ML20084N716
| ML20084N716 | |
| Person / Time | |
|---|---|
| Site: | Fort Calhoun  |
| Issue date: | 05/09/1984 |
| From: | William Jones OMAHA PUBLIC POWER DISTRICT |
| To: | John Miller Office of Nuclear Reactor Regulation |
| References | |
| RTR-NUREG-0654, RTR-NUREG-654 LIC-84-053, LIC-84-53, NUDOCS 8405170178 | |
| Download: ML20084N716 (13) | |
Text
___
r---
Omaha Public Power District 1623 Harney Omaha, Nebraska 68102 402/536-4000 May 9, 1984 LIC-84-053 Mr. James R. Miller, Chief U. S. Nuclear Regulatory Commission Office of Nuclear Reactor Regulation Division of Licensing Operating Reactors Branch No. 3 Washington, D. C.
20555 i
Reference:
Docket No. 50-265
Dear Mr. Miller:
Class A/B Site-Specific Atmospheric Diffusion Model N
Please find attached a description of Omaha Public Power District's site-specific Class A/B atmospheric diffusion model. This description is be-i ing submitted pursuant to Appendix 2 - Annex 1(7) - of NUREG-0654. This model is a combination of Class A (to be used out to the plume exposure EPZ) and Class B (to be used out to the plume ingestion EPZ) models. Please note the District's Class A/B model is currently in operation and was uti-lized for dose assessment functions during the 1983 annual exercise.
Sincerely, I
dh p'U b l} j,L v.ff h
- W. C. Jones r
Division Manager f
Production Operations l
l WCJ/JJF/nh l
Attachments f
i cc: LeBoeuf, Lamb, Leiby & MacRae Mr. E. G. Tourigny, NRC Project fianager Mr. L. A. Yandell, NRC Resident Inspector i
8405170178 840509
/
nk/
l PDR ADOCK 05000285 l
F PDR V
i I
i
_=,-
Description of Atmospheric Diffusion Model The site-specific atmospheric diffusion model is designed to provide diffu-sion estimates for gaseous effluent releases from the Fort Calhoun Station using actual 15-minute average, on-site meteorological data. This base model is a Gaussian " segment-plume" model. This model is a combination of Class A (to be used out to the plume exposure EPZ) and Class B (to be used out to the ingestion EPZ) models as outlined in Appendix 2 (page 2-3) of NUREG-0654.
The " segment-plume" model is a variable trajectory model based on the stat-istical approach to diffusion. The model allows the oeformation of a contin-uous plume by dividing the plume into a number of contiguous segments. Each plume segment describes a portion of plume behavior between successive time intervals with advection by the local wind field and diffusion in a Gaussian fashion. Ground-level concentrations and depositions are calculated by de-termining the contribution each plume segment makes to the grid of receptors over which it passes.
The " segment plume" model is one class of the plume element models identi-fied in Regulatory Guide 1.111. The basic model has the capability to account for both ground and elevated modes of effluent releases, buoyancy and momentum plume rise, building wake influence, terrain interaction, and dry deposition. Detailed descriptions of the " segment-plume" concept and modeling algorithms for the Fort Calhoun Station emergency preparedness ac-cident assessment are presented below.
A.
BASIS OF THE SEGENT-PLUE MODEL The fundamental basis for the " segment-plume" model is an integral mass balance over a finite plume segment. Mathematically, the conser-vation of mass over a plume segment of length can be expressed by the following mass balance equation:
AQ =
/,I,{G(s.r,z)drdz)ds
/,,/,uxdrdzls+as- /, f. uxdrdzls (I)
+
Where s, r, and z are the longitudinal, lateral, and vertical plume co-ordinates, G(s.r,z) is the rate of change (gain-loss) of effluent con-centration X(s,r,z) by decay and removal processes, AQ is the result-ant rate of change of effluent mass, and u is the wind speed. Assum-ing a quasi-steady state, G(s.r,z) and u are considered to be constant from s to s+As, where s is the currert distance of a plume segment end-point from the release point, measured along the plume axis.
4 Two possible vertical concentration distribution functions can be chosen in the model for diffusion estimates: Case 1; a vertical Gaussian pro-file, ignoring any effects of the mixing depth; or Case 2; a uniform ver-
' tical distribution below the mixing lid.
For Case 1, the ground-level concentration (s.r) of a plume segment is calculated by:
2 x(s.r)=
0(s) exp[
-r
] ex [
-h
]
(2) 2 2
vuo(s)o(s) 2a 20 z
y y
z
~
\\
s
... N.,
m Whefe Q(s) is the effluent mass flux, he is the effective plume
- height, and oyfs) and a (s) are, respectively, the lateral and verti-z cal dispersion coefficients at downwind distance s.
Full reflection from the ground is assumed.
s For Case 2, if the effecti've plume hotsht lies below the mixed lid Hme the ground-level concentration of the' plume segment is calculated by
(
i the expression for unifom vertical mixing:
1
,N cx(s.r) 2]
(3) 0(s) exp[
-r
=
2suHmoy(s) 2ay 2 t
s sWhere Hm is the mixing depth encountered by the plume segment.
If the hiume center 11nt lies above the mixing lid, no ground-level concentra-
{
tions' are calculated.
t s
The computational scheme of lhe t" segment-plume" nadel'has three dis-l t
tinct functional elements:
(1)'a Lagrangian plume traje'ctory func-i tion, (2) a plume dispersion function, and (3) a plume sampling func-tion.
Th'e t.agrangian plume trajectory function is used to advect the end-s points of each plume segment during'a basic time step; the resultant n
y distance betwee'n c,onsecutive ernipoints defines the length of each
, plume segment.
In a temporally-varying but spatially-homogeneous wind field, the position of the endpoint fran the source can be expressed u
T mathematically by the fo.' lowing equatiopt-s(t+at)=s(t)+u(t+At)at (4)
The plume dispersion function detemines the horizontal spread of a i
plume along its trajectory by the following equation:
oy(s+AS)a oy(s) + As[.!! Y.]
ds s+As/2 (5) az(s+AS) = Oz(s) + As[.!!0z,]
ds s+As/2 (6) 1 These tems allow for spatial and temporal changes in atmospheric sta-l bility to be included, without violating the entropy principle (center-i
-line concentrations cannot increase with downwind distance).
i t
The set of equations (1) through (6) provides the theoretical frane-work of the " segment-plume" model under a temporally-varying but spa-tially-homogeneous meteorological field. Under steady-state condi-tions, the model should yield the same results as the straight-line Gaussian dispersion model.
B.
MAJOR FEATURES OF THE MODEL The site-specific atmospheric diffusion model is based on the Gaussian i
h
" segment-plume" concept. The model combines and enhances various dis-persion modeling algorithms into the computer program that can be used to provide near real-time diffusion estimates during a radiological emergency condition and to detemine the dimension and location of the plume trajectory, and the location and magnitude of maximum concentra-tions and depositions for each 15-minute time interval. The model j
t
'i
- } :<
5 l
uses a time-dependent, homageneous wind field as meteorological input.
(
Major features of the m6 del are described below.
i
~
i I
B.1 jtefeaseMode t
Two types of' source release modes are considered:
elevated re-j
^
lease and ground-level release. The elevated release consists of any release in which the effective release height is highar than twice the height of an adjacent solid structure. The
't i
ground-ledl release includes all releases with a release height e
of less than twice the height of the tallest adjacent structure.
f
[
In practice, the release height for the ground-level release is l
assumed to be at the 10-meter level above the ground.
l A
B.2 Atmospher'ic Dispersion Coefficients j
j The atmospheric dispersion coefficients used in the model calcu-l 1ations are based on the Pasquill-Gifford's curves presented on 1
Figures 1 and 2 of Regulatory Guide 1.145. As indicated in these two figures, the horizontal and vertical dispersion coef-i ficients, oy and oz, withoat building wake effects are restrict-ed to no greater than 10,000 and 3,000 meters, respectively.
i i
The following approximations are used to estimate the atmospher-ic dispersion coefficients under extremely stable atmospheric
. conditions (Pasquill stability class G):
ay(G) = 2/3 oy(F)
(7) oz(G) = 3/5 oz(F)
(8)
The atmospheric dispersion coefficients are calculated for each l
plume segment in a given time period using Equations (5) and (6), based on the values they had for that plume segment in the i
previous time period. Since the atmospheric stability may change during real-time dispersion, the derivative tenns in j
these two equations account for the growth of a plume as a func-i tion of travel distance along the plume trajectory for each spec-ified time interval.
i 1
I B.3 Building Wake Adjustment
(
i The building wake adjustment is applied to ground-level releases only and follows the fonnulation stipulated in Regulatory Guide l
1.111. This fonnulation is expressed by:
I oy' = ( y + CA/w)1/2 (9) az' = ( z + CA/w)1/2 (10)
Where:
l I
l oy', oz' are nodified horizontal and vertical disper-sion coefficients accounting for building-induced turbulence, respectively;
[
ay, az are Pasquill-Gifford horizontal and vertical dispersion coefficients without building influ-f ence, respectively; L
9
..r-..,
,.c-
,-m.-----,-
u,.
=- _
C is building shape factor (C = 0.5); and i
A is minimum vertical-plane cross-sectional area of the reactor building.
i B.4 Effective Plume Height l
For elevated releases, the effective plume height is calculated fran I
i he = hs + Ah - ht-c (11) i Where:
i i
he is the effective plume height; ah is the final rise of the plume above the release l
point; i
hs is the physical stack / vent height; l
ht is the terrain height between the release point and l
the point for which the calculation is made (ht must I
be greater than or equal to zero); and t
i l
t l
c is the correction term for stack downwash i
r L
. ~ _ _,.. - _ _, -.
When the vertical exit velocity is less than 1.5 times the hori-zontal wind speed, a correction for stack downwash is subtracted l
t from Equation (11) by the following equation:
c = 3(1.5 - W/u)d (12)
Where:
i e
d is the inside diameter of the stack / vent release l
point; i
u is the mean wind speed at the height of release; and W
is the vertical exit velocity of the plume.
i Both the momentum and buoyancy aspects of the plume rise is in-corporated into the model using Brigg's formulae. The higher P
final plume rise due to either its momentum or buoyancy effect is used in the atmospheric diffusion calculations.
For neutral or unstable conditions, the final momentum plume rise is calculated from:
ahl = 3.0(W/u)d (13) ah
=
m For stable conditions, the final momentum plume rise is cal-I culated from:
min (ah, ah, ah )
(14) ahm l
2 3
=
With 4(Fm/s)l/4 Ah2
=
1.5(F /u)1/3 s-1/6 Ah3
=
m L
0.25(Wd)2 l
Fm
=
i t
I g/T cQ/oz s
=
Where i
Ahm is the final plume rise due to the momentum of the i
plume; Fm is the momentum flux parameter; i
s is the stability parameter; i
T is the ambient temperature; g
is the acceleration of gravity; and 9
is the ambient potential temperature.
For unstable or neutral conditions, the buoyant plume rise, Ab e b
is calculated from:
l 1.6 F /3 (5x}2/3 u-l (15) 1 Ahb
=
4.3 x 10-3 Qh With F
=
0.52 F.4 h 0.6 O
and x =
3 Where Ahb is the final plume rise due to buoyancy of the plume; hs is the stack or vent height; Qh is the effluent heat flux; and j
F is the buoyancy flux parameter.
Under stable conditions, ahh is defined by:
2.9(F/us)1/3 (16) abb
=
B.5 Vertical Variation of Wind Speed l
The vertical variation of wind speed with height is calculated l
by the following power-law relationship:
u = um(2/Zm)P (17)
I i
Where u
is the wind speed at height z; I
I i
is the wind speed at the sensor height z ; and um m
P is the power law exponent whicit is stability dependent.
l Values of P as a function of atmospheric stability are taken
[
from: DeMarrais, G.A.,1959; " Wind Speed Profiles at Brookhaven National 1.aboratory", Journal of Applied Meteorology, Vol.16,
[
t pp. 181-189.
[
i B.6 Plume Depletion i
l Radioactive material may be removed when the plume touches vege-tation or other surfaces. These physical removal processes are included in the plume dispersion calculation by means of a cor-l rection factor for plume depletion. This correction factor is I
i expressed as the fraction of material released which remains in l
the plume.
The plume depletion correction factors shown in Figures 2 I
through 5 of Regulatory Guide 1.111 are used to assess plume-I depletion effects for all distances from the source and atmos-l pheric stability classes for both ground and elevated release modes. The relative concentration calculated at a given recep-tor us' Cquations (2) or (3) is multiplied by the fraction remaining in the plume, as detennined from these figures, to arrive at the corresponding depleted relative concentration.
l B.7 Dry Deposition 1
Dry deposition of elemental radiotodines and other particulates i
is calculated for both ground and elevated releases. The rela-tive deposition rate as a function of release height and atmos-pheric stability, shown in Figures 6 through 9 of Regulatory Guide 1.111 is used in the model. The relative deposition rate f
i is deposition rate per unit downwind distance divided by the source strength. To obtain the relative deposition per unit area at a given receptor in a given sector, the relative deposi-tion rate is divided by the arc length of the sector.
It is l
noted that the figures in Regulatory Guide 1.111 are based on I
the assumption that the effluent concentration in a given sector j
is unifonn across the sector at a given distance. For the "seg-ment-plume" model where concentration at a given distance is not l
unifom across the sector, the relative dry deposition at a spec-i ific receptor is calculated by the following procedure:
1.
Detemine the relative deposition per unit area at the re-ceptor as a function of release height and atmospheric sta-i bility;
{
l 2.
Multiply the undepleted relative concentration at the re-l ceptor by the relative deposition per unit area; and i
3.
Divide the resulting value by the undepleted sector-aver-I age relative concentration across the sector.
t t
.., _. =
,.--J