ML20097F690
| ML20097F690 | |
| Person / Time | |
|---|---|
| Site: | Hope Creek  |
| Issue date: | 08/31/1984 |
| From: | Castelli F, Matthew Smith METEOROLOGICAL EVALUATION SERVICES, INC. |
| To: | |
| Shared Package | |
| ML20097F667 | List: |
| References | |
| NUDOCS 8409190071 | |
| Download: ML20097F690 (19) | |
Text
- - _ _ - _ _ _ _ - . . . . - . _ . _ _ _ - . _ _ _ . _ _ . .
l l
l J
i HOPE CREEK GENERATING STATION EXTREME EVENT SITE FLOODING SUPPLEMENTAL REPORT ON METEOROLOGY i
AUGUST 1984 !
i i
i J
l r i
l
! l r::a:: .;c :;
, ::+ a *: ;
! ' N U I ' =' !
0409190071 040917 MatronotocicAt avAtuAtlON BEHVICES,INC, i
', PDRADOCK05000g c .
t
r-S c.
7,. ,- .MES liOPE CREEK GENERATING STATION EXTREME EVENT SITE FLOODING SUPPLEMENTAL REPORT ON METEOROLOGY August 31, 1984 Prepared by:
Meteorological Evaluation Servicon, Inc.
Amityville, New York Maynard E. Smith Frank P. Cantelli Prepared for Public 11ervico Electric l. Unn Company METEOROLOGICAL EVALUATION DERVICES, INC.
134 n l10 A D W A Y , A M i f Y V i l L l' N O W Y O f4 M 11101 2 7 U U e S 10.n 91 3 3 0 %
August 31, 1984 NOPE CREEK GENERATING STATION EXTREME EVENT SITE FLOODING SUPPLEMENTAL REPORT ON METEOROLOGY I
- 1.
SUMMARY
On July 31, 1984, PSE&G staff members and consultants met with the NRC staff in Bethesda MD, to consider various aspects of possible-flooding of the llope Creek site. During the courne of that discussion several points were raised about the meteorological estimates
. involved. This report responds to thoco points, and provides supplementary data that should be helptul in evaluating the probabilities of austained high wind events. For clarity, it also restaten come of the reasoning used in our initial report on the subject (July 6, 1984).
Two reports that we had not used in our evaluation of hurricano charactoristics and probabilities wore brought to our attention by the NRC staff. We have reviewed those reports in detail, and find that they do not ,
substantially alter our conclusions, largely because they are based on annontially the came data that we ourselvon utilized in analyzing the problem. Ilowever, the reporta do contain indupondent otatistical analynon that wo havo unod to chock our own findings.
We have alno looked into the probability of having nua-tained winda of 80 mph or greater from the Sl: aoctor, an would bo needed to creato a nurge of 3-4 ft. abovo plant grado at the nite. In preparing thin ntudy, wo havo ME f t;Of tot o(HC AL KVALUA floN 868MVC88,1NC,
made use of the original chart records, which permitted-us to estimate the fastest mile winds during the periods in which we observed the maximum six-hour and maximum one-hour winds over an 11-year period at the site. -
These estimates, coupled with some of the detailed sta-.
tistical estimates in the documents recommended by NRC, permitted us'to tie the peak wind data to that of longer-term averages.
Our conclusions are:
A. A typical hurricane, designated as the "model" hurricane in our July 1984 report, may be expected to cause a major tidal surge in the area with an
. annual probability of 10-2, B. _The annual probability of the PMH (probable maximum hurricane) moving into the-site area on a trajectory capable of causing a very large tidal surge is 10- 5, C. The annual probability of sustained (one-hour or longer average) winds of 80 mph or. greater from-the SE sector, caused by any meteorological' event, is 5 x 10-5, s
II. REVIEW OF THE REFERENCES SUGGESTED BY NRC The' initial MES report to PSE&G on the' subject of maxi-mum winds at the Hope Creek site (July 6, 1901was based 'in-part on the original hurricane data summarized
- in '" Tropical _ Cyclones of the North Atlantic Ocean". 2 The two documents suggested by NRC were also based 3
- largelygon these data,_and it' is not surprising that their conclusions and ours are in ~ substantial agreement.
~
'g.
- ) -
~
METEOROLOGICAL EVALUATION SERVICES,INC. g, an _
-- + _ . .-. - _ ._
The1 major conclusions that one can draw from these re-ports (NWS-15 and NBS-124) are: p e That because of the relation between the configuration of the coastline between Cape Hatteras and New York City and the typical recurving hurricane. tracks, there, has been only one tropical storm directly entering the coastal area in the vicinity of Delaware Bay in such a way as to cause a large surge at Hope Creek during the 1871-1983 period. , e That the small number of hurricanes which have af-fected the Delaware Bay area have passed over land be. fore-reaching the area, and they have also lost contact with the very warm moist air that constitutes their source of. energy. Consequently these storms have been weakened significantly by the time the Hope Creek' area has been affected. e The authors of both documents have recognized that the small n' umber of storms affecting the area make. it difficult to develop reliable statistics. Pertinent points covered in the documents are summa-rized below: NWS-15, Some Climatological Characteristics of Hurri-canes and Tropical Storms, Gulf and East Coasts of the United States - Ho, Schwerdt and Goodyear, May 1975. Page Comments 40 The authors, in describing the high minimum central barometric pressure at the Delaware Bay METEOROLOGICAL EVALUATION SERVICES,INC.
w- - t Page' Comments location-(an indication of'a weak storm), note
. that the-shape of the coastline plays an impor-tant'part in the establishment of this maximum.
They identify one weak hurricane as having moved inland on the Jersey Coast in 1903, and one tropical storm as having penetrated the Delmarva Peninsula in the last 74 years. (Only the latter was on a track that could have pro-duced a_ major surge for Hope Creek). 15 Figure 2 shows a minimum in the number of hur-ricane entries /100 years /10 nautical miles at
. .- location 2400-(the entry to Delaware Bay).
150 Figure 19 provides estimates of the probability of the radius of maximum winds exceeding cer-tain. values at location 2400. The authors also
~ note that these data are' based on only seven known values from Virginia northward.
58 Figure 23 shows the probability of forward
, speeds estimated for various locations.
NBS - 124, Hurricane Wind Speeds in the United States-E Batts,_Cordes,. Russell, Shaver, Simiu, May 1980. p_. This document is based on actual hurricane climatology
.- - -and statistics only insofar as the frequency of occur-rence of hurricanes in various locations is concerned.
The remainder of the study was developed from theo- _ retical hurricane characteristics and a Monte Carlo METEOROLOGICAL EVALUATION SERVICES,INC. c
l simulation of their possible effects at various shore locations.- It is important to keep this in min in using the document. Page Comments 10 The authors note that research results con-cerning the ratios of hurricane wind speeds corresponding to different averaging times do not appear to be available. They therefore suggest using ratios obtained from non-tropical storms. 27-28 It is noted that conclusions & bout effects north of Cape Hatteras are not very reliable. 14-15 Figures 5 and 6 give fastest mile estimates at various locations near the coast and 200 km inland, without regard to wind direction. F ig '. In the Appendix, estimates of fastest mile A-9 -speeds as a function of wind direction are.
-given for location 2400 (Delaware Bay).
III. EXTENSION OF THE ANALYSIS OF HURRICANE WIND DATA Because of the specific point developed in our discus-sion with the NRC with respect to the possibility of flooding.the site to a level 3-4 ft. or more above plant
-grade, it was necessary to prepare additional estimates of the probability of wind speed-direction combinations that might cause such an event. Also, it appeared to be 1 desirable to make the best use of the more elaborate statistics in NBS-124 and NWS-15 to develop a second METEOROLOGICAL EVALUATION SERVICES. INC.
m -
+
. estimate of maximum wind events for comparison with that' prepared ~ earlier by MES. The steps involved in develop-
-ing this estimate are described below:
' JL ' Ratios of Fastest-Mile and Sustained Winds Both in the above references and in others, sta-tistics and estimates of the-short-term peak winds are commonly given in terms of either one-minute or fastest-mile wind speeds, but these. data are seldom joined with information about the longer-term aver-ages that are critical to the generation of a major surge in water levels. On page 10 of NBS-124, the authors note this deficiency and suggest that the best. solution is to use peak / average ratios based on
.non-tropical storms.
Although short-term peak data were not extracted for the routine summaries-of the Salem and Hope Creek site records, it was possible to determine them from 4 the original charts,'using the chart time scale indications and the ink trace width as guides. Table 1 summarizes these data, showing the ratios between fastest mile and the sustained wind averages of one hour and six hours. -As is apparent in'the table, the ratios tend to become larger with in-
~
creasing wind speed, and' Figure 1 shows this tenden-cy very clearly. On the basis of this summary we have chosen a.value of 1.7 for the FM/1-hour ratio
~
and 1.9 for the FM/6-hour ratio to be used in the subsequent analysis. METEOROLOGICAL EVALUATION SERVICES, INC.
q 4 TABLE'1-RELATION BETWEEN PEAK AND AVERAGE WIND SPEEDS (79* - 170* Direction' Sector) ft. Winds (MPH) Ratios [. Fastest . l' l FM-l Mile . 'One-Hour 'Six-Hour FM (6H) 1H. 6H Date 'FM (1H) 24 23.0 1.29 1.35 6/5/69 31
.29 23~ 20.8 1.26 1.39 4/30/70 '
22 17 15.8 1.29 1.39 6/18/71 41 30 21.8 1.37 1.88 11/26/72 39 25 21.7 1.56 1.80 2/24/77 59 27 23.7 2.19 2.49 1/26/78 1/24/79 52 30 30.0 1.73 1.73 t 1.76 1 44 28 25.0 1.57 3/21/80 33 23 20.1 1.43 1.60 5/11/81
-42 26 24.3 1.62 1.73 4/3/82 50 30 26.5 1.67 1.89 3/27/83 (Data obtained from the Salem - Hope Creek 30-ft. instrumentation)
~
RELATION BETWEEN FASTEST MILE AND 1-HOUR & 6-HOUR AVERAGES 2.6 O 2.4 - ~o _ FM 6H
- 2.2 - e-RATIO ,
2.0 - FM o o [ } - o O l(AVERAGE LONG-TERM);1.8 o *@ 1.6 - o e e e f 1.5 - l 1.4 O O
. *?
1.2 -
' ' ' ' I i r 1.0 60 20 30 40 50 FASTEST MILE WIND SPEED (MPH)
Figure 1
r -y 4 B. Estimates of the Fastest Mile for the Delaware Bay Area The National Bureau of Standards study (NBS-124) includes an Appendix containing their best estimates of fastest-mile wind speeds as a function of wind direction and return period. A separate plot is given for each of the sequence of locations ranging from southern Texas to New England. For clarity we have redrawn the plot given for Mile Post 2400, the entrance to Delaware Bay (Figure 2). From this plot we have obtained the maximum fastest-mile data for
'the critical 79-170* sector, and these are listed
. below: Fastest Mile Wind Speeds (79-170* Sector) Return Period Wind Speed (years) (mph) 100 70 50 57 10 25 A comparison between these fastest-mile estimates and the observed fastest-mile data-in Table 1 seems to show an inconsistency, because the values ob-served in-our 11-year period-exceed the NBS-124 estimates for both the 10-year return period and the 50-year return period. This result is not actually
' inconsistent because the observed data reflect the effects of extra-tropical storms rather than hurri-canes, and at more modest wind speeds one would METEOROLOGICAL EVALUATION SERVICES, INC.
_ . _ . . . . _ . ,. . , _ ~ ,. _ - . - - _ _ _ . _ .
120 _ FASTEST MILE WINDS AT MILE POST 2400 ( DELAWARE BAY ) 100 - FASTEST l-MILE 80 - SPEEDS 100 Yr. Ret. Period (MPH) 60 - ~.~. . '/ p ,.- s
, ,/.-
/
40 - 50 Yr. Ret. Period e --- ,,
. N '
/
10 Yr. Ret. Period m' 20 -
/\
s i I t ! l %J t l t t I I t I ! 0 90 180 270 360 WIND DIRECTION (DEGREES) ( Obtained from NBS-124, Appendix A ) Figure 2
expect the normal, non-tropical, coastal storms to dominate,the statistics. Figure 9 of NBS-124 (re-Lproduced here as Fig. 3) shows a similar result, although in the case they cite, the crossover be-tween extra-tropical and hurricane winds occurred at a lower return interval. Because.of our interest in winds at the extreme end of the distribution, it is important to extrapolate the data obtained from the NBS report to return periods of several thousand years. One could use any one of a number of statistical techniques to accomplish this, each of which is subject to the same data limitations. However, extrapolating the plot:used in-Figure 3 is a reasonable approach, and
-the results obtained will be similar to those found with other statistical methods. Figure 4 shows this extended plot, with the three wind speed estimates made by the NBS staff serving'as the basis for the extrapolation. From this plot, one derives fastest-mile speeds, from the critical direction sector, of 130-150 mph at a 20,000 year return period, or an annual probability of 5 x 10-5 C. Comparison with Earlier Analysis In our July 1984 report, we examined two. hurricanes that might affect the site, and estimated their probaility of occurrence. One, a so-called "model hurricane" is a storm of relatively modest attri-butes that one can reasonably expect to affect the site in the forseeable future. On the basis of reference 2 and an exhaustive study of the litera-ture on hurricanes affecting the United States METEOROLOGICAL EVALUATION SERVICES,INC.
)
e -
!y
@ 9 9 6 e 4 e e, e N O ;
** e e e g g ,p g ,
S @ m ns .e
- O N. e naJ f t I h t I f I a t t a *
- b. ^
W 4 k Q. O e v m 03. ci G 0 kd O. m C nJ
\ I > I C # g S i- \ 2 ?. '; m e
> w 3
\. . 2 u a z e a:
U b q h
+
kt M y*N-\ . j d a
>- \
- e e u
o ._
'. I N e a . c
- a I'
- u
> m
. .a e-w w -
\-\-N.\
.*I
, w w
c as l
@ o R m o
2
\
- g M
\. . w O
0
= g-
- u 4 t
\ G 3 3
'U.
w
- w F p e4 u k sd a
. a. .g -
m .
; . T 7 \,
\ ._ . a _ W h U .. ' '
M d
\.
- M d
\. t. .4
\.'. .a to o
.g G o g 9
. g o.
.\ H
-N, 8, e
. A
\ h h_ @ H y
e o e e e ru < > e e en e
> Cs lhp e S N rt 4 , G 8T. N 3 O us '9 @ M % r*J 4 3 @ @ 9 g E LA ? CQ m G4 r*J 4/ e Sa & q td & :M @ C% La i ts % i e e, f.e C4 - d% ,
*J
. , D. , LM [*
A. CC. LA. d A. . LA. . W N a' w f 1 1. i i i f I O, a J H M f. x w Q Q O Q Q Q O G G L;J
* @ Q O O C3 O O @ G 3 . . . . . . ,
o c. e- m v- n tu - e - 8 u g he
.h sa 3
tb Y ,4 A Ehe Ll...
REDUCED VARIATE EXTRAPOLATION OF NBS FASTiST - MILE DATA RETURN-PERIOD
#R*fUENCY SE SECTOR - DELAWARE BAY
' e
-2E 10.00 3 10000
- 999800 [ [- 5000 8'00
/ -
2000 ' 7.00 ,999000 p " - 1000 6.00 v PROBABLE REGION
,995000 - OF PEAK WINDS -
200 100 50 7980000 /G 3.00 20 i 2.00 l 5 l 1.00 .
,500000 2
[ 0.00
-1.00 20 30 40 50 60 70 80 90 100 110 120 130 140 FASTEST - MILE WIND SPEED ( MPH )
Figure 4
w-before 18713, we concluded that such a storm is
-likely-to pass the site on a trajectory slightly w
west:of it approximately once each hundred years (an
. annual probability of 10 2).
.We had also_ concluded that the PMH (probable maximum
; hurricane)4r5 would have an annual probability of occurring anywhere north of Cape Hatteras of no more
'than 10-3 To produce a huge tidal surge in Dela-ware Bay,_it would have to be on the very narrowly circumscribed trajectory passing just west of the site, as described for the model hurricane. The combination of these two probabilities is 10-5, Maximum gradient winds in such a storm would be
.. expected to reach 142 mph.
D. Sustained High Winds For..the purposes of this analysis, one is interesteo in sustained :(one hour or longer) high winds from the_SE sector, not the fastest-mile or maximem gradient' winds. Therefore,:to translate the fa'stest-mile estimate to one-hour or six-hour aver-age wind speeds associated with a. surge in Delaware
' Bay, we combine'd fastest mile' data from Fig. + .ith the ratios in Table 1 to obtain the relation between return. periods and sustained winds shown in the table below. In doing so, we used a fastest-mile estimate of 140 mph as a starting point, so our analysis is consistent with the PMH value.
METEOROLOGICAL EVALUATION SERVICES,INC.
m Return Fastest- One-Hour Six-Hour Period Mile Average Average (years) . 100 70 41 37 500 88 52' 46 1,000 97 57 51 20,000 140 82 74
-The foregoing. analysis is specifically directed
.toward hurricanes, but from the standpoint of the floating missile problem, one is interested in the overall probability.of sustained high winds, not just that associated with hurricanes. In the inter- ,
mediate ranges of wind speeds (30-70 mph), it is
~
legitimate to add the hurricane and extra-tropical probabilities together, especially on the northern oceanic coasts. However, a problem arises in making extrapolations of extra-tropical events to very improbable but very strong winds. The physics of the large-scale extra-tropical storms is such that 4 one simply does not find sustained high winds of 100 mph or more. The hurricane is the only storm of
~
significantly large scale that is capable of pro-ducing such winds. Therefore, as shown in Figure 3, above approximately 80 mph, the joint probability-becomes solely the hurricane probability. IV. CONCLUSIONS We'have completedftwo analyses of the hurricane winds that might be experienced at the Hope Creek site, one METEOROLOGICAL EVALUATION SERVICES,INC. L
in our July 6, 1984 report, based directly on available hurricane'-data, and a-second using a combination of
-fastest-mile hurricane data developed by others and
~
fastest-mile /long-term average wind speed ratios from site. data. The two reports are consistent-in indicating that th'e ' annual probability of fastest-mile winds reach-ing-or' exceeding 140 mph and sustained winds of at least 80 mph is no greater than 5 x 10-5,
\ ,7.
MCTEGT4GLCCiCAL EVALUATION SERVICES. INC.
.- \
l l References L1;. Smith M.E.;and-Castelli F.P.: Hope Creek Generating Station, Extreme Event Site Flooding, Meteorology, MES,JInc. . Report to PSE&G, July 1984.
- - 2.1 Neumann C.J., Cry G.W., Caso C.L. and Jarvinen B.R.
!:_ Tropical Cyclones of the North Atlantic Ocean, U.S. ! Department of Commerce, Environmental Data Service, - NOAA-(1981). 4- 3. .Ludlum'D.M.: Early American Hurricanes 1492 - 1870, American Meteorological Society (1963).
-4. . Meteorological-Criteria-for Standard Project Hurricane and
< l D Probable Maximum Hurricane Wind Fields, Gulf and East Coasts of the United States, NOAA Technical Report NWS 23 ( 1979 ) . -
- 5. _ Meteorological Characteristics of the Probable Maximum
-Hurricane, Atlantic and Gulf Coasts of the United I
. States, Weather Bureau Memorandum HUR 7-97 (1968).
'h 4
METEOROLOGICAL EVALUATION SERVICES,INC.}}