ML20097F699

Revised Analysis of Likelihood of Waterborne Traffic & Other Floating Objects on Delaware River Impacting Hope Creek Generating Station in Severe Storms

ML20097F699
Person / Time
Site:	Hope Creek
Issue date:	09/30/1984
From:	ARTHUR D. LITTLE, INC.
To:
Shared Package
ML20097F667	List: ML20097F666 ML20097F690 ML20097F699
References
C-50918, NUDOCS 8409190074
Download: ML20097F699 (51)

Text

AN ANALYSIS OF THE LIKELIHOOD OF WATERBORNE

-TRAFFIC AND OTHER FLOATING OBJECTS ON THE DELAWARE RIVER IMPACTING THE HOPE CREEK GENERATING STATION IN SEVERE STORMS REPORT TO PUBLIC SERVICE ELECTRIC AND GAS COMPANY NEWARK, NEW JERSEY 07101 BY ARTHUR D. LITTLE, INC.

CAMBRIDGE, MASSACHUSETTS 02140 REVISED REPORT SEPTEMBER 1984 A

C-50918 l

l hMkh 0 j .

E A Arthur D.Little,Inc.

TABLE.0F CONTENTS P*&S

1. INTRODUCTION -

1

1.1 INTRODUCTION

TO REVISED REPORT 1

1.2 BACKGROUND

1 1.3 THE NRC REQUEST FOR CLARIFICATION 2 1.4- OBJECTIVE OF THIS REPORT 2 1.5 APPROACH 3 1.6 ASSUMPTIONS IN THE ANALYSIS 4

2. DEFINITIONS OF PARAMETERS 5 2.1 WEATHER EVENT SCENARIOS 5 2.2~ NORMAL VESSEL TRAFFIC AND POPULATION 8 2.3 VESSEL POPULATION AND TRAFFIC IN HEAVY WEATHER 10 2.4 RESISTANCE TO DAMAGE OF CATEGORY I STRUCTURES TO RECREATIONAL BOAT-IMPACTS 15 2.5 TOTAL TRAFFIC OF CONCERN 17 2.6 IMPACTS OF OTHER FLOATING OBJECTS 17 2.7 SITE IMPACT PROBABILITY ASSESSMENT MODEL 21

3. PROBABILITY ESTIMATES 24 3.1 COOLING WATER INTAKE STRUCTURE 24 3.1.1 EXTREME WIND EVENTS 24 3.1.2 HURRICANES 26 A. PROBABLE MAXIMUM HURRICANE 26 B. INTERMEDIATE HURRICANE 27 C. MODEL HURRICANE 27 3.1.3 ' TOTAL COOLING WATER INTAKE IMPACT PROBABILITY 28

-3.2 HOPE CREEK SITE IMPACT , 28 3.2.1 PROBABLE MAXIMUM HURRICANE (PMH) 28 3.2.2 INTERMEDIATE HURRICANE 29 3.2.3 TOTAL HOPE CREEK SITE IMPACT PROBABILITY 30 3.3 SUMMATION OF PROBABILITIES 30 3.4. CONSERVATIVE NATURE OF THE PROBABILITY ESTIMATES 30

,4. REFERENCES 33 APPENDIX A - COMMUNICATIONS WITH U.S. C0AST GUARD CAPTAIN OF THE PORT, PHILADELPHIA 34 APPENDIX B - COMMUNICATION WITH PHILADELPHIA NAVAL SHIPYARD 40

. APPENDIX C - POISSON MODEL TO ASSESS CONDITIONAL PROBABILITY OF IMPACT GIVEN A MARINE CASUALTY 44 A Arthur D. Little,Inc. - i

LIST OF TABLES Eagg TABLE 1 ' HOPE CREEK GENERATING STATION ANNUAL PROBABILITIES OF EXTREME SIX-HOUR 33-FT. WIND SPEEDS BASED UPON 11 YEARS OF ARTIFICIAL ISLAND WIND DATA 6 TABLE 2 INCREASED WATER DEPTHS AT HOPE CREEK WATER INTAKES DURING SELECT EXTREME WIND EVENTS 7 TABLE 3 TOTAL RUNAWAY VESSEL ESTIMATES FOR STORMS 12 TABLE 4a EXPECTED NUMBER OF RUNAWAY VESSELS MOVING NORTH FROM THE SOUTH OF ARTIFICIAL ISLAND IN A PROBABLE MAXIMUM HURRICANE 14 TABLE 4b EXPECTED NUMBER OF VESSELS INADVERTENTLY MOVING UPRIVER FROM THE SOUTH OF ARTIFICIAL ISLAND IN A PROBABLE MAXIMUM HURRICANE 14 TABLE 5 TOTAL TRAFFIC OF CONCERN 18 LIST OF FIGURES Eagg FIGURE C.1 IDEALIZED SCHEMATIC OF THE DELAWARE RIVER, HOPE CREEK AND THE IMPACT GEOMETRY 38 I-A Arthur D.Little,Inc. 1i

, _ _ = _ . ___ - . _ . _ _ _ . ___

e

1. INTRODUCTION 1.1 Introduction to Revised Report BasedL on a ' request for clarification regarding the impact of

~

waterborne objects on the Hope Creek Generating Station, Arthur D.

Little, Inc. prepared a report dated July 1984 entitled, "An Analysis of s

the Likelihood of Waterborne Traffic on the Delaware River Impacting the Hope Creek : Generation Station in Severe Storms." This report was formally submitted to the NRC in July and subsequently discussed with

-the NRC staff. .In the course of that discussion the NRC requested that the report be expanded to include floating objects such as utility poles and l trees in addition to waterborne traffic since such objects could compromise the leak-tightness of certain doors in the Hope Creek power block. In addition, whereas the previous report addressed the Probable Maximum Hurricane and the model hurricane, the NRC requested that an

. intermediate hurricane capable of flooding the grade at Hope Creek with water of depth three feet or more also be addressed.

This revised report with a slightly expanded title was prepared to

' address the expansions requested by the NRC. This revised report dated

' September 1984 replaces in -its. entirety the previous report of July 1984.

'1.2 Background The Public Service Electric and Gas Company (PSEG) is in the process of constructing the Hope Creek Generating Station (HCGS). The i

Hope Creek site is located on the Delaware River estuary near the southern end of an artificial peninsula known as Artificial. Island. The site is located.in Salem County, New Jersey.

- As f a part of the overall safety evaluation for the plant, the potential effects of waterborne traffic on the control room and water intake structure :at HCGS were analyzed by Arthur D. Little, Inc. (ADL) in 1974 and described in a report (Reference 1) to PSEG. This study considered risks to the intake structure and the control room from barge and ship / tanker related spills. Significant findings of that study are contained in the Hope Creek FSAR.

L A Arthur D. Uttle,Inc. 1

e

}

. In their review of certain portions of the Hope Creek FSAR, the Nuclear Regulatory Commission (NRC) raised certain new questions and requested clarification regarding the discussion of storm related high water events in the FSAR. In order to respond to those questions, PSEG asked ADL to assist them with the answers to those questions.

1.3 The NRC Request for Clarification During their review of the FSAR, the NRC staff noted that the postulated maximum hurricane could result in a situation where there is as much as 12 feet of water on the Hope Creek site. Whereas the plant has been designed to safely withstand up to 12 feet of water above grade, such water depths could allow small draf t marine vessels and l, other floating objects such as utility poles to enter the site shoul,d they lose power and steerage or are uprooted in a storm. The NRC requested an evaluation of this scenario from an overall plant safety perspective.-

1.4 Objective of This Report This report addresses three major objectives:

1. Although the NRC question addresses the postulated maximum hurricane, one objective of this report is to assess the probability of occurrence and the level of high water associated with these storm related events as follows:

( o extreme wind events i

[

o typical or "model" hurricanes l

o a more severe hurricane leading to 3 ft. of water or more on grade o- postulated maximum hurricanes

2. The second objective is to profile the marine vessel traffic on the Delaware River and to estimate the likely population of i

! . runaway or out-of-control vessels.

\

- A Arthur D. Little,Inc. 2 e

n

3. Another objective is to develop a profile of floating objects other = than marine . vessels and assess their impact on the overall integrity of the plant.

4. The final objective is to utilize the information relating to storms and vessels to assess the overall probability of marine traffic on the Delaware River impacting the HCGS.

'The approach taken in achieving these objectives is described below.

I.5 Approach An evaluation of.the likelihood of occurrence of the extreme wind events . of concern, the model hurricane and the postulated maximum hurricane was conducted through an analysis of the site meteorology.

This evaluation was performed by Meteorological Evaluation Services, Inc. for-PSEG and is reported in Reference 2. The levels of.high water associated with the storm situations of concern were determined using a site specific Delaware River Storm Surge Analysis model. This analysis was performed by Dames & Moore, Inc. .for PSEG and is reported in

~ Reference 3.

' The profiling of traffic was based on information obtained in meetings and written communications with the U.S. Coast Guard Captain of the Port of- Philadelphia (see Appendix A) and the United States Navy (see -Appendix B)., In addition,. use was made of previous studies t

(Reference 1, 4, 5) and information obtained from contacts with the Philadelphia Maritime Exchange and the Pilots Association. Finally,

,- information contained in the document " Waterborne Commerce of the United States" (Reference 6) issued annually by the Army Corps of Engineers was ,

also analyzed to assess marine vessel distributions by draft.

The! approach to assessing-the likelihood of vessel impact during a storm was to utilize the characteristics of the storm events, the vessel

i. population and marine casualty data from the U.S. Coast Guard Computerized Casualty files (Reference 7) and previous studies

. (Reference 1,14, 5) to assess the overall probability of concern. Use e

i 5

i A Arthur DL Litde,Inc. ~3 N 1

was made of a Poisson failure rate model tied to the river / plant

-geometric configuration (see Appendix C) in the conduct of this evaluation.

-Finally the non-marine vessels considered in this study included utility poles, houses, automobiles, fuel tanks and trees. These lighter, low kinetic energy per unit area objects were analyzed for their ability to strike metal doors in the power block with sufficient force to compromise the leak-tightness of the doors.

1.6 Assumptions in the Analysis

'In performing the analysis and arriving at quantitative estimates of the probability of impact it became necessary to make certain assumptions regarding physical situations and failure rates. When such assumptions became necessary they were made in a conservative manner so

~

as to ensure that the assumption led to an over-statement of the probability. As such, the final probability estimates reported here are deemed to be conservative over-estimates of the likelihood of occurrence of the events of concern.

A listing of the conservative estimates and assumptions utilized in this study is presented in Section 3.4.

i

! A Arthur D.Uttle,Inc. 4

_ _ . . .- . = _ _ _ _ . . __ . . - _ _

~

2. DEFINITIONS OF PARAMETERS 2.1 Weather Event Scenarios There are three meteorological events of concern to the current

investigation. These include

o Extreme winds.'(no water on grade) o- Typical (model) hurricanes (no water on grade) o Intermediate hurricane (3ft. of water or more on grade) o Probable maximum hurricanes (12 ft. of water on grade)

In each ; case, it was required that estimates be made of the probability that the event of concern would generate high tidal surges

~

at the Hope Creek site. The meteorological analysis was conducted by Meteorological Evaluation Services, Inc. (MES) and is de' tailed in the f

.. July and August 1984 - reports entitled " Hope Creek Generating Station Extreme' Event . Site Flooding Meteorology," (see Reference 2). Dames &

Moore,' Inc.,'was given the task to estimate peak surge water levels associated with various wind velocities and directions.- Its findings t

are presented .in detail in a report entitled " Storm Surge Calculations for Hope ~ Creek Generating Station" (Reference 3). 1 Table--I nummarizes findings of.MES with respect to extreme six-hour average - wind speeds in the Artificial Island area as a function of a

-specific wind direction sector. The wind direction sectors noted thereon are those required for high water levels at the Hope Creek site.

The analysis considers six-hour averages, - since substantially shorter time periods would not permit the tidal surges of interest. Table 2 couples the MES findings with those of Dames & Moore- to indicate the I

-probabilities. associated with various water depths above mean low water level at.the' Hope Creek site due to extreme wind effects. Since grade level at.the site is approximately 11 feet above the National Geodetic 4- . Vertical Datum (NGVD), and since the NGVD is about three feet above the Mean Low Water - (MLW) Level, it is apparent that there is negligible

. . probability of significant flooding of the Hope Creek site due solely to 4

-extreme wind effects. However, it is also evident that water depths at A Arthur D.Little,Inc. 5

TABLE 1

' HOPE CREEK GENERATING STATION ANNUAL PROBABILITIES OF EXTREME SIX-HOUR 33-FT. WIND SPEEDS BASED UPON 11 YEARS OF ARTIFICIAL ISLAND WIND DATA Wind Direction: 79 - 170 (*AZ)

Annual Probability: 0.002 0.001 Corresponding Wind Speed (mph): 52 57 i

I A Arthur D. Little,Inc. 6 b

n TABLE 2 INCREASED WATER DEPTHS AT HOPE CREEK WATER INTAKES DURING SELECT EXTREME WIND EVENTS i

Approximate Increased Maximum 6-hr Average Water Depth

-Annual Probability Wind Speed (mph) (ft) o

~3 2 x 10 52 9-12 1 x 10 -3 57 10-12 Note: 1. Data for maximum 6-hr average wind speeds as a function of

?

annual probability were developed by Meteorological Evaluation Services, Inc. (MES)

2. Water depths presented are based upon results of Dames & Moore analyses.

3. Water : depths herein are peak surge levels above mean low water. These are surge levels at the site considering the effect of open coast surge. Surge levels with 10 percent

. exceedance high tide are in the range of 6-7 ft for all cases.

4. An increased water depth of 12 ft over Mean Low Water at the service water intake results in a water level which is about 3 ft below plant grade.-

e s

A Arthur D Utde,Inc. 7

the service water intake structure could approximate 28 feet under such conditions.

With respect to the typical (model) hurricane, MES has predicted an annual occurrence rate of'10 -2 . Dames & Moore, in turn, has predicted a surge water level' of slightly less than 14 feet above Mean Low Water.

Thus, it is concluded that the model hurricane is also incapable of causing flooding at grade. It is, however, also capable of causing

water depths on the-order of 28-30 feet at the water intake structure.

The MES analysis predicts a 1 x 10

-5 per year occurrence rate for the probabis maximum hurricane. Dames & Moore predicts a surge level at the site that would produce a maximum 12 ft or so water depth over

, grade, with the time for initial flooding above grade to depletion of high -water spanning a period of 6-8 hours. Marine vessels with less than 12 feet draft could enter the plant site under these conditions.

! Also the water . depth at the intake structure could reach 40 feet for this unlikely, probable maximum hurricane, i

f 2.2 Normal Vessel Traffic and Population Appendix A describes a meeting and correspondence between Arthur D.

Little, Inc. and the U.S. Coast Guard Captain of the Port, Philadelphia.

These contacts permit estimates that:

o There are about 50 large commercial vessels on the Delaware River and in port each day with draf ts in the range of 18-40 fc.

o Less than 80 cuss would be operating in the area on any

given day.

i o There are about 4,500 recreational boats moored in the area.

l o There are roughly 150 barges on the Delaware River on any given day. Drafts of these range from a few feet when empty to 35 feet when loaded.

i i

A Arthur D Utste,Inc. 8 I

The intermediate hurricane was also analyzed by Dames & Moore i as well as MES. The intermediate hurricane, one with sustained wind speeds of 80 mph or more, is capable of producing flooding on grade. Such a hurricane could result in the grade being flooded with 3 ft. of water with an estimated annual probability of 5 x 10 -5 ,

o Larger vessels in the Philadelphia area generally use Marcus Hook or Mantua anchorages. There are 6-12 vessels in these locations on any given day.

o There are contingency plans in place to increase vessel security in heavy weather.

Appendix B describes a meeting and correspondence be.eween Arthur D.

Little, Inc. and the Operations Officer of the Philadelphia Naval Shipyard. These contacts revealed that:

o The shipyard has only one active ship (draft 25 ft).

This would go out to sea or be more securely moored in the event of a hurricane.

o There are typically 4-5 ships in the shipyard for overhaul, o There are 29 small craft, such as tugs or barges, in addition to camels which service overhaul efforts.

o There are three cruisers, seven destroyers, and several submarines mothballed at the base.

o Naval traffic on the Delaware River is relatively minimal.

o There are no " Newport", " Anchorage", and "DeSoto" class Navy vessels at this yard.

The above data, together with information obtained from the Pilots Association and the Philadelphia Maritime Exchange, permit the following consolidated estimates with respect to the number of vessels travelling or moored fn the overall area of interest on any given day.

/h Arthur D. Uttie,Inc. 9

g $ ;.f x ,

y y yj

e

~'

.a

.q p $

4 . o $ 60-70 self-propelled Tommercial vessels (excluding tugs)

.100- 150 non-self-propelled? barges

. o-

M o .70-80 captive (i.e..'. Local) tugs a= ;_ ~

- . i\

o y 10-20.other tugs.  ;, .

.4 -+

os P o4;6-12 large self-1,ropelled ~ vessels in anchorages to the M A D t/ south of Artificial Island

^

IIo 6L 12 31arge self-propelled vessels in anchorages near the

-\ ~

-r -port of' Philadelphia (noetE of the Hope Creek site) ~

3 4

_ o. .2500-3000 recreational vessels actually in the water flci - _ _ .About 50 Naval vessels l

, 1;

. 2.3 Vessel Population and Trdffic in Heavy Weather

-IU.S.CoastGuardcontingenhplans,aswellasthedesireofvessel owners / operators . to safeguard - their investments, indicate that vessal n;  ; ,

traf.7fic.l,upon the Delaware ; River would be greatly curtailed, if not

. - s completely halted, is? the'evest of a severe storm. . Larger vessels would have a live bridge' watch'andi. standby engine room personnel as mandated

!- m-

..by the Coast Guard.. All major ve'ssels would be secured by additional anchSrs, longer anchor :chainsIl and/or additional mooring lines.

It follows that . the._ key hazard to' the Hope Creek plant would be from l ~

vesselsJhat break / mooring ~ lines and become runaways, or in the case of

=

Slarger' vesselen 'simultaneobs.;1oss ,g of power and steering capabilities

[

af ter loss of moorinh 'Nevertheless, there . may be a few vesss3 that failf tMre'ach ,a safe [ anchorage or mooring area in time and these must

! abo .be given,special attention.

JIt 1 should be ~ noted' thac^ bohh the model and probable . maximum lc  % hurricane would be tracked from~their, initiation either in the Caribbean

's or the. South Atlantic for several; days prior to arrival in the vicinity

-of'the State of New Jersey.

~

At lebst twelve hours of warning would be available to Delaware River marine-vessel operators of the arrival of an impending. hurricane. Sufficient time is available to implement USCG l plans and it is highly.unlikely that there would be any vessel movement

--on the river.

-w o o .,

v l- .,

l' .? ib' A ArthufIlLSkInc.

10 N : %-.- .x. . _ - , __ - -.__-- _.-.- - -._..-_.

, The postulated high wind events are not as severe as the postulated scenarios for hurricanes. As such, at least six hours of persistant high winds are required to cause appreciable high water surges at Hope

' Creek. Once again six hours is sufficient time for marine vessel operators on the Delaware to seek shelter and secure mooring lines.

Very few large ships may be underway with tug escort but it is highly unlikely that small'eraft or barge tows would be operating under these conditions.

Runaway Vessels It is difficult to determine precisely the fraction of vessels in various size categories that might break loose of moorings during

j. extreme wind or hurricane conditions. It is, however, feasible to formulate conservative estimates for the purposes of the current analysis.- Such estimates are presented for the probable maximum hurricane in Table 3 for the total vessel population in the Delaware Bay to Philadelphia area.

It is highly significant that weather conditions associated with high water levels at Hope Creek, these being the conditions of specific interest to the current analysis, require winds blowing from a generally

-easterly to southerly direction, and that any vessels drif ting in such weather on the Delaware River will travel in a generally northerly direction. This indicates that only vessels to the south of Hope Creek are of ~ concern as potential missiles impacting Hope Creek facilities.

Since the shoreline to the south- of Hope Creek is relatively devoid of highly populated or developed areas (in comparison with northern reaches l of the river) , it becomes necessary to account for the fact that the

! vast majority of runaways will occur north of the subject site and proceed in a direction away from the site.

l i

J

• j Thase estimates for the probable maximum hurricane are conservatively assumed to apply to the less severe intermediate and model hurricanes and the extreme wind events of concern.

a l

l-A Arthur D.Little,Inc. Il-

TABLE 3 TOTAL RUNAWAY VESSEL-ESTIMATES FOR STORMS Assumed Total Total Percentage Number Vessel Type Number Runaways Runaways Self-propelled 70 2% 1.4 Non-self-propelled. 150 5% 7.5 Tugs 100 2% 2 I

Recreational boats 3000 25% 750 Navy vessels 50 1% 0.5 Note that most of these vessels are substentially north of the Hope Creek site. In any postulated high wind /high water scenario any

' runaway vessel would be pushed north. As such only vessels to the

--south of Hope. Creek are vessels of possible concern.

Conservative estimates based on discussions with the U.S. Coast Guard, the Pilots Association and the Philadelphia Maritime Exchange.

i' A Arthur D. Little,Inc. 12

. .. - - - . = . .. --

V' As noted earlier, it has~been determined that there are typically 6-12 large self-propelled vessels in the Bombay Hook Point anchorage  ;

approximately 10 miles to the south of Hope Creek. For barge cows, it

~is relatively conservative to assume that no more than one tow of 4 barges mi8 ht be forced to moor south of Hope Creek in the Bombay Hook

. Point' anchorage due to a lack of time for reaching a safer location. In

' making this assumption, it is noted that there are no barge terminals or

ship moorings within a'20 mile' distance to the south of Hope Creek. Tow

~ operators would not wish to be found'in this area in foul weather unless forced by unavoidable circumstances. The one tow would be associated with one . tug. For reasons similar to those given above, no more than y three. additional tugs would be expected in the area.

g The Delaware Bay and River sections to the south of Hope Creek have relatively unpopulated coastlines with few roads and few facilities that

.might be described as marinas. Given this fact, and the fact that the vast majority of recreational boats are found in populated areas far to the north of Hope Creek, it is assumed that no more than 10 recreational boats would somehow be forced to. find shelter by mooring on the Delaware River in the area immediately to the south of-Hope Creek.

The U.S . ' Navy has indicated that it has but one active ship in l Philadelphia Navy yard and that this ship goes out to sea only twice a

month, unless ordered to do so to avoid being in port in a hurricane.

It is therefore considered conservative to assume that this vessel would l not be-in the Hope Creek area during those times it would be vulnerable f

to the effects offa storm.

Table ' 4a summarizes the results of the evaluation L for vessels to the south of Hope Creek with the potential for becoming runaways.

Vessels Enroute There is always a. chance that a few vessels may attempt to outrace H a storm to their ultimate destination. It is therefore assumed thera

'might be two recreational boats and one large self-propelled vessel

T actually moving on the river intentionally under storm conditions.

These are shown in Table 4b.

A Arthur D.Little,Inc. . 13

h TABLE 4a EXPECTED NUMBER OF RUNAWAY VESSELS MOVING NORTH ,

FROM THE SOUTH OF ARTIFICIAL ISLAND IN A PROBABLE MAXIMUM HURRICANE Assumed Number Number Percentage Southern,,

Vessel Type To South Runaways Runaways Self-propelled 6-12 2% 0.24 Non-self-propelled 4 50% 2.00 Tugs 4 2% 0.08 Recreational boats 10 25% 2.50

' Navy vessels 0 1% 0.00 TABLE 4b EXPECTED NUMBER OF VESSELS INADVERTENTLY MOVING UPRIVER ,

FROM THE SOUTH OF ARTIFICIAL ISLAND IN A PROBABLE MAXIMUM HURRICANE Vessel Type Number Large self-propelled 1

-Recreational 2 I

These estimates for the probable maximum hurricane are conservatively assumed to apply in the event of less severe events such as the model hurricane and the extreme wind events.

The non-self-propelled barges and unmanned recreational boats are true runaways in that they will move as directed by the wind and surface currents. The self-propelled vessels, tugs, and Navy vessels may break mooring but could still be controlled using their own power and steerage. Should they subsequently lose both power and steerage they would be classified as' runaways as well.

1 A Arthur D.Little,Inc. I4

.=. . ..

2.4- Resistance to' Damage of Category I Structures to Recreational Boat Impacts One . of the possible results of a probable severe storm on the Delaware River'could be'that anchored, moored, or underway recreational boats could become unsecured or lose control under the action of the

-wind and waves and, as a runaway, impact the Category I structures (e.g., the service water intake structure) at the Hope Creek Generating i:

Station. The question is whether, under such impact conditions, the Category I structures could be damaged to the extent-that their ability to function is compromised. To examine this question, an evaluation is

made of recreational boat impacts on Category I structures under the I most severe storm postulated -- the probable maximum hurricane.

l

The Category I structures are designed to withstand severe design loadings based on extreme external and natural' hazard conditions. These include seismic effects to the entire structure and to major components and. structural elements within the structure, tornado loads, and i hurricane and storm winds and flooding conditi'ons. As a result,.the Category- I ' structures are typically of heavily-reinforced concrete-construction with wall structures of thicknesses of two feet or more.

One of the structural design requirements of these structures is its- ability to . resist the impacts of tornado-generated missiles.

Several types of missiles must be considered in this regard, including L wood planks, utility poles, steel pipe, and even entire automobiles.

?

! The total . kinetic energy of these - design missiles range from 5000 to -

1,800,000 f t-lbs. On an impact area basis, the design missiles are

[

L typically in the 150,000 to 700,000 ft-lbs per ft 2 of impact area. In

[ one ' series -of tests (see Reference 8). using a utility pole missile, i

~

13.5" in diameter, 35 f t long, and weighing about 1500 lbs, was driven r,

P against a 12" thick reinforced concrete wall panel at 140 mph. The I result was the splintering of the end of the pole into many small pieces and negligible structural damage to the concrete.

The kinetic energy of these utility pole missiles was about one h million ft-lbs, and also was about one million ft-lbs per ft2 of impact l

t ,

h A Arthur DL Little,Inc.- 15 '

L

4 s

J p

area. A very large recreational boat'might be of the order of 10 tons.

-It would' generate 100,000 ~f t-lbs of kinetic energy travelling at about 12 aph, or 10 knots. It's kinetic energy per unit area would be in the range of 1,000 ft-lbs per ft2 of impact area.

Thus, although the total

energy . of a boat - impact - could be about - the same as a utility pole

missiler_the unit load on the wall structure would be much less. The j ' A . pole, furthermore is dense and strong in axial end-loadings, while a typical - boat - bow-structure is 4 not designed for major head-on impacts.

On this basic, _ we - believe that~ is it reasonable to conclude that the j . impact of -large pleasure- boats, wind and wave driven, against the

! = concrete walls of Category I structures would result in severe damage to r

+. .

the boat and negligible damage to the concrete structure, either locally

.or over an extended structural area. Such events would be similar to impacts of boats or sailing vessels against sea walls or breakwaters.

i: ImpactsLof boats wind and wave driven by squalls or storms against sea walls or other shoreline structures is a fairly common event.

Invariably, the results in such accidents is severe damage to the boat

and negligible effects to the shore structure, i :-

There is other; evidence to support this conclusion. Model studies (see Reference -9) carried out on collisions between two ships of

[ differing impact strength have shown that the distribution of the structural damage between the two ships is quite sensitive to the relative strength or structural resistance of the two ship structures, l with the weaker'of the two absorbing most of.the impact energy and hence

~being destroyed one-sidedly. On this basis, it would seem reasonable to F conclude that if a boat collided with the relatively massive reinforced f

i concrete Category I structure, most of the impact energy could be b dissipated in damage-to the boat. The Category I structure would not l experience significant damage and would continue to meet its functional requirement.

'As a result of the above considerations, recreational boats are not considered an issue of concern in this analysis and are not considered g potential vessels of concern.

I_

l A Arthur D.Little,Inc. 16 L

2.5 Total Traffic of Concern Based-on the previous discussions and the data in Tables 4a and 4b, a summary table is shown in Table 5 which identifies the total vessel population of ! potential concern in a storm related situation on the Delaware River. In viewing Table 5 it should be kept in mind that the non-self-propelled vessels are barges with drafts in the range of 3 feet (empty barges) to up to 40 feet (loaded ocean going tanker barge). The commercial'self-propelled vessels have a draft range from 5 feet to 49 ofeet. The barges (non-self-propelled vessels) are considered true runaways and will move vectorially in a vector that is typically the sum of the surface current vector and three to five percent of the wind

~ velocity vector (Reference 10). The self-propelled vessels are potential drifting objects only if they lose both power and steerage.

2.6 Resistance of Doors to Other Floating Objects GENERAL The safety class structures and components of a nuclear facility must be designed to remain functional following the possible exposure to floating missile = impacts resulting from- the extreme environmental conditions associated with storm winds and floods. Such events can generate potentially damaging missiles from a variety of objects which are in the path of the storm winds.

The characteristics of objects which define their behavior as missiles are their shape, density, surface area, and the maximum velocity they attain. The effects of impact of missles on a target structure depends on these missile characteristics and on the geometry and material of the target. In some cases, the impact process may be local, or take place so rapidly that it can be considered to have only I local effects. In other cases, the duration of the load is long enough, i

or- the impact areas is large enough relative to the structural dimensions, so that the target structure will experience a gross response over the entire structure.

The misailes considered were utility poles, automobiles, houses, fuel tanks and trees. This listing of potentially damaging missiles i

r L

(

. A Arthur D.IJttle,Inc. 17

L

).

TABLE 5 TOTAL TRAFFIC OF CONCERN Vessel Type Potential Runaway Runaway Self-propelled 1.3 -

Non-self-propelled - 2 Tugs and Navy vessels negligible negligible A distinction is made between self-propelled and non-self-propelled vessels. A self-propelled vessel is a potential runaway since it first must lose power and steering prior to becoming a runaway vessel moving with the wind and current.

."1 l-

.c i

e A ArthurD.Little,Inc. 18

l l

i l

during floods and severe storms was developed in discussion with the United States Army Corps of Engineers in Washington, D.C. In particular, Mr. Henry Campbell, Operations Branch, was most helpful in developing this list.

DESIGN MISSILES At the present time, design criteria for environmentally-damaging missiles have not been established by the NRC of by any industry agency.

The missiles considered for the analysis of the door structures at HCGS included the following, based on Bechtel Design Guidelines:

o Telephone Pole 1490 lbs, 13)" round cross-section, impact velocity of 20 mph o Automobile 4000 lbs, 20 ft frontal area, impact velocity of 20 mph 2

o House 4000 lbs, 50 ft frontal area, impact velocity of 20 mph In addition to these missiles, the possibility exists at the HCGS for a wind- and water-driven marine vessels, such as a recreational boat *, impacting the door structures at times of extreme high tidal and wind conditions. Such a design missile was assumed to have the following characteristics:

o Boat 25,000 lbs, 10" round cross-section, impact velocity of 20 mph It should be noted that floating objects are driven by both the wind and the surface current (or wave break effects near shorelines).

The' wind driven speed is between 3% and 5% of the wind speed. For the PMH this would translate to a maximum velocity of 7.5 mph. The current and/or wave effect could add up.to 5.5 mph to this value for a maximum missile speed of 12 mph. A highly conservative value of 20 mph is used in this report.

Since recreational boats were not found to impact the plant in a manner as to compromise plant integrity, they are examined here. All other marine vessels are considered in the probability analysis in the next section.

/h Arthur D.Little,Inc. 19 L

l l

Finally, trees and floating fuel tanks were also considered and found to be structurally waak from a " battering" point of view or having a kinetic energy per unit impact area smaller than the missiles already considered.

DOOR CONFIGURATIONS The size of the doors considered in this analysis ranged from 3 ft by 7 ft for the smallest to 14 f t by 18 f t for the largest. Each door is fitted with double inflatable seals which control the leakage around the door periphery to a constant amount over an extension of the seal from 0" to 7/16". For som:: doors, the double seals are arranged in the door jamb, such that lateral deformation of the door compresses the inner seal, thereby maintaining leak tightness over a wide range of door deformations. For other doors, the double seals are outside the door opening, in the plane of the - doors, with the deors larger than, and overlapping, the door opening. For this geometry, lateral deflection of the door will increase the gap at both of the seals as the door structure pivots about the edge of the door opening.

ACCEPTANCE CRITERIA Two conditions were required for the acceptance of the door design under the postulated missile loadings:

o Structural Integrity of the door structure, based on a maximum 4 permissible ductility ratio of 10, and o Leak Tightness, with a maximum permissible displacement at the seals of 7/16".

ANALYSIS METHODS o Structural Integrity: Each of the missiles was assumed to impact the door structures at the center of the door, and the entire kinetic anergy of the missiles was equated to the strain energy of the door as it deformed under the impact loading. The structural behavior of the door was assumed to be elastic-perfectly. plastic, and the maximum deformation at the center was limited to 10 times the elastic deflection at yield stress (i.e., ductility ratio of 10).

A Arthur D.Little,Inc. 20

o Leak Tightness: A simple geometric model was used to evaluate the increase in the gap at the seals in relation to the deformation of the center of the door. The elastic component was based on a . deflection curve appropriate for a center-loaded beam, and the plastic component was based on the rotation of the door about the pivot at the door edge.

RESULTS Bascd on these analysis procedures, all of the individual doors were determined to meet the above acceptance criteria.

CONCLUSIONS The adequacy of the various door structures on the safety class structures at HCGS in resisting a variety of missiles generated under

. extreme environmental storms and flood conditions has been evcluated.

The design missiles included such obj ects as wind-and-water-driven telephone poles, an automobile, a small house structure, and a pleasure boat.- The door structures were evaluated for their resistance to such missiles for criteria based on the stress and deformation limits of the doors as centrally-loaded plate s truc ture's , and for the displacement j- limitations of the seals around the periphery of the doors which control the leak tightness.

The. results of this evaluation indicated that all the door structures were acceptable as currently designed.

2.7 Site' Impact Probability Assessment Model Based on information contained in previous sections the Hope Creek site impact assessment model is based on the following:

1. All' critical safety related structures at the Hope Creek site are contained in the 750 foot radius circle with the radius connecting the water intake structure and the control room.

2. The water intake structure is approximately 120 feet long and is assumed to be parallel to the shore line.

-T A Arthur D. Uttle, hw. 21

3. For the case of the probable maximum hurricane, it is assumed that all vessels of concern regardless of draft (see Table 5) can potentially strike the intake.

4. For the case of the probable maximum hurricane, the water levels over plant grade are auch that only barges with drafts under 12 feet can enter the 1500 foot diameter circle and therefore be of potential concern. However, to maintain conservatism it has been assumed that all vessels, regardless of draft, shown in Table 5 can enter the site.

5. For the case of the extreme wind events and the model hurricane, regardless of draft, it is assumed that all vessels shown in Table 5 can strike the water intake.

6. Self-propelled vessels which move up towards Hope Creek from the south are tracked for purposes of this model once they are within 10 miles of Hope Creek and should they lose power and steerage they could potentially strike the intake or enter the plant site.

7. Non-self-propelled vessels (barges) moving towards Hope Creek from the south are tracked (for purposes of this model) once they are within 10 miles of the plant.

8. Runaway vessels along a river may ground, capsize, sink, or remain floating free depending upon a complex function of wind and current velocities / directions, vessel characteristics, and river characteristics.

9. Appendix C addresses the probability that a vessel enroute in severe weather would lose power and steering and become a potential missile that impacts the service water intake structure or enters the Hope Creek site It should also be noted that the service water intake structure is approximately 120 feet in length along its single potentially vulnerable surface and is a massive structure constructed of reinforced concrete slabs, most of which are two to three feet thick, with the reinforcement A Arthur D.1.ittle,Inc. 22

. . . . , _ ~ . .. .- . . -- ..

I cover generally . equal to two or three inches. Only two of the four water intake pumps are required to be fully operational to permit safe

. plant shutdown.

i f

4 4

r i

i t

l l

l

[

L 1

1 i

l i-t

A Arthur D.Uttle,Inc. 23 L

l

3. PROBABILITY ESTIMATES Based on 'information contained in ::he previous sections at2 Appendix C it is now possible to estimate the probability of various classes of marine vessels impacting either the Hope Creek site or the j water intake structure during postulated storm situations. The two major areas of concern, the water intake structure and the 1500 ft diameter area _ which constitutes the power block area of the plant are discussed in turn.

3.1 Cooling Water Intake Structure The cooling water intake structure is about 120 ft long and during any of the four postulated storm events the water depth at the structure

-could exceed 28 to 30 ft. The two discrete (and independent) events of

}

concern which are evaluated here are extreme winds and hurricanes. The hurricane is a single discrete event of concern but, with some conditional probability, may occur at one of three levels of severity:

model; intermediate; and probable maximum. Each discrete storm event is examined in turn for probability of impact.

. 3.1.1 Extreme Wind Events Extreme wind events are more common than the model hurricare but at least six hours of high winds blowing from certain key directions is necessary to create a high water situation at the intakes. In the event of the postulated high wind event the water depth at the water intakes could-be as much as 30 feet. It is assumed, however, that all vessels of concern (with draf ts up to 30 or even more) can potentially strike the intake.

.w A Arthur n Little,Inc. 24 ,

Self-Propelled Vessels Which are Potential Runaways Probability of occurrence of the

-3 high wind event: 2 x 10 Number of potential runaway vessels: x 1.3 Probability of intake impact given the vessel loses power and

~0 steering: x 2.1 x 10 Annual probability of a potential runaway impacting the Hope Creek

~9 water intake structure: 5.2 x 10 /yr pon-Self-Propelled Runaway Vessels (Barges)

Probability of occurrence of the

-3 high wind event: 2 x 10 Number of runaway vessels: x 2.0 Conditional prcbability of runaway vessel entering the vicinity of Hope Creek (i.e., within 10 miles) prior to grounding and sinking: x 0.1 Probability of impacting the x 1.2 x 10

-5 water intake structure:

l Annual probability of a runaway I

vessel impacting the Hope Creek f water intake structure: 4.8 x 10 '/yr

~

i l

l A Arthur D.Uttle,Inc. 25

r 3.1.2 Hurricanes A. Probable Maximum Hurricane Self-Propelled Vessels Which are Potential Runaways Probab'ility of occurrence of PMH: 1 x 10 /yr

-5 Number of potential runaways: x 1.3 Probability of runaway impacting the Hope Creek cooling water ~ intake structure given the vessel loses power and steering: x 2.1 x 10

-6 Annual probability of a potential runaway impacting the Hope Creek cooling water intakes: 2.7 x 10 "/yr ~

Non-Self-Propelled Runaway Vessels (Bargec)

Probability of occurrence of PMH: 1 x 10 -5 j, Number of runaway vessels: x 2.0 Conditional probability of runaway vessel entering the vicinity of Hope Creek (i.e., within 10 miles) prior to grounding or sinking: x 0.1 Probability of impacting the cooling water intakes once the vessel is within ten miles of Hope Creek: x 1.2 x 10 -5 Annual probability of a runaway impacting the cooling water intakes during a probable maximum hurricane: 2.4 x 10 -11 /yr r

A Arthur DL Little,Inc. 26

B. Intermediate Hurricane The probability calculation for this case is similar to that for the PMH except the initial probability of occurrence for an

' intermediate or larger hurricane is 5 x 10 /yr. As such, for this

-5 integration process where the ultimate probabilities will be added, the initial term is 4 x 10

-5 /yr.* The final probabilities are as follows:

Self-Propelled Vessels Which are Potential Runaways Probability of a potential runaway impacting the cooling water intake

-10 structure: 1.1 x 10 /yr Non-Self-Propelled Runaway Vessels (Barges)

Probability of a runaway impacting

-10 the cooling water intake structure 1 x 10 /yr C. Model Hurricane Once again the probability calculation is similar to that for the other hurricanes except that the initiating event probability

~ **

is 10 /yr . The probability calculations are as follows:

Self-Propelled Vessels Which are. Potential Runaways Probability of a potential runaway impacting the Hope Creek cooling

~0 water intake structure: 2.7 x 10 /yr Non-Self-Propelled Runaway Vessels (Barges)

Annual probability of a runaway vessel impacting the Hope Creek cooling

-8 water intake structure: 2.4 x 10 /yr This represents the exceedance probability of occurrence for the intermediate hurricane less the probability of occurrence of the PMM.

This value is the exceedance probability of occurrence for the model hurricane less the probability of occurrence for the intermediate hu.tricane .

A Arthur D.Uttie,Inc. 27

3.1.3 Total Cooling Water Intake Impact Probability Adding the above estimated probabilities of impact, the annual probability of a marine vessel on the Delaware River impacting the~ Hope Creek cooling water intake structure is 6.1 x 10-8/yr.

3.2- Hope Creek Site Impact In the context of this study, the Hope Creek site is defined as the 1500 f t diameter -cicele with the center at the control room and the radius extending the cooling water intake structure. This circle includes the entire power block.

Based on the discussion in Section 2 of this report, the plant site can only be impacted provided there . is water on grade. Of the two

, discrete storm events considered here, high winds and hurricanes, the high wind event does not result in any flooding of the grade at Hope Creek. The hurricane event is analyzed at three levels of severity.

The model hurricane does not result in flooding of grade and a more severe hurricane is required to flood the Hope Creek site with 3 f t or more of water depth. As a result, only the intermediate hurricane and the PMH are of concern to the plant site impact analysis.

3.2.1 Probable Maximum Hurricane (PMH)

During a probable maximum hurricane, only vessels with draft of less than 12 feet can potentially enter the 1500 foot diameter plant site but most vessels on the river can potentially strike the 120 foot intake structure. For reasons of conservatism it has been assumed that all vessels can potentially enter the plant site.

l i

l I

A Arthur D. Uttle,Inc. 28

(

y:-

Self-Propelled Vessels Which are Potential Runaways Probability of occurrence of PMH:

-5 1 x 10 fy, Number of potential runaway vessels: x 1.3

-Probability of.' runaway entering the Hope Creek Site given the-

~4 vessel loses power and steering: x 2.5 x 10 Annual probability of a potential runaway entering the Hope Creek

-9 3.2 x 10 /yr Plant Site:~

Non-Self-Propelled Runaway Vessels (Barnes)

-5 Probability o( occurrence of PMH: 1 x 10 /yr Number of runaway vessels: x 2.0 Conditional probability of runaway vessel entering the vicinity of Hope Creek (i.e., within 10 miles) prior to grounding or sinking: x 0.1

. Probability of entering the Hope Creek site once the vessel is within ten miles of Hope Creek: x 3.1 x 10

-3 Annual probability of a runaway entering the Hope Creek site

~9 during a probable maximum hurricane: 6.2 x 10 /yr 3.2.2 Intermediate Hurricane The intermediate hurricane can be exceeded with an annual

-5 probability of 5 x 10 /yr. The probability of the more severe PMH is 1x 10 -5/yr. As a result, for purposes of integrating all hurricane events the likelihood of a hurricane at least as severe as an

-S intermediate hurricane but less severe than the PMit is 4x 10 jyy, This hurricane, should it occur, could result in the grade at Hope Creek being flooded with water to a depth of greater than 3 ft. Clearly, most A Arthur D Uttle,Inc. 29

large draft marine vessels would ground prior to entering the site. For reasons of conservatism, however, it has been assumed that all vessels can potentially enter the site. With the above factors and a probability calculation similar to that for the PMH except for a larger initiation event probabilit'y, the estimates for impact probability are as follows:

j Self-Propelled Vessels Which are Potential Runaways Annual probability of a potential runaway entering the Hope Creek

-8 plant site: 1.3 x 10 jy,

! Non-Self-Propelled Runaway Vessels (Barges)

! Annua'l probability of a runaway

-8 entering the Hope Creek site: 2.5 x 10 7y, 3.2.3 Total Hope Creek Site Impact Probability M

Adding the above estimated probabilities of impact, the annual

, probability of a marine vessel on the Delaware River impacting the Hope

-8 Creek site itself is 4.7 x 10 /yr.

3.3 Summation of Probabilities Based on.the above, the combined probability of the service water intake structure at Hope Creek being impacted by any vessel for any

~

postulated storm condition is 6.1 x 10 occurrences / year.

Similarly, the combined probability of the Hope Creek site being

-8 impacted by. any vessel for the case of a hurricane is 4.7 x 10 occurrences / year.

3.4 Conservative Nature of the Probability Estimates The probability estimates presented in this report for the combined probability of a vessel impacting the service water intake as well as the combined probability of a vessel entering the Hope Creek site during storm conditions are conservative in nature. Each combined probability is composed of several initiating events and conditional events and many A Arthur D Uttle,Inc. 30

of these sub-elements are overestimated. The net result is that the combined probabilities of interest are also conservative, overestimates of the likelihood of occurrence. Some of the conservative estimates include the following:

1. The lines fitted to the Frechet distribution plots of the extreme wind speeds were drawn in a conservative manner. The higher observed wind speeds were given more weight in the distribution.

2. The large width of the 79-170 (*Az) sector used as a persistence criteria for the six-hour wind specd analysis.

3. For the typical or model hurricane producing very serious surge effects, the intensity, course and transport speed would all have to be synchronized with the normal tidal oscillation.

Therefore the value of 10 is conservative by a significant amount, probably by half an order of magnitude.

4. The great rarity of the PMH is emphasized by the fact that between 1899 and 1982, no storm having the calculated maximum wind value of 142 mph or greater (NOAA Classes 4 or 5) has made a landfall anywhere north of Cape Hatteras. A

-5 conservative probability of occurrence of 1 x 10 /yr has been utilized.

5. The surge calculations are based on steady-state conditions which result in an overprediction of build-up of water at the Hope Creek site. Such ideal steady-state conditions do not occur in actual storms and such storms would not cause the degree of high water predicted and would in actuality result in lower water levels than has been used in this report.

6. The well developed United States Coast Guard plan for Delaware River traffic under severe storm conditions should preclude any large vessel from becoming a runaway. Yet it has been conservatively assumed that some runaways of significant draft would be found during the storms of concern.

A Arthur D.Little,Inc. 31

7. In the event of an intermediate hurricane, the water depth on grade at Hope Creek would only be about 3 ft. In spite of this fact it has been assumed that the total population of potential marine vessel " missiles" could enter the site.

8. In the event of a PMH, the grade at Hope Creek could be covered by about 12 ft of water. Vessels with draft in excess of 12 ft would not be able to drif t onto the site. Yet, in order to be conservative it has been assumed that all runaway vessels, regardless of draft, are a potential concern and could enter the site.

-9. Similarly, for the extreme wind events and model hurricanes it was assumed that all vessels regardless of draft would be of concern and could potentially reach the service water intake structure. In actuality many large vessels would ground or sink prior to reaching the structure.

10. The damage potential of missiles is related to its kinetic and as such depends on the square of the velocity. Although it is highly unlikely that a floating missile would exceed a speed of 12 mph, a value of 20 mph was used in this analysis.

This introduces a degree of conservatism of a factor of 3 in the damage estimates.

A Arthur DL Little,Inc. 32

f

[n s

4. REFERENCES

1. " Analysis of Potential' Effects of Waterborne Traffic on the Safety

-. of the - Control . Room and . Water Intakes at ' Hope Creek Generating Station", Arthur D. Little. Inc. - Report :77289 to Public Service Electric and Gas Company,' September 1974.

.2. '" Hope. Creek = Generating Station E::treme Event Site Flooding Meteorology", Meteorological Evaluation Services, Inc., to Public Service Electric and Gas Company,' July 1984. Also see Supplemental Report, August 1984.

3. '? Storm Surge Calculations for Hope Creek Generating Station", Dames

& Moore, Inc., to Public Service Electric. and Gas Company, July 1984.

4. " Monitoring LNG and LPG Shipping and Construction Activity in the Vicinity of the Hope Creek Generating Station", Arthur D. Little, Inc., Report 83202. to Public Service Electric and Gas Company, October.1979.

5. "An Update of the Analysis - of Potential Effects of Waterborne Traffic on , the Control Room and Water Intakes at Hope Creek Generating Station", Arthur'D. Little, Inc., Report 88536 to Public Service Electric and Gas Company, March 1983. _

6. " Waterborne commerce of the United States", Part 1 - Waterways and Harbors - Atlantic Coast, 1981, Department of the Army, Corps of Engineers.

-7.- ." Marine Casualty Computer Data", Office of Merchant Marine Safety, United States Coast Guard, Washington, D.C., 1968-1982.

8. 'EPRI HP-440, " Full Scale Tornado-Missile Impact Tests", Feb. 1978.

9. "A Study on Collision by an Elastic St.im to a Side Structure of Ships" by Y. Akita and K. Kitamura, Journal of the Society of Naval Architects of Japan, Vol. 131, June 1972.

10. " Search and Rescue" Manual prepared and utilized by the United States Coast Guard. kevised Edition 1976.

J A Arthur D Uttle,Inc. >

33

u

's-

__t' b

i 1

Y APPENDIX A f j COMUNICATIONS' WITH U.S. COAST GUARD j CAPTAIN OF T E PORT, PHILADELPHIA l

1; i

+; s a

5' J

i k

i s

t s

A Arthur D, Little,Inc. 34

_4 Acorn Port Cemhrtege. Messeehusetts 42140 687 864 8779 Teles 9214M A Arthur D.Little,Inc.

April 12, 1984 Lt. Robert Francis United States Coast Guard King and Cumberland Streets G1cuosster City, New Jersey 08030 -

Subject:

River Traffic and Contingency Plans

Dear Lt. Francia:

] Many thanks to you and your colleagues for taking time to meet with me on 1

April 10, 1984. As you recall, we are under contract to Public Service

Electric and Gas Company to perform a study which requires us to further our understanding of the marine vessel traffic in the Port of Philadelphia /n=1==re River and to learn about the contingency plans that

.have been formulated in case of hurricanes. 'this wm .--A.cs will confirm the details of our conversation.

Marine Vessel Traffic. i, o There are typically no more than 50 large vessels in port each day on evenen

:ial business. 'these vessels include oil tankers, container ships, etc., which have drafts of w mimately c 18-40 ft. .

e As thre are 20 tug conpanies in the Philadelphia area which operate an average of four tugs each, there are approximately 80 tug boats. They have a draft of 12-13 ft.

e It is difficult to estimate the exact nunter of barges that

, are typically found on the Delaware River on any given day. A rough estimate is approximately 150 on an average day. While the draft of these barges can range from 3 to 35 ft, the

majority have a draft.of 4-5 ft when anpty and 12 ft when

-loaded.

e 'there are about 130 marinas 'on the Delaware River with an average of 100 moorings wach. Thus, it is estimated that there

' are approstimately 13,000 smaller recreational boats with lengths less than 45 ft and an average draft of 3-4 ft.

I I peoneets Madrid Me Psein W heden

! Havotan Path T$a Rio de Janeirn Imrunsa 35 1.nndon g, 3,,, ,,, 3,, y,,,, ,,, w ,,n6,,,,,

=

A ArthurEluttle,Inc.

Lt~. Robert Francis United States Coast Guard .

Apn,1 12,1984 w

e 'there are 16' anchorages on the Delaware River The average.

raaber of 6 hips /per anchorage is variabis and a function of the size and ty the =4 % .pe of the ships. Selection of anchorages is made by

. jw M Cont _4!MLenM Planning e Ikuvicenes in' the Port of PM1mklphia are typically "not that trasmtic" as their comtercloclorian winds " lose their punch" on ,tas menrLitory travt& over land as they Weech philadelphia. 'Itns,-vessels up the river are better y.we then those near the mouth of the Delaware River or those out to sea. With at least one day 8s notice for a hurricane, there is ,

' - adequate times to seek a protected berth.

e 'the contingency plan for t$ Port of Philadelphia, which includes a plan for heavy weather, is presently whing a . ;mejor revisions and updates. It will be completed during the next year. f'

,- e At present,Ithere are gorgency procedures for heavy weather

.g, y that states,'"Have boats renoved frem the water or anchor in a safe anchorage area es di.Gi by the Consending officer."

Furthermore, there arte yshes for reconnendations made on radio broadcasts. Wtan winds are in excess of 25 knots, broadcasts ake made that reconmend a live bridge watch and a 306 starmby fs e the main propulsion machinery. With 4'

winds in excess of 0 knots, the latter should be on immediate stanty. On an avr it adecificJasis, telephone calls might be

?- made to reonmend , doubling'up."" Similarly, the Captain of the Port might'make urther rect.mmndations on the location, monning or securj y of vessels if the situation warranted it.

/

e- #4t is kirly rratine that be ueUal anchor length of 5 times

' the depth i.? .Jetended' to */ 'thnes the depth in a storm

, situation, wh.ch is particuhely inportant in this area where there is a ecft botten.

//

/

~

, r i i-

~

r /

.*'fy

.6 a

?

• J 36 ig

-.r---c ._,,,..----e,e%,,,,,,---,-.w,w ,y

A Arthur n Uttle,Inc.

Lt. Itabert Francis

• thited States Coast Guard ,

April 12, 1984 Page 3 We would appreciate your confirmi$g'the accuracy of the information presented in this document in a letter to be sent to me.

Again, our sincere thanks for your assistance.

Yours truly, Mi h Marian H. Icng Ehclosure e

e 9

9 37

r y .:.

$ ,)[b ' . Ctptein of the Port, U.S. Coact Guard Basa

- , . 7,- . '" ' / W 'f Philadelphia Gloucester City, NJ y;-N I; ;; a j Q.' / 08030 Cart CCcm ib, i .

16000

~

Ms. Marian H. Long Arthur D. Little, Inc. MAY I 8124 Acorn Park

-Cambridge, MA 02140

Dear Ms. Long:

As requested in your letter of 12 April 1984, this letter will clarify and confirm the information exchanged during our meeting on 10 April 1984.

Marine Vessel Traffic

1. Excluding the vessels that would be in the Delaware Bay and the C & D Canal, the 50 ship average is a reasonable maximum figure.

2. Of the estimated 80 tugs homeported in the Philadelphia Port, a varying percentage will be away from this port on business, and a number will be in drydock at any given date. Additionally many of these tugs are relatively small, low powered vessels that would have a very limited role in contingency planning.

3. .The number of barges on the Delaware River includes oil, chemical and construction barges.

4. -There are approximately 60 marinas on the Delaware River with an average of 75-100 moorings each. This would make an average of 4,500 recreational boats in the area, excluding trailered boats.

Anchorages l'. The larger vessels would generally use the Marcus Hook or Mantua anchorages.

On a typical day there would be between 6 and 12 vessels utilizing the anchorages.

Hurricane Contingency Planning

1. Although the statement regarding hurricanes is generally correct, I would l

not count on storms losing their punch for contingency planning purposes. As l

I' recall, the strongest winds ever recorded in the continental United States area'were at Mt. Washington NH in the September 1938 storm. This storm went inland at Connecticut.

2. The statement about the heavy weather plan revisions is correct.

, r f

- ' 38

16000

. MAY I 81984

3. In the third statement, the sentence "have boats removed from the water er anchor in a safe anchorage area as directed by the Commanding Officer" p;rtains strictly to our own bases boats. The remainder of this statement which-pertains to commercial vessels is correct. It should also be noted that under existing regulations, the COTP may require vessels to use two or more anchors when deemed necessary.

4. The statement regarding anchor chain length is correct.

Sincerely, W1 D. B. CHARTER JR.

Captain, U.S. Coast uard Captain of the Port, i Philadelphia 39

APPENDIX B COMMUNICATION WITH PHILADELPHIA NAVAL SHIPYARD A Arthur D.Little,Inc. 40

Acuro 14rk Cambridge. Massachmetts 02140 617 864 5770 Teles 921436

/ i Arthur D. Little,Inc. .

April 12,1984 Lt. Richard Oftedal Attention: Code 810 Philadelphia Naval Shipyard Philadalphia, Pennsylvania 19112 .

Subject:

Naval Traffic and Severe Weather Ccmtengcy Plans i

Dear Lt. Oftedal:

Many thanks to you for taking tin.a to meet with me on April 10, 1984. As 1 you recall, we are under contract to Public Service Electric and Gas i Chapany to perform a study which requires us to further our understanding of the marina vaaa*1 traffic in the Port of Philadelphia / Delaware River

^

and to learn about the contingency plans that have been fornulated in case of hurricanes. This corre W will confirm the detailn of our conversation, e The Philadalrhia Naval Shipyard has one active ship, which is the USS Patterson, a 438 ft frigate with a draft of 25 ft. It travels up and down the Delaware River twice each month.

e There are typically four or five ships in for overhaul. The

length of stay ranges fran approximately seven months to over two years. Frigates and cruisers that have a draft of 22-29 ft are serviced in approximately one year. It is taking two and one-half years to service the USS Forrestal, an aircraft carrier with a draft of 37 ft. In addition to entering and leaving the Port, the ships undergoing service make one trial run out to sea and back.

i e There are 29 small craft, such as tugs and barges, in addition

to camels which service the overhaul efforts.

1

e. In the event of .a hurricane, the USS Patterson, which is i typtrally secured with six standard mooring lines, would possibly go out to sea. If it did stay in port, extra mooring lines would be used, an anchor would be dropped on the foot and a chain might be used to secure it to the pier. The service crafts would be moved off the windward sides of the piers and secured.

L

+

Brussels Aladrid San Paulo Wiesbaden Hoessee Paris Tobo 41 Leadon itso de Janeira Torimin Las Angeles San Franciwn Washingiam

d Arthur D.Uttie,Inc.

l Lt. Richard Oftedal Philadelphia. Naval Shipyard April 12, 1984 Page 2 e The Philadalphia Naval Shipyard is no longer a construction

• yard. " Newport" class LST's (LST-ll79 series) have not been built there since 1973. "Desoto County" class IST's (LST-1173 series) and " Anchorage" class ISD's (ISD-36 series) were never txtilt there. Construction of shnarines ended before 1970.

'Ibese types of vessels are not present at the base nor do they '

oczne in for overhauls.

e There are three cruisers, seven destroyers, and several sutznarines mothh=11ad at the base. -

We would appreciate your confiming the accuracy of the informaticm presented in this 6 % in a letter to be sent to me.

Again, our sincere thanks for your assistance. --

Yours truly, .

, Yk&

Marian H. Icng MIL /rs l

d o

l l .

^

l l

l-

, s I

e 42

~

<<IR

, p' ' p,. hQ- DEPARTMENT OF THE NAVY 9,, . ' P*iLAOEAiA N AVAL $=19YAR3 M

g PM LAOCLPH A. PA 1932 lN REPLY REFE9 TO 1 Code 810:RT0:jlr "4

/ 24 April 1984

. Arthur D. Little, Inc.

c- Acorn Park ,

Cambridge, Massachusetts 02140 Attn: Ms.. Marian H. Long

Dear Ms. Long,

I received your letter of 12 April 1984 concerning Naval Traffic and

-Severc Weather Contengency Plan for the Philadelphia Naval Shipyard.

> The information as stated in your letter is correct.

,- : If I can b'e of further assistance feel free to contact me.

] Y chard T. Oft al

, LT, USN -

Operations Officer

'h S

T J*

6 A

? '

43

'4 APPENDIX C POISSON MODEL TO ASSESS CONDITIONAL PROBABILITY OF IMPACT GIVEN A MARINE CASUALTY There are two basic situations -in terms of marine casualty which are of concern to this analysis.

First, a manual self-propelled vessel such as a tanker, dry cargo ship or a recreational boat could be moving north on the Delaware River from the south of Hope Creek. If at some point it loses power and steering (a remote possibility) it would move under the action of wind

,and waves to the east bank of the Delaware. If the direction of wind and currents are just right, the vessel would strike the water intake structure (a 120 foot target) or in the event of the probable maximum hurricane, enter the plant (a 1500 foot target). The latter requires that the vessel have a draft of under 12 feet.

The second situation, which a special sub-set of the first, more general case, involves a vessel which has already lost power and steering (or has none to start with). The classes of vessels of concern to this study which fit this category are non-self-propelled barges and

- unmanned recreational boats. Once again, if they are coming from the south of Hope Creek and get sufficiently close to Hope Creek, the wind and surface currents can, in some cases, cause these vessels to either impact the water intake structure or in the case of the probable maximum hurricane enter the Hope Creek site. -

In modeling this situation it should be noted that under the postulated storm conditions vessels moving north from the mouth of the Delaware River are quite likely to ground substantially before they reach the vicinity of Hope Creek. For purposes of this analysis,

" vicinity of Hope Creek" is defined as distance up to 10 miles south of Hope Creek. Based or. considerations of wind and current directions during the postulated storms and the geometry of the river the chances

-of a runaway, unmanned vessel approaching within 10 miles of Hope Creek without a prior grounding is less than ten percent. Should a runaway A Arthur D.Little,Inc. "

4 vessel or a self-propelled vessel approach the vicinity of Hope Creek, the probability of the vessel striking the water intake structure of entering the site can be estimated utilizing the model developed below.

With respect to Figure C.1, the probability that a vessel underway moves a distance x north without loss of power and steering and then loses d' precisely at x is given by:

=

P Ae where A is the probability per mile of simultaneous loss of power and steering.

Should this fcilure occur, the vessel will strike the target of concern (either a 120 foot diameter circle or a 1500 foot diameter circle) only if it moves within the sector described by the angle 20.

Geometric considerations indicate that 0 = tan"

!(10-x) + (W+R) - R where R = radius of the target W = half width of the river ,

If all movement directions were equally likely the overall probability that a vessel will enter the vicinity of Hope Creek, lose power and stecting and strike the intake is given by Q where 0

A -Ax -1 R di q ,

0

/(10-x)2 + (W+R) -R The integral Q can be evaluated for the Hope Creek situation where:

l R = 60 feet and 750 feet W = 1 mile

-5 A = 10 M le A Arthur D. Little,Inc. 45

AN 2W Hope Creek site Target Radius R x=10 -

R = 750 ft for the site

= 60 ft for intake structure x=x _

-1 6= can R

[(10-x)+(R+W)-R l

Delaware .

River x=0 __,________

FIGURE C.1 IDEALIZED SCHEMATIC OF THE DELAWARE RIVER, HOPE CREEK AND THE IMPACT GEOMETRY l'

A Arthur D.Little,Inc. 46

The failure rate A for simultaneous loss of power and steering in a c

self-propelled vessel is based on historical data contained in Reference f1 and 7.

With the above parameters it can be shown that:

If a self-propelled vessel reaches a point 10 miles south of Hope Creek and is moving north, then its probability of simultaneously loosing power and steering and impacting Hope Creek is:

~

2.1 x 10 for the water intake structure

~0

] 2.5 x 10 for entering the Hope Creek site

'However, the above calculations assume that once power and steering is lost the vessel is equally likely to move in any direction. In fact, for any of the three postulated storm situations the vessel is at least S to 20 times more likely ' to head towards the targets of concern.

Accounting for the relative sizes of the targets, it is conservatively assumed that corrections need to be made for the above estimates

-involving a factor of 10 increase in the intake impact probability and a factor of 100 in the site impact probability.

It is concluded,- then, - that in ' any of the postulated storms, the likelihood of a self-propelled vessel arriving in the vicinity of Hope Creek, simultaneously losing power and steering and striking the water int'ake structure is 2.1 x 10 /-6vessel. The related probability of

-4 entering the Hope Creek site is 2.5 x 10 / vessel under 12' foot draft.

4 In the -special case of non-self-propelled vessels such as barges and 1 recreational boats, based on their location of origin, size and river geometry, : there is a high probability the vessel will ground and/or sink prior to entering the zone within 10 miles of the Hope Creek site. Nonetheless, if the vessel enters the vicinity of Hope Creek, it is very likely to ~ head. in a northeasterly direction under wind and current action in the postulated storms. If it enters the vicinity of Hope Creek it is far more likely to drift eastward and ground several

. miles.to'the south of Hope Creek rather than hit the targets of concern.

A Arthur D.Little,Inc. 47

, . _ , _ . . _ - - _ ~ , . _ . - _ _ . _ _ _ _ _ . _ , . _ _ ~ - . . _ . , _ _ _ _ . _ ,

Ut'ilizing this fact and the impact integral discussed earlier overall.

probabilities for non-self-propelled vessels were estimated.

The findings are that in the event of the postulated storms the likelihood of a non-self-propelled vessel already within ten miles of

-5 Hope Creek striking the water intake structure is 1.2 x 10 / vessel and the corresponding probability for entering the Hope Creek site is 3.1 x

-3 10 / vessel.

i i

I~

l t

i L

I i

A Arthur D Little,Inc. 48

- _ . _ . . . , _ . . . _ _ _ , _ _ . _ _ _ . _ - - _ _ .-.__ _ _ __ __._--_- -_ __ _ --- ..--__,.__ ___ .___ ..