earliest peak inflow to a reservoir and ending once a maximum flow was established, were evaluated.

No outflow from the reservoirs was assumed during the precipitation event to maximize the storage behind the dam and produce conservative peak flows downstream during the dam breach. All of the modeled reservoirs have a primary function for flood control and are expected to pass some flow during the precipitation event.

In order to determine the peak flow based on a seismic dam failure, seven different breach times were evaluated to determine the ultimate peak flow rate associated with a 50% PMP and a seismic dam failure.

Multiple model dam breach scenario runs were performed with all dams failing simultaneously at various (b)(3) 16 USC timesteps.-from ~ hrs after the start of the rainfall. Table 17 shows the results of these multiple

~ ll?4n 1(rl) CTr I\ /Lllf\lf'I model dam brea~ los.

Table 17: Results of the seismically-Induced dam failure scenarios THREE MILE ISLAND NUCLEAR POWER STATION Page 42 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 (b)(3) 16 USC § 8240-1 (d) (b)(4) (b)(7)(f-)

Figure 14: Peak flow results at Three Mlle Island from the dam breach analysis The results of this analysis reveal the most significant contribution to Impacts at TMI Unit 1 comes from the

~~~~.~~.~~~. J.. .

l and the addition of com oslte dam C-6. In HEC-HMS, a conservative approach was

§'s~t~!ft&f;(~ . . . . . . . t.ak en..to modeLcom.poslte damC-6and Dam in series. Composite dam C-6 includes three dams that are 34 to 56 feet high. (b)(3) 16 us c § 824o-1(d) (b)(4). (b)(7)(F)

( '.l1fl t JS,§ 4o-1(,(h) ,() Thepeak outflow from the breach of com osite dam C-6 is 113 000 cfs which is directed into * * .. Oam.Jb)(3) 16 USC Since C-6 and (b)(3) 16 U SC § 8240 - 1(d), (b )( 4), (b)(7 )(F) ' ~ime with no reach modeled in between the§ 8240:11(d), (b)

~~~~~~~}!(~ . . . .to.talbreachoutflow from*~==:: _ _,..Dam increases to approximately I * !cfs.Thisp.ea.kJl~w is(b)(3)16 ~S C attenuated by the time it mee s e con uence with the Susquehanna River to approximately 596,000 cfs. § 8240-1 l,I\

(d),

IL.\/7\/r\

(b)

An additional model run was performed using the same configuration and with all dams (including (b)(3) 16 us c corr,p(,)~lt~clil,:n~J,withtheexceptionofl ........... ~ reaching. The peak flows resulting from this

§ 8240-1{d}, (b ) *

4 7

( ) (bJ( )(r) ~v~~~;~:da;~e~h~~~ t~: ~~~s~~~:i~~i:;s!~e:~u~;e;~~- ~~c~:P:~t~e~:~n2-6. .. ** * !D.amw~~ Qpt~b~~1~:~~):(~

IA \ ll\11\lr""\

Based ao tbe results from the dam breach scenarios the eak flow associated with . the breaching of (b)(3)* 16 USC

§ 8240 1(d):'(6j I-

Dam accounts for (b)(3) 16 USC § 824o-1(d), (b)(4). (b)(7)(F) s, while the peak flows

,., ,u,-,"r, from breaching of the remaining dams accoun or ,

(b)(3) 16 USC lt Based on the stage-discharge relationship developed in Calculation 1101-122-E410-003 and provided in Section 4.i (Figure the r timated peak water surface elevation at TMI Unit 1 ISPH as a result of the

~~i~mi!: damJallure.i . ..... . .. ft NGVD-29. Also, the flood warning time for seismic dam failure is treated as

§ 8240-l(d), (b) 1-'\ ILIM\lr\

THREE MILE ISLAND NUCLEAR POWER STATION Page 43 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 similar to sunny-day failure. The TMI dike was designed for a discharge of 1, 100,000 cfs including wind driven waves and, therefore, the current licensing basis bounds t he reevaluated flood hazard. No further analysis was performed to determine the combined effect of wind driven waves on top of the seismic dam failure peak water surface elevation. However, using a conservative estimated maximum wave amplitude of 2.5 feet (4.9 ft maximum wave height) on top of the PMF from Calculation C-1102-122-E410-014, the est imated combined effects water surface elevation due to seismic dam failure and waves generated by the (bl\3) 16 USC 2 ::y eacwind speed in the-Gritic-al direction is Dtt NGVD-29 at t he ISPH. This elevation is less than the

§8?.du f(<lrM minimum height of t he dike (304 ft NGVD-29 at the south end) and bounded by the current license basis hazard.

d. Storm Surge St orm surge is the r ise of offshore water elevation caused principally by the shear force of t he hurricane or t ropical depression winds acting on the water surface. Storm surge is not considered In the current design basis for TMI Unit 1. The head-of-tide on the Susquehanna River is approximately 3 miles downstream of the Conowingo Dam (Reference 10), which is located approximately SO miles downstream of TMI Unit 1 (Reference 13). Since TMI Unit l is located approximately 53 miles upstream from the tidally influenced reach of the Susquehanna River, storm surge was not considered an applicable flood causing mechanism for evaluation at TMI Unit 1.

e. Seicbe Seiche is an oscillation of the water surface in an enclosed or semi-enclosed water body Initiated by an external cause. Seiche is not considered in the current design basis for TMI Unit 1. The Susquehanna River is not an enclosed or semi-enclosed water body and, therefore, seiche was not considered an applicable flood causing mechanism for evaluation at TMI Unit 1.

f. Tsunami Tsunami is a series of water waves generated by a rapid, large scale disturbance of a water body due to seismic, landslide or volcanic tsunamigenic sources. Tsunami is not considered in the current design basis for TMI Unit l. TMI Unit 1 is approximately 59 miles upstream from the mouth of the Susquehanna River at the Chesapeake Bay (References 10 and 13). The downstream dams, elevation difference, and topography of the Susquehanna River would likely attenuate a tsunami wave generated in the Chesapeake Bay.

Therefore, tsunami was not considered an applicable flood causing mechanism to be evaluated for TMI Unit l.

g. Ice Induced Flooding Ice jams and ice dams can cause flooding by impounding water upstream of a site and subsequently collapsing or downstream of a site impounding and backing up water. There is no method to assess a probable maximum ice jam or ice dam. Therefore, historical records are generally used to determine the most severe historical event in the vicinity of the site.

The Ice Induced Flooding evaluation was performed per NUREG/ CR-7046 (Section 3.7). A historic review was performed to identify the most severe historical ice jam event in the vicinity of t he site. NUREG/CR-7046 states: "In the HHA framework, it may be possible to determine whether a flood caused by another flood-causing mechanism at or near the site may exceed that resulting from an ice event. If such an alternative and bounding flood is found, no further analysis for the ice-induced flood is necessary."

(Reference 6)

THREE MILE ISLAN D NUCLEAR POWER STATION Page 44 of 64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl Flooding at TMI Unit 1 caused by an ice jam event is considered similar to flooding caused by an instantaneous dam failure in that these floods would occur with little to no warning at the site. Therefore, the peak flow from the most severe ice jam event and the dam failure caused by a seismic event peak discharge (Section 4.e) were compared to determine if the ice jam event is a bounding event. If the dam failure caused by seismic event peak discharge exceeds the most severe ice jam event, or there are no historical ice jam events identified, then no further analysis for the ice-induced flood will be required since it would be screened out of being a bounding event.

Date, location, and description of ices jam events were obtained from the Ice Jam Database provided by the USACE Ice Engineering Group at Cold Regions Research and Engineering Laboratory (CRREL). Stream gage data was obtained from gage stations at Harrisburg (USGS Gage 01570500) and at Marietta (USGS Gage 01S76000) using the USGS National Water Information System (NWIS) Web Interface.

Twenty-eight (28) ice jam events were identified upstream at the Harrisburg gage station between 1784 and 2009, and six (6) ice jam events were identified downstream at the Marietta gage station between 1934 and 2005. The Marietta gage station is located approximately 13.7 miles downstream of the site and therefore these events were not identified as causing backwater flooding at the site. The ice jam events that occurred prior to 1890 are not included in the full evaluation due to the lack of corresponding gage information. A maximum average daily discharge was determined for each ice jam event identified in the CRREL database by reviewing the USGS gage records five days prior to and five days following the ice j am event.

Once the three largest ice jam event discharges were identified, the USGS gage records for 30 days leading up the ice jam event and 30 days following the ice jam event were evaluated to determine the average daily maximum discharge of each event. The peak flow for the most severe ice jam event was compared to the dam failure peak discharge calculated in C-1101-122- E410-012 (Reference 17) to determine if the ice jam required further modeling analysis.

The ice jam events for the three (3) largest maximum average daily discharges are presented in Table 18.

Figure 15 provides the corresponding stream gage discharges for the January 1996 ice jam, which was determined to be the most severe event.

Table 18: Three most Severe Ice Jam events Marietta Harrisburg Date Gage Gage (CFS)

(CFS)

January 20, 1996 518,000 556,000 March 8, 1936 424,000 457,000 March 3, 1902 449,000 N/A To further investigate the discharge and gage height associated with the January 1996 ice jam, gage data from the USGS Instantaneous Data Archive for the Harrisburg and Marietta gages was reviewed. For the above-mentioned gages, the archive provides flow rates at intervals of 5 to 60 minutes. The data from the 1 th archive was retrieved for a period from December 19 \ 1995 to February 20 , 1996, as shown in Figure 15.

The average daily discharge was used for dates for which discharge data were not provided in the instantaneous report.

THREE MILE ISLAND NUCLEAR POWER STATION Page45 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Table 19: Peak Flows and gage heights for the January, 1996 Ice Jam Harrisburg Marietta Peak Flow (CFS) 5*67,000 588,000 Rating Curve Gage Height (ft) 24.64 56.53 Annual Water Report Gage Height (ft) 25.08 600,000

- Harrisburg - Marietta 500,000 400,000 i

QI

.,,~ 300,000

.c u

Q 200,000 100,000

.... .... .... .... .... N N N

N

\IJ

N N

.... ....... .g 0 N

10 10 10 10 10 10 10 10 10 10 10 10 10 10 V,

°' °' °' °' °' °'

Date Figure 15: 60-Day Discharge, January 1996 Ice Jam th th The most severe historic ice jam event occurred between January 19 and 20 , 1996 with a peak discharge of 567,000 cfs at the Harrisburg gage station and 588,000 cfs at the Marietta gage station.

NUREG/CR-7046 states that "In the HHA framework, it may be possible to determine whether a flood caused by another flood-causing mechanism at or near the site may exceed that resulting from an ice event.

If such an alternative and bounding flood is found, no further analysis for the ice-induced flood is necessary." (Reference 6) Flooding at the TMI Unit 1 caused by an ice jam event is similar to flooding caused by an instantaneous dam failure in that these floods would occur with little to no warning at the site.

Therefore, the peak flow from the most severe ice jam event and the dam failure peak discharge calculated in C-1101-122-E410-012 (Reference 17) were compared to determine If the ice jam event is a bounding event.

THREE MILE ISLAND NUCLEAR POWER STATION Page 46 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 The comparison in Table 20 shows that the dam failure peak discharge is greater than the ice jam peak flow. Therefore, no further evaluation of ice jam flooding will be required for the NUREG/CR-7046 flood hazard evaluation at TMI Unit 1.

Table 20: Peak Flow Comparison Peak Flow (cfs)

Most Severe Ice Jam Event I 588,000 cfs Dam Failure Peak Discharge I 778,026 cfs 1

1 From Calculation C-1101-122-E410-012 (Reference 17)

Per the UFSAR, TMI Unit 1 dike is designed to provide protection for a flow of 1,100,000 cfs (Reference 13).

Based on this evaluation, the dike has significant margin and will protect TMI Unit 1 from the most severe ice jam event peak flow of 588,000 cfs.

Based on the stage-discharge relationship developed in Calculation 1101-122-E410-003 and provided in Section 4.i Figure 18, the estimate peak water surface elevation at the TMI Unit 1 ISPH from the historic most severe ice jam event would be 292.0 ft NGVD-29.

This evaluation indicates that water level increase due to this hazard would not approach site grade or thereby challenge plant safety. Therefore, this hazard is considered bounded by current design.

h. Channel Migration or Diversion Flood hazard associated with channel diversion is due to the possible migration either toward the site or away from it. For natural channels adjacent t o the site, historical and geomorphic processes should be reviewed for possible tendency to meander. For man-made channels, canals or diversions used for the conveyance of water located at a site, possible failure of these structures should be considered.

As indicated in NUREG/CR-7046 Section 3.8, there are no well-established predictive models for channel diversions and, therefore, it is not possible to postulate a probable maximum channel diversion event (Reference 6). As part of the TMI Unit 1 analysis, historical records and available studies were reviewed to evaluate whether the adjacent Susquehanna River has exhibited the tendency to meander towards the site.

Available sources including the TMI Unit 1 UFSAR, current and historical topographic maps, and geologic information were reviewed.

TMI Unit 1 UFSAR Section 2.7.3.3 Depositional History of Three Mile Island (Reference 13) states:

"Three Mile Island, along with Shelley Island and other small islands in the vicinity, were formed as a result of fluvial deposition by the Susquehanna River. The carrying capacity of the river sharply decreases as the channel width increases after the stream crosses on east-west trending, very resistant diabase dike just downstream from Middletown. At this point the channel hos been incised into the more easily erodoble Gettysburg shale, resulting in a conspicuously wider channel. In addition, a change of flow direction occurs ofter the stream cuts through the diabase dike which crosses the river at that point (represented by Hill Island), and is directed along N l0°E striking vertical joints until it is again restricted and deflected by a second diabase dike just south of Three Mile Island.

Boulders carried by the glacial meltwater or transported downstream by ice rafts were first deposited in this wide-channel, low-velocity section of the river and became the nuclei for subsequent deposition of smaller materials. This gradual accretion of river sediment resulted in the growth of most of the islands in this area.

THREE MILE ISLAND NUCLEAR POWER STATION Page 47 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl Three Mile Island is made up of two such nuclei which eventually merged. This area between the two nuclei is presently represented by fine-groined deposits which were deposited in a low energy fluviol environment between the two growing depositional islands."

In addition, the TMI Unit 1 UFSAR suggests minimal changes in channel geometry of the heat sink pool adjacent to the site: "Hydrographic surveys of the pool performed from 1974 to 1993 have shown no substantial changes in the bathymetry of the pool." (Reference 13)

Historic and current USGS Topographic maps for the Middletown, PA Quadrangle (Quad) spanning 1908 to 2010 were reviewed to identify changes in configuration of the Susquehanna River that have occurred over this period. Historic maps bounding the 1936 and 1972 floods were reviewed to identify changes that may have occurred due to extreme flooding events.

The March 1936 flood was caused by a rain-on-snow event over the entire basin, with a gauged peak flow at Harrisburg of 740,000 cfs (Reference 53). This flood was established as the flood of record and exceeded the previous 1889 flood of record (gauged peak flow of 699,000 cfs) (Reference 53). Quad maps were available reflecting the 1908 and 1943 configuration of the Susquehanna River and were reviewed to evaluate potential changes following the 1936 flood. This review revealed sediment accumulation occurred near the northern tip of Three Mile Island, as well as near the smaller intermittent islands between Shelly Island and Three Mile Island. In addition, an area of sedimentation is shown along the eastern channel by Three Mile Island, just downstream of the bridge. Both maps show the York Haven Dam but no development on the island at t he time. The 1943 Quad map shows the construction of the East Channel York Haven Dam and an area of sediment accumulation in the east channel downstream of the dam. The width of Three Mile Island in the vicinity of the Each Channel York Haven Dam appears to have narrowed slightly since the 1908 mapping but the site's overall width remains unchanged. Overall, there is very minimal if any visible channel migration seen in this comparison. A comparison of the 1908 and 1943 USGS Quad maps are provided in Figure 16 below.

The June 1972 flood was caused by Tropical Storm Agnes with a gauged peak flow at Harrisburg of 1,020,000 cfs (Reference 13). This flood exceeded the previous 1936 flood of record. Quad maps were available reflecting the 1963 and 1999 configuration of the Susquehanna River and were reviewed to evaluate potential changes following the 1972 flood. Based on the Quad maps, t he overall channel configuration of the Susquehanna River adjacent to Three Mile Island appears mostly unchanged between the 1963 and 1999. In addition to showing conditions before and after the 1972 flood, this comparison also shows the construction of TMI Unit 1 and the northern and southern access bridges to the island. The width of t he island downstream of the East Channel York Haven Dam appears to be mostly unchanged. The northern portion of the island appears to be unchanged between the 1963 and 1999 Quad maps. A comparison of the 1963 and 1999 USGS Quad maps are provided in Figure 17 below.

THREE MILE ISLAND NUCLEAR POWER STATION Page48 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 M iddletown, PA 1908 (Reference 54) M iddletown, PA 1943 (Reference 55)

Figure 16: Comparison of 1908 and 1943 Middletown, PA USGS Quadrangles

=:;. ,:*;:;;

\.l\l t l =-~ -~

. '4 .

Middletown, PA 1963 (Reference 56) Middletown, PA 1999 (Reference 57)

Figure 17: Comparison of 1963 and 1999 Middletown, PA USGS Quadrangles THREE MILE ISLAND NUCLEAR POWER STATION Page 49 of 64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 The 2010 USGS Quad map (Reference 58) was also reviewed and compared to the previous historic maps.

Overall, the comparison of the previous historic maps have revealed very minimal changes to the Susquehanna River configuration adjacent to TMI Unit 1 even with the occurrence of two floods of record and construction of a dam and two bridges adjacent to the site. While areas of sedimentation and aggregation have occurred adjacent to the site, It appears the sediment transport processes have not caused any instances of channel migration or diversion. Based on this evaluation of available data, the Susquehanna River has not exhibited the tendency to meander towards the site, and "channel migration or diversion" is not a significant contributory factor for determining the flood hazard at TMI Unit 1.

This evaluation indicates that water level Increase due to this hazard would not approach site grade or thereby challenge plant safety. Therefore, this hazard is considered bounded by current design.

I. Combined Effect Flood For flood hazard associated with combined events, ANS 2.8 (Reference 9) provides guidance for combination of flood causing mechanisms for flood hazard at nuclear power reactor sites. In addition to those listed in the ANS guidance, other plausible combined events should be considered on a site specific basis and should be based on the Impacts of other flood causing mechanisms and the location of the site.

The combined effect flood was evaluated for the critical combination of flood causing mechanisms, which was determined to be Scenario 3 of NUREG/CR-7046, Appendix H.1 Floods Caused by Precipitation Events (Reference 6). This scenario includes a combination of mean monthly base flow; a 100-year snowpack; snow-season PMP; and wind generated waves caused by the 2-year wind speed in the critical direction. This scenario results In a critical peak discharge of 1.481,745 cfs. A dam break analysis during the critical PMF was evaluated to determine the governing PMF peak discharge of 1,491,618 cfs. A hydraulic analysis using the governing peak discharge was performed to determine the combined-effect peak water surface elevatlons at TMI Unit 1. A wind generated wave analysis of the 2-year wind speed In the critical direction on top of the governing PMF water surface elevations was performed to determine the resultant water surface elevations and hydrodynamics during the combined effects flood.

Govcrnln~ PMF Hydraulic Analysic; A hydraulic analysis was performed to determine the relationship between the Susquehanna River flow rates and the water surface elevation at TMI Unit 1 (I.e., st age-discharge analysis) to determine the water surface elevation and the flood duration associated with the governing PMF (also referred to as the NUREG/CR-7046 PMF). This evaluation was performed with the hydraullc parameters, Inputs, and assumptions recommended in NUREG/CR-7046. Hydraulic parameters, inputs, and assumptions based on Federal regulatory guidance from other agencies (i.e., NRCS, USGS, and USACE), previous studies, and engineering judgment were used to develop parameters where NUREG/CR-7046 did not provide guidance.

The methodology in this calculation package is consistent with the following guidance documents unless otherwise noted.

NRC Office of Standards Development, Regulatory Guide: RG 1.59 - Design Basis Floods for Nuclear Power Plants, Revision 2, dated August 1977.

American National Standard for Determining Design Basis Flooding at Power Reactor Sites (ANSI/ANS 2.8-1992).

NUREG/CR-7046 "Design-Basis Flood Estimation for Site Characterization at Nuclear Power Plants ln the United States of America", publication date November 2011.

THREE MILE ISLAND NUCLEAR POWER STATION Page 50 of 64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl The governing PMF is the maximum estimated flow rate calculated in accordance with NUREG/CR-7046 Appendix H.1, Floods Caused by Precipitation Events, and is presented in Calculation C-1101-122-ElO-Oll.

Per NUREG/CR-7046 section 3.3.1, Estimating the PMF and Its Effects, the hydraulic parameters of the PMF should be obtained from a hydraulic model. Hydraulic models use the PMF hydrographs estimated by hydrological models at key locations to estimate the PMF flow velocities and depths. AMEC utilized two hydraulic models, HEC-RAS and RiverFL0-20 in estimating the stage-discharge relationship, and consequently the water surface elevation associated with the dike design flow rate, NUREG/CR-7046 PMF flow rate, and the Original Licensing Basis flow rate.

Topographic information for the hydraulic models was obtained from the PA OCNR liDAR data for Dauphin, York, and Lancaster Counties, additional liDAR survey performed by Photo Science for the TMI Unit 1 site in 2012, and supplemental ground survey performed by AMEC in 2011 and 2012. AMEC validated the liDAR data for use on this project through AMEC's commercial grade dedication process.

Bathymetric data along the Susquehanna River around TMI Unit 1 upstream of the York Haven dams, developed by Ocean Surveys in April 2005 and provided by Exelon, was combined with the liDAR data.

Additional survey data for the top of dams and t he river bottoms, representing the East and Main Channels below each dam (the "Conewago Rapids" ), were incorporated into the topography from the 1999 York Haven Hydrostation Fish Passage Project. For all other cross sections upstream and downstream of the site, bathymetric data was obtained from FEMA's effective (1978) HEC-2 hydraulic model. All elevations from the 1999 Fish Passage report and the HEC-2 model were adjusted by -0.8 feet to convert from NGVD 1929 to NAVO 1988. The river bottom was interpolated using ArcGIS between the downstream-most Ocean Surveys cross section and the York Haven headrace entrance in the 1999 York Haven Fish Passage Report.

Additional parameters (such as Manning's n-values, ineffective flow areas, and blocked obstructions) were added based on review of PAMAP orthophotographs for Dauphin, Lancaster, and York Counties as well as a site visit conducted by AMEC in July 2011.

Structures, including the East Channel York Haven Dam, West Channel York Haven Dam, Shocks Mill Railway Bridge, upstream TMI Unit 1 access bridge, and downstream TMI Unit 1 access bridge, were added to the HEC-RAS model. The bridge geometry for the Shocks M ill Railway Bridge was obtained from the 1904 (original bridge construction) through 1972 (reconstruction of center span after Tropical Storm Agnes) as-built plans as provided by Norfolk Southern. The bridge geometry for the upstream TMI Unit 1 access bridge was interpreted from plans attached to the 1990 Gilbert/Commonwealth Study. The downstream TMI Unit 1 access bridge geometry was estimated using aerial photography and using the upstream TMI Unit 1 access bridge as a template for parapet heights and support pier widths. The geometries of West and East Channel York Haven dams were estimated using the 1999 Fish Passage report, adjusting for the datum conversion. In addition, the geometries of these dams were qualitatively verified through field observation, although no survey was conducted. The dams provide a low head drop in the channel and are overtopped during high flow events.

A stage-discharge rating curve for the Mariet ta gage station was obtained from the USGS and was used to establish the downst ream boundary conditions of the HEC-RAS model.

The results of the governing PMF hydraulic analysis are presented in Table 21. The river stage discharge relationship at the TMI Unit 1 Intake Screen and Pump House is shown graphically in Figure 18 below.

THREE MILE ISLAND NUCLEAR POWER STATION Page 51 of64 RCN: TMl-140

NTTF Recommendation 2.1 {Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 320.0 i 315.0 l~ 310.0

..s

~ 305.0 E

D..

-u 300.0 C

C

.)i 295.0

.¥

~ 290.0 C

0 i~ 285.0 w

l ::,

Ill 2ao.o oi J 275.0 Surface Elevation at ISPH lC Expected RiverFLO-2D WSE at the ISPH 270.0 0 2 4 6 8 10 12 16 18 20 Discharge Rate (cfs) x 100000 NUREG /CR 7046 PMF=l,491,618 cfs Figure 18: Flood Stage versus Discharge Curve at Intake Screen and Pump House Table 21: Results :summary from the NU REG/CR 7046 Still Water Analysis WSE at North End ofTMI WSE at ISPH Unit 1 Ft NGVD-29 Ft NGVD-29 Event Flow (CFS) (NAV0-881 (NAVD-881 NUREG/CR-7046 Probable 1,491,618 312.6 [311.8) 311.0 (310.2)

Maximum Flood Figure 19 shows the result of the NUREG/CR-7046 PMF water surface elevation versus time hydrograph.

Based on the results presented in Figure 19, the water surface elevation is estimated to rise at an average rate of 0.4 ft/hr with a maximum estimated 1-hour rate of 1.1 ft/ hour. The water surface elevation will be above the dike for approximately 37 hours4.282407e-4 days 0.0103 hours 6.117725e-5 weeks 1.40785e-5 months . Hydrodynamics of the governing PMF were calculated as part of the evaluation of w ind-generated waves to capture the combined hydrodynamic effect s.

THREE MILE ISLAND NUCLEAR POWER STATION Page 52 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 0.000 315 fi 0.100 - 1, 313

§. 0.200 311 l ... I

-- - -;r - - i * \ - - - - - .. - .

..c

o. 0.300 ~ .

0 GI

.~ ~

309 g 0.400 307

-~ 0.500 ' \ 305

'\

C.

'ij

£ 0.600 303 GI Ill j 0 0.700 .I. 301 :c I *- 299 C.

E Cl.

\, 297 "0

C nl C

Scenario 3.4 Precipitation (March 295 ..

GI GI

,)!

PMP + Snowmelt) (IN) J \ / 293 GI -

V ~~

- - Intake Screen and Pump I \ .. 0 291 ..!:

GI

<(

House Dike Elevation (ft l £Z NAVD-88) -* J ,_,.... 289 .. .t:

I- *- 287 nl -

C 0

- WSEL ft NAVD-88 I \ 1GI 285 iii I

GI u

283 ~

I \.. ::,

j 281 279 VI 3

GI nl

~

277 275 0 20 40 60 80 100 120 140 160 180 200 220 240 260 Time from Begining of Rainfall (hrs)

Figure 1:9: NUREG/CR 7046 Analysis PMF Water Surface Elevation vs. Time Hydrograph at the Intake Screen and Pump House Wind-Generated Waves An evaluation was performed to determine the height of waves induced by the 2-year wind speed applied along the critical direction on top of the governing PMF water surface elevation and evaluate effects of wave propagation for the site. This evaluation calculated the maximum water surface elevation at flood barriers for TMI Unit 1 safety related structures, and maximum flooding depth, maximum velocity, and THREE MILE ISLAND NUCLEAR POWER STATION Page 53 of 64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl maximum hydrodynamic pressure at TMI Unit 1 safety related structures. The analysis was performed in Calculation C-1101-122-E410-014 (Reference 19).

NUREG/CR-7046 requires that the waves generated by the 2-year wind speed be applied along the critical direction on top of the PMF water surface elevation (Reference 6). The water surface elevation caused by the governing PMF peak discharge was determined in C-1101-122-E410-003 (Reference 14).

ANSI/ANS-2.8-1992, Section 7, (Reference 9) addresses the issue of surges, seiches, and wave flooding.

Section 7.4 references the use of the USACE Shore Protection Manual (1984) for conducting wind-generated wave analyses. However, according to the US Army Engineer Research and Development Center (ERDC) website, the Shore Protection Manual has been superseded by the USACE Coastal Engineering Manual (EM) 1110-2-1100, dated August 2008 (Reference 59). For a riverine system such as the Susquehanna River, wind-generated wave heights were computed using the simplified wave prediction method in EM 1110-2-1100 (Reference 59). The maximum wave height was determined in accordance with section 7.4.3 of ANSI/ANS-2.8-1992 (Reference 9).

The 30 ft Above Ground, 2-year Recurrence Interval wind speed of SO mph was obtained from Figure 1 of ANSI/ANS-2.8-1992 (Reference 9). Section H.1 of NUREG/CR-7046 specifies that waves induced by a 2-year wind speed along the critical direction be applied to the governing PMF stillwater elevation to estimate the highest flood elevation at the site. Since NUREG/CR-7046 does not specify the wind speeds, ANSI/ANS-2.8-1992 was used as a reference for the 30 ft Above Ground, 2-year Recurrence Interval Wind Speed for the United States.

The critical fetch length was estimated to be 4.94 km (3.08 mi) based on a PMF inundation map developed reflecting the NUREG/CR-7046 PMF analysis floodplain. The critical fetch length is the longest straight line distance estimated from the safety related buildings to the edge of the NUREG/CR-7046 PMF water surface contour. Fetch lengths to safety related structures from various directions were evaluated to determine the critical fetch length. The fetch length (4.94 km) was determined from the longest possible open water path to the Intake Screen and Pump House. The path is based on an open water path to the north of TMI Unit 1.

The maximum open water fetch length to the west is 2.13 km. In the analytical model, the fetch length of 4.94 km was applied in the most critical direction towards the site, from the west, to provide a conservative result. The maximum wave height in the critical fetch direction before the wave is propagated through the site was estimated to be 1.5 m or 4.9 ft .

The propagation of the waves through the site was modeled using the FLO-2D hydrodynamic modeling software. The FLO-2D model was used to reflect the wave dissipation effects from the flood protection dike, which is submerged during the governing PMF.

The FLO-2D software solves the continuity and the dynamic wave equation to route flow throughout the site. FLO-20 depicts site topography using a DEM to characterize grading, slopes, drainage divides, and low areas of the site. The DEM is a grid model developed from LIDAR and survey data. LIDAR data was assembled using validated portions of the 2008 PASDA LiDAR data set and t he 2012 LIDAR data set (collected by Photo Science) to produce a composite data set for the TMI Unit 1 site. The composite LIDAR data was accepted by AMEC for use on this project through AMEC' s commercial grade dedication process.

Bathymetric data along the Susquehanna River around TMI Unit 1 upstream of the York Haven dams was developed by Ocean Surveys in April 2005 and provided by Exelon.

Manning's n-values were used to characterize the site's surface roughness and calculate effects on flow depths and velocities. Land cover for the site was evaluated using interpretation of orthoimagery that was verified in the field by AMEC.

THREE MILE ISLAND NUCLEAR POWER STATION Page 54 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 The wave runup at the interior safety related buildings (Air Intake Pagoda & Tunnel, Unit 1 Auxiliary Building, Unit 1 Control Building, Diesel Generating Building, Unit 1 Fuel Handling Building, Unit 1 Intermediate Building, and Unit 1 Reactor Building) is estimated to be equal to the wave height, which is a conservative assumption for a wave runup of broken wave on a vertical structure. Chapter Vl-5 of EM 1110-2-1100 (Reference 59) calculates the wave height as a function of t he Stillwater depths at the point where the wave breaks to the depth at the structure, and the breaking wave height. Using equation Vl 174 from EM 1110-2-1100 (Reference 59), the height of the wave at the vertical wall of the Air Intake Pagoda would be 1.6 ft compared to the characteristic wave height of 2.0 ft.

The wave run up along the west side of t he Unit 1 ISPH is calculated using Equation 13 from USACE CERC 4 (Reference 61). The Unit 1 ISPH is located on the bank of the Susquehanna River adjacent to the flood protection dike slope, which could result in breaking waves on the structure toward the north side of the dike. The longest fetch length of 4.94 km is based on a fetch to the north side of the ISPH. There are no openings on the north or west exterior walls of the ISPH above elevation 275 ft NGVD-29. The path for water from the intake channels (i.e., below elevation 275 ft NGV0-29) to the unprotected area of the ISPH is through small gaps around intake rakes and screens. Wave effects from this path would be attenuated.

There is a much larger open pathway (min. pathway is 9 ft wide above elevation 308 ft NGV0-29) between this interior area of the ISPH and the entrance on the south side of the ISPH. The roll-up door from the south entrance to this area will be open in the event of a flood. The water level in the unprotected interior area of the ISPH is controlled more by wind generated waves at the south entrance than waves on the north or west exterior walls.

The maximum predicted water surface elevation at the north or west exterior walls of the ISPH is 318.2 ft NGVD-29 due to wave runup. The hydrodynamic pressures from this wave are addressed in the next section of the report.

A summary of the predicted water surface elevations at limiting pathways to safety related structures is provided in Table 21. The maximum predicted water surface elevation is at the flood gates (TMI-FG-El, 2A, 28, 2C & E4) on the interior of the ISPH. That maximum elevation is 313.2 ft NGVD-29. The calculated maximum water surface elevations resulting from the characteristic wind generated wave with a wave runup on the site do not exceed the acceptable water surface elevation of 313.5 ft NGVD-29.

THREE MILE ISLAND NUCLEAR POWER STATION Page55 of64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1 Table 22 : Wind Generated Wave Predicted Flooding Elevations and Depth Results at Pathways to TMI Unit 1 Safety Related Structures.

safety Related Acceptable MaKtmum Maximum Characteristic Wave Runup Maximum Maximum Building Water level Water Surface Water Surface wave Height Water Surface Wave height Elevation due to due to due to Stillwater Characteristic Maximum Wave Wave NAVD-88 (ft) NAVD-88(ftl NAVD-88 (ft) ft NAVD-88 (ft) NAVD-88 (fl) ft fNGVD-29 (ltl] INGVD-29 lftll INGVD-29 (ftll INGVD-29 (ftll INGVD-29 (ftll Air Intake Pagoda & 312.7 310,4 311.4 312.4 312.0 2.0 3.2 Tunnel 1313.S) 1311.2) [312.2I 1313.2) [312.8)

Auxiliary Building, unit 1 312.7 310.6 311.4 312.2 312.0 1.6 2.8

[313.5) (311.4) 1311.2) 1313.0) (312.11)

Control Building Unit 1 312.7 310,4 311,3 312.2 311.9 1,8 3.0

[313.5) {311.2) 1312.11 (313.0) (312.7)

Diesel Generating 312.7 310.8 311.5 312.2 312.0 1,4 2.4 Building [313,5) [311.6) 1312.3) 1313.0) 1312.8)

Fuel Handling Building 312.7 310.4 311 .3 312.2 311 9 1.8 3.0 Unit 1 [313.51 [311.21 1312.1) 1313.0J (312.7)

Intake Screen &

Pump House, Unit l 312.7 3104 312 4 d NA1 NA1 NA 1 (Flood Gates 1313.SJ [311,21 [313.2)

(TMI-FG*El, 2A, 28, 2C & E4)1 Intermediate Building 312 7 310 8 311.S 312.2 312,0 1~ 24 Unit l 1313 .SJ [311.6) [312.3] 1313.0) [312.8)

Reactor Building Unit 1 312 7 310,6 311.4 312.2 312,0 16 2.8

[313.S) [311.4) [312,2] 1313,0) [312.8)

The maximum water surface due to Stlllwater and maximum wave are based on FL0-20 model results along the south exterior entrance to the Interior o f the ISPH. This Is the limiting pathway for t he water elevat ion at flood gates (TMI-FG-El, 2A, 2B, 2C & E4).

2 The Interior flood gate of the ISPH ls protect ed from t he occurrence of wave runup effect and Is, t herefore, reported as Not Applicable (NA). The maximum water surface due to characteristic wave based on FL0-20 model along the north eicterlor wall of the ISPH Is 311.8 ft NAVD-88 (312.4 ft NGV0-29). The maximum water surface due to wave runup at t he exterior of the building at this location Is 318.2 ft NGVD-29 (317.4 ft NAVD-88), This con dition Is t he basis for the maximum hydrodynamic pressure In Table 23.

Hydrodynarmc Load A summary of the maximum predicted hydrodynamic pressure from the combined river flow and waves generated by the 2-year wind speed in the crit ical direction wave om t op of the governing PMF are provided in Table 23. The table provides a comparison of t he acceptable hydrodynamic pressure versus t he maximum calculated hydrodynamic pressure. The calculat ed maximum hydrodynamic pressures do not exceed the acceptable hydrodynamic pressures for structures at any of the safety related buildings.

THREE MILE ISLAND NUCLEAR POWER STATION Page S6 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl Table 23: Predicted Hydrodynamic Results at TMI Unit 1 Safety Related Structures.

Acceptable Max. Velocity due Max. Hydrodynamic Hydrodynamic 1 to Maximum Wave Pressure Safety Related Building Pressure psf ft/s psf Air Intake Pagoda & Tunnel 279 2.C, 16.8 Auxlllary Building, Unit 1 2,029 55 75.1 Control Building 8,289 2.8 19.5 Unit 1 Diesel Generating Building 2,029 2.8 19.5 Fuel Handling Building 1,439 2.6 19.5 Unit 1 Intake Screen & Pump House, 6,633 7.37 134.9 Unit l Intermediate Building 42,965 15 5.6 Unit 1 Reactor Building 14,706 l.l 3.0 Unit 1 This Is an acceptable hydrodynamic pressure In addition to the static pressure equivalent of a water level of 313.5 ft NGVD-29.

Debris Load The potential movement and impact of large debris on the site was evaluated for the governing PMF event combined with waves Induced by 2-year wind speed applied along the critical direction. The potential sources of debris during an extreme flood event are from the upstream watershed or the immediate site.

The greatest movement of debris from the watershed occurs during t imes of high flow events, when the flow depths and velocit ies are high enough to float large debris from overbank areas and carry debris downstream to the lower Susquehanna River {Reference 52). DebrJs in the Susquehanna River and its many tributaries comes from both natural and manmade sources. The vast majority, however, comes from natural sources {Reference 52). Natural debris includes branches from streamside t rees and vegetation

{sometime whole t rees and shrubs) that fall into the streams or onto stream banks and flood plains, which often Is carried downstream during high flow events (Reference 52). A significant percentage of the natural debris comes from the more heavily-forested upper reaches of the Susquehanna River Basin, where the Susquehanna River and its tributaries meander, creating conditions for streamside trees and brush to fall into the waterways (Reference 52).

The area within t he TMI Unit 1 owner controlled area is either paved or covered with gravel and has little vegetation or loose materials available t o generate enough debris during the PMF event. However, the parking lot in the northern/upstream portion of t he parking lot is expected to be Inundated during the PMF THREE MILE ISLAND NUCLEAR POWER STATION Page 57 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl and vehicles left in the parking lot could potentially become buoyant and transported toward the safety related buildings.

The hydrodynamic modeling analysis of the governing PMF event combined with waves induced by 2-year wind speed applied along the critical direction provides calculated maximum velocities and maximum hydrodynamic pressures at flooding pathways to safety related buildings. A comparison between the acceptable and calculated maximum hydrodynamic pressures shows that each of these pathways can withstand the estimated flood loading. The topography of the site will concentrate the higher velocities to the north east perimeter of the site away from any safety related buildings. In addition, the layout of non-safety related buildings to the north and east of the safety related structures will limit the available pathways for debris to strike safety related buildings. The concrete security blocks with a top elevation ranging from 302.9 ft to 306.9 ft NAVD-88 north of the diesel generator building will be submerged by 4 ft to 8 ft at the still water elevation. While lighter debris will be able to travel over these blocks, larger debris such as large trees and vehicles may be trapped or redirected away from the safety related structures centrally located on the site. The movement of large debris (i.e., large trees and buoyant vehicles) from the northern parking to the safety related structures will be further limited by the upslope change in grade of around 8 to 10 feet. Low velocities of approximately 1.1 fps to 5.5 fps around the interior portion of the site will further limit any potential movement of debris. The low velocities on the site adjacent to safety related structures and the site configuration (grading, concrete security barriers, and non-safety related buildings) will limit large floating debris from impacting safety related structures.

THREE MILE ISLAND NUCLEAR POWER STATION Page58 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl

5. COMPARISON WITH CURRENT BASIS FLOOD HAZARD The current and reevaluated flood causing mechanisms at the site were compared to assess whether the reevaluated flood hazard Is bounded by the current design basis flood elevation. The comparison is provided in Table 24.

Table 24: Summary Comparison with Current Licensing Basis Flood Hazard Current Licensing licensing Basis Flood Causing Basis Flood Hazard Flood Hazard Reevaluation Bounds Reevaluation Mechanism Elevation Elevation Flood Hazard?

Local Intense Not addressed by Varies from 305.1 ft NGVD-29 to Not Bounded Precipitation design basis. 305.4 ft NGVD-29 at pat hways to TMI Unit 1 Safety Related Structures.

(b)(3) 16 USC He>e>c:liri&...1.ri...Streams. .. 313.3.ftNGVD-29-at-* * *l =:.Jft NGVD-29 at the ISPH. This Bounded

§ 824o-1(d): (b/*-**- and Rivers the ISPH. includes the combined effects of PMF, dam break, and wind-generated waves.

(b)(3) 10 U.S.C .Oam&eachesand ,.,..,. 1

§ 824o-1(dl:"'(br *- Failures lftNGVD~29at ... l=__jtt NGVD-29 at the ISPH. Bounded IH fl::,A\#F\ ~ PH.

This includes the combined effects of seismic dam failure during a high discharge event and wind-generated waves.

Storm Surge Not addressed by Not an applicable flood causing Not Applicable design basis. mechanism.

Seiche Not addressed by Not an applicable flood causing Not Applicable design basis. mechanism.

Tsunami Not addressed by Not an applicable flood causing Not Applicable design basis. mechanism.

Ice Induced Not addressed by 292.0 ft NGVD-29 for the historical Bounded Flooding design basis. most severe ice jam event peak flow of 588,000 cfs.

Channel Migration Not addressed by Based on a historic review, there is Bounded of Diversion design basis. no tendency for channel migration and diversion that would affect the flood hazard.

THREE MILE ISLAND NUCLEAR POWER STATION Page 59 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl

6. REFERENCES

1. Exelon Letter to U.S. Nuclear Regulatory Commission. Exelon Generation Company, LLC's 90-Day Response to March 12, 2012 Request for Information Pursuant to Title 10 of the Code of Federal Regulations 50.54{!) Regarding Recommendations 2.1 and 2.3, of the Near-Term Task Force Review of Insights from the Fukushima Dai-ichi Accident (Flooding). June 11, 2012.

2. Nuclear Energy Institute (NEI), Report 12-08. Overview of External Flooding Reevaluations. August 2012.

3. Nuclear Energy Institute (NEI), [Draft Rev E]. Supplemental Guidance for the Evoluation of Dam Failures. November 2012.

4. U.S. Nuclear Regulatory Commission. Letter to Licensees. Request for Information Pursuant to Title 10 of the Code of Federal Regulations 50.54(!) Regarding Recommendations 2.1, 2.3, and 9.3 of the Near Term Task Force Review of Insights from the Fukushima Dai-ichf Accident. March 12, 2012 .

5. U.S. Nuclear Regulatory Commission (NRC). 2007. NUREG-0800, "Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants: LWR Edition - Site Characteristics and Site Parameters {Chapter 2)," M L070400364, March 2007.

6. U.S. Nuclear Regulatory Commission {NRC). 2011. NUREG/CR-7046, PNNL-20091, Design-Basis Flood Estimation for Site Characterization at Nuclear Power Plants in the Unfted States of America.

ML11321A195, November 2011.

7. U.S. Nuclear Regulatory Commission {NRC). 1977. Design Basis Flood for Nuclear Power Plants.

Regulatory Guide 1.59, Rev. 2, Washington, D.C.

8. U.S. Nuclear Regulatory Commission {NRC). 1976. Flood Protection for Nuclear Power Plants.

Regulatory Guide 1.102, Rev. 1, Washington, D.C.

9. American Nuclear Society (ANS). 1992. American National Standard for Determining Design Basis Flooding at Power Reactor Sires. Prepared by the American Nuclear Society Standards Committee Working Group ANS-2.8, La Grange Park, Illinois.

10. Susquehanna River Basin Commission. Susquehanna River Basin Ecological Flow Management Study Phase I, Section 729 Watershed Assessment. April 2012.

11. Susquehanna River Basin Commission. 2011 Nutrients and Suspended Sediment in the Susquehanna River Basin. Publication No. 284. December 1, 2012.

12. Susquehanna River Basin Coordinating Committee. Susquehanna River Basin Study, Appendix D -

Hydrology. June 1970.

13. Three Mlle Island Nuclear Power Station (2013). Updated Final Safety Analysis Report {UFSAR).

[Update 21 and all approved changes under 50.59 as of Feb 1, 2013]

14. AMEC (2013). Calculation C-1101-122-E410-003, River Stage Discharge and Frequency Analysis Calculation Package.

15. AMEC {2013). Calculation C-1101-122-E410-010, HEC-HMS Model Calculation Package.

16. AMEC (2013). Calculation C-1101-122-E410-0ll, Precipitation Driven Discharge Calculation Package.

17. AMEC (2013). Calculation C-1101-122-E410-012, Dam Failure Peak Discharge Calculat ion Package.

THREE MILE ISLAND NUCLEAR POWER STATION Page 60 of 64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl

18. AMEC (2013). Calculation C-1101-122-E410-013, local Intense Precipitation Calculation Package.

19. AMEC (2013). Calculation C-1101-122-E410-014, Wind Generated Wave Analysis Calculation Package.

20. AMEC (2013). Calculation C-1101-122-E410-015, Ice Induced Flooding Evaluation Calculation Package.

21. U.S. Army Corps of Engineers, Hydrologic Modeling System HEC-HMS computer software. Version 3.5. Hydrologic Engineering C;nter, Davis, CA. August 2010.

22. U.S. Army Corps of Engineers, Hydrologic Modeling System, HEC-HMS, Technical Reference Manual, CPD-74B, Hydrologic Engineering Center, Davis, CA. March 2000.

23. U.S. Army Corps of Engineers, Hydrologic Modeling System HEC-HMS. User' s Manual. Version 3.5.

Hydrologic Engineering Center, Davis, CA. August 2010.

24. U.S. Department of Agriculture (USDA) and Natural Resources Conservation Service (NRCS). Part 630 Hydrology. National Engineering Handbook. Chapter 15. Time of Concentration. May.2007.

25. Larry W. Mays. John Wiley & Sons, Inc. Water Resources Engineering. 2005.

26. U.S. Army Corps of Engineers Baltimore District, " Probable Maximum Flood Estimate for Susquehanna River at Harrisburg, Pennsylvania." Letter. August 3, 1973

27. U.S. Army Corps of Engineers Baltimore District. " Hydrologic Study, Tropical Storm Agnes."

December 1975.

28. U.S. Army Corps of Engineers Baltimore District. Tropical Storm Agnes June 1972" prepared by Gannett Fleming, Corddry, and Carpenter. November 1974, revised February 1975.

29. United States Department of the Interior. "A Rainfall-Runoff Simulation Model for Estimation of Flood Peaks for Small Drainage Basins." Geological Survey Professional Paper 506-B. 1972.

30. Buckler, S.J., 1968. Probable maximum snowpack spring melt and rainstorm leading to the probable maximum f lood Elbow River, Alberta. Canada Meteorological Branch, Prairie Hydrometeorological Centre, 37 pages.

31. Froehlich, D.C. 1995a. Embankment dam breach parameters revisited. Proceedings of the 1995 ASCE Conference on Water Resources Engineering, San Antonio, Texas. August. p. 887-891.

32. Froehlich, D.C. 1995b. Peak Outflow from Breached Embankment Dam. Journal of Water Resources Planning and Management, vol. 121, no. 1, p. 90-97.

33. U.S. Bureau of Reclamation (USBR). 1982. "Guidelines for defining inundated areas downstream from Bureau of Reclamation dams." Reclamation Planning Instruction No. 82-11, June 15.

THREE MILE ISLAND NUCLEAR POWER STATION Page 61 of64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1

34. U.S. Weather Bureau, " Meteorological conditions for the probable maximum flood on the Yukon River above Rampart, Alaska", Hydrometeorological Report No. 42, 1966.

35. U.S. Weather Bureau, "Probable Maximum Precipitation Susquehanna River Drainage Above Harrisburg, PA," Hydrometeorological Report No. 40, May 1965.

th

36. U.S. Weather Bureau, "Seasonal Variation of the Probable Maximum Precipitation East of the 105 Meridian for Areas from 10 to 1,000 Square Miles and Durations of 6, 12, 24 and 48 Hours,"

Hydrometeorological Report No. 33, April 1956.

37. Wahl, Tony L., et al. (2008), "Development of Next-Generation Embankment Dam Breach Models,"

United States Society on Dams, 28th Annual USSD Conference, Portland, OR, April 28-May 2, 2008, pp. 767-779.

38. Y. Xu and L.M. Zang, "Breaching Parameters for Earth and Rockfill Dams", Journal of Geotechnical and Geoenvironmental Engineering, December 2009.

39. U.S. Army Corps of Engineers, " Runoff from Snowmelt," Engineering Manual 1110-2-1406. March 1998.

40. U.S. Army Corps of Engineers, "A Synopsis and Comparison of Selected Snowmelt Algorithms," Cold Regions Research & Engineering Laboratory (CRREL) Report 99-8. July 1999.

41. Ward, Andy D. et al. "Environmental Hydrology" Chapter 10 - Hydrology of Forests, Wetlands, and Cold Climates. CRC Press LLC. 1995.

th

42. Lutgens, F.K. and Tarbuck E.J., 2007. " The atmosphere: An introduction to meteorology", 10 Ed.

520 pp, Pearson Prentice Hall.

43. U.S. Department of Agriculture (USDA) Natural Resources Conservation Service (NRCS} Technical Release No. 60 - Earth Dams and Reservoirs. 2005.

44. Anderson, E and Larson, L. "The role of Snowmelt in the January 1996 Floods in the Northeastern United States." Proc. 53d Eastern Snow Conference, Eastern Snow Conference, Williamsburg, VA, 141-149. 1996.

45. Leathers, Daniel J. et al. "The Severe Flooding Event of January 1996 across North-Central Pennsylvania." Bulletin of the American Meteorological Society, vol. 79, Issue 5, pp.785-797. 1998.

46. R.W. Katz et al., 2002. "Statistics of extremes in hydrology", Advances in Water Resources pp.

1287-1304.

47. U.S. Army Corps of Engineers, 1996. Engineering Manual EM-1110-2-1619,"Engineering and Design, Risk Based Analysis for Flood Damage reduction Studies."

THREE MILE ISLAND NUCLEAR POWER STATION Page 62 of64 RCN: TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Rev 1

48. Nuclear Energy Institute, 2012 (REV. E Draft). Post-Fukushima Near-Term Task force Recommendation 2.1, Supplemental Guidance for the Evaluation of Dam Failures.

49. White, Kathleen D. Ice Events In the Susquehanna River basin, Ice Engineering Information Exchange Bulletin Number 21, April 1999.

SO. National Weather Service Middle Atlantic River Forecast Center Historic Floods by Events - Flood Events in 1936. National Oceanic and Atmospheric Association website.

http:ljwww.erh.noaa.gov/marfc/Rivers/FloodClimo/MARFCHistoricFloodEvents/1900sFloods/1936

/ 1936-March 17.pdf

51. U.S. Nuclear Regulatory Commission (NRC) NRC (2011). "Design-Basis Flood Estimation for Site Characterization at Nuclear Power Plants in the United States of America", NUREG/CR-7046, Washington, D.C. November, 2011.

52. Susquehanna River Basin Commission. " Managing Debris in the Susquehanna River Watershed" Information Sheet. June 1999.

53. U.S. Department of the Interior. "The Floods of March 1936 Part 2 Hudson River to Susquehanna River Region." Geological Survey Water-Supply Paper 799. 1937.

54. U.S. Geological Survey. Pennsylvania Middletown Quadrangle. Map. Edition of September 1908.

Reprinted June 1913.

55. U.S. Army Corps of Engineers. Middletown Quadrangle 15 minute Series. Map. First Edition 1943.

56. U.S. Department of the Interior Geological Survey. Middletown Quadrangle Pennsylvania 7.5 Minute Series. Map. 1963.

57. U.S. Department of the Interior, U.S. Geological Survey. Middletown Quadrangle Pennsylvania 7.5 Minute Series. Map. 1999.

58. U.S. Department of the Interior, U.S. Geological Survey. Middletown Quadrangle Pennsylvania 7.5 Minute Series. Map. 2010.

59. United States Army Corps of Engineers (2008), Engineering Publication EM 1110-2-1100.

60. Holmes, Patrick (2001), Professional Development Programme: Coastal Infrastructure Design, Construction, and Maintenance. A Course in Coastal Defense systems I. Chapter 10: Coastal and Offshore structures. Department of the Civil Engineering, the University of the West Indies, in conjunction with Old Dominion University and the Coastal Engineering Research Center. US Army Core of Engineers, Vicksburg, MS.

61. United States Army Corps of Engineers (1990), Miscellaneous Paper CERC-90-4, "MaJCimum Wave Run up on smooth slopes".

THREE MILE ISLAND NUCLEAR POWER STATION Page 63 of 64 RCN : TMl-140

NTTF Recommendation 2.1 (Hazard Reevaluations): Flooding March 11, 2013 Exelon Corporation Revl

62. GAi drawing 744-057, Grading plan and details, storm drains and flood control structures.

63. Anderson, Eric A. "A Point Energy and Mass Balance Model of a Snow Cover." National Oceanic and Atmospheric Administration (NOAA) Technical Report 19, February 1976.

64. Exelon (2013). Technical Evaluation 1467688-04 Local Intense Precipitation Impact Evaluation.

65. United States Department of Congress National Oceanic and Atmospheric Administration, National Weather Service Easter Region Headquarters. Remnants of Tropical Storm Lee and the Susquehanna River Basin Flooding of September 6-10, 2011. May 2012.

66. Anderson, Eric A A Point Energy and Mass Balance Model of a Snow Cover. National Oceanic and Atmospheric Administration (NOAA) Technical Report 19, February 1976. U.S. Department of Commerce.

THREE MILE ISLAND NUCLEAR POWER STATION Page 64 of64 RCN: TMl-140

Enclosure 2 TMI Site Topography Map

C-1101-122-E410-013 APPENDIX TMl- 1 Diesel Generator Butld1ng TMl-1 Intake Screen Pump House TMl -1 Air Intake Structure Legend Model Boundary 1111 Building Outline

- - Security Wall

- - 1O ft Contour 2 ft Contour Three Mile Island Nuclear Power Station TMl-13-017 Enclosure 2 - TMI Site Topography NOTE Elevauon NGVD-29=-NAVD-88+0 8 FT Ground Surface Elevations (NAVD 88)

WV 0 300 600 900 Feet Figure A-01 , C-1101-122-E410-013 amecf3

NRC-2017-000688 (Formerly FOIA/PA-2017-0690) - Resp 5 - Final, Agency Records Subject to the Request Are Enclosed, Part 12 of 15
Person / Time
ML20366A020
Issue date:	12/29/2020
From:	NRC/OCIO
To:
Shared Package
ML20366A007	List: ML20366A008 ML20366A009 ML20366A010 ML20366A011 ML20366A012 ML20366A013 ML20366A014 ML20366A015 ML20366A016 ML20366A017 ML20366A018 ML20366A019 ML20366A020 ML20366A021 ML20366A022 ML20366A023
References
FOIA, FOIA/PA-2017-0690, NRC-2017-000688
Download: ML20366A020 (66)
v • d • e

ML20366A020

Text

SUMMARY

SUMMARY

Navigation menu

Search