ML19269C481

Forwards Info on Wedge Model Wind Speed Parameter & Population Density Model Used in NRC Rn-222 Assessment,As Requested at 780830 Hearing

ML19269C481
Person / Time
Site:	McGuire, Mcguire
Issue date:	01/04/1979
From:	Hoefling R NRC OFFICE OF THE EXECUTIVE LEGAL DIRECTOR (OELD)
To:	Jeffrey Riley CAROLINA ENVIRONMENTAL STUDY GROUP
References
NUDOCS 7902030056
Download: ML19269C481 (14)

Text

{{#Wiki_filter:C 4 UNITED STATES gh. NUCLEAR REGULATORY COMMISSION g %(..~ a j WASHINGTON, D. C. 20555 J e s7 %, *N /a January 4, 1979 %p e q Y ,#3 9 N O* M[ Mr. Jesse L. Riley, President D [* g++"[ '] Carolina Environmental Study Group 854 Henley Place Charlotte, North Carolina 28207 r, p In the Matter of M ,p ^ DUKE POWER COMPANY (William B. McGuire Nuclear Station, Units 1 & 2) Docket Nos. 50-369 & 50-370

Dear Mr. Riley:

As promised by Dr. R. Gotchy (Tr. 2493) at the August 30, 1978 session of the hearing in this proceeding, Dr. Keith Eckerman (developer of the RABGAD code) has responded to your questions regarding the population density junction used in our Rn-222 assessment. In addition, Dr. Eckerman has provided further discussion on the question you raised regarc'ing the 2 m/sec wind speed used by the Staff and the sensitivity of une population doses to wind speed in the attached document entitled " Wedge Model Wind Speed Parameter". Sin ly, f i ard K. Hoefl hg Counsel for NRC Staf .. Enclosure as Stated cc (w/ encl.): Robert M. Lazo, Esq. Dr. Emmeth A. Luebke Dr. Cadet H. Hand, Jr. J. Michael McGarry, III, Esq. William L. Porter, Esq. Shelley Blum, Esq. Atomic Safety & Licensing Board Panel Atomic Safety & Licensing Appeal Panel Docketing and Service Section 79020300$b

Wedae Model Wind Speed Parameter The wind speed parameter of the wedge atmospheric transport model of GESMO is assigned a default value of 2 meter /second ( 4.5 meters / hour) in the RABGAD computer code. This parameter assignment has at various times been suggested as being non-conservative (overestimating decay in transit) and overly conservative (overestimating plume residence over the land mass). The following examines the assign-ment of a numerical value to this parameter. The wind c. peed parameter of the wedge model is often compared with u observed wind speed data. For example, the average surface wind speed for the U.S..is about 9.7 mph ( 4.3 m/s)(Ref. 1). Comparison with this value provides some insight into the numerical range of data but the comparison not valid. The Edge model is a one-dimensional transport model and thus the one dimensional wind .. speed parameter can not be compared with the observed scalar wind speed data determination regardless of direction. It should be clear that the observed wind speed data will overestimate the appropriate value for the one-dimensional parameter of the model. An appropriate wind speed value, appropriate lid tilallit c6nform[s]tio the considerations of the model development,can be generated as the magnitude of the frequency weighted sector sum of the wind rose data. This quantity yields the net direction and magnitude (speed) of airborne transport of material. Such data have been compiled _ _ _ _ _ _ _. _ _. m

2 on a monthly basis and are presented in Ref. 1. Attached are figures from Ref. 1 showing the mean resultant surface wind vectors (direction and speed) for a typical month in each of the four seasons. Analysis of the scalar speeds shown in the figures for the region east of the Mississippi River suggested a wind speed value of about 4.5 mph or 2 m/s. Model Parameter Sensitivity For a non-depositing nuclide and a constant population density the ratio of the population dose computed at wind speed V to the GESMO model, V = 2 m/sec, is given as n 1 - exp (-AR/V) 1 - exp (-AR/2) where A is the decay constant (sec-I) and R'fhe' plume pathway in meters. Tabulated below are the ratios as a function of wind speed for various half-lives and a plume path length of 1000 miles as used in GESMO with a constant population density. Population Dose Ratio (I) Half life Wind Speed (m/sec) (days) 1 2 3 4 0.5 1.0 1.0 1.0 0.99 2.0 l.0 1.0 0.92 0.83 4.0 1.2 1.0 0.82 0.69 10.0 1.5 1.0 0.74 0.58 20.0 1.7 1.0 0.70 0.54 (lIRatio of population dose assuming tabulated wind speed values _ to that estimated with a wind speed of 2 m/sec.

3 As can be seen from the tabulated data the population dose is inversely related to the wind speed, use of a higher wind speed would reduce the populatic.n dose estimate. However, the population dose is not highly sensitive to the wind speed parameter assignment, e.g., if one increased or decreased the 2 m/sec value by a factor of two the population dose estimate would be changed by about 50%. The radon evaluation of the mining and milling industry is scmewhat further complicated by the spatial dependence of the population density. For these considerations the population dose is proportional to: \\ % 1 - exp (-( A-a)) V (A - a) where V is the wind speed parameter (m/s), a,the. exponential increase ~ ~ a ~r -7 in population with distancc 9.6 X 10 1/m, and A is given or X = A/V-a; -0 ~ A is the Rn-222 decay constant (2.11 x 10 sec ). 7abulated below is the ratio of the radon population dose for various wind speed values to the value computed'using the 2 m/s value assigned to this parameter in GESMO. d M 'e F MM4g

4 II) Radon Population Dose Ratio Wind Speed Population Dose - (m/s) Ratio 1.0 0.56 1.5 0.78 2.0 1.0 2.5 1.17 3.0 1.29 3.5 1.37 4.0 1.42 u 4.5 1.44 5.0 1.45 5.5 1.44 6.0 l.'N ~s O) Ratio of the populi. tion dose assuming tabulated wind speed to that estimated with a wind speed of 2 m/sec. The aoove ratios exhibit a rather broad maximum in the neighborhood of 5 m/sec ( 10 mph). This is the neighborhood of the observed scalar value of the wind speed which is not applicable to the model considerations as discussed above. Furthermore, note that less than 50% increase in the estimated dose is indicated even at the maximum. In conclusion, wind speed value based on the magnitude of the vector sum of the wind vectors is appropriate in the wedge model. Considera-tion of eastern U.S. wind rose data suggests.a numerical value of

5 2 m/sec. In general, increasing or decreasing the 2 m/sec value by a factor of two would result in population dose change of about 50%. This pe7centage change is somewhat sensitive to the halflife of the nuclide, approaching no change as the halflife approaches zero. The sensitivity analysis presented here considered only non-depositing nuclides; those nuclides experiencing atmospheric deposition are even less sensitive to the wind speed parameter. is Ref. 1" Climates of the United States", U.S. Dept of Commerce, Washington, D. C. (1978). d O j an e qq + _ y ew r e e e a wee e ee+ww+

...- 3 L -. =.u. l,,,,. '* S c:: D %., n..Cr.

• J h..{l v:.:. 1 l.u -

t . n.. a r- - T ..t .. s.... ,, y ygn.. -.gn m.;p,g. ,7;.,--i gy \\ . *~-o ~.7, j 4: t .......-..c. s s. 4 e r%:r. .., l m t i f. -- ~,.; f $. t .I c[. A hI J r. c ,.d g 4 g,c n -r i, h, j - >+ ~ 7 ~~..]'y... lz,. -n G.A f e. ', 4 .. 6 - 'r'- J \\ ..,_g ; - -'g' f..!". n. 1 4 ~ :> 4,- 'g l s l ~. ~;..-- y:r-su>. k,> .:::P~L-:;~".f s Q g.,,.) .b Y;. .. 5., g. i 'y '., ~ r,;~ J+ \\+I,l- /j. .,,,. t .f.d..,..g )~ y~,g g. w h. %l '. t-

?' !.,

w y,;.t

• I '.*y*,}.,,,,

.'aci.-.-7 '.'.'.7  :\\ --- - --~L',-,..-c.!._ 1 -.. g _. 9. 1,

7.

s. a l i 71 ~, / '; ,\\ )= :; - s ..n.

n.,;-y

%y }^l~ J s s' i. t /%'N ~C,.e, {! y., -..g...- l %.; ;,- ,f" g (/ w. n e- / ;4 .} }; t ) s ..--.-. h 1.. & - *-~ y.r.4.., x ) T I ~J ~2h"I'r5 k.,,. s rs ~ = ' ' ' .Qg.. -s- ..a. _ \\ t' s s. 4,- Y g'., a' t =- l % f f.h~&'gi L. ~.: '- =( ~ \\ W' - VJr.s ?.... \\ ~ - ---- jjf c.r' ,Ll--~i

k. /"# '

U t _.L ":.T:".*:* :O= ,i. 7 ", g~ '* :;-~4 ' 7.":::.' U ' _ ll [l ~~~) Ld }l ..g y.., g L _, 13.. , ? ,'e-- -.-.M 'M 'fD'[.g N RESULTANh SUR[ ACE WEND DfRECTI0li AND SP ';.:A,,, \\ MIDS? RING M0 NTH - -APRIL. e ' ' ' .,;:m ( ., T '*"mi : [

---

c-74'.

z \\ ' bl g,, ( k 1' k h. . ' =k

f 8y.b.,,;%..

'\\ ' - l. 1 = m f r s lo h. i k.g - ll. e f ~ 't f I ll n, - a ., i s ; ~,,j~. L L. f ,t ' ~~\\s f Wj:

• i #.-. } c__..~.m V

?" *.'.' ~~*:,v. . M k~ (T T P(7.. y.b : ). Q sD\\ ,1 l .r +J H% b yy i 1 t- .\\ ~. v = -" g,..~9.; yh \\.g . j. j j. o s e-. x ( 8

v. A- ~

_.h ..'~a.

s. -.w.4.,N..'..-

k. l; ~ a .i. u . / n..y -s"* ; +* i, s. \\ / - "*'.j s .m. jy' '.j i l s ,i i 'y ' 7.;s _ l %. f s j / l p ~ -' ~- r ~~ c - - jg. s-

M.

n)..; i <, Z*lty, \\ 4 ?

..--='****

_ Y \\ -,~ %,..., %. 1 k.c.4. ~. s 1 '.D,' d .7o \\ 1~-lay.d g %y. -.tt 1 9 ,X,,..., > \\ =..=..=.. [. s.c.; ' 1 pA s4%

c. :.=:==t

y C i iA't 3

. /j.7.. l .- Lr. @.;~ 1 Pl {,pg i ~

u:=.::e :7. c.v-t t,,.

. ' !, y,. Y. p =..m .d.- .g g p 'yy...j. y m osern +. ~ --- :#

.. s.n. A .>..w

p..,t;

,.,, / y...-.... w.>....us ., es. / - ,t e -.-..c. ps '.I,' f c. ~--.,,-c~'/ m ~.: ,r, .s : u -~. ;.,.. s, .J,., . p,f - ?- ....--}.~f:(f'\\f.:;.)I ')'.' j.. - -. 5-I i~ r _3 y fg 4/.:..l~~..'~

II.

~ \\ Q~.f ' -fy..mq j.; j o 'l. , ~.. l l 't I If----- - 7.,,f,^,ll^ r t .) M. +,'Q. yt-O'.s.-Q::; 3'_/ i' t)'*' ~

• • ^~~%. y

\\. ,y " k -1 F,i -F .......-, M,sv -.;s. .\\ l :'. ' N I j

9..L

~ > . ~ ~J.- .,n n

• \\

< s '- h)l. ' I' \\* '..,,N e / T...,' 3 i

/ k

/ 4., N / \\ ~ -s..' } ',c..' \\ l - ,1 l l \\ G 'f'"l-,f"\\ r j i .~ g s..,..... 5 - $ ~~~~L{~}Y- -jl - f & e ----- Ii

r

.. l' ~ h~ p ~., l t ),. H i ~j L y-j :- a s g .I s .j p,%.. :'.s\\. ~ ='? ? I \\ ?I s {c %j - 1 L r i t ~ ~~' ~ ~[ :.- \\ l,:-s,,7.v ?a@e-9. j I v i T

• \\,.L j

eb+I I -

~. hN k r.

J --

R '.-T* I'~%,.M'4.p=. =^*"f"" \\ 'g .n k.kL .,...y - 'h \\f'% ~~~ \\. ,\\f V. ',P "4 N,l, f s o x o g f l ..\\ 1.e -.+ x1 .,,.c. \\ c.p. bs., / ..... ~.. ..q g u ._ f'$[i ' 2,. ,~ 1j. '.-f ~f ..y ~ -~ ~~ - ~~-"' p,. p'.- *g'.__. ;i l , g, r'% I ~_ .l 1*- ui i. i. . p; => _ a ri_ g.....- 1 r----. a. t: i ~ _ I ~~- MEAN RESULTANT SURFACE WIND DIRECTION AND SEEED j/ '9 i ,L..MID..Al.m._2_ N MNTH -- OCTOBER r" "" M .;, g.'!. ~ T > f..'*i 4 '~ u. l p n I \\ /' - Wg... -~ h *** ') (% l 's. \\ 7' 'l ~~' s .I

.5.

Y.,, jN '.. t. E 'v N ') \\.:27";;T' h I \\ a '" ~ 4.i \\ l !\\ 1TE Di'~ rr{~k'j! \\ f -M. g.$p -- l y v '%.! J.'I'"' (..L.. V / . ~. -( 3 -w'=% .e n ,,1.wl..d. h M t '\\, r "'1~I I i 1W,D.,.=.=. s i \\ " *7T _.w L y P r- , i. s r \\ a ( ,,)- sov. K j r . ****

• f n,.. u..W g.

g Y* 'I ~,,..f f[ \\ - % ua < Y I .I a l .M. g

t. _

~... =de.~ ,J,_,li. a. - %. *=. t,j g *7 '"' I g ,=.*;ue

t..

/e. i '\\I

/

l \\- Q...-.g; u m..,r .-.V,.-{, w. s. %/ - q,,,,; .A.- - u N u. p .,Lj-tg. ,.g -+

n..,

..1Vc .....n_-- x - gL*2.*.~ % "t,., \\

• w a***'....

+...-.m._.._..- e t_ s g j.4 g-(/ ,,gg *%rE> e g 1l.N.6-.' 3r,. uM). 1 " o N ~. . \\ j \\,\\ .o 1 f y;:.::_ -.;. f i UE. c, i = = = =:- --' W,h.-- l J d.

7.::::".::.* '7.'.".'

~ . _i,,,,.

• }' l ~g -> ;.,.. - l g."

o ~' ' K., t,,- ., ^* i j p t# j

t., t,, yg.>

--. y.g _u 1=J m a e e = e n ---eem.i-. += *-*= pee,

GESM0's flining and Milling Populations In the GESMO radiological assessment (NUREG-0002) a population density of 160_ persons per square mile was used for all fuel cycle facilities other than mining and milling. The 160 figure (1970 population census) represents the population density of the region east of the Mississippi River and was typical of reactor siting experience. The mining and milling industry was considered separate from the other fuel cycle facilities, as these facilities were primarily associated with western states of much lower population density. However as the effluent from the mines and mills, Rn-222 being of primary interest, is transported eastward by tHe prevailing westernly winds it will enter regions of higher population density. Thus the assessment model must reflect consideration of the spatial dependence of the populat-ion density. The following is an account ofthedevelopmentofthTEESMOpopulaticndensitymodeland further staff consideration in the past GESMO period. In the course of the GESMO analysis estimates of the population density in the immediate vicinity (50 mile region) of various existing mines and mills were prepared. These data were compared with various western state densities and a numerical value of 7.5 persons per mile square was determined. A density model was then proposed which considered an caponential increase in the density with distance such that the above cited figure of 160 persons per square mile was reached at the end of the plume path length. The plume path length was established by considering the release

2 point to be located on the Colorado plateau, about 2000 miles from the east coast. The expression for the population density is then p(x) = 7.5 exp (0.00153x), where p(x) represents the population density per square mile at a location x miles from the assumed release point. This model was compared with the 1970 population census data to confirm that it was a reasonable reflection of the population distribution. This is to say that no detailed analysis but a subjective judg-ment was made. The model as formulated above was employed in the GESMO radiological analysis. Following completion of the analytical effort for GESMO a further evaluation of the density,model was undertaken. The effort ~ r con:,isted of sampling ther U.S. population density ~ever.$.two degrees of longitude eastward from the 108 longituda line for three latitude lines at 37, 39, and 410N. The sampling was under-taken by placing a 70 mile radius circle centered at the above location and deriving the inscribed population through use of the 1970 census computer tapes. Table 1 summarizes the data obtained. These data were then fitted to the exponential function to yield p(x) = 7.2 exp(0.00219x). e e 9 w.

,s 3 2 The correlation coefficient, r, was 0.80 indicating a reasonable fit to the data. Other common analytical functions, e.g., the 2 power functioy, were considered but exhibited lower r values -- suggesting " poorer fits". The intercept of the fitted expression (7.2) compares quite well with the GESMO model value (7.5). However, the exponential coefficient of the fitted expression (0.00219) is somewhat different from the GESMO model parameter (0.00153). The significance of this difference is not readily apparent due to'the appearance of this parameter in the exponential, however, it' tan be judged through its potential impact on the population dose estimate. The population density will exhibit a maximu3 nfluence on the i population dose estimate n consideration of a non-depositing nuclide with a halflife long relative to the transnort time. For such nuclides, e.g., Kr-85, the population dose associated with the plume exposure is proportional to the spatial integral of the population density function, i.e., R man rem d p(x)dx, o where R is the plume pathlength. In the GESMO model a plume pathlength of 2000 miles was conservatively assumed for use with the density function, note in the GESMO model the growth parameter -g e. .-summ. m- -wa

4 was in fact determined by the plume pathlength parameter. In the case of the fitted function the parameters are derived from observa-tion of the population density which extend out to about 1800 miles. Beyond this distance, the fitted expression is not valid since farther locations would be in the Atlantic Ocean. The ratio of the integrals of the fitted function to the GESMO function is 7.2 x 0.00153 x exp(0.00219 x 1800)-1 7.5 x 0.00219 x exp(0.00153 x 2000)-1 or about 1.7. Thus, the fitted population density function would yield a population dose estimate a factor of about 1.7 higher J[i,s is the maximum 'FTtid and is -applicable than the GESM0 model. to long-lived nuclides not experiencing atmospheric deposition. In the case of Rn-222, consideration of decay during transport (resultant wind speed of 2 m/sec), lowers the above ratio to about 1.2. If one had not represented the population density with an analytical function, but simply numerically integrated the data of Table 1 a similar ratio would be obtained, i.e., about 1.2. In conclusion, the GESMO population density function em,loyed in evaluation of radon releases from the mining and milling industry yields results comparable (within a factor of two) with

5 those obtained using observed population densities. The density functica employed is not considered to be a major source of uncertainty i.D.the analysis nor a potential source of bias. It should be noted that in the above discussion the numerical values cited were based on the 1970 census data. In recent analysis of radon releases the population densities have been scaled up to a total U.S. population of 300 million which results in the initial density estimate being 11 people per square mile rather than 7.5, the terminal density becomes 240 vs 160. n N 7 98/ d = '9 es up m +-edh.. ee nu m A e + ese ee w= e ee e

TABBE 1 1970 POPULATION DENSITY Longitude Distance (miles)II) Density (#/mi )(2) 2 108 0 4.8 106 107 12.5 104 215 20.9 102 322 4.2 100 430 7.8 98 537 28.3 96 645 53.5 94 752 66.7 92 859 32.1 90 967 106.8 88 1870 183.9 86 1180 121.9 84 1290 136.1 82 1400 155.2 80 1505 142.4 78 1611 135.7 76 1720 233 footnotes II) distance measured from 105 longitude line (2) density is average for the 37,39 and 41 N latitude lines Ex,aanential curve fits 2 p(x) = 7.2 exp(0.00219x); r = 0.80 _}}