ML20070Q680

Evaluation of Procedures Proposed by Detroit Edison Co in Emergency Plans to Consider Meteorological Conditions in Enrico Fermi Site Coastal Environ

ML20070Q680
Person / Time
Site:	Fermi, Diablo Canyon, 05000000
Issue date:	10/08/1982
From:	Van Der Hoven COMMERCE, DEPT. OF, NATIONAL OCEANIC & ATMOSPHERIC
To:	Office of Nuclear Reactor Regulation
Shared Package
ML19302F581	List: ML20070Q680 ML20070Q688
References
CON-NRC-03-81-099, CON-NRC-3-81-99 NUDOCS 8301270051
Download: ML20070Q680 (4)

Text

. -- = : _: . - _ _ - . .

- *; s.

• i Evaluation of the Procedures

' Proposed by the Detroit Edison Company in their Emergency Plans to Consider lieteorological Conditions in the Enrico Fermi Site Coastal Environment by Isaac Van der Hoven Plational Oceanic and Atmospheric Administration Air Resources Laboratories October 8,1982 h

4 9

. .a .

I. The review which follows'is in response to Task Order No.1 of Inter-agency Agreement No. NRC-03-81-099 which requires an evaluation of the procedures proposed b/ the Detroit Edison Company in their emergency plans to consider r.eteorological conditions in the coastal environment at the Er.ri:o Ferr.f Atenic F:wer Plant, Unit 2.

!!. The documents reviewed which were submitted under NRC Docket No. 50-341 are as folicws:

1. Lake Breeze Analysis Methodology for the Fermi 2 Nuclear Power Plant Project, October 29, 1931.

2. Preliminary Methodology for Manual Dose Assessment for Emergency Preparedness, November 1981,

3. Development of Sector-Specific Recirculation-Stagnation Factor for Meteorological Data Collected at the Enrico Fermi Atomic Power Plant, Unit 2, September 13, 1978.

III. The reviews of the documents listed above are as follows:

1. " Lake Breeze Analysis Methodology for the Fermi 2 l Nuclear Power Plant Project" The empirical expressions describing the shape and extent of the TIBL, in-cluding the Raynor model (page 5), have very little observational verification.

In addition, the Raynor expression is based on a rather unique situation since Long Island can have sea breezes on both the south and north shore with the scuth shore being further complicated because of barrier reefs with extensive bodies of water between the reefs and the mainland. In addition, parameters in the expression such as the AT/AZ value over the water and the friction velo-city over the land are not easily obtained. Moreover, the inland extent of the

-lake breeze is not predicted by the equation on page 5 For emergency deci-sions, the inland extent and trajectory of the radiological effluent would be of prime concern. It is doubtful that calculational determination of the height of the TIBL, which is an important parameter in trapping and fumigation models, can be made to distances beyond the site boundary since the inland, 60 m tower is about 1 km from the shore and still within the site boundary.

2. " Preliminary Methodology for Manual Dose Assessment for Emergency Preparedness" The basic concept behind the applicant's methodology to determine concentra-tion isopleths is the so-called straight-line Gaussian model expressed as follows:

2 2 Q + H )),

X" **PE~I{y uncyoz g For a specific radionuclide, the methodology specifies a dose conversion tactor as a function of time of exposure (e.g., page 5). It is also assumed tFat a conitor reading is available downwind at a known distance and direction from the e

1. 4,~..

- s ..

2 source. The r.!onitor reading is then multiplied by the conversion factor as shown in the worksheet on page 4. ahich gives the activity released (Q) From a measurement of stability one can determine the Pasquill class and therefore cy and c2 as a functi:n of downwird distance. The wind speed, u, can be ceasured directly. Hopefully one also knows the effective height of release (H). However, in the .~ethodology there does not seem to be any way to deter-mine y, the arc line distance of the monitor from the location of the plume cen terl ine . Unless one has a number of monitors spread out at downwind loca-tions or a single mobile r.onitor making a number of plume traverses such that a plot of the monitored readings as a function of distance along the arc shows a reasonable Gaussian distribution and thus the location of the peak or center-line, there is no way to determige y by the suggested methodology. Thus, X.

the downwind concentration (Ci/m ) cannot be determined and there is no' basis for the determination of the red, yellow, or white emergencies.

3. " Development of Sector-Specific Recirculation-Stagnation Factor for Meteorological Data Collected at the Enrico Fermi Atomic Power Plant, Unit 2" The Puff Advection Model (PAN) described in this report is driven by a single station, time-varying wind field. A somewhat more sophisticated model called MES0DIF (Start and Wendell,1974) uses a 20-station wind measuring net-work over a 80x130 km grid. This allows the model to incorporate a two-dimensional wind field as computed from measured temporal and spatial varia-bility of the wind as compared to only the temporal changes in PAM. Neverthe-less, the results of ccmparing the straight-line Gaussian model with the wind field models are surprisingly similar when averaged over a year. The peak Gaussian / Wind Field sector recirculation ratio at Fermi on an annual basis was 1.79 at a distance of 2.4 km while the Idaho annual ratio was 1.79 at 14 km (Start and Wendell,1974) from the release point.

Both MES0DIF and PAM use winds at a particular height. Thus, return cir-culations at another height, with a reversal in wind direction such as would happen in a lake breeze situation, would not be detected. The Puff Advection Model would be particularly affected because it does not simulate the spatial changes occurring at the inland lake breeze " front". This could be detected by MES0DIF but on the other hand, it could not detect the rising air at the

" front" and the return flow towards the shore of a coastal site.

Reference:

Start, G.E., and L.L. Wendell, " Regional Effluent Dispersion Calculations Considering Spatial and Temporal Meteorological Variations." NOAA Tech.

Memo ERL ARL-44, May 1974.

IV. Conclusions regarding the documents as listed and numbered above are as ,

follows:

1. The TIBL height expression listed on page 5 needs to be tested with regard to its applicability to the Fermi site. This requires measure-ment of TIBL heights concurrently with measurement of the appropriate e

e

3 frput parareters to 'the expression. Also, a methodology to determine the inland extent of the 1ake breeze needs to be developed.

2. A nonitoring methodology needs to be developed so that at any given distance from the source, the distance to the plume center-line can be deter?.ined.

3. The Puff Advection Model is not adequate for a real-time, emergency response capability which will delineate the tra-jectory of the plume centerline.

b ddA !

Isaac Van der Hoven Air Resources Laboratories, NOAA October 8,1982 4

e 4

9

• e 9

e I

l .

• ~ . - . . _ , . . . ... ..