ML19207B808

Summarizes Atmospheric Transport & Diffusion Consultation & Evaluation of Field Support Activities

ML19207B808
Person / Time
Site:	Crane
Issue date:	04/17/1979
From:	Start G Metropolitan Edison Co
To:	Dickson C Metropolitan Edison Co
References
NUDOCS 7909050305
Download: ML19207B808 (3)

Text

,

DodcEf No

~

60-320 April 17, 1979 tienorandun To:

C. Ray Dickson bE.db+r From:

G. E. Start

Subject:

'D4I:

Summary of Atmospheric Transport & Dif fu11on Consultation

& Evaluation of Field Support Activities A.

General Activities 1.

Basic support provided to the DOE Command Post (at Capitol City -

Airport), and to the supporting ceasurement groups coordinating through DOE, was adequate and well received. Approximate plume trajectories and relative plume concentrati 7a estimates (x/Q) were providad (usually based on ARAC runs of ADI C).

Peric.dic es timates of soure utrength (Q) were calculated using aerial measurements of plume autoactivity, calculated x/Q values, and a bulk conversion constant.

In addition, past conditions (trajectories, X/Q, etc.)

and expected winds and weather were presented at daily DOE CP briefings, periodic critiques of sampling operations, and to 3pecific participating groups as the need arose.

~

L

~ ' ' '

2 The level of ARAC activity necessary to support the DOE CP needs vas evaluated and LLL personnel were guided in adjustments of the level-of-staffing and intensity of operation.

i-l 3.

Professional judgment was exercised in the usage of all N9S and i

ARAC products prior to dissemination. Adaptations and redetc.rmi-

[

nations by alternate methods (to evaluate local forecasts, trajectories, atmospheric diffusion, etc.) were made as needed for these products.

[

4.

Support groups and other federal agencies conducting field c

measurenents and environmental monitoring activities were advised I

(on an individual basis as needed) concerning the role of the -

[

atmosphere in transport and diffusion of airborne materials.

Archived meteorological data and the possible types of (af ter-the-i fact) calculations which may be useful to the interpretations and

}

comparisons of various data were discussed.

In this manner a

{'

more completely integrated description of the atmospheric related L-behavior of emissions and their fate should be reconstructed for a final report.

I B.

Evaluation of Field Support Methodolcafes E

I 3.

ihe field support activitics (for atmospheric transport and

?

diffusion) were usually adequate for the situation as it f

occurred. When LLL personnel arrived upon the scene (N2 days later) and telephone and telecopier connunications links vere established fjjl}]!M,.,

0Mf h

E

• K C. Ray Dickson April 17, 1979 in a local DOE Command Fost (N3 days later), the support was acceptable.

Ilowever, if a nassive discharge of radioisotopes had occurred during the first tuo or three days, I doubt that the methodology would have been suf ficient.

Certain communications and ARAC computational products were being relayed to Emergency Operation Centers of either NRC, or DOE, or both.

These EOCs were in the Washington, D.C. area.

The content and 2dequacy of these LLL communications is not known.

The ARAC system was not established to be a full-tine operations center and as the days passed, the limited staff (of mostly researchers) became taxed by long shif ts day-after-day.

Since the discharges to the environment from the TMI plant were small and quickly diminished, the field support system was never put to the ultimate test; consequently a critique of performance does not identify possible short comings if a major atmospheric emission were to have occurred.

2.

NWS products (teletype transmissions of r surf ace observa*ic 9 and NAFAX charts) were available about 500' away frot the taE connand post at the FAA Flight Service Station in the capitol City Airport Building.

This source of information was overlooked unti3 the second week of field support operations. These products could have been utilized to manually deduce atnospheric trajectories in an emergency if no other data were available; this source of information should not be overlooked in any future field support operations.

3.

For a relatively modest cost, a higher quality communications and graphics terminal nay be used to receive and transmit inf o rmation at the field command post. A more legible and higher quality product could be transmitted in less time with a graphics scope 1

and video hard copy unit in the field. / tother possibility would be a terminal system capable of receivint, AFOS products from NWS, and trajectory and diffusion model outpet from an external source i

or from a simple model run in the field.

4.

A simpler diffusion model which may ba calculated more rapidly should be considered.

The ARAC model run at Livermore provided helpful guidance I

but some practical drawbacks were noted. Model output ( ARAC) was i

obtained for 1 to 1-1/2 hours following input of latest wind data.

Wind 3

persistence for 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> was assumed and the calculated concentrations j

provided were for data time plus 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />.

The model, as executed for i

TMI support, utilized a grid cell size of about 2-1/2 kan.

This resolu-

[

tion is too coarse for estimation of plune peak concentrations within E

the first few to ~10 kn downwind of the release area.

For back calcula-

[

tions of source terms, especially for aerial monitoring within the first t.

several km, the calculated normalized concentrations should be specified j

with better resolution; the ADPIC output contours appeared to be about an order of magnitude too small due to limited grid cell resolution in the presence of smaller plume dimensions and larger concentration t

{

gradients. The product, as available for TMI field support, required i

h-0$3]CG t

\\

2L C 3ay t h na April 17, 1979 9

guessing an approximately correct peak concentration near the plant (within r"10kr... Again, due to the need to use 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> vind per-sistence and with a requirement for better definition of normalized plume concentrations at the shorter distances, a simpler (even Gaussian or uniformly mixed) plume model seems soaewhat appropriate.

If a massive release were to occur, both the shorter distance and longer distance trajectories and plune concentrations would be desired.

In this situation either a simpler fine resolution model, two dif ferent models (either nested runs of the same model or 2 distinctly dif ferent models), or a model with non-uniform grid cell size is needed.

5.

ARAC products have not been shown to really be any better or more useful than simpler models calculations. ADPIC does have a limitation upon the number of grid cells usable in finite-sized cocputer memories and the resulting trade-off between spatial resolution and spatial coverage.

The concept of ARAC seems sound and an emergency operations center vould need a regional plune description for postulated massive atmospheric discharges of radionuclides.

Fewever, one or more simple scoping estimates of plune concentrations within a few minutes at for the initial 3-10 km downwind are most urgently needed. The present or a modified ARAC type of system should provide this initial information.

6.

The ADPIC model used by ARAC, as well as most other models, lacks a sound objective method for specification of.taospheric rates of diffusion. For each model type, a stabil A6y category is estimated and then Ty & T: values are determined or e K (dif fusivity) value is derived from these signas. ADPIC shapes a vertical profile for the derived K value although there seems to be little basis for selection of a particular profile at any particular time.

Here again, due to a lack of suf ficient initial and boundary condi: ions, and limited updating of atmospheric parameters, the magnitude of computational uncertainties of the complicated models (such as ADPIC) may easily contain (or be included in) the uncertainty envelope of acceptatle and sinpler plume diffusion modele.

~~

7.

A "saving" feature of the transport and diffusion supporting guid-ance provided to DOE (& NRC) during Three } Ele Island was the "on-site" or in command post presence of one or more highly experienced atmospheric transport and dif fusion specialists.

The standard 1

products which were provided required a careful professional evalu-

{

ation to assure their proper interpretation and use.

In the event that a rapid change of conditions resulted in a massive aerial dis-charge of radionuclides, probably only the quickly calculated and/or graphical presentations of atmospheric relative diffusion (assembled by the "on-site" specialist) would be timely enough to guide rapid decisions of potentially critical importance. The interpretative, professional guidance was probably of equal importance (or more) corpared to whether or not a complicated or a simplified atmospheric dif fusion model were utilized in a regional advisory center, because both the timeliness of t'.e guidance and profecsional (somewhat qualitative) avaluations must (and could) be included.

313.JCG r

r-

.