ML18086A753

Forwards Meteorological Info Per OL Requirements

ML18086A753
Person / Time
Site:	Salem
Issue date:	06/30/1981
From:	Mittl R Public Service Enterprise Group
To:	Miraglia F Office of Nuclear Reactor Regulation
References
NUDOCS 8107020418
Download: ML18086A753 (27)

Text

/.

...... ~. 'l e

0 PS~G Public Service Electric and Gas Company 80 Park Plaza, T160 Newark, N.J. 07101 201/430-8217 Robert L. Mittl General Manager - Licensing and Environment Director of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission 7920 Norfolk.Avenue Bethesda, MD 20014 Attention:

Mr. Frank J~ Miraglia, Chief Licensing Branch 3 Division of Licensing Gentlemen:

METEOROLOGICAL INFORMATION NO. 2 UNIT SALEM NUCLEAR GENERATING STATION DOCKET NO. 50-311 June 30, 1981 PSE&G hereby submits, in the enclosure to this letter, certain meteor-ological information requirements in accordance with paragraphs 2.c (24) (d) (iii>' and (iv) of Facility Operating License DPR-75.

Should you have any questions, do not hesitate to contact us.

Very truly yours, Enclosure CC Mr. Leif Norrholm Senior Resident Inspector Bl 070 201//1~

I The Energy People

Item -24.d.iii PSE&G shall provide substantiation that the back-up source of meteorological information from the NWS Office, Greater Wilmington Airport adequately characterizes the site condi-tions with respect to wind direction and wind speed by July 1, 1981.

RESPONSE

1.

DATA ANALYZED Wind direction frequencies at the thirty three-ft. level at Artificial Island and the twenty-ft. level at the Wilmington Delaware National Weather Service (N0S) Sta-tion were compared for two periods, May 1976 -

April 1977 and May 1976 -

August 16, 1976.

The shorter time period was chosen because it is the most likely season for "sea_breeze" effects.

Figures l and 2 are local and regional maps of the area surrounding the site, including the location of the Wi+mington Station, designated as ILG.

Notice that the river orientation is somewhat different at the two meteorological locations, the axes being generally NNE-SSW at Wilmington and NNW-SSE at Artificial Island.

M P81 125 05/l

2.

• WIND SPEED ANALYSIS An hour-by-hour comparison was made between wind speeds at the 33-ft. level of the Artificial Island Tower and the 20-ft. level at the Wilmington NWS Station for the period 5/76 through 4/77.

The comparison shows that wind speeds at the two sites are comparable and the majority*of hours are *within 3 mph of each other.

How-ever, beyond this general agreement, there are certain trends evident in the data.

When wind speeds at Art if ic ial Isl and are low <5_"7 mph), there is a general.

tendency for the winds at Wilmington to be 1 or 2 mph higher.

This tendency reverses itself during higher*

wind speeds (>7 mph at the site), when Wilmington wind speeds tend to be 1-3 mph lower than Artificial Island during corresponding hours.

The closest correlation between the two sites occurs when winds at Artifici,al Island are 6-7 mph, which is also the most predominant speed group at Artificial Island.

3.

DIRECTION ANALYSIS The key role of the backup wind measurements for an emergency survey team is to assure that the team begins its monitoring search in the area most likely to be M P81 125 05/2

affected.

This in turn means that the wind direction at the backup site should be quite representative of that at the site.

Perfect agreement is, of course, not to be expected for many reasons.

Research groups who have specialized in the conduct of field experiments have long recognized that no measurement short of a visual tracer released in an actual effluent plume at the time of the release will define the trajectory precisely.

Even wind instruments on towers or poles almost at the point of release will not repiesent the plume or puff behavior perfectly.

The reasons for such discrepancies are many.

First, the source is seldom at the exact height of the wind instru-ment,. and frequently large chang*es_in both direction and speed are found as the height changes.

Secondly, even the most minor variations in terrain will introduce variations in the wind-flow patterns.

Finally, all wind instruments are supported on some sort of struc-ture, and the structure itself introduces aberrations in the wind flow.

An excellent example of the type of wind variability to be expected was described by Brennan et al. (1977); Ap-pendix A, attached.

He found that the wind statistics M P81 125 05/3

on two towers within 1 km of each other p~oved to be slightly different, even though the instruments were located at identical heights above mean sea level.

This study showed that comparative instruments had wind roses which appeared to be nearly identical, but on an hourly basis the standard deviations of the wind direction ranged from 14 to 320.

A backup instrument therefore may be considered satis-factory if it represents the reference instrument well enough so that a search team in an emergency would pro~

ceed within the correct 600 sector most of the time.

It is also important that the reference instrument seldom offer truly misleading information, such as a wind di-rection 1800 different than that G-f the reference in-strument.

An exception to this requirements should be made whenever the w~nds are very light (3 mph) or less.

In very light winds one expects and finds marked dif-ferences in measurements of instruments very close together.

A.

PROBABILITY STUDY The analysis most pertinent to the question of representativeness is the overall annual probability distribution shown in Table 1.

This table indicates M P81 125 05/4

that the Artificial Island wind direction is within

+300 of the Wilmington direction 74.1% of the time on a weighted average basis.

This seems a credit-able comparison when one notes the standard devia-tions of 14-320 for stations <l km apart in the Brennan (1977) study.

B.

SYSTEMATIC WIND SHIFTS Because of the difference in orientation of the river at the two sites, one might anticipate a tendency for systematic shifts in the win~ direc-tion, especially when the general flow tends to be nearly parallel to the river.

The data in Table 1 would not necessarily refle9t such shifts because the data in the.:_300. comparison might be largely positive or largely negative.

We conducted a more detailed study to reveal such shifts and found that they exist.

The most prominent change occurs with wind directions from SE-SW at Wilmington.

Simultaneous observations at Artificial Island are typically turned 20-300 counterclockwise.

M P81 125 05/5

The shift is shown very prominently in the table below where one can see the *skewing of the Artificial* Island directions toward the negative (counterclockwise) direction.

Wilmington Hourly Direction Comparison-Artificial Island Del.

140 150 160 170 180 190 200 210 220 230

-300

-200

-100 oo

+100

+200

+300 Percent Occurrence 8.5 17.0 20.6 12.8 10.6 2.8 2.1 6.9 25.7 20.8 11.9 5.9 3.0 2.0 23.0 33.2 15.0 4.3 2.7 4.8 1.1 3 6. 5 20.1

12. 9 1.6 2.8

1. 2 2.0 18.1 12.2

12. 5 10.5 5.9 2.6

1. 3 7.8

12. 4 16.6 17.1 6.7 4.1 2.6 10.4 16.8 18.8 11. 9 7.9 4.5 4.0 13.2 28.9 15.7 13.8 5.0 2.5 3.8

9. 1 23.8 18.9 10.2 5.3 5.3 1.9 10.0 21.1 15.3 9.2 5.7 5.7 3.4 There is slight evidence for a shift when the flow is downriver (generally northerly) but the aberration is too slight to be significant.

M P81 125 05/6

C.

QUINTON WIND INSTRUMENT A three cup Anemometer and Wind Vane were installed on February 13, 1981 at the PSE&G Quinton Training Facility, Quinton, N. J.

This building, which is 7.8 miles from the Artificial Island site, also serves as the interim Emergency Operations Facility for the Salem Nuclear Generating Station.

Data from this location will be compared with meterological data available from the Salem site and nearby NWS stations when the processed records are complete enough to make the comparison meaningful.

4~

INFLUENCE OF THE DELAWARE BAY The NRC has correctly recognized that coastal sites*

often display very specialized flow patterns arising from the differential heating and cooling of the land and water surfaces.

These patterns are usually characterizd in the daytime by abrupt changes in wind, temperature and stability along a mini-front, and at night by more diffuse differences between overland flow and flow over the water.

In such circumstances, evalua-tion of the trajectory of a coastal release could be M P81 125 05/7 I

~

confusing, because a plume of effluent might change di-rection very abruptly and fail to appear where the search team thought it should be.

Even more confusing is the fact that in such onshore circulations the ef-fluent may rise vertically after a time and move in a very different direction from the near surface flow.

Having examined the Artificial Island site with respect to such problems we find that the typical textbook sea breeze very seldom exists.

As can be seen in Figure 2, the Delaware Bay narrows significantly as one proceeds' northward into the Delaware River and there is no well-defined shore-?ay interface at some relatively fixed angle.

Instead, the onshore flow tends to move northward along the river with slight deviations from the general flow aloft but without any well-defined shift.

The primary effect of the bay on the local

• circulation is to increase wind speeds at all of the sites along. the river, but not significantly change the direction.

Yet another complexity that exists which may explain the undefined sea-land circulation in the vicinity of Artificial Island is the rather extensive wetlands which surround the site.

These wetlands extend from 3 to 5 miles of the New Jer~ey side and for approximately 2 miles on the other side of the river.

M P81 125 05/8

To some extent this type of circulation is apparent in the overall wind statistics, but it is better shown in the ten specific cases in Table 2.

These tables repre-sent ten days on which a well-defined sea breeze existed on the eastern side of the New Jersey Coast near Atlan-tic City.

The Atlantic site tower showed winds typical of the normal sea breeze.

Half of the days the sea breeze penetrated to the Atlantic City (NAFEC) location which is 10 miles inland.

However, Artificial Island, Wilmington and Philadelphia continued to show nearly identical non-sea breeze directions in all but two instances.

On June 9, 1976, in the last two observa-tions of the case, Artificial Island did show southerly directions while the other-two stations showed winds between west-southwest and northwest~* On June 27, 1976, during the last set of observations the wind directions.

were confused at all three sites and there would have been no way to define the flow intelligently from any one of the instruments.

We would conclude therefore that there is no particular need to introduce auxiliary wind instruments for the purpose of defining a "sea breeze" circulation because the typical flow patterns do not exist.

The routine observations will give good guidance to an emergency team.

M P81 125 05/9

I tern 24.. d. iv PSE&G shall provide substantiation that uncertainties associated with plume trajectory prediction, associated with the occurrence of sea-land breeze circulations within the plume exposure pathway zone, are compatible with the planned recommendations for protective actions that would be based

~pan such projections by July 1, 1981.

RESPONSE

Determination of recommendation for protective action are provided for in the Salem Generating Station Emergency Plan Procedure Manual.

The procedures that specifically address protective action requirements are EP I-1 through EP I-4 and EP I-12 "Site

• Evaluation and Recommended Protective Action."

The classi-fication procedures are consistent with the recommendation of NUREG-0654 "Criteria for Preparation and Evaluation of

• Radiological Emergency Response Plans and Preparedness in Support of Nuclear Power Plants."

Associated Emergency Procedures which impact on Protective Action Recommendations are: EP IV-108 "Protective Action and Emergency Exposure Guides" and EP IV-110 "Field Monitoring."

M P81 125 05/10

• e TABLE 1 HOURLY WIND DIRECTION COMPARISON Wilmington, Del. NWS Wind Direction 1 0 20 3 Ci 40 so 60 70 80 90 100 11 0 120

• 13 0 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 5/76 -

4/77 Artificial Island

%_Coincidence With.in + 30° 7 3. 1 60.4

71. 3 72.0 69.4 73.4 63.6 57.9 "66.6 65.4 65.7

71. 6 6 6

• 1 74.4 76.2 84.1 7 7

• 1

63. 1 -

67.3 74.3 82.9 74.5 7 0. 4 67.6 69.2 73.4 80.5 85.2 7 5. 5 79.7 8 1

• 2 7 6. 1 64.3 62.9 72.2 72.0 Total Hours 7,431

% of Total Hours 0.6 0.6 0.6 0.9 1

• 0

1. 9

2. 1

1. 1 0.6

0. 4" o.s.

0.6 o.. 8 1

• 9 1.4 2.5 3.4 4

• 1..

2.6

. 2. 7

,2. 1 3.6 3.5 4.3 3.2 4

• 1 4.3 4.* 2 5.7

8. 8 9.4 7.8 3.5 2.8

1. 3 1.3 Weighted average of all hours within +30° = 74.1%

, 'lf rteo~ofoo1ca/ [,L~aluat1t:t"?._\\rlt*rCll, l.9r:c

T/\\TJril~ 2 COMPARISON O~ WIND DATA DURING 1EN DOCUMENTED SEA BREEZE DAYS AT ATLANTIC MAIN TOWER May to August 1976 Three wind directions were observed per day at 1300, 1600 and 1900 LST respectively.

Philadelphia Wilmington Artificial Island Atlantic City Atlantic Main Tower Time Pennsylvania Delaware New Jersey New Jersey

. New Jersey e LST Elev.:

5 ft.

Elev. : 20 ft.

Elev. :

33 ft.

Elev. :

64 ft.

Elev.:

210ft.

(

0

)

(

0

)

(

0

)

(

0

)

(

0

)

Date:

May 22, 1976 1300 300 300 305 320 160 1600 310 310 290 290 295 1900 330 320 290 320 310 Date:. Jur:ie 9, 1976 1300 250 300 305 270 150 1600 260 300 205 180 185 1900 270 240 100 200 215 Date:

June 26, 1976 1300 340 310 345 290 270 1600 310 310 340 270 195 1900 300 300 325 320 330 Da-te:

June 27, 1976 1300 290.

300 320 270 190 1600 290.

300 290 310 285 1900 120

. 20 275 240 27 0

TABLE 2 (Cont'd. )

Philadelphia W i ~ming ton Art i f i*c i al Island Atlantic City Atlantic Main Tower*

Time*

Pennsylvania Delaware New Jersey New Jersey New Jersey LST Elev. :

5 ft.

Elev. : 20 ft.

Elev. :

33 ft.

Elev. :

64 ft.

Elev.:

210 ft.

(

0

)

(

0

)

(

0

)

(

0

)

(

0

)

• Date:

July 1, 1976 1300 240 280 235 170.

175 1600 240 290 270 190 180 1900 290 300 300 240 2 1.0 Date:

July 9, 1976 1300 350 360 340 000 120 1600 320 330 355 180

• 170 1900 300 330 330 16 o*

185 Date:

July 1 8, 1976 1300 320 310

.. 280 260 295 1600 290 29 0 260 270 190 1900 270 230 270 260 275 Date:

July 1 9, 1976 1300 240 230 235 240 175 e

1600 2130 240 2130 260 190 1900 230 230 230 210 2.15 Date:

July 25, 1976 1300 340 340 340 340 60 1600 360 20 340 340 80 1900 000 3"3 0 350 130 150 Date:

Au9ust 2, 1976 1300 340.

350 345 330 330 1600 310 300 345 010 1.1 0 1900 310 310.

. 330 130 135

... ~Jlrtro10/ng1caf Gt1al11all~~:5twiw, $11c.

I

___ _/

I I ~W l_-1i\\~

.---~ N-E WA_R_K_j I

\\

DEL.

~* *.I

)

\\

• ~~ \\

'.~

I I

~

I

\\

I I

\\

LOCAL MAP PENNS GROVE SALEM

• L N.J.

0 5

MILES 0

5

. - 10 I I I I I I -

J KILOMETERS

PENN.

/

I I

I I

I

"'IDDLETOWN

• I I

\\

PHILADELPH§ WILMINGTON

\\

DELAWARE MD.

I I

\\

\\

I I

I

\\

\\

I NEW JERSEY CAPE MAY METEOROLOGICAL SITE

.A.

N

-t 0

!'.i 10 20

~lullul~lulc__~-'-'~~-~-'I MILES.

0

!'.i 10 20 11111 I I

I KILOMETERS 1*

APPENDIX A Reprinted from Preprint Volume: Joint Con!erence on Ap;:ilicotion8 on l.ir Pollution H.eteorology.

Nov. 2't-Dec. 2, 1977; S.elt Lake c:.tcy, J.J;.

Publ!eh~

eC by Ei~ric.an MeteorologicBl Society, Boston, MA.

Patrick T. Brennan, Frank P. Castelli, and :Maynard E. Smith Meteorological Evaluation Services, Inc.

Amityville, New York

1.

Th"Tro!XJCTION AND SUMW\\RY

~"henever a meteorological facility is designed to provide represe..ritative data for a site,* questions arise about the number and loca-tion of the instruments required.

The answers to such questions becare rrore difficalt as the o::rnplex.ity of the terrain increases.

Since the environrrental programs associated with rrodern electric generating facilities are J::oth costly and time-consuming, it is inportant to limit the instrumentation to that 'Which is really necessary.

The investigation of the rnicra:reteor-ology at the Lirrerick Generating Station in south-eastern Pennsylvania offers a unique opportunity to ariswer questions about the rreasurements need-ed for representative data at a typical inland "Site with m:rlerate, but variable terrain.

The site is equipped with two meteorological tc:M>ers, one on high terrain where the nuclear station*

is being constructed, and the other in the valley close to it. The study provides interesting and fX)tentially valuable infonra.tion on wind and tatpe.rature profiles as func'-c.ions of height and l:X)Sition in reference to the terrain.

These data. could be use:l. in conju.riction with infol'.Illa-tion fran other sites to establish better methods for adjusting available data to fit sources whose elevations and fDSitions differ frcrn those of the rreteorological instruments.

2.

SITE DE.SCRIPITCN The two meteorological tCM>ers US--.c<l in t.'1is study are located in southeastern Penn-sylvania. at the site of the Lirrerick Genera.ting Statim of the Philadelphia. Electric eropany.

This site is in rolling terrain along the Schuyl-kill River, 34 km nor..hwest of Philadelphia.

As Figure 1 indicates, the terrain variability is not severe, with a rise of only 43 rreters fran the nol'.Illal river level to the base of the

• higher tcwer.

Figure 2 shCM-'S a split =ss-sec'-....ional vie-w of the tq:ography ta1<.en along the line A-A' in Figure 1.

Tne unique feature of the installation is that certain instrurrents were located at match-ing heights above sea level 'While their heights 15_1 above ground level differ.

As can be seen in Figure 2,the mid level on TCM>er 2 and the lowest level on To.ver 1 are J::ot.l-t 85 meters above sea level.

For purposes of this analysis, we have designated this MSL height as "Level One. "

The u~....r level on To.-.>er 2 and the mid level on To.ver 1 have iden-tical* M.SL heights, and this elevation has been designated "Level Twcl" in the subsequent discussion.

Wnenever we ref er to the height aJ::ove ground on the ta... -ers, t.lie nurrerical value of the height is used.

For example, the lc:FM:!st level on Tc:wer 1 is called the 9 meter level.

The exact elevations of all wind and ternoerature sen-sors on J::oth to,.;o....rs are shown in Fi~e 3.

The prefixes W and T in Figure 3 are used to differ-entiate wind and terrperatlire sensors respectively.

One year of data, fran April 1972 throuqh March 1973, was selected for.analysis.

This ~

a period of early site preparation prior to the co."1StruC--c.ion of any building structures.

Ccnparison of data s=ies fran this period wi. th those of longer re=rds fran l::oth towers indicates t.l-tat this is a representative year.

3.

WIND DIRB:'I'ICN CCMPARISONS The wind directions on the two towers have been =ipared in several ways.

The first is a simple illi.-pect.ion of the direction frequency distributions at identical MSL heights (Levels One arid Tw:J).

The distributions are so nearly eaual when =roared on this lY.....sis that a figure i~ u:mecessary*.

Evidently the wind direc'-c.ions at the 4Bm arid 93m elevations of Tower 2 are alrrost CXI11?letely free of any tendency tCM--ard channeling in the river valley.

Scne differences do bea:rre evident when the 9 neter levels on the two towers are =ipared, as shewn in Figure 4.

These directional roses are separated by stability classes according to the syst5n developed by Singer and Smith (1953) wi.th the A, B2 and B1 (very unstable, m:rlerately un-stable, Md uristable) group2d together in* one un-stable category.

Tne only sectors in v.hlch sig-nificant differe.rices are found are those parall-eling the river valley, centered on SSE and l\\'NW.

0 Vi" 0:

w t-w 175

~

_J w

150 w

_J

<l:

125 w

(fJ z

100

<l:

w

~

w 75 0

(l)

<l:

50 z

0 t-25

<l: >*

w

_J w

0 A

00

- TOWER 2

~

.5 KILOMETERS

(~ ~

Figure 1.

Locations of the meteorological towers.

All contour elevations are in feet above mean sea level.

METEOROLOGICAL TOWER N0.2 METEOROLOGICAL W14 TOWER NO. l T13 Wl8 1

-~~e_!_~2._(1_3.Q~i:..!_e~~S.h.) ___ Wl3 T17 Tl2 W17, __ ~vel Oni:J. 85 Meler~YlSLL _

Wll Tl6 1 Tll

\\(;(iw;:.,;',i*'>;p1?:~@J)~'1f !&tif~J~}f~tJ);~~t,,)ti:~~~~J~~~~~1 500 1000 1500 DISTANCE {METERS)

A' Figure 2.

Cross-sectional view of the meteorological towers taken along the line A-A' in Figure 1.

155

• ~.

In these* sectors the valley wind is slightly more channeled than that over the higher ground, but the effect is not large.

Figure 3 NOMINAL HEIGHTS OF METEOROLOGICAL SENSORS Instrument Number Wll Wl3 Wl4 Tll Tl2 Tl3 Instrument Number Wl6 Wl7 Wl8 Tl5 Tl6.

Tl7 Tower No. l Elevation (meters)

(Above MSL)

(Above Grade) 85 130 158 84 128 157 Tower No. 2 9

53 82 8

52 81 Elevation (meters)

(Above MSL)

(Above Grade) 46 85 130 45 84 128 Figure 4 9

48 93 8

47 91 COMPARISON OF WIND DIRECTION.FREQUENCIES(%}

9 METERS ABOVE GROUND Unstable Neutral Stable All Hours Sector Tower No. Tower No.Tower No. Tower No.

NNE NE ENE E

ESE SE SSE s

SSW SW WSW w

WNW NW NNW N

12 12 12 12

1. 9 1.5 3.1 3.8 2.0

1. 2 2.1 3.6 3.1 1.8 2.7 4.1 6.7 3.4 3.1 2.7 2.4

.7 2.2 3.1 2.7 2.2 5.3 4.5 2.4

.6

1. 4 2.5 5.9 6.4 6.3 3.6

.7

.3

.5.l 1.3 1.1 1.5 1.0

.2

.7

.2

.2

.4

.9

.7

.8

.6

.3

.3

.1

.4

.3

1. 2

.6 3.1 2.7 1.8 2. 7 1.2 1.4

.6

.3 1.9 1.8 2.6 3.4 2.9 2.3 2.6 2.5 "1.9 1.4 1.4 2.3 3.8 3.0 2.1 2.1

1. 7 1.5 1.4

l. 7 2.1 2.0 4.1 2.6

l. 7

l. 5

l. 5 1.9 2.3 2.5 3.6 2.3 4.5 4.4 3.8 2.3 7.1 4.7 8.? 5.8 5.2 5.5 3.6 4.4 5.1 10.2 6.8 7.9 5.7 4.4 3.6 2.2 4.6 3.2 7.5 5.0 13.6 10.9

8. 3 11. 6

6. 5 11. 3 5.4 6.2 Another way of examining the wind direc-tion differences involves the comparison of hour-ly average directions.

Hourly direction differ-ences between pairs of wind sensors, one on Tower l and another on Tower 2, were developed.

The standard deviations of these differences were computed for all hours when the direction at the instrument selected as a reference remained with-in a given 10 degree sector.

The results of this review were inconclusive in revealing direction-dependent differences, although some modest cor-relation between the magnitude of direction var-iability and the general direction of the flow were observed.

The study, however, was useful 156 in specifying the variability between instrument readings.

Hourly wind directional differences did prove to be a function of sensor elevation above ground.

Figure 5 shows that the standard devia-tion of the directional differences was greatest when the two 9 meter levels were compared.

The difference decreased significantly at the high-est elevation of comparison.

Figure 5 STANDARD DEVIATION OF HOURLY WIND DIRECTION DIFFERENCES Level Tower l 9 meter versus Tower 2 9 meter Tower l 9 meter versus Tower 2 48 meter (Level One)

Tower l 53 meter versus Tower 2 93 meter (Level. Two)

Figure 6 Standard Deviation (0 Az) 32 20 14 ANNUAL AVERAGE WIND SPEEDS (M/SEC)

BY HEIGHT l".BOVE GROUND Stabilitv Class Moderately Unstable Unstable Neutral Stable All Hours Stability Class Moderately Unstable Unstable Neutral Stable All Hours Tower l 9m 53m 82m 3.3

l. 7 5.0 1.2 2.7 4.5 1.8 7.1 2.5 4.1 5.0 1.9 7.9 3.1 4.6 BY Y.SL HEIGHT Level One (85m MSL)

Tower l Tower 2 3.3 3.3 1.7 1.5 5.0 5.6 1.2 1.2

2. 7

2. 9 Tower 2 9m 48m 93m 2.5 3.3 1.3 Ls 3.9 5.6

• 7 l. 2 2.0 2.9 Level Two (130 MSL) 4.4

l. 7 7.3 2.3 4.1 Tower l Tower 2 4.5 4.4 1.8 1.7 7.1 7.3 2 :5

2. 3 4.1 4.1

4.

WIND SPEED COMPARISONS Several comparisons were made of the wind speeds at the two towers.

These included hourly speed comparisons, as well as compari-sons of wind speed distribution by directional sector and atmospheric stability class.

The comparison of annual average wind speeds from all levels at both towers is shown in Figure 6.

These speeds have been categorized by stability.

The wind speeds from the two in-struments on Level One are almost all within one-half meter per second.

This indicates that terrain effects on the flow are minimal at this elevation.

The agreement between sensors is even better at Level Two.

However, average wind

speeds at sirr~lar heights above ground are lower at Tower 2, as one would expect in a valley.

To find any terrain induced effects on the low level wind flow which may not have been evident when comparing annual averages, a study of wind speed ratios was made*.

Ratios of the Tov;er 2 wind speed to the Tower l wind speed for Level One and *Level Two were computed hourly.

This was an attempt to discern any convergence or divergence effects on the wind flow as it en-countered the sloping valley walls.

While dif-ferences did appear among various directional sectors, it was determined that the ratio tech-nique was too sensitive to small speed differ-ences, and no conclusions were made.

A more detailed representation of the wind speed in the surface layer can be de~ived by sU!:lllla.rizing annual* average wind speeds by directional sector.

At Level Two the speeds are equal, regardless of direction, but distinct differences appear at Level One.

During un-stable hours, the Level One wind speeds are equal only with up-valley or down-valley directions.

In all other directional sectors, the Tower l and Tower 2 speeds differ, but no consistent pattern exists.

During stable hours, Level One wind speeds at Tower 2 are higher than those at Tower l with up-valley or down-valley flow.

Large vertical shear can be expected during stable diffusion conditions.

5.

VERTICAL PROFILES OF WIND SPEED In order to assess the variation of wind speed in the vertical, we have analyzed the conformity of each tower to the standard power law equation used to define wind speed changes with height.

where u z b

u (ff z

wind speed at height h wind speed at height z stability dependent exponent In this analysis, wind speed values from the various heights above ground were used.

Figure 7 WIND PROFILE POWER LAW EXPONENTS BY STABILITY CLASS Brookhaven Stab.

BNL Tower l Tower 2 Class Exp. 9-53m 9-82m 53-82m 9-48m 9-93m 48-93m Mod.

Unst.

.16

.03

. 04

.ll

.10

.ll

.13 Unst.

.25

.18

.19

.24

.17

.25

.45 Neut.

.32

.19

.20

.23

.22

.27

.41 Stab..

.so

.42

.43

.47

.35

.-53

.98 157 e

Figure 7 Cont.

Pasquill Stab.

EPA Tower l Tower 2 Class Exp.

9-53m 9-82m 53-82m 9-48m 9-93m 48-93m A

.10

.09

.ll

.20

.21

.25

.34

.B

.15

.09

.10

.16

.16

.21

.34 c

.20

.ll

.ll

.ll

.17

.21

.31 D

.25

.22

.23

.26

.26

.36

.61 E

.25

.30

.32

.38

.25

.31

.47 F

.30

.43

.44

.45

.30

.44

.80 G

.30

.45

.45

.49

.13

.35

.95 Figure 7 is a summary of these power law exponents computed from the average wind speed values in Figure 6.

In t..,e upper part of the figure, separate values were computed for each of the Brookhaven National Laboratory turbulence classes, with power law exponents derived at Brookhaven included for comparison.

The lower part of the figure shows the same information for the Pasquill stability classes derived from the USNRC (1972) lapse rate categor-ies.

The power law exponents used by EPA,. Busse et al (1973), have also been included for compari-son.

The data are quite straightforward in their implications.

Tne tower exponents are app-roximately what one would expect over rough, flat terrain.

The exponents do increas*e slightly with the elevation of the height interval above ground.

Over the valley, most of the increase of wind with height occurs between the 48 and 93 meter levels, especially during stable conditions.

This is quite consistent with the idea that a relatively stagnant layer exists in the valley during noc-turnal hours, but the flow above behaves much like that over the higher terrain.

The profile exponents compare reason-ably well with those found in th~ il5 meter height interval at Brookhaven, but the overall range of exponents is less.

The compariscn with the EPA values is not quite as close, and here the range of observed values is significantly greater.

Hour 0200 0800 1400 2000 Figure 8 WIND PROFILE POWER LAW EXPON1:1\\'TS BY HOUR OF DAY Tower 1 Tower 2 9-53m 9-82m 53-82m 9-48m 9-93m 48-93m

.35

.36

.41

.37

.47

.75

.25

.27

.34

.22

.32

.59

.13

.14

.20

.17

.21

.32

.31

.32

.36

.32

.42

.68 Figure 8 is a potentially useful t.able of wind profile exponents by hour of the day.

Often one does *not have reliable stability in-formation, and this table could serve as a sub-stit~te.

The data show what one would expect; daylight hours tend to be unstable, a,nd noctur-nal hours are stable.

6.

STABILITY Considerable controversy over methods of estimating stability has appeared in recent literature and conferences (USEPA 1977).

With this in mind it is interesting to study compari-sons of different stability systems at this rather typical inland site.

Three types of estimates have been com-pared:

A.

Brookhaven National Laboratory -

The stability estimates based on wind direction gustiness (Singer and Smith 1953) have been made routinely at this site.

B.

Pasguill-Gifford (NRC)- Pasquill-Gifford stability cla?sifications as derived from lapse rate information using the NRC (1972) specifi-cations.

C.

Pasauill-Gifford (Modified by MES)- Similar to B but based on a slightly different set of temperature-difference categories which are be-lieved more consistent with the early Brookhaven experience.

A comparison of the lapse rate categor-ies used to represent the Pasquill-Gifford cate-gories in both the NRC and MES systems is shown in Figure 9.

Pas quill Stability Class A

B c

D E

F G

Figure 9 TE!1PEP.ATURE LAPSE RATE STABILITY CLASSIFICATIONS

(°C/100 METERS)

MES Modified NRC s::i::stem system

s -1.9

s

1. 9

-1.8 - -1. 7

-1.8 -

- 1. 7

-1.6 - -1.5 -1.6 -

- 0.7

-1.4 - -0.5 -0.6 -

0.0

-0.4 1.5 0.1 -

1.5 1.6 4.0 1.6 -

4.0

> 4.0 4.0 A comparison of stability estimates in Figure 10, shows the remarkable differences in the percentage frequencies of occurrence for the classes sho~~ by the three systems.

The most striking feature is that the NRC system predicts far more neutral and stable hours than either of the other two, primarily at the ex-pense of unstable hours.

Figure 11 shqws a similar comparison of the lapse rate stability in the layer from Level One to Level Two.

While it is apparent that this layer is more unstable near the ground at Tower 1, the same differences exist between stability systems.

These figures illustrate clearly the problems faced when using any stability system.

Estimates of stability using any of the three systems may give differences in si?ecific cate-gories of as much as 35%.

For example, at 15~

Tower 1 the NRC system underestimates the class C frequency by 35% when compared to the MES system, but overestimates class D by 20%.

If the Pasquill-Gifford categories are grouped together to roughly approximate the Brookhaven categories, an even larger discontin-uity can be seen at Tower 2.

The NRC system defines less than 5%* unstable, compared to over 55% unstable by the Brookhaven system.

This difference is narrowed when the MES lapse rate breakdown is used, but the Brookhaven unstable category is still approximately 20% greater.

Clearly, when such large differences exist between specific categories, there is a fundamental incompatability a.~ongst these systems, and one or both approaches must be wrong.

There seems little doubt that a careful review of sta-bility systems is needed.

Figure 10 COMPARISON OF STABILITY SYSTEMS TOWER l VERSUS TOWER 2 FREQUENCY ( % )

Pas quill Tower l' Tower 2 Stabilit;t 81-8rn AT 91-8rn AT Class NRC MES NRC MES A

1.0 1.0

.4

.4 B

1.0 1.0

.4

.4 c

4.7 40.0 1.2 27.1 D

46.3 26.0 36.7 30.6 E

32.7 17.7 38.8 19.1 F

10.l 10.l 13.4 13.4 G

4.3 4.3 9.1 9.1 Brookhaven Stabilit::i:: Class Tower 1 Tower B2 Mod. Unstable 3.3 3.7 Bl Unstable 44.8 54.0 c

Neutral 14.6 13.6 D

Stable 37.3 28.7 Figure 11 COMPARISON OF LAPSE RATE STABILITY CATEGORIES FROM LEVEL ONE TO LEVEL TWO FREQUENCY ( % )

Pas quill Tower 1 Tower 2 Stabilit::t 52-8m AT 91-47rn AT Class NRC MES NRC MES A

5.9 5.9

.9

.9 B

4.9 4.9

.7

.7 c

8.2 35.1 2.0 34.6 D

38.6 25.1 45.3 30.7 E

29.3 15.9 38.8 21.0 F

8.5 8.5 9.5 9.5 G

4.5 4.5 3.0 3.0 2

7.

CON:LUSICNS

l. Wind - There is little question that winds at the 48 and 93 neter levels of Tower 2 are unaffected by the rrcderate rolling terrai..r1 at the site.

Wind data fran these inst.rurrents agree

• well with data fran their Level One and Level Tu'O =unterpart.S on TcMer 1.

This indicates that th<>..xe is little lateral variation in wind speed and direction above the general level of highest terrain.

Hcwever, lower wind speeds and slight channeling are found at the 9 r:eter level of TOM:r 2 in the valley.

Tne vertical profiles of wind speed indicate that both ~s fit the Brookhaven profiles reasonably v.>ell, while the EPA values te'!d to underestimate the vertical shear at Tower 1.

2.

Stabili tv - The stability cx:mparisons fran these to;.>ers serve only to point out the diffi-culty in choosing a representative classifica-tion systeLl.

TcMer 1 is generally IIDre unstable by lapse rate ~isons, while Tower 2 is IIDre stable, as one would expect in a valley.

Ho,.;iever,

• the large differences bet:wee..'1 these classification systems, especially that between lapse rate and gustiness at Tcwer 2, indicate a basic need for a new system to estimate atrrospheric stability.

Acknow ledge.-rents Tne authors would like to acknowledge Frank Lawre.'lce of Meteorological Evaluation Ser-vices f= his help with the ccmputer surnnaries, as well as Mary Ellen Reeves who typed the manu-script and Ronald Thc:mpson for the artwork.

We v.uuld also like to thank Mr. A. J. Hogan of the Philadelphia Electric Crnpany for his many sug-gestions.

References Busse, A.O., and J.R. Zimremlan (1973):

Users Guide f= the Climatological Dispersion M::del, EPA Report R4-73-024, available fran NTIS.

Singer, I.A., and M.E. Smith (1953) :

Relation of Gustiness to Other Meteorological Pararreters, J. Meteor., 10, 121-126.

u. S. Envirornrental Protection Agency (19 77) :

Report to the U.S. EPA of the Specialists Con-ference on the EPA M::deliiig Guidelines, Februarj 22-24, 1977, Chicago, Illinois.

U.S. Nuclear Regulatory Carrnission (1972) Reg-ulatory Guide 1.23: Onsite Mete=ological Pro-grams.

159

,.~*

-=-*- --

Item 24.d.iv PSE&G shall provide substantiation that uncertainties associated with plume trajectory prediction, associated with the occurrence of sea-land breeze circulations within the plume exposure pathway zone, are compatible with the planned recommendations for protective actions that would be based upon such projections by July 1, 1981.

RESPONSE

Determination of recommendation for protective action are provided for in the Salem Generating Station Emergency Plan Procedure Manual.

The procedures that specifically address protective action*

requirements are EP I-1 through EP I-4 and EP I-12 "S{~e Evaluation and Recommended Protective Action."

The classi-fication procedures are consistent with the recommendation of NUREG-0654 "Criteria for Preparation and Evaluation of Radiological Emergency Response Plans and Preparedness in Support of Nuclear Power Plants."

Associated Emergency Procedures which impact on Protective Action Recommendations are: EP IV-108 "Protective Action and Emergency Exposure Guides" and EP IV-110 "Field Monitoring."

M P81 125 05/10

However, as discussed in Response to Item 24.d.iii our analysis shows that the typical "Sea Br~eze Flow Patterns Do Not Exist at Artificial Island."

Protective action recommendations contained in the Salem Emergency Plan Procedures are based on: (1) Dose assessment calculations, (2) Predetermined effluent monitoring set-points (which are not based on realtime meteorological con-ditions), (3) Field survey assessment, and (4) Predetermined protective action recommendations based on actual degraded plant conditions.

In order to implement any of the more significant. protective actions (i.e., sheltering or evacua-tion) it is necessary to define the affected area (Emergency Planning Areas) within the confines of definable bound-aries.

In order to define these boundaries, physical or political boundaries are chosen (see Figure 3).

Consistent with the requirements of NUREG-0654 Appendix 4, the em~r gency planning areas are configured to zones representing sectors goo, 1800, and 3600 about the site.

It is antici-pated that all protective actions initiated by the stat~s(s)

(i.e., evacuation or sheltering) would be consistent with these broad planning areas.

M P81 125 05/11

In addition, predetermined protective action zones consis-tent with this philosophy are identified as part of Salem Emergency Plan Procedure*EPI-4 (Figure 4, attached).

Should a major ~ccidental*release occur at either Salem unit, pro-cedures are available to initiate augmented monitoring sup-port including plume tracking by helicopter (EP IV-109).

• It, therefore, cari be seen that uncertainties associated with exact wind direction are accounted for when initiating protective actions.

Further, during the clarification meeting held May 5, 1980*

in Philadelphia, Pa. on the topics of Emergency Response Facilities (ERF's) and meteorological monitoring, we were tnformed that additional guidance on sea breeze identifica-.

tion and modeling would be forthcoming.

This additional in-formation would be based on studies on which the NRC is, 9ur-rently working.

M P81 125 05/12

e*

-T-(

Figure 3

Plume Exposure Emergency Planning Zone

\\

90° Section C Evacuation Sections New Jersey Salem Generating Stations Hope Creek Generating Stations 8' 5 Miles