ML20101Q874

Rev 1 to Description of Perry Nuclear Power Plant Emergency Offsite Dose Calculations

ML20101Q874
Person / Time
Site:	Perry
Issue date:	12/31/1984
From:	Friedman L, Mitchell A, Wilkinson M NUS CORP.
To:
Shared Package
ML20101Q864	List: ML20101Q862 ML20101Q874
References
NUS-4336, NUDOCS 8501080379
Download: ML20101Q874 (71)

Text

_ _ _ _ _ _ _ _ _

I I

NUS-4336 I

Revision 1 I

DESCRIPTION OF THE PERRY NUCLEAR POWER PLANT EMERGENCY OFFSITE DOSE CALCULATIONS I

I I

Unit No. 1 Unit No. 2 Docket No. 50-440 Docket No. 50-441 Prepared for The Cleveland Electric Illuminating Company I

Revised December 1984 (April 1983)

M.J. Wilkinson E

A.E. Mitchell B

L.A. Friedman l

l NUS Corporation 910 Clopper Road Gaithersburg, MD 20878 5

pausa8ajia

I Table of Contents Section No.

Page I

l 1.0 Introduction 1-1 2.0 Hand-Calculated Emergency Offsite Doses 2-1 l

2.1 Preliminary Estimate of Dispersion 2-2 I

from Onsite Data 2.2 Related Meteorology 2-3 2.3 Tabular Estimate of Dispersion 2-4 2.4 Dose Assessment Based on Effluent 2-4 I

Monitor Reading 2.5 Dose Assessment Based on Effluent 2-5 Analyses I

2.6 Dose Assessment Based on Containment 2-5 Monitor Reading 2.7 Dose Assessment Based on Containment 2-5 g

Analysis 3

2.8 Dose Assessment Based on Offsite 2-6 Measurements 2.9 Dose Projection Based on FSAR Analyses 2-7 1

2.10 Calculation of Accumulated and Pro-2-7 jected Doses from Releases 3.0 Automated Emergency Offsite Dose Calculations 3-1 3.1 General Model Description 3-1 3.2 Lake Breeze Model Capabilty 3-9 I

3.2.1 Detection of the Lake Breeze 3-10 3.2.2 TIBL Trapping 3-10 3.3.3 Lake Breeze Helical Circulation 3-11 3.3 Meteorological Basis 3-12 3.3.1 Results of Lake Breeze Case 3-12 Study 3.3.2 Lake Breeze Climatology for Perry 3-14 3.4 Input Meteorological Data 3-20 a

3.4.1 Meteorological Tower and Valida-3-20 g

tion 3.4.2 Determination of Stability Class 3-22 3.4.3 Other (Supplemental) Local Meteo-3-24 rological Data 4.0 Summary 4-1 5

ii Revision 1

I Table of Contents (Continued)

Section No.

Page 8

5.0 Cited References 5-1 6.0 Cross-Reference to NRC Coments 6-1 a

6.1 Round 1 Questions 6-1 E

6.2 Round 2 questions 6-2 6.3 Contractor Evaluation Findings on Meteorology 6-3 I

I i

I I

E I

~ I I

I I

i

5 iii Revision 1

=

I I

List of Tables Table No.

Page 1

Algorithm for Detection of the Lake Breeze 3-26 2

Model Values of Surface Temeprature for Lake Erie 3-27 1

3 Algorithm for the Maximum Inland Penetration Distance of the Lake Breeze 3-28 4

Algorithm for the Return Flow of the Lake Breeze 3-29 5

Mesosynoptic Characterization for April through i

July 1980 at Perry 3-30 6

Wind Speed at 10-m level of the Perry Meteo-I rological Tower During lake Breezes in April through July 1980 3-31 1

7 Histogram of Observed Wind Direction at Perry During Lake in April through July 1980 3-32 8

Pasquill-Gifford Stability Class (60-10m Delta T)

I at Perry During Lake Breezes in April through July 1980 3-33 9

Land-Water Temperature Difference During Sea I

Breeze Hours at Perry Nuclear Power Plant, April-July 1980 3-34 I

10 Conditions Observed at the Perry Nuclear Power Plant in the Hour Before the Onset of a Lake Breeze Regime 3-35 11 PNPP Meteorological System Equipment Specifica-tion 3-36 1

12 Summary of Variables Reported from the Perry Meteorological Tower 3-39 I

13 Alternate Data Sequence in ERIS for Onsite Perry Meteorological Data 3-40 14 Classification of Atmospheric Stability by Temperature Change with Height 3-41 I

iv Revision 1

I I

List of Tables (Continued)

I Table No.

Page 15 Classification of Atmospheric Stability by i

Sigma Theta 3-42 16 Modified Sigma Theta Method to Estimate Stability Class for Sigma Z from Sigma Theta 3-43 17 Pseudo Sigma Method to Estimate the Atmospheric I

Stability Class Applicable to Sigma Y on the Basis of Delta T 3-44 18 Sources of Local Meteorological Data 3-45 5

I I

I i

t R

I

' I l

l l 8 v

Revision 1

l I

List of Figures I

Figure No.

Page 1

Lake / Sea Breeze Circulation and Its Effect I

on Near-Coastal Releases 3-46 2

Locations within 100 Miles of the PNPP 3-47 3

Local Sources of Meteorological Data 3-48 4

TIBL Depth as Function of Inland Distance 3-49 I

1 5

I I

I I

I I

I i

l vi Revision 1

I

1.0 INTRODUCTION

I This revised report is issued (1) to reflect the installation of the selected dose assessment model at Perry and (2) to respond to the NRC's review comrrents (NRC letter, Youngblood to Edelman, August 22, 1984: NRC Contractor Evaluation Findings of Use of Meteorology in Emergency Response at Perry Nuclear Power I

Plant (Units 1 and 2)). Accordingly, Section 3 has been totally revised, as well as references in Section 5.

The cross-references in Section 6 have been modified to correspond to the revised Section 3 and to correspond to NRC recom-mendations. Change bars are added to all sections except Section 3 that is completely new.

The Cleveland Electric Illuminating Company (CEI) will adopt for use at the I

Perry Nuclear Power Plant (PNPP) two methods for determining offsite doses during an emergency:

a computerized method and a hand-calculated method.

In I

this document the methods of making these emergency offsite dose calculations will be discussed in the context of the Perry Emergency Response Facilities (ERFs) that are described separately (CEI,1983). The MONICORE computer system incorporates a meteorological information and dose assessment system (MIDAS) that will run in conjunction with the Emergency Response Information System (ERIS) computer. MIDAS incorporates (1) a straight-line Gaussian approach as well as (2) a plume segment approach that is able to simulate the I

lake breeze meteorology of this coastal location.

MIDAS is the dose projection methodology of choice because it provides rapid l

dose assessments based on up-to-the-minute treteorological and radiological data. Provisions, however, are being made for dose projections in the event that MIDAS is inoperable or unavailable.

In the event that MIDAS is inoperable l

or unavailable, dose projections will be calculated by hand. Hand calculations I

do not account for such things as lake breeze effects and time-varying source terms.

This document provides CFI's response on the technical bases of the dose calcu-lational methodology used to assess the impact of an accidental airborne release.

While the bases of both the automated and backup manual methods are generally described in Section 7.3.11 of the Emergency Plan, details are provided here lI 1-1 Revision 1 l

!I I

of the assumptions, models, and technical bases used in developing these calcu-I lational procedures.

The remainder of this document is organized into five sections that involve the following-I o

Hand-calculated emergency offsite doses I

o Automated emergency offsite dose calculations I

o A surmiary o

A list of references that were cited in the text o

A cross-reference of sections of this document that respond to NRC I

comments.

I I

I I

I I

I I

I I

1-2 Revision 1

I I

2.0 HAND-CA!.CULATED EMERGENCY OFFSITE DOSES Dose projections will only be calculated by hand in one or both of the follow-ing situations:

I o

MIDAS is non-functional o

The Technical Support Center (TSC) or Emergency Operations Facility (EOF) have not yet been activated.

In the second situation, dose projections are carried out in the Control Room I

until either the TSC or E0F have been activated (CEI, 1983). The hand calcula-tion of offsite doses is therefore a backup method for use in the event that MIDAS cannot be used to generate computerized dose projections.

I The method to be selected for hand calculation of offsite doses is based on the availability of data and on the time constraints for performing a dose assessment. These methods are discussed in Section 7.3.11.2 of the Emergency Plan.

This section contains descriptions of the assumptions, methods, and technical I

bases used in generating the hand-calculated dose projection procedures. The instructions for the hand calculations are contained in the Emergency Plan Implementing Instructions. Each dose projection method is contained in a separate attachment to that instruction; the basis of each attachment is described below.

In the 10 subsections that follow, the first three deal with obtaining the I.

atmospheric dispersion parameters that are required for dose calculations.

The next five subsections use the atmospheric dispersions parameters and avail-able monitoring parameters to calculate offsite dose rates. One subsection is available as a quick method to determine offsite dose when monitoring data are not available. The last subsection is used to determine accumulated and pro-jected offsite doses based on dose rates calculated in previous sections and on the estimated duration of the accident.

2-1 Revision 1

I I

A number of the dose projection procedures require an identification of accident type before the analysis can proceed. Accident identification provides a source term as the starting input for the dose projection.

It is the m asibility of the operator to identify the accident.

For all of the dose calculation procedures addressed in the remainder of this chapter, the standard methodology of multiplying a release rate by a dispersion factor and a dose factor is used; this methodology is employed in Regulatory Guide 1.109, Revision 1 (USNRC, 1977).

In all of these methods, the Chi /Q values for the site boundary and downwind distances are obtained from the more appropriate of the two methods contained in the first three subsections. Next, the appropriate dose factor (s) are selected for whole body, child thyroid, or both. Selection is based on the source term that results from each incident, i.e., the amount of noble gas and iodine released. Finally, the main difference from one calculational method to another is the manner in which radioisotope release rate is determined. Using actual grab sample analysis and release flow rates, an actual release rate can be calculated. Otherwise, the release rate must be inferred from available data. Obviously, actual isotopic analysis is the most accurate means of assessing the release. Once the concentration and dose factor have been determined, the difference in dose rate (R/hr) or dose projection (rem) at each of the downwind locations is the result of the l

differing amount of atmospheric dispersion at these locations.

2.1 Preliminary Estimate of Dispersion from Onsite Data One part of this method describes the automatically determined dispersion infor-mation. A " Preliminary Estimate" of the information is prepared by each of the independent systems at the Perry meteorological tower. This information includes normalized concentration (Chi /Q), the direction of plume travel, the speed of plume travel, the travel time, and the plume width. Both the Main and Backup Systems have a microprocessor (MDPS, Meteorological Data Processing System) which uses validated, realtime, 15-minute meteorological data to prepare the Preliminary Estimate.

(Same-tower substitutions are obtained if data are I

missing; see Section 7.3.7 of the Emergency Plan for further discussion.) Each I

2-2 Revision 1

I I

MDPS routinely sends the Preliminary Estimate information to the Contrei Room (as well as ERIS) so it is always immediately available. The Preliminary Estimate can also be obtained by interrogating either MDPS by telephone or by going to the Meteorological Tower for a hard copy.

The Preliminary Estimate calculations are less sophisticated than those for the Model A' that are performed in MIDAS. The Preliminary Estimate uses the l

FSAR approach to atmospheric dispersion estimates. A straight-line Gaussian dispersion model, as described in Regulatory Guide 1.145 (USNRC, 1979), is used for consequence assessment; release characteristics are the same.

Input meteorological data are wind speed, wind direction, and atmospheric stability class.

2.2 Related Meteorology I

This part provides the methodology for acquiring the meteorological information needed to obtain an estimate of atmospheric dispersion (Chi /Q) at selected I

distances from the site. This nethod is only used when an automated Preliminary Estimate (discussed in 2.1) is not available or when values for other distances are desired. This method is used to generate wind speed, wind direction, and stability. This information is then used with the methodology described in the next section to generate an estimate of atmospheric dispersion.

Of course, onsite data are preferred for this method because the Perry meteoro-l logical tower location is representative of the site region. However, provision l

is made, too, for using offsite sources should they be needed. The Cleveland National Weather Service is open 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> a day. Closer high-quality sources are preferred when they are available to supply wind speed, wind direction, l

and have the necessary observations to eventually yield stability class. Stabil-ity classification schemes include delta T (USNRC, 1972), modified sigma theta (Mitchell & Timbre, 1979; USEPA, 1981), and Turner-Pasquill sky conditions l

(Turner,1970),

i 2-3 Revision 1

I I

2.3 Tabular Estimate of Dispersion I

This method describes the means of generating the preliminary estimate of dis-persion from the output of the method described in Section 2.2.

Seven tables of dispersion parameters are presented; they are organized by stability class.

From each table the normalized concentration (Chi /Q), plume travel time, and plume width can be determined.

I These tabular estimates are based on the same technique used for making the Preliminary Estimate (Section 2.1).

Input data required are current wind speed, wind direction, and atmospheric stability class. The straight-line Gaussian model used is in accordance with that described in Regulatory Guide 1.145 (USNRC, 1979). Release characteristics are the same as used in the Perry FSAR. Disper-sion values are generated for the Exclusion Area Boundary and for each incre-mental mile out to 10 miles.

2.4 Dose Assessment Based on Effluent Monitor Reading I

This method can be used to project offsite dose and release rates when the release is monitored by an effluent monitor, the release flow rate is known or can be estimated, and the accident (incident) that causes the release can be correlated to an accident type analyzed in the FSAR. This method is only used when actual analyses of the release are unavailable. Since actual analyses are not availab'.e, the source terms from the FSAR are used. An identification I

of the accident must, therefore, be made first so that the appropriate source terms can be determined.

In this method, the effluent monitor reading is combined with the effluent release ficw rate to obtain a release rate. However, an initial identification of the accident type (FSAR) must be made before determining the effluent release rate. In the event that the flow rate out this path is zero, this method cannot I

be used because the effluent monitor readings will be invalid. AFter the release rate is determined, it is multiplied by the appropriate dose factors and Chi /Q values as detailed above to obtain dose rates in R/hr at each of the four down-wind locations, i

2-4 Revision 1

I I

2.5 Dose Assessment Based on Effluent Analyses This method can be used to determine dose rates at selected downwind locations using a known isotopic release rate. This method is appropriate when sample results provide a radionuclide breakdown for the release.

I This method is the most accurate of any of the methods described in this pro-cedure since it is based on:

o An actual measurement of the radionuclide mix and concentration being released I

o Actual measurements of the flow rates from the event.

In this method actual concentrations of noble gases and iodines are determined from analyses of effluent samples. The actual concentration is multiplied by the actual release point flow rate, appropriate dose factors for each identified isotope, and Chi /Q to obtain dose rates in R/hr.

I 2.6 Dose Assessment Based on Containment Monitor Reading This method can be used to project offsite dose rates and release rates based on the high range containment monitor reading. This method assumes that the containment activity is being released at the design leak rate of 0.2 percent per day and that 96 percent is collected and filtered by the filtration system and 4 percent is released directly (FSAR source assumptions). This method is used only for accidents inside the containment when containment ventilation is not operating. Release rates in this method are inferred from the containment I

release ratas dcccribed above and the readings of the hign range containmeni, monitor.

2.7 Dose Assessment Based on Containment Analysis I

This method can be used for projecting offsite dose rates based on a measured isotopic concentration in containment. This method is appropriate when sample I

I 2-5 Revision 1

I Draft 11/29/84 results provide a radionuclide inventory that could leak from containment.

In the event of a LOCA, this method accounts for child thyroid dose rates due to releases from both direct leakage to the environment and indirectly through a filtered pathway.

Using the containment activity release rate described in Section 2.6, as pro-vided in the FSAR, and the isotopic analyses as actual source terms, the l

designed cor.tainment leakage release rate is calculated.

This method would only be used when a LOCA has occurred and the containment ventilation is not o,.c sting.

2.8 Dose Assessment Based on Offsite Measurements This method can be used to project offsite dose rates and release rates from offsite measurements of dose rates or iodine concentrations. An estimate of the atmospheric dispersion factor is required for the sampling location. This method assumes that the offsite isotopic composition for dose rate measurements corresponds to a representative FSAR accident type (for estimates based on external dose rate measurements). For gross iodine measurement this method makes the conservative assumption that all iodine is I-131.

I For the calculation of doses using an offsite dose rate reading, the reading in R/hr is divided by the Chi /Q at that location to obtain a release rate.

Calculations for dose rates at each of the downwind locations then proceed as described at the beginning in Section 2.4.

g For the calculation of doses using measured offtiM i: dire concentrations, the o,eaa,eu co.n.erite4tiun is divided by the Chi /0 at the sample location to get a release rate. Calculation then proceeds as described at the beginning in Sec-tion 2.4.

I 2-6 Revision 1

I I

2.9 Dose Projection Based on FSAR Analysis This method can be used to project offsite dose when the accident (incident) can be correlated to an accident type which has been analyzed in the FSAR.

This method is very approximate and should be used only when parameters are I

not available to perform other methods or when an offsite estimate is needed very quickly.

Af ter the accident is identified, the FSAR-calculated offsite dose factors are multiplied by the site-related Chi /Qs (atmospheric dispersion factor) to obtain an offsite dose estimate.

2.10 Calculation of Accumulated and Projected Doses from Releases This method is used to determine accumulated dose and projected dose based on the results of previous methods.

Accumulated dose is simply obtained by multiplying the dose rate obtained from the methods in Sections 2.4 through 2.9 by the elapsed time between whatever (previous and present) onsite or offsite monitor readings (used for the calcula-tions) and suming this product for each subsequent period.

I The projected dose is obtained by multiplying the current dose rate by the projected duration of release and adding to it the accumulated dose.

I I

I I

I 2-7 Revision 1 l-

I Draft 11/29/84 3.0 AUTOMATED EMERGENCY OFFSITE DOSE CALCULATIONS Automated calculation of offsite doses during emergencies are described in I

this section under the following headings:

o General Model Description o

Lake Breeze Capability o

Meteorological Basis o

Input Meteorological Data I

The first two address the model itself. The second focuses on the site-specific capability to represent the lake breeze with input of realtime onsite data and use of climatologically-based adjustments. The third addresses the information used to develop the climatological basis. The fourth describes the available meteorological information.

These discussions describe the ability of Perry's dose assessment model to esti-mate offsite doses within the 10-mile EPZ and beyond.

3.1 General Model Description

Background

Dose assessments in the event of an emergency at the Perry Nuclear Power Plant will be accomplished utilizing a computerized system that receives data auto-matically from the meteorological tower and plant radiation monitors. Two I

plume dispersion models are available--one that utilizes the straightline Gaussian approach and a second that estimates plume trajectory utilizing a plume segTent approach. The plume segment model incorporates methodology for approximating dispersion conditions during lake breeze situations. Results are displayed as dose rates or dose projections for comparison with the EPA Protective Action Guides (PAGs). The software package that performs these functions was provided by I'ickard, Lowe and Garrick, Inc. (PLG).

It is referred to as MIDAS (for I

Meteorological Information and D_ose A_ssessment S_ystem).

I 3-1 Revision 1

I I

Hardware The software package is installed and operational on a VAX 11/780. A magnetic tape drive is provided for data archive functions and system utilities. A 300 line per minute printer is available for development and system management use. The computer runs in a multi-tasked environment under a virtual memory operating system which can accorrrnodate many users simultaneously via local or remote terminals. Tektronix 4113 19" color graphics terminals with hard copy devices are provided in the E0F and TSC facilities, and a Tektronix 4107 13" color terminal is provided in the Control Room.

I General Software Characteristics The software is entirely menu driven and is configured to supplement the emer-gency plan for rapid calculation of offsite doses. Site-specific parameters are stored in disc files under system manager control. Security is provided by log-on procedures under control of the system manager. Dispersion and dose I

calculations can be initiated by a single operator on the CRT keyboard.

The Gaussian model is used for rapid initial dose estimates and projec. ions while the plume segment model can be utilized for more refined estimates of plume location and dose history. Results are displayed on site maps extending to a 50-mile distance from the plant. Once the maps are loaded into the terminal from the VAX, they are stored in the terminal to enable more rapid display of results. Printed information is provided to supplement the graphics output.

The software provides for manual entry of both meteorological and radiological effluent data if monitor data are unavailable.

Files are available for operator entry of simulated data for use during drills.

This enables trainees to practice using predetermined scenarios in a mode that does not interfere with the online emergency mode that would be used during a real emergency.

All software is written in FORTRAN. Listings are maintained onsite for use in interpretation or problem solving.

I 3-2 Revision 1

I I

File Structure Meteorological and radiological data bases are stored on disc for use in making I

the required dose calculations. Files containing (" constant") information specific to the site-plant situation are changed by privileged edit routines provided in the package. A series of routines perform these calculations using both the " fixed" and " time-dependent" data and in many cases stores the results of the calculations in files used by other routines for system output. The user can schedule runs that automatically read and display results from these files without operator intervention.

I Meteorological and Effluent Data Maintenance Tasks A series of tasks is provided to inspect, maintain and archive the data bases created by the system. Examples follow:

o A task is provided to print the hourly or 15-minute meteorological I

parameter averages over any specified time period (within the bounds I

of the file).

o The " bad data" task can display the areas of bad data recovery for quick inspection.

o The " joint frequency" task categorizes and prints the meteorological data (in joint frequency form) by direction, speed group and stability class for use in Regulatory Guide 1.21 reports, o

Summaries of total release by isotope can be printed.

I o

The " trend plot" task can be used to plot meteorological or radiologi-cal effluent data which enables checking for problem areas in the data.

o Other tasks can be used to sumarize the delta-T and wind rose data.

I I

3-3 Revision 1

I I

Input Data Requirements The computer interrogates microprocessors periodically to determine 15-minute averages of meteorological and radiological effluent monitor signals. Wind speed and wind direction at the 10- and 60-meter level along with vertical temperature difference between the 10- and 60-meter levels is derived from redundant instruments on the meteorological tower. Digital signals are sent to the computer via redundant data links. Radiological data are received from the monitors in the plant vents, the condenser offgas system and the heater l

bay / turbine ventilation systems. Releases can be assigned to any of four release points from which a release rate (microcuries/sec) is computed. Since these effluent monitors do not provide an isotopic breakdown, the fraction of the total release for each isotope is determined from default isotopic mixes as a function of accident type. Manual entry of the isotopic breakdown is also provided.

Accident Dispersion and Dose Calculations Results of real-time atmospheric dispersion and dose calculations for accidents are available in printed and graphical form. MIDAS software is available for the modals referred to in NUREG-0654 (USNRC,19800), Appendix 2 as Class A and/cr enhanced Class A.

The following two sections describe these models.

The Class A Model The Class A model used for real time assessment of dispersion is the standard Gaussian model. The graphical isopleth output, representing a straight-line Gaussian-shaped plume, was designed to replace the plastic overlays (for maps of the same scale) currently found in the emergency kits in many control rooms.

A background map of the site is plotted along with the isopleths so that both appear on the same plot.

All accident calculations are under menu control for ease of use by the opera-tor. The map scale, release point (along with vent flows), level for data on the meteorological tower and terrain height can all be pre-selected or selected l

I 3-4 Revision 1

during the run by the user. Any previous hour (or the last 15-minute average) can be selected for the calculation.

Certain self-checks are provided to warn the user of problems. For example, if meteorological data are " bad" the user is notified and asked if data from some other source are available. If so they are entered by the operator.

Likewise, if dose results are selected and there are no effluent release values present (from real time effluent monitors) or the data are bad, the I

user will be prompted for input. Beta, gamma and/or thyroid-inhalation doses are computed after all input data have been entered. A calculation will not be completed without contemporaneous meteorological and effluent data.

Results are in printed or plotted form.

Several choices are available to the user for the source term.

If the accident classification is known, but the release is unknown, preset release scenarios can be used for up to ten accident categories. Otherwise, real time data from effluent monitors can be used.

Enhanced Class A Model NUREG-0654, Appendix 2 also refers to a more complex model for estimating dif-fusion and exposures out to greater distances. The model currently programmed and operational in the Perry Plant MIDAS package is a plume segment model based on a program developed by CLG called CRACIT (For Calculation of Reactor Acci-dent Consequences Including Trajectory) which is similar in concept to that of the CRAC program which was written for the Reactor Safety Study (WASH-1400, USNRC,1975). The " front end" source term and "run" menu options provided for the Class A model are also used to drive the enhanced version, thus the operator interface is essentially unchanged.

The Plume Segment Model The basic functions of the plume segment model are the calculation of meteoro-logical dispersion of the released radioactive material as it travels downwind and the estimation of the resulting doses from this material. The meteorologi-3-5 Revision 1

I I

cal dispersion is modeled assuming Gaussian diffusion and variable trajectory transport.

The transport portion of the dispersion model allows the plume travel direction to vary as the wind direction varies. The model divides the plume into segments called spatial intervals according to the travel distance for each quarter-hour. The standard Gaussian model is used to estimate plume dispersion based on the wir.d speed, wind direction and delta temperature measured on the weather tower. The plume, therefore, is represented by a series of segments, each of which has different characteristics based on the meteorology at the time the I

segments are in their respective locations.

The model simulates plume rise, building wake effects, dry deposition and wet deposition as a function of rain rate. The model is run using quarter-hourly wind averages.

Short-term Releases The dose calculation in the plume segment model provides information necessary for use in making immediate protective action decisions. Projected integrated organ doses for the whole body, thyroid, and lung are computed for each plume segment for a given short-term (usually 15-minute) release. Three pathways are used including plume shine, inhalation and ground shine. The whole body dose consists of the sum of plume shine dose due to plume passage overhead, inhalation dose due to inhaling airborne radioactive material and to ground shine dose from particulates deposited on the ground. The thyroid inhalation I

dose is reported separately for use in comparison with the PAGs, although the plume shine and ground shine components are available in printed tables. The average dose rate (rem / hour) to each of the three organs is also estimated from the three pathway components.

I The model can be run in a predictive mode using the most recent data from the tower. After the first hour, persistence is used for all dose projections.

g W

The lake breeze processing, if in effect, will cause changes to the otherwise straight-line plume trajectory.

I 3-6 Revision 1 I

I I

The results of the plume trajectory and dispersion modeling and the calculated doses can be plotted on the graphics CRT overlaid on the site map. The widening of the plume as it moves away from the site is a function of the atmospheric stability. The changing plume direction is controlled by the changing wind directions based on information fron the meteorological data files. Characteris-tics of each spatial interval can also be printed in tabular form.

Long-term Releases The same calculation routine is used for making longer-term dose estimates (for more than one 15-minute release).

In this program a release is simulated as several short-term releases. Each of the releases is treated as a separate plume moving away from the site according to the meteorological conditions for that time. Therefore, each successive release is controlled by a different weather sequence. The dose over the area is accumulated separately on a fine mesh grid for each release. The total dose over all releases for each of up to four requested projection time periods can be displayed graphically as iso-pleths on the CRT.

Lake Braeze Processing in the Plume Segment Model I

During warmer seasons, wind patterns caused by " lake breeze" effects can occur at the Perry site. The lake breeze phenomenon is well-known along the shores of the Great Lakes and has been the subject of extensive field studies for many years. Several lake breeze characteristics are of note which affect atmo-spheric dispersion and plume trajectory (Figure 1). During onshore flow, a parabolic shaped boundary or lid (referred to as a TIBL) can be formed starting near the shoreline and increasing in height with distance inland. This lid I

can result in plume trapping or fumigation with associated high ground level concentration. The flow from the lake penetrates inland to a certain distance where turning and also uplifting occurs. These phenomena are the result of air density effects caused by temperature differences between the warm land and colder water. A reverse condition can occur at night causing a " land breeze" that flows out over the lake.

I 3-7 Revision 1

I I

Since releases are from the plant would be essentially at ground level below I

the TIBL boundary only, modelling of the fumigation conditions is not appropriate for the Perry site.

Plume trajectory changes may occur when the plume reaches the " convergence zone" or the inland location where the lake breeze stops. Thermal convection can actually form a " cell" which results in a return flow aloft back toward, and generally moving (spiraling) parallel to, the shoreline. This 3-dimensioral I

phenomenon, after reaching the convergence zone, is extremely difficult to characterize. However, plume dispersion under these circulating conditions is I

very good. To account for these phenomena, methods for estimating plume disper-sion and location are applied in the MIDAS model. They are based on algorithms which use conditions measured on the meteorological tower at the site as well as lake water temperatures and time of day as described below.

The lake breeze submodel is incorporated in the MIDAS plume segment model using a series of preprocessors in the software which provide three basic functions as follows:

o Determine whether meteorological conditions mect the criteria estab-lished for the existence of a lake breeze.

o Estimate future meteorological input parameters using existing meteoro-logical conditions.

o Estimate the inland distance of lake breeze penetration and estimate changes in wind flow patterns as a function of time.

These functions are accomplished by a series of logic checks on the available data to categorize current weather conditions.

These checks require use of the previous few hours of data which are stored in the computer to determine I

when lake breeze conditions started. Since the plume segment model is time-dependent (i.e., it steps into the future) and is used to project doses, lake breeze input to the model must be prepared for the future as well. For example, it would not be correct to assume lake breeze conditions exist after sundown.

I I

3-8 Revision 1 m

I I

Projections beyond that time would, by necessity, be made using the tower data I

without lake breeze preprocessing or dispersion equations.

The following data are used in the preprocessors:

o Date(season) o Time of day o

Lake temperature I

o Air temperature o

Wind speed o

Atmospheric stability o

Wind direction Remote User Emergency Reports (the Broadcast Function)

Software is provided which automatically sends reports to predefined remote terminals. The " Broadcast" reports include meteorological data in the format specified originally in NUREG-0654. The remote terminal operator does not have to constantly monitor the termiaal and schedule tasks to receive 15-minute updates.

3.2 Lake Breeze Model Capability The model accounts for the lake breeze circulation. The lake breeze is the I

most significant site specific effect of the terrain and lake-shore location of the Perry Nuclear Plant. The lake breeze occurs a quarter to a third of the days during the lake breeze season of the year when the daytime air temper-atures often rise above the Lake Erie surface temperature.

Because the Perry meteorological tower is located sufficient 1/ inland, it is representative of the overland conditions even during the onshore flow of the lake breeze with one exception. The model accounts for this exception:

during the lake breeze a release from the plant would be transported only as frr as the lake breeze front before being moved upward into the halical circulation.

Because the model has this site-specific capaollity to account for the lake I

3-9 Revision 1

I I

breeze, the Perry dose assessment system provides representative estimates of dispersion for airborne releases from the Perry plant.

In the following subsections, the various characteristics of the lake breeze are described as to how they are accounted for in the model.

3.2.1 Detection of the Lake Breeze I

The lake breeze must first be detected before it can be represented in the dispersion parts of the model. The detection algorithm used for Perry is sum-marized in Table 1.

The method for detecting the presence of the lake breeze was based upon the local and regional information provided in Section 3.3, Meteorological Basis for Model.

The meteorological variables required for evaluation of the criteria are observed on the Perry Meteorological Tower (discussed in Subsection 3.3.4, Input Meteoro-logical Data). Observation for these are continually available (updated every 15 minutes) for routine assessment as to the presence of the lake breeze. A climatological data base of Lake Eric is needed for surface water temperature.

Table 2 indicates the weekly values that are used based on a study by Webb (1974).

In addition, an operator can manually enter an observed lake surface temperature.

3.2.2 TIBL Trapping In the presence of the lake breeze, material released into the air from a ground-level source may be slightly restricted in its dispersion.

(See Figure 1.)

If the air over land is sufficiently turbulent, the airborne material may spread to the top of the TIBL (actually by the stable air above the TIBL restricts material from moving higher).

If some of the material reaches the top of the TIBL, it becomes trapped.

In time, this may result in ground-level concentra-I tions that are slightly higher than on a day when vertical dispersion is not restricted by the TIBL.

3-10 Revision 1

I I

However, because the possible effect of the trapping is relatively small in comparison to changes in concentration due to the direction of transport (helical circulation), no direct account is made of trapping.

3.2.3 Lake Breeze Helical Circulation The lake breeze circulation as described in Section 3.3 is characterized by a helical circulation. A parcel near the ground moves with the inflow to the lake breeze front, moves up in the front, enters the return flow back past the coast, until it descends and starts an over-water reversal back into the inflow.

I At this point a parcel has usually moved up or down the coast such that its continuing path would represent a helix.

Inland Penetration I

The distance that the sea breeze (front) penetrates inland is represented in Table 3.

The table shows the maximum inland penetration as a function of the start time of the lake breeze.

The table provides values for both the " classic" and the " parallel confluence" types of lake breeze. The " classic" type occurs when the inflow is quite direct to the shoreline (directions 280 through north to 20 ).

The " parallel con-fluence" type occurs when the inflow is less direct to the shoreline (directions 0

U 250 through 280 and 40 through 70 ) and if before the lake breeze the wind 0

speed was greater than 4.5 mph for directions of 220 through 250 or 70 through 100.

I Return Flow The speed and direction of the wind in the return flow of the lake breeze is determined by the algorithm described in Table 4.

The component of the wind parallel to the shore in the inflow is maintained.

The perpendicular component of the inflow is reversed and cut in half. These two new components are combined to yield the return flow speed and direction.

3-11 Revision 1

I I

This return flow is used to transport airborne material from the frontal zone (inland penetration) of the lake breeze to the point of over-water reversal.

The lake breeze model uses a Pasquill stability class of E to determine plume dispersion in the return flow. Stability E dispersion is used from the time the plume turns back toward the lake and out over the lake until the plume turns back toward the land. At this point, the stability goes back to the classification used at the start of the lake breeze.

The use of Class E stability was tested by Dr. Lyons (1978) during the late I

1970s. His field tests using airplanes to trace the path of plumes in a lake breeze showed the return flow layer to be stable.

Over-Water Reversal I

The offshore distance at which there is an over-water reversal of the lake breeze is selected. At this distance, airborne material in the return flow subsides and re-enters the inflow. The algorithm used to select the distance is that the offshore distance is one half of the inland penetration distance.

3.3 Meteorological Basis The site-specific lake breeze characteristics of the model for real time dose assessment were just described in the previous section. The meteorological basis for the model is provided under two headings in this section:

o Results of Lake Breeze Case Study o

Lake Breeze Climatology for Perry 3.3.1 Results of Lake Breeze Case Study I

This section describes a study of the lake breeze that was conducted for the 1980 season of the Lake Breeze. This was a local study that included the use of observations from the Perry Meteorological Tower. The results of this study were used along with the lake breeze climatology as the basis for the site-specific capabilities of the model that are described in Section 3.3.2.

3-12 Revision 1

I Draft 11/29/84 The site-specific factors developed for the Perry lake breeze model were deter-mined from a climatography study performed in 1984 by Dr. Walter Lyons. The study wns of the site area using all available local data and available liter-c, ature concerning lake breeze from the Great Lakes region. The climatography involved attaining hourly meteorological data from all sites in the Perry area and an in-depth analysis of the data for all lake breeze days in 1980. The following sites were used in the study:

I Surface Data Perry Plant Eastlake Plant Avon Lake Plant Cuyahoga County Airport Cleveland Hopkins Airport Ashtabula Plant Burke Airport Youngstown Airport Upper Air Data Buff alo Airport Pittsburgh Airport Refer to Figures 2 and 3 for geographical locations.

Each day from April 1 through July 31 was characterized as to the mesosynopite climatology. Table 5 shows the final classification of the 4 months of data, i

Next the lake breeze days were examined to test criteria critical to the forma-tion and termination of a lake breeze. Tables 6 through 10 show the results of tests performed on categories wind speed and direction, stability, ambient temperature and onset time versus penetration. As a result of this climato-logical study and our literature search, the following site-specific criteria were developed for determining existence of conditions for a lake breeze to exist at the Perry site.

\\

o Calendar date: between 3/1 and 10/15 o

Time of day: one hour after sunrise to sunset 3-13 Revision 1

o Temperature comparison:

ambient temperature (10m) minus lake water temperature is greater than or equal to -2.0 F o

Wind speed: wind speed (10m) is between calm and 13.4 mph o

Stability:

using delta temperature (60-10m) is a Pasquill Category A, B, C, or D U

0 o

Wind direction: wind direction (10m) is from between 250 and 68.

All criteria must be met for two 15-minute periods before the model will initiate use of the lake breeze algorithm. Once the lake breeze has started, it will persist for at least four 15-minute periods. The criteria are checked every 15 minutes and if they are not all satisfied during four consecutive 15-minute periods, the lake breeze is terminated. The lake breeze is automatically termin-ated at sundown.

If a lake breeze is terminated during the day, it may be started again using the same criteria.

I The extent of inland penetration was difficult to accurately assess given the available data. However, it was determined that the penetration distance could be related to the start time of the lake breeze and the angle of the wind direc-tion in relationship to the shoreline.

3.3.2 Lake Breeze Climatoloay for Perry This section describes the lake breeze circulation as it applies to Perry.

The description is a climatology for the Great Lakes Region and focused on Lake Erie. The climatology was used along with the lake breeze case study as the basis for the site-specific capabilities of the model that are described in Section 3.2.

A stylized drawing of the lake breeze is presented in Figure 1.

As reported in Section 2.3.2.3.2 of the PNPP FSAR (p. 2.3-24,-25):

The major local effect on site meteorology is the presence of Lake Erie and the resultant occurrences of lake and land breeze circulations. The fact that water has a higher thermal capacity I

3-14 Revision 1

I I

I than the land mass, and therefore responds more slowly to changes in radiation intensity, implies that temperature / density gradients between the water and land will occur with diurnal and seasonal periods. Turbulent mixing within the lake, effecting a downward I

transport of surface heat through large masses of water, also contributes to the land-lake temperature variation. Lake breezes (surface wind blowing from lake to land) form when the water I

temperatures are colder than the land temperatures, i.e., during spring and summer on a seasonal scale and late morning to late afternoon on a diurnal scale. The air over the land will be more buoyant than the lake air, and as it rises, a horizontal density gradient will form causing the colder air over the water to flow underneath the warmer air. Land breezes are the converse of lake breezes and occur when the water is warmer than the land, such as during the fall and winter or during the night in the summer. The lake breeze is generally stronger and occurs more frequently than the land breeze due to the fact that the I

buoyancy of the warmer air is the driving mechanism, and this is accomplished more effectively by heating the land mass rela-tive to the water as in summer, than vice versa. This phenomena becomes most pronounced when synoptic scale motions are weak, such as when a large high pressure system is centered in the region. When synoptic scale motions are strong due to larger horizontal pressure gradients, the land / lake breeze circulation is effectively masked.

During onshore wind flows, such as a lake breeze, cool air flowing off the lake is modified by thermal (surface heating) and by surface roughness effects as the air flows over the land. The air from the lake is modified significantly as it flows over the land especially during the spring and early summer. The I

air is heated from below resulting in an unstable vertical temper-ature gradient and hence enhanced diffusion conditions. Surface roughness effects over the land increase atmospheric turbulence (also resulting in enhanced diffusion conditions), although low I

level wind speeds will decrease. The thermal and roughness effects occur at the shoreline and form a " boundary layer" which increases in depth with distance inland. Within this boundary I

layer the air is unstable with more stable air (suppressed dif-fusion) above the boundary layer.

In continuing discussion, the following characteristics of the lake breeze are I

discussed:

Frequencies and inland penetration Inflows and outflows Helical recirculation Wind speed Penetration speed Time dependency 3-15 Revision 1

I I

Air-water temperature differences TIBL 3.3.2.1 Frequencies and Inland Penetrations It is estimated that about a quarter to a third of the spring and summer have an associated lake breeze circulation. The Lake Erie frequency is estimated to fall within the range of occurrences on Lake Michigan and Lake Ontario.

In the Great Lakes region to the west, Lyons and Olsson (1973) reported that approximately 35 percent of the days during May and August were associated with a true lake breeze on the Lake Michigan-Chicago shoreline.

In 1974, there were two separate studies on Lake Ontario, to the east of Lake Erie. On the north shore, Bennet and List (1975) reported 21 lake breeze events for April to July. On the south shore of Lake Ontario, Mitchell (1975c) reported 25 lake breeze events for the same period. Based on an average of 23 events I

over the 122-day study period for the two studies, 19 percent of the days were associated with a lake breeze. This frequency is about half of that for Lake Michigan for the whole season.

In another study for Lake Ontario, Guski and Miller (1980) found a 20 and 30 percent frequency of occurrence for the same April to July period in 2 differ-ent years, respectively. However, for the entire season (March to September, the average lake breeze frequency was 32 percent of the days.

The maximum inland penetration distance of Lake Erie lake breezes is estimated to be about 50 miles. This is based on Guski and Miller's (1980) findings on Lake Ontario. They found, for example, that penetrations as far inland as 28 miles occur with 43 percent of the lake breezes. They concluded that their results were consistent with the distance of inland penetration for Lake Mich-igan (Lyons and Olsson, 1973; Moroz, 1967).

I I

I 3-16 Revision 1

I I

3.3.2.2 Inflows and Outflows The depth of the inflows is greatly varied but could be typified as approxi-mately 500 m for Lako Erie. On Lake Ontario, Mitchell (1975a) reported depths of 240 to 500 m, and Moroz and Koczkur 1967) reported depths of 500 to 750 m.

Guski and Miller (1980), in their 2-year study, observed depths of 100 to 3000 m.

On Lake Michigan, depths of 500 to 1000 m were reported by Lyons and Olsson (1973); 400 m, by Olsson (1969); and 400 to 750 m, by Moroz (1967). Lyons (1975) identified depths of 100 to 1000 m with 500 m as typical.

I The classical outflow--or return flow--is probably best described as " typically about twice the depth and half the peak speeds found in the inflow" (Lyons and Olsson,1973). The reports for Lake Ontario and Lake Michigan lake breezes (Guski and Miller, 1980; Lyons and Olsson, 1973; Mitchell, 1975a; Moroz, 1967; Moroz and Koczkur, 1967; and Olsson, 1969) point towards this. They suggest an outflow height of up to 500 to 2000 m, perhaps typically 1500 m.

However, I

these reports also indicate that the return flow is frequently indistinguishable or not observed because of the influence of large-scale winds.

3.3.2.3 Helical Recirculation I

The typical two-dimensional picture (Figure 1) of the lake breeze circulation suggests the probable return of shoreline releases to the shoreline. A ground-I level release is transported inland to the frontal or convergence zone, it moves upward, moves toward the lake in the return flow, subsides once more to the inflow, and thus turns to the shoreline.

In reality, the flow is three-dimensional and the synoptic, large-scale winds influence the actual trajectory of a release. The recirculation usually results I

in an along-shore drift. Trajectories resemble flattened helices (Lyons,1975).

Only rarely might a release return to its point of origin near the shoreline.

I I

3-17 Revision 1

I I

3.3.2.4 Wind Speed I

The wind speed in the inflow region is typically about 7 mph at the 10-m level (Mitchell, 1975c; Guski and Miller, 1980). This data suggests a typical range of hourly wind speeds of 3 to 12 mph. These values are consistent with a typical description by Lyons (1975). Thus an onshore speed of 16 mph or more is likely not associated with a lake breeze and characteristic recirculation.

3.3.2.5 Penetration Speed The penetration speed of the lake breeze front can be determined from previous studies. As reported in the next subsection on time dependency, the average duration of the lake breeze is about 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br />, with about a 3 hour3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> standard deviation.

In Guski and Miller's (1980) study the distance of inland penetra-tion 50 percent of the time was about 38 km. Relating the average duration to the distance of inland penetration yields an average frontal speed of 1.5 m/s.

Another study of the south shore (Mitchell, 1975c) of Lake Ontario showed a similar duration (7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br />). Field measurements (Mitchell, 1975a) for two cases within 1.5 km of the shoreline showed penetration speeds of 0.6 and 1.3 m/s can be used for Perry. The value could easily vary by + 50 percent.

I 3.3.2.6 Time Dependency I

Lyons (1975) reported a typical onset time for lake breezes as 0800 to 1000 LST (local standard time). This was supported by the Lake Ontario study by Mitchell (1975c) for which monthly average onset times of 0930, 0830, 1030, and 1000 LST were reported. However, as Lyons (1975) described, onsets as late as sunset have been reported. Usually, the lake breeze is best developed about 1600 LST (Lyons, 1975).

I The duration of lake breezes is about 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br />, as reported from Lake Ontario studies. Mitchell (1975c) reported monthly average durations of 5 to 9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br /> and monthly maximum durations of 9 to 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />. Guski and Miller's (1980) study implies durations of 1 to 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br />.

I I

3-18 Revision 1

I I

3.3.2.7 Air-Water Temperature Differences I

The lake breeze circulation is related to the differential temperature of air over land and water; the land temperature is normally warmer. However, it may happen for a time that at a particular point over land the temperature is about the same as that of the water.

In this instance the lake breeze circulation may well be continuing; it is just that a cloud has shadowed the particular point and the upwind over land fetch and unmodified (unwarmed) marine air has I.

reached that particular location.

A typical air-minus-water differential temperature as about 110F as reported in the two Lake Ontario studies. Mitchell (1975c) reported differentials in categories ranging from less than 20F up to 270F, with a mean of approximately 110F. Guski and Miller (1980) reported an average of 110F with a standard deviation of 90F.

3.3.2.8 TIBL The TIBL (T;rbulent, or Thermal, Internal Boundary Layer) forms near the ground as unmodified lake air moves over the warmer, rougher land (downwind) in the inflow of the lake breeze.

(The same happens during a stable onshore flow, although no return circulation occurs.) Although the turbulent characteristics of the air flowing over the land start to be modified immediately, some distance of travel is required in order for the complete modification to occur. See I

Figure 1.

Over the lake, in a lake breeze, the air tends to be stable before it moves inland. This is caused by subsidence and by conduction cooling by the water.

As air moves onshore in the inflow, a shallow TIBL forms as a result of the warmer and rougher land. As the inflow continues, more of it is modified by the increasing depth of the TIBL. The depth of the TIBL grows faster near the I

shoreline than it does later in the downwind (over land) fetch. The growth rate has been approximated as being proportional to the square root of travel distance or of travel time. Raynor, et al. (1974) provided a more complete formulation for the TIBL depth; it is referred to as Equation (1) in NUREG/CR-0936 (Raynor, et al.,1979).

3-19 Revision 1

I I

Septoff, et al. (1976) reported a calculational estimate of the TIBL depth at I

1.5 km inland of 109 m: several methods from the literature were used to make the composite estimate. The Raynor method reported in NUREG/CR-0936 yielded TIBL heights at 1.5 km of 134 to 300 m for a variety of conditions associated with an 11oF (60C) air-minus-water temperature difference (which is typical for Lake Ontario during a lake breeze situation). Mitchell (1975b) demonstrated the applicability of Raynor's formula slightly modified for Lake Ont0rio lake breezes; it provides an estimated 100 m depth for the TIBL at 2 km during a I

typical situation of B stability (unstable) and a 120F air-minus-water temper-ature difference.

Observed TIBL depths at 2 km inland include the Lake Ontario reports by Mitchell (1975a) of 100 to 150 m; the Lake Michigan repcrt by Lyons (1975) of approxi-mately 150 m; and the Lake Erie report at FERMI (NUREG/CR-0936) of 61 to 183 m.

These values averaged together indicate a typical TIBL depth of approximately 130 m at 2 km inland.

I Figure 4 illustrates both these observed and estimated TIBL depths as a func-tion of inland distance (downwind fetch from shoreline). Characteristically, the Perry tower will be completely in the TIBL because the overland fetch is 1.8 km--or more, if the wind comes onshore at an angle rather than perpendicular.

I 3.4 INPUT METEOROLOGICAL DATA I

Data that can be routinely used for the preparation of dose assessments are described in this subsection. This includes a discussion of the following:

Meteorological Tower and Validation Determination of Stability Class Other (Supplemental) Meteorological Data I

3.4.1 Meteorological Tower and Validation The tower is an open lattice structure with sensors at the 10 and 60-m levels.

It is 60 m tall and is located in terrain similar to that of the site region that is of relatively low relief. Even during stable onshore flow, the tower I

3-20 Revision 1

I l

I l

is representative of the over-land conditions because it is well within the I

TIBL (Thermal Internal Boundary Layer). The shoreline is oriented approximately NE-SW as illustrated in Figure 2.

[

Table 11 describes the instrumentation with manufacturers, model number, report-ing range, location of the tower, and performance characteristics.

l1 Table 12 summarizes the meteorological variables measured. There are two inde-pendent systems on the tower, Main and Backup. Each systems has its own pro-cessors, recording equipment, and microcomputer (MDPS-Meteorological Data Pro-j I

cessing System) that are housed in separate shelter near the tower. Every 15 minutes, each MDPS transfers validated data and preliminary estiaates of dis-persion to onsite--Control Room, TSC ERIS computer, and the Plant Process computer--by various multiple communication links.

I For each parameter, the MDPS normally develops hourly values that are derived from 15-minute values. The 15-minute values are developed from sub-second I

sampling. The MDPS automatically performs electronic and status checks. Daily, it makes calibration zero and span checks for each data channel, except for the dewpoint reading.

It uses the information to refine subsequent ob;ervations for any normal electronic drif t that might be detected. Nominal, small, and large drifts are reported. The MDPS continually monitors for reduced air flow in temperature aspirators, for the dewpoint system in auto-balance, for manually initiated bypass codes (used during weekly checks or calibrations), etc.

In addition, samples are taken in 5-second groups (primary and validity) and then screened at the end of 15 minutes before being accepted for use in process-ing calculations.

If the sample is out of acceptable limits, it is rejected.

At the end of each 15 minutes, all of the screened values are used to determine the 15-minute value (if there were an insufficient number of the potential

samples, i.e., less than 80 percent, then the 15-minute value would be reported as missing).

Before the 15-minute value is finally accepted by MDPS to store, print, or transmit, a realtime validation is performed. This is done by performing meteoro-logical reasonability checks on the 15-minute value and concurrently generated 3-21 Revision 1

I I

statistics. For parameters of prime importance, a dual statistics approach is I

used that employs signals from colocated, redundant sensors (Table 12). The dual statistics approach (Mitchell, et al.,1984) is based upon NOAA compar-ability techniques (Hoehne; 1971,1977). For other parameters, a single statistic approach is used. The single statistic approach involves an evaluation of various single statistics for their relation to known ranges of statistics, to climatological extremes, and (in limited cases) to other parameters.

If the validation is not positive, the data is reported as missing along with a validity I

indicator.

In the event that a particular variable is missing, the ERIS computer can seek out a substitute within the reported tower data from each of the two system, Main and Backup. See Table 13.

I The meteorological data collection program at PNPP is subject to detailed quality assurance and quality control procedures which are supplemented with site-spacific plans and procedures. Data are reviewed regularly. A Site Observer I

performs a weekly inspection to verify proper system operation, routine opera-tions (change charts) and minor preventive maintenance. Calibrations and routine preventive maintenance are conducted at four-month intervals by trained personr.el according to set procedures. Repairs requiring emergency maintenance between regularly scheduled calibrations are performed promptly.

6 3.4.2 Determination of Stability Class I

Atmospheric stability class is determined for use in estimating dispersion parameters of airborne material.

In the split sigma Gaussian modeling concepts, there is an horizontal dispersion parameter for the y direction and a vertical dispersion parameter for the z direction. There is a different growth rate with distance for each that is associated with each of seven atmospheric stability classes.

I Four methods of classifying atmospheric stability are used in the microcomputer at the Perry Meteorological Tower:

o delta T (for vertical, z, dispersion) o sigma theta (for horizontal, y, dispersion l

3-22 Revision 1 L_

I E

o modified sigma theta (for vertical, z, dispersion) o pseudo sigma (for horizontal, y, dispersion)

As a result there are a variety of sources of stability class information (see Tables 12 and 13).

E o

Delta T (Table 14)

I The method reported in Regulatory Guide 1.23 (USNRC 1972, 1980a) uses the vertical temperature (delta T) gradient as the basis. This method is normally used for dispersion estimate submittals to the NRC.

In ERs, FSARs, Appendix I assess-ments, and semi-annual operating reports, usually, delta T is used to determine both the horizontal and vertical dispersion factors. However, it is generally a better description of vertical rather than horizontal dispersion. The latter may be significantly altered by meander, especially during low wind speeds and I

stable conditions, as is recognized in the methodology of Regulatory Guide 1.145 (USNRC, 1979). The delta T classification ranges are presented in Table 14.

o Sigma Theta (Table 15)

This method reported in Regulatory Guide 1.23 (USNRC, 1980a) uses as its basis the horizontal wind direction fluctuation, sigma theta. This method has been used historically to estimate both vertical and horizontal dispersion. However, I

at night it is generally a better description of horizontal rather than vertical dispersion. During low wind speeds and stable conditions (common at night),

horizontal meander may be large and result in large time-averaged horizontal dispersion, while vertical dispersion is small. The sigma theta classification is reported in Table 15.

Modified Sigma Theta (Table 16) o This method has been reported in the literature and considered for monitoring by EPA: Mitchell and Timbre, 1979; Irwin, 1980; Mitchell and Snell, 1981; and Mitchell, 1982.

I l

B 3-23 Revision 1

I I

The method uses as its basis the horizontal wind direction fluctuation, sigma I

theta, during the day; however, it also uses wind speed at night. The method takes into account the increased horizontal meander associated with low wind speeds and stable conditions at night. As such, this method yields the same stability classes as does the sigma theta method during the day and thus is a good description of horizontal and vertical dispersion. At night, however, this method better describes vertical dispersion than the standard sigma theta method since this method yields classifications to mimic those yielded by the I

delta T method. This classification is reported in Table 3.3-6.

o Pseudo Sigma (Table 17)

This method utilizes delta T measurement to approximate the stability class estimated by sigma theta. As indicated in Table 3.3-5, this method provides an adjustment to incorporate the effect of horizontal meander of the wind direc-tion. This adjustment is consistent with Regulatory Guide 1.145 (USNRC, 1979) which prescribes an enhancement of horizontal plume growth during stable, light-I wind conditions. And, it is this enhancement described in Regulatory Guide 1.145 that is also the basis for the Modified Sigma Theta method. The approach of the method presented in Table 17 is to make the adjustment to the stability class controlling horizontal, y, dispersion.

I 3.4.3 Other (Supplemental) Local Meteorological Data I

While the Perry dose assessment system does characterize lake breeze dispersion through the use of input from the meteorological tower (as described in 3.2),

there are local (supplemental) sources of meteorological data. These sources could be used to verify current conditions (extent of inland penetration of the lake breeze).

Regional sources of meteorological observations are presented in Figure 2.

I These sources are 17 or more miles from Perry and are generally operated by agencies, like the National Weather Service, for example.

In addition, there are a variety of generally non-governmental stations that are close to the site (Figure 3). These stations can be contacted for supple-3-24 Revision 1

I I

mental information, especially during a lake breeze event. Although all the I

stations are not open 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> a day, the lake breeze itself is limited to the daytime. Therefore, it is reasonable to expect that several of these could provide wind information to confirm the extent of penetration of the lake breeze.

These local stations are listed in Table 18 that includes a description of their inland distance as well as their relationship to the plant.

I I

I I

I I

I

' I I

E I

I I

I 3-25 Revision 1

I I

Table 1.

Algorithm for Detection of the Lake Breeze I

I Onset 1.

Calendar Date:

Between 3/1 and 10/15 2.

Time of Day:

One hour after sunrise to sunset 3.

Temperature Comparison:

Ambient temperature 10m - Lake I

watertemperaturgisgreaterthan or equal to -2.0 F 4.

Wind Speed:

Wind speed (10m) is between calm I

and 13.4 mph 5.

Stability:

Stability using delta temperature I

60-10m is a Pasquill category A-D and the stability has shifted at least one category towards I

unstable (i.e., from E to D) after sunrise 6.

Wind Direction:

Thewinddiecti"6;05)iSfrm 5

I between 250 and 70 and the winddirectionhasshiftedfrog angver-landsectorbetween80-I 240 to an over-water sector 0

(250-70 ).

The shift in direc-tion must occur between one hour before sunrise and thirty minutes I

before the onset of the lake breeze I

7.

Rainfall:

No rainfall during the 15-minute period All criteria must be met for two 15-minute periods before the model will I

initiate use of the lake breeze algorithm. The shift in wind direction from over-land to over-water need only have taken place before the onset of the lake breeze as described above.

Cessation Once the lake breeze has started, it will persist for at least four 15-minute I

periods. Each criterion will be checked every 15 minutes.

If they are not all met for four consecutive 15-minute periods the lake breeze is terminated.

The lake breeze is automatically ended at sundown.

Restart When a lake breeze is ended during the day, it can be started again by meeting I

the two 15-minute period checks, above, but without any wind direction shift.

3-26 Revision 1

I I

Table 2.

Surface Temperature for Lake Erie I

Values are interpolated from the six-year study by Webb (1974).

Date Temperature (O )

F March 1-7 33 8-14 33 15-21 34 I

22-28 34 29-4 35 April 5-11 36 12-18 38 19-25 40 26-2 42 May 3-9 43 I

10-16 45 17-23 48 24-30 51 I

31-6 53 June 7-13 55 14-20 59 21-27 62 I

28-5 65 July 6-12 67 13-19 69 I

14-20 71 l

21-27 72 l

28-3 73 l

August 4-10 74 11-17 74 18-24 73 25-31 72 I

September 1-7 71 8-14 69 i

I 15-21 67 22-28 65 I

29-5 63 October 6-15 58 l

l I I

lg l

3-27 Revision 1 l

I I

Table 3.

Algorithm for the Maximum Inland Penetration Distance I

of the Lake Breeze Distance (miles *) by Lake Breeze Type I

Start Time Parallel (EST)

Classic Confluence 0600 20 5

0700 20 5

0800 20 5

0900 16 4

1000 16 4

1100 12 4

I 1200 12 3

1300 10 3

1400 10 3

1500 8

2 1600 8

2 1700 5

1 1800 3

1 1900 0

0 i E Inland penetration rate for the classic type is 2.66 mph and W

for the parallel confluence type is 0.75 mph.

l lE I

lI l

l3 E

l 3-28 Revision 1 1

I TABLE 4 Algorithm for the Return Flow Wind of the Lake Breeze I

Calculation of Return Wind Speed and Direction N

I i

l l

1 15 Angle of e

pot *go I

\\ min Wind Shoreline Direction

's M

g1

[ } O 's Sp I

g is y iG I

g/

9oT*pt" S

3 g/

i t

a i gg BSPD = Return Speed ggg ph N

BDIR = Return Direction i

s Y

BDIR Picture l

DIR = Initial Direction SPD = Initial Speed l

e

= Angle of Shoreline - DIR Speed Calculation X

= SPD i Cose I

3 Y

= (SRD x SIUo)/2.0 s

2 BSPD = X

+Y s

3 X s y

Cos!

I

=

BSPD Direction Calculation Y

= Arc Cos (Xs/BSPD)

BDIR =, Angle of shoreline - Y l

!I 3/29 Revision 1

I I

Table 5.

Mesosynoptic Characterization for April through I

July 1980 at Perry I

(Days with Events Indicated)

Code Characterization N*

0 Steady Synoptic Influences Predominate All Day 12 6.0 1

Synoptic Discontinuity, _ 50% Change /2 Hours 10 5.0 2

Gradient Onshore Flow (Cloudy / Night) with Plume 10 5.0 Trapping 3

Gradien'. Onshore Flow (Sunny) with Fumigation 31 15.5 4

Gradient Offshore Flow - Stable Air Over Colder 8

4.0 Water 5

Near Calm (Pooling) 1 0.5 6

Land Breeze 21 10.5 7

Classic Lake Breeze - Onset Past Hour Not Used 8

Classic Lake Breeze in Progress 39 19.5 9

Parallel Shore Confluence - From West 8

4.0 10 Parallel Shore Confluence - From East 0

0.0 11 Ridge / Trough Passage Lake Creeze 5

2.5 I

12 Gradient Onshore Flow - Lake Warmer (Night)

Not Used 13 Gradient Onshore Flow - Lake Warmer (Day)

Not Used 14 Thunderstorm Mesosystem in Area 40 20.0 15 Poorly Defined Mesosynoptic Regimes / Inertial 15 7.5 Flows 200 100 I

• N - Number of events (more than one event can occur in a day).

I I

I I

I 3-30 Revision 1

I Table 6.

Wind Speed at 10-m Level nf the Perry Meteorological Tower during Lake Breezes in April through July 1980 I

Wind (m/sec)

Observations Number of 0.0 - 1.0 6

2.4 1.1 - 2.0 28 11.4 2.1 - 3.0 63 25.7 3.1 - 4.0 75 30.6 4.1 - 5.0 47 19.2 I

5.1 - 6.0 23 9.4 6.1 - 7.0 4

1.6 7.1 - 8.0 0

0 8.1 - 9.0 0

0 9.1 -10.0 0

0 10.0 0

0 246 100 I

I

I

.I I

I I

I l

lI 3-31 Revision 1

m m

W W

m m

M M

M M

M M

M m

TABLE 7 HISTOGRAM OF OBSERVED WIND DIRECTION -- FROM 10H PERRY TOWER Time 06 GMT 12 GMT 18 GMT 00 GMT Directio; 1 XXXXXXXXXXX (11)

XXXXXXXXXX (10)

XXXXX

( 5)

XXX

( 3)

(18From) 19 XXXXXXXXXXXXXXXXXX (18)

XXXXXXXXXXXXXXX (15)

XXXXX

( 5)

XXXXXXXXXXXX (12) 20 XXXXXXX

( 7)

XXXXXXXXXX (10)

XXXXXX

( 6)

XX

( 2) 21 XXXXXXXXX

( 9)

XXXXXXXXXXXXXXXX (16)

XXXXXX

( 6)

XXXXXX

( 6) 22 XXXXXXX

( 7)

XXXXXX

( 6)

XXX

( 3)

X

( 1) 23 XXX

( 3)

XXXX

( 4)

XXXX

( 4)

XXX

( 3) 24 XXXXX

( 5)

XX

( 2)

XX

( 2)

XXXXXXX

( 7) 25 XXXXXX

( 6)

X

( 1)

XXXXXX

( 6)

XXXXXX

( 6) 26 XXX

( 3)

XXXX

( 4)

XXXXXXXXX

( 9)

XXXXXXXXXXXX (12) 27 XXXX

( 4)

XX

( 2)

XXXXXXXXXXXXXX (14)

XXXXXXXXXXXXXXXX (16) 28 XX

( 2)

XXXX

( 4)

XXXXXXXXXXXXX (13)

XXXXXX

( 6) 29 XXXX

( 4)

XXX

( 3)

XXXXXXXXXXXX (12)

XXXXXXX

( 7) 30

( 0)

XXXX

( 4)

XXXXXXXXXXXXXXXXXXX (19)

XXXXXXXX

( 8)

Y 31 XX

( 2)

XX

( 2)

XXXXXXXXXXX (11)

XXX

( 3)

N$

32 X

( 1)

XXX

( 3)

XXXXXXX

( 7)

XXXX

( 4) 33 XX

( 2)

XXX

( 3)

XXXX

( 4)

XXXX

( 4) 34

( 0)

( 0)

XXXXXXX

( 7)

XXXXXXX

( 7) 35 XX

( 2)

XX

( 2)

XXXX

( 4)

XX

( 2) 36/00~

XX

( 2)

XX

( 2)

XXXXXXXXX

( 9)

XXXXX

( 5) 1

( 0)

XXX

( 3)

XXXXXX

( 6)

XX

( 2) 2 XXXX

( 4)

XXXX

( 4)

XXXX

( 4)

XXXX

( 4) 3 XXX

( 3)

X

( 1)

XXXXXX

( 6)

X

( 1) 4 XXX

( 3)

XX

( 2)

XXXX

( 4)

XXXXXXX

( 7) 5 XX

( 2)

X

( 1)

X

( 1)

XXXXX

( 5) 6 XX

( 2)

X

( 1)

XX

( 2)

XXXXXXXXXXXXX (13) 7 XXX

( 3)

XXX

( 3)

( 0)

XXXXX

( 5) 8 XXX

( 3) XXXX

( 4)

( 0) XX

( 2) 9 XXXX

( 4) XX

( 2)

X

( 1) XXX

( 3) 10 XXXXXXX

( 7) XXX

( 3)

( 0) XXXX

( 4) 11 XXXX

( 4) X.4 ;XXXX

( 7)

X

( 1) XXX

( 3)

o 7

12 XXXXX

( 5) XXXXXXX

( 7) X

( 1) X

( 1) i g

13 XXXX

( 4) XXXXX

( 5) X

( 1) XXXX

( 4) 3 14 XXXXXXXXXXXXX (13) XXXXXXX

( 7) XXXX

( 4) XXX

( 3) 15 XXXXXXXXXXXXXXX (15) XXXXXXX

( 7) XX

( 2) XXX

( 3)

(

i 16 XXXXXXX

( 7) XXXXXXXXXXXX (12) XX

( 2) X

( 1)

~

j 17 XXXXXXXX

( 8) XXXXXXXXXXXX (12) X

( !) XXX

( 3)

( 0)

( 0)

( 0)

( 0)

( 5),

Hissing

( 8)

( 8)

(6)j

(

1 I

I Table 8.

Pasquill-Gifford Stability Class (60-10m I

Delta T) at Perry Du-ing Lake Breeze Hours in April through July 1980 j

I Number of I

P-G Class Observations A

4 1.6 8

14 5.7 C

47 19.2 D'

159 64.9 E

14 5.7 I

F 5

2.0 G

2 0.8 245 100 I

I I

I I

l l

3-33 Revision 1

f{~

~

I TABLE 9 LAND-WATER TEMPERATURE DIFFERENCE DURING SEA BREEZE HOURS AT PERRY NUCLEAR PLANT, APRIL-JULY 1980 I

DELTA T (*F)

N 5

-10 X

1 0.4 I

-09 XXXXX 5

2.0 2

-08 X

1 0.4

-07 X

1 0.4

-06 X

1 0.4

-05 XXX 3

1.2

-04 0

0

-03 XXXX 4

1.6 I

-02 XXXXXXXXX 9

3.7

-01 XXXX 4

1.6 0

XXXXXXXXXXXX 12 5.0 I

+01 XXXXXXXX 8

3.7

+02 XXXXXXXXXXXXXXXXXXX 19 7.8

+03 XXXXXXXXXXXXXXXX 16 6.5

+04 XXXXXXXXXXXXXXXXXXX 19 7.8

+05 XXXXXXXXXXX 11 4.9

+06 XXXXXXXXXXXXXXXXXXXXXXX 23 9.4

+07 XXXXXXXXXXXXXX 13 5.3

+08 XXXXXXX 7

2.8

+09 XXXXXXXXXXXXXX 14 5.7

+10 XXXXXXXXXXXXXXXX 16 6.5

+11 XXXXXXX 7

2.8 I

+12 XXXXX 5

2.0

+13 XXXX 4

1.6

+14 XXXXXXX 7

2.8 I

+15 XXXX 4

1.6

+16 XXXXXX 6

2.4 t

+17 XXXXXX 6

2.4

+18 XXX 3

1.2 I

+19 XXX 3

1.2 l

+20 X

1 0.4

+21 0

0 l l

+22 0

0 E

+23 0

0

+24 0

0

+25 0-0

+26 0

0

+27 0

0

+28 0

0

+29 X

1 0.4 I

+30 0

0

+31 X

1 0.4

+32 XXXX 4

1.6 I

+33 X

1 0.4

+34 0

0

+35 0

0 I

l Revision 1 3-34

M M

M M

M M

M M

M q

q TABLE 10 CONDITIONS OBSERVED AT Tile PERRY NUCLEAR POWER PLANT IN Tile IIOUR BEFORE Tile ONSET OF A LAKE BREEZE REGIME linur Penetration PNP (10m)

PNP (60m)

T*

Honth Day (GMT)

(km)

TD-Vr 1 0 - vv-P-6 Solar Precip.

1 April 6

15 8

24*

3.0 24*

3.2 D

M 0

+16*

April 22 16 3

23*

2.9 23' 3.4 D

H 0

+29' Hay 15 13 30 23' 3.2 24' 4.2 0

M 0

-0l*

0

+0l*

May 16 13 30 09' 2.7 10' 3.8 D

H Hay 27 12 30 19*

2.3 06*

3.1 F

If 0

-07*

June 12 14 20 25' 3.9 25' 6.0 0

H 0

+05*

June 13 18 2

18*

3.8 17*

4.2 C

H 0

+18' June 14 14 9

21' 3.8 21' 4.8 D

M 0

+14*

June 17 13 35 22*

1.3 30*

2.1 F

H 0

-14*

L, June 18 14 15 24*

3.0 24*

3.2 0

M 0

+04*

o, cn June 21 12 25 22*

2.5 24' 5.0 E

M 0

-05*

June 22 14 9

20*

2.1 21*

2.2 D

H 0

+06*

June 24 14 10 22*

1.5 23*

1.5 0

11 0

+08' June 25 13 12 23' l.9 23*

2.9 D

ll 0

405*

June 26 14 8

24*

4.0 24*

4.8 D

H 0

+10' 0

+04*

June 29 12 8

25*

4.8 25' 7.0 0

July 3

13 30 19' O.9 27*

2.3 F

M 0

-11*

July 14 11 9

14*

2.6 16' 6.8 G

0

-13*

July 18 14 18 11' l.1 10' l.2 D

M 0

-02*

July 24 13 35 10' l.2 07" 2.1 E

11 0

-06*

July 25 14 16 22*

2.7 23' 3.3 0

M 0

40l*

July 26 15 10 24*

2.9 25*

'3.6 D

11 0

+05*

• T = Average Ambient Temperature - Lake Water Temperature.

j{

3 D7 DD = Wind Direction VV = Wind Speed

{

P-6 = PG Stability Class

E E

E E

E E

E E

E Table 11 PNPP METEOROLOGICAL SYSTEM EQUIPMENT SPECIFICATION (Page 1 of 3)

System Manufacturer Model Number Range Location Characteristics MAIN SYSTFM Wind speed system Teledyne Geotech Cup I M-41 0 to 100 mph 10m (primary) 9' W of tower Threshold 0.60 mph Sensor 15648 60s (primary) 9' W of tower Distance constant 5.0 ft includes cups, sensor, 10m (validity) 9' W of tower Error + 0.29 aph less Processor 21.11 tad processor 60m (validity) 9' W of tower than 5 mph

+ 1.12% from 5 mph to 50 mph u

Ct Wind direction system Teledyne Geotech Vane 53.2 0 to 540' ide (primary) 9' W of tower Threshold 0.70 mph eW 60s (primary) 9' W of tower Damping 0.4 sensor 15655 includes vane, sensor, 10m (validity) 9' W of tower Distance constant 3.7 f t Processor 23.22-1 end processor 60s (validity) 9' W of tower Error

• 3*

Temperature system Teledyne Ceotech RTD T-200 T -20 to 100 F 10m (primary) 6' W of tower T accuracy + 0.!!*F 10m (validity) 6' W of tower Time constant I sin Processor 40.35 RTDs and processor 60m (primary) 6' W of tower 327C aspirated T

60s (validity) 6' W of tower shield Delta T (60-10m)

Delta T range card Delta T -4 to 8*F Delta T accuracy + 0.!! F 20.42X Precipita6 tion Belfort 5-40$u rain gauge 0.01" increments Ground level Accuracy +_ 11 (0.01" for 1"/ hour)

Weather Measure P565 wind shield Teledyne Geotech Processor 21.52

-20 to 100*F 10m 6' W of tower Accuracy + 0.7 F y

Dewpoint EC&C 220 Accuracy + 0.02" of Hg 2m p

Station pressure Teledyne Geotech BP-100 Processor 40.68 28 to 32" of Hg (Main shelter) sensor and processor o

3 e-*

)

e

m

'M M

M M

M M

M M

Table 11 (Cont.)

PNPP METEOROIOCICAL SYSTEM EQUIPMENT SPECIFICATION (Page 2 of 3)

System Manufacturer Model Neber Range Location Characteristics MAIN SYSTEM (CONTINUED)

Multipoint recorder Temperature Esterline-Angus 4.!!24E

-20 to 100*F Main shelter Accuracy 1 0.25Z of Delta T (60-10m)-

12 channel

-4 to 8 F full scale

-20 to 100*F Dewpoint Pressure 28 to 32" of Hg Precipitation 0 to I" Speed Servo 11 Recorder Esterline-Angus Lil S2 S 0 to 100 mph (10m)

Main shelter Accuracy + 0.25% of Y

(3 ea) (we/wd)

O to 100 mph (60m) full scale 0 to 540* (10,60s) w i

N Microprocessor Dagical Equipsent LSill/23 CPU Nain shelter Accuracy of analog Corporation EFDil-AA to digital converter is better than Analog to Digital Convarter 1012

+ 0.10% of full scale BACKUP SYSTEM Wind speed system Teledyne Ceotech Cup 170-41 0 to 100 mph 10m (primary) 13' W of tower Threshold 0.60 mph includes cups, sensor.

Sensor 15648 10m (validity) 10' W of tower Distance constant 5.0 ft end processor Processo-40.12CX Error + 0.29 aph less than 5 mph

+ 1.122 from 5 mph to 50 mph

D cu<

Wind direction system Teledyne Ceotech Vane 53.2 0 to 540*

10m (primary) 13' W of tower Threshold 0.70 mph Includes vane, sensor, Sensor 1565B 10m (validity) 10' W of tower Damping 0.4 o

end processor Processor 40.22-1 Distance gonstant 3. 7 f t Error + 3 e-*

/

E E

E E

E E

g g

Table 11 (Cont.)

PNPP METEOR 01.OCICAL SYSTEM EQUIPMENT SPECIFICATION (Page 3 of 3)

System Manufacturer Model Number Range Location Characteristics BACKUP SYSTEM (CONTINUED)

Temperature system Teledyne Geotech RTD T-200

-20 to 100 F 10m (primary) 6' W of tower Ambient f 0.20*F Processor 21.32 10m (validity) 6' W of tower Time constant I min.

RTDs and processor 327C Aspirated

)

shield Servo recorder Esterline-Angus 6 channel recorder

-20 to 100 F Backup shelter Accuracy 1 0.5%

full scale MS426C 0 to 100 mph i

Temperature vs/wd O to % 0" 7

i Microprocessor Digital Equipment LS111/23 CPU Backup shelter Accuracy of analog uO to digital converter i

Corporation KFDil-AA is better than l

Analog to Digital 3 0.10% of full scale Converter 1012 l

}

l M

I Y

a O

3 w

I l

5 I

Table 12. Summary of Variables Reported from the Perry Meteorological Tower Main System Backup System 10m wind speed

• 10m wind speed

• 10m wind direction

• 10m wind direction

• 10m sigma 10m sigma Stability classes Stability classes (for y and z)

(for y and z) 10m temperature

• 10m temperature

• Delta temperature

• Stability classes (for y and z)

Dewpoint Precipitation Station pressure 60m wind speed

• 60m wind direction

• 60m sigma Stability classes (fcr y and z)

• Dual instrumentation for validation 3-39 Revision 1

I I

Table 13. Alternate Data Sequence in ERIS for I

Onsite Perry Meteorological Data Variable and First Second Third Primary Source Alternate Alternate Alternate Main System Backup System Main System N/A 10m Wind Speed 10m Wind Speed 60m Wind Speed (3)

Main System Backup System Main System N/A 10m Wind Direc-10m Wind Direc-60m Wind Direc-tion tion tion Stability for Backup System Main System Backup System Dispersion (l)

Horizontal (Main System 10m Sigma Delta T 60m Sigma 10m Sigma)

(60-10m)

Stability for Main System Backup System Main System Dispersi n 9

Vertical (2)

(Main System 10m Sigma 10m Sigma 60m Sigma Delta T,60-10m)

Main System Backup System N/A N/A Temperature Temperature (10m) 1.

Stability classification based on sigma theta method or equivalent l

pseudo sigma method.

l 2.

Stability classification baed on delta T, or equivalent modified sigma theta method.

3.

Power law wind profile applied.

lI I

l 3-40 Revision 1

I I

Table 14 I

CLASSIFICATION OF ATMOSPHERIC STABILITY I

BY TEMPERATURE CHANGE WITH HEIGHT Stability Pasquill Temperature Change Classification Categories with Height ('C/100 m)

Extremely unstable A

AT/Az 5 -1.9 Moderately unstable 8

-1.9 < AT/Az 5 -1.7 Slightly unstable C

-1.7 < AT/Az 5 -1.5 Neutral 0

-1. 5 < AT/Az $ -0. 5 Slightly stable E

-0.5 < AT/Az 5 1.5 Moderately stable F

1.5 < AT/Az 5 4.0 Extremely stable G

4.0 < AT/Az I

f Based on Regulatory Guide 1.23 (1980)

I

I I

a l 'J l I 3-41 Revision 1

I I

Table 15 CLASSIFICATION OF ATMOSPHERIC STABILITY 8Y SIGMA THETA Stability Pasquill o*

Classification Categories (decr$es) e > 22.5 Extremely unstable A

c Moderately unstable B

22.5 > oe 1 17.5 Slightly unstable C

17.5 > c, > 12.5 Neutral D

12.5 > oe > 7.5 Slightly stable E

7.5 > oe > 3.8 I

Moderately stable F

3.8 > c 12.1 6

Extrerely stable G

2.1 > og I

lI I

I I

lI l

l l

" Standard deviation of horizontal wind direction fluctuation over a period of l

15 minutes to 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />.

t Based on Regulatory Guide 1.23 (1980) 3-42 Revision 1 e

I Table 16 I

I Mo'dified Sigma Theta Method to Estimate the Stability Class for Sigma Z from Sigma Theta I

I This table is followed during the night which is defined as the period of one hour prior to sunset to one hour af ter sunrite. During the day the stability is determined directly from Sigma.

Then the If the Sigma And if the wind speed u is stability class for I

stability class is (m/s)

(mi/h) the vertical (z) is I

A u LT 2.4 u LT 5.3 G

2.4 LE u LT 2.9 5.3 LE u LT 6.4 F

2.9 LE u LT 3.6 6.4 LE u LT 7.9 E

3.6 LE u 7.9 LE u D

B u LT 2.4 u LT 5.3 F

2.4 LE u LT 3.0 5.3 LE u LT 6.6 E

3.0 LE u 6.6 LE u D

C u LT 2.4 u LT 5.3 E

2.4 LE u 5.3 LE u D

D no restriction D

E no restriction E

F no restriction F

G no restriction G

I I

I I

I 3-43 Revision 1

I Table 17 Pseudo Sigma Method to Estimate the Atmospheric Stability Class Applicable to Horizontal (y) Dispersion on the Basis of Delta T I

If it is daytime, the delta T stability class is used directly as being representative for sigma y.

If it is nighttime, apply the following:

Then the If the Delta T And if the wind stability class for stability class is speed (mi/h) is the horizontal (y) is G

u LT 5.3 A

5.3 LE u LT 6.4 B

I 6.4 LE u LT 7.9 C

7.9 LE u D

I F

u LT 5.3 B

5.3 LE u LT 6.6 C

6.6 LE u E

u LT 5.3 C

5.3 LE u D

any speed D

C any speed C

B any speed B

A any speed A

GT = greater than GE = greater than or equal I

LT = less than LE = less than or equal I

I I

I 3-44 Revision 1

I I

Table 18. Sources of Local Meteorological Data Location Station Distance Inland Number Name from PNPP Distance I

1 Eastlake Plant, CEI 17 miles, WSW 0 mile 2

Ashtabula USCG 19 miles, ENE 0 mile 3

Perry (PNPP)

On site 1 mile 4

Lost Nation Airport 16 miles, WSW 1 mile 5

Woodworth Airport 5 miles, E 2 miles 6

Casement Airport 6 miles, SW 2 miles 7

Woerner Airport 12 miles, E 2 miles 8

Lake County Health 8 miles, SW 3 miles District, Painsville 9

Germack Airport 12 miles, E 6 miles 10 Concord Airport 10 miles, SSW 7 miles 11*

Eckhard Airport 10 miles, SE 8 miles 12 Birdland Airport 10 miles, S 9 miles 13 Ashtabula Co. Airport 23 miles, E 9 miles 14 Armington Airport 15 miles, ESE 10 miles 15 Whispering Pines 11 miles, SSE 11 miles t

l (Fielitz) Airport l

16 Thompson Airport 13 miles, SSE 13 miles I

No longer in operation - 1/84 I

l l

3-45 Revision 1

W W

m M

M M

M M

M M

M M

M M

M M

l

.i l

i 4

'O ~

~ LITTLE TURBULENCE RETURN FLOW

~

7-HEATED LAND

'~~~'%

N AIR

"~

1000

_,,_ _,_ l l -

g l

N i

\\

HE HT CONSIDERABLE TURBULENCE

{~~

LITTLE TURBULENCE ij$j;p::'1 HEATED INFLOW !?', '

.z:f

HEATED -

LAND dAKEMCEAN.

AIR f

'; At

\\

I COOL

/

V 3

~

LAKE / OCEAN 7

NSIDERABLE;;

g i

AIR TURBULENCE FRONT l

/

$gyz;::

l 4

l g

a w

{

Figure 1 Lake / Sea Breeze Circulation And its Effect On Near-Coastal Releases l

1 I

\\

h

-l'Q

][Aonto

{'

'~

{

s

~ <-

~

,c

\\.

/

M 0

g

w. / NN

/

a.n A'g j

N

/-

YHM' YU he.h

... n / N. ~pr. a

_-..___t.

• . ~.

V h

.r ig t

g 7?

~y y

urr I

n.\\ W3.. 4*

i -

5' 'm =-,-e, WMK

..x f#

i

+-

y

\\

).

e-y x k??

%- ~ > 's '/. 0#ue i estion %

.?

o

'N. I'.A.,

6 k:,._..ms

/gr,.

• ~).

e

/

yoo r

E 3.-, 2%g P

E

\\p.t.:

i t

y R

{.

[

V-E

, i.

>/

,,3 -

} [,,, h c-

-c

.,,-~

{ p,.,

.--..He.

..+

\\,

I ry,

led *" Lot. Nation

'LhI, L,u

.. [*

~

+

t,Ashtabula

.n w

...-.a g

o g t '<' EastEe

• i

!Y s P E N N'SYl1V 4 N 0-I

- 3._,

. * ~ y s. _-

-yn-

~ //

us.,

~ ~ '

~

n.

,,~

., ' s _,aaf..,r ~ ~ [

o.. -

PM -

-=

~

[M*.h 3pg

,. cye

. -. f..y',

I

[g i. _J,J

-- ~

,/CLE,t.

.g 5*

,s g

l

,s a.

! l

^

k *,,,,"

$*'% j 77-*

7 m. - v% +

w.,gy N

.$ llI hy b

~5~~ g pQ /Q { mplY' g h-d [tg g.

l

~~

m s,

s

'l'% Pa.i

\\*

jh' ?

L

• C, ?

Y g--

c~

Map reference: S. B. Cohen,19 73; Os/orcr Worg Arfat LEGEND

-)( Perry Nuclear Power Flant (PNPo; S

'E u

Meteorological Data Locations (ABC = Station identifier)

I

@ National Weather Semce (U. S.)

@ Atmospheric Environmental Service I

(Canada)

S USCG Figure 2 Locations within 100 miles 3 FAA of the PNPP A Other 3-47 Revision 1

\\

h 1

EASTLAKE PLANT 2

ASHTABULA USCG j

3 PERRY (PNPP) 4 LOST NATION AIRPORT 5

WOODWORTH AIRPORT 6

CASEMENT AIRPORT 7

WOERNER AIRPORT 8

LAKE COUNTY HEALTH DISTRICT i

9 GERMACK AIRPORT 10 CONCORD AIRPORT i

11 ECKHARD AIRPORT J

12 BIRDLAND AIRPORT 8 *, u 13 ASHTABULA COUNTY AIRPORT 0

14 ARMINGTON AIRPORT 15 WHISPERING PINES (FIELITZ) AIRPORT

/

- rq 16 THOMPSON AIRPORT.

G',Y y

&g sis-. - -

s 1

i

~ o P/~

L

• -L

a y

\\3,. 'Q[ ' ] TA.

7

~

g.

, s.

.e I.

~.A;_

'5 i,,,'

v l~,

'~"

' f,

%2 x,..,

F

.m g"~i, j

_..L.c.

t.

~

j,.

lU

[

r uu Na I

7

/

• • x

{

_,2

.~.

1 O

w-3 12A.

w.

,j' h

g.;

. F[,

5)#,

s-

}

p,N

~

PD 4*

q

,i

.s.

g k

.u..

ls r

~

~'.w 6

i

- ~[ 9

• j tg u u u...

2

.l

\\

- r !,T,, -

t N>, P

~'

9 K

.t ip6 ' '

-n.

/

[,9

' }l;; _

jy 'n j ' \\A ei

fy v

7 h!

k f

T

$T -

1. "i Y

3.ac.c.ut.

C -

'1 J

't

?

e l

1

! (1 l [j 3-48 Revision 1

F EWM

,/

t 2

p

/,

,- s--

7

/

.,Y I

I'-

,. (_

.N y

.b

jj y,.g 7

- (Q" r

j

i. -

g b

!./

)

d M

,cy 3

RQ== f f-1 l

f%

\\ ) 13

..O

,r

Il)

,I 91 b

Y' l._ W 3

N f

A

..r.WE&

l Y

~,

\\

[

I4 n.

i. =

9_ >; /1,

)

,I 7

QA

~.

~

r ' " iT"' ' ' I

\\

\\\\

~-

m j

4 s

N

~,

4l

,p G:=(

~

i s x

• I 0

1 2

3 5

10 3

MILES j

i

</ 'g am O'

O 2

4 8

16 l.

i k

l M KILOMETERS l

N

_t a

\\-N, il Tri-County Planning Area 5e y

Perry Nuclear Power Plant Also Available On Figure 3.

Aperture Card Map of Sources for Meteorological Data 85 0108 0 3 7 9 -b

^

E E

E E

E E

E E

E E

E M

E E.

W W

M g

i

)

Figure 4 TIBL Depth as Function of inland Distance i

Legend:

Observations Estimates

+ Lake Michigan (Lynet 1975)

O Composite Methods (Septoff et al.,1976)

X Lake Ontarse (Mitchell, itI5a)

O Reyner et at Method (Septoff et al.,1976)

J

• Lake Erie (NUREU/CR-0936)

A Lake Dataria Method (Mitchell,1975b) a l

i'l l

I 8

l 8

l l

g g

i l

e l

l e

l i

e s

g l

~

T 300 - -

g i

HEIGitT I

200- -

g,g 4

j l

x 1

o l

100--

i 2

~

i

/

~

l t

l t

i I

I I

t 0

0 1

2 3

4 h

6 h

8 9

ls Il l2 13 14 15 16 DOWNWINO FETCH FROM SHOREt INE (km)

Relative Location / Height of l

PNPP Meteeseleycal Tower Note: Tuus of the sarse symbols et a yvem distance imply the s age of values observed.

I I

4.0

SUMMARY

The methods and technical bases have been presented for making emergency offsite dose calculations at the Perry Nuclear Power Plant. Both a hand-calculated and a compatible automated method will be adopted.

I The automated method is the more sophisticated one.

It uses real-time source term and release characteristics information, as well as real-time meteorology that takes into account the coastal location of the PNPP. The system is menu driven to enhance the man-machine interface. The system will provide for rapid dose assessment for the Perry EPZ in the event of an accidental atmospheric release.

I I

I I

I I

I I

I 4-1 Revision 1

I I

5.0 CITED REFERENCES Bennet, R.C. and R. List,1975: Lake Breeze Detection with an Acoustic Radar.

Preprints of 16th Radar Meteorology Conference, April 22-24, American Meteoro-I logical Society, 260-262.

CEI, March 1983: Description of Perry Nuclear Power Plant Emergency Response I

Facilities. Units 1 and 2 Docket Nos. 50-440 and 50-441. The Cleveland Electric Illuminating Company.

Cohen, S.B., 1973: Oxford World Atlas. Oxford University Press. New York, I

NY.

Guski, M.E., and P.L. Miller, 1980: A Study of the i.ake Breeze Circulation in the Proximity of Lake Ontario. Second Conference on Coastal Meteorology.

Paper 5.1, American Meteorological Society, January 30 - Feburary 1, Los Angeles, CA.

Hoehne, W.E., 1971: Standardized Functional Tests. NOAA Tech Memo NWS T&EL-12, U.S. Department of Commerce.

Hoehne, W.E., 1977: Progress and Results of Functional Testing. NOAA Tech Memo NWS T&EL-15, U.S. Department of Commerce.

I Irwin, J.S., 1980: Dispersion Estimate Suggestion No. 8.

Environmental Appli-cations Branch. NOAA/ EPA Meteorology and Assessment Division, Environmental Science Reserach Laboratory.

Lyons, W.A., 1975: Turbulent Diffusion and Pollutant Transport in Shoreline Environments. Lectures on Air Polution and Environmental Impact Analyses.

American Meteorological Society, Boston, MA.

Lyons, W.A. and L.E. Olsson, May 1973:

Detailed Meso-Meteorological Studies of Air Polution Dispersion in the Chicago Lake Breeze. Monthly Weather Review, 101:5, 387-403.

Mitchell, A.E., Jr., June 1975a: Case Stuides of the Lake Breeze on the South Shore of Lake Ontario. NUS-TM-S-205. NUS Corporation, Rockville, MD.

Mitchell. A.E., Jr., June 1975b: Growth of the Thermal Internal Boundary Layer During the Lake Breeze and Stable Onshore Flow. NUS-TM-S-206. NUS Corporation, Rockville, MD.

Mitchell, A.E., Jr., June 1975c: General Study of Stable Onshore Flow at Somer-set. NUS-TM-S-207. NUS Corporation, Rockville, MD.

Mitchell, A.E., Jr. and W.G. Snell, 1981:

Effect of Four Stability Classifi-cation Methods on Dispersion from an Elevated Source. Paper 3.3, Fifth Sympo-sium on Turbulence, Diffusion. and Air Pollution. American Meteorological I

Society. March 9-13, Atlanta, GA.

I 5-1 Revision 1 l-

I I

Mitchell, A.E., Jr. and K.0. Timbre, 1979: Atmospheric Stability Class from I

Horizontal Wind Fluctuation. Paper 79-29.2, 72nd Annual Meeting, Air Pollution Control Association, June 24-29, Cincinnati, OH.

Mitchell, A.E., Jr., 1982: A comparison of Short-Term Dispersion Estimates I

Resulting from Various Atmospheric Stability Classification Methods. Atmo-spheric Environment, 16:4, 765-773.

I Mitchell, A.E., Jr., R.W. Jubach, A. Malkoc, 1984: Use of Operational Comparability Techniques to Determine Realtime Acceptability of Meteorological Measurements. Manuscript submitted to Journal of ' Atmospheric and Oceanic Technology.

Moroz, W.J., July 1967: A Lake Breeze on the Eastern Shore of Lake Michigan, Observations and Model. Journal of Atmospheric Science. 2$:4,337-355.

j I

Moroz, W.J. and E. Koczkur, September 1967: Plume Rise and Dispersion Year the Shoreline of a large Lake When Flow Patterns jre Dominated by the LLL6 Breeze. USAEC Meteorological Information Meeting.

Olson, L.E., 1969: Lake Effects on Air Pollution Dispersion. University Microfilms, Ann Arbor, Michigan.

Pasquill, F., 1976: The Gaussian Plume Model with Limited Mixing. Report EPA 600/4-76-042, Environ. Monitoring Service, NTIS, Springfield, VA.

Raynor, G.S., et al, September 9-13, 1974: A Research Program on Atmospheric Diffusion from an Oceanic Site. Brookhaven National Laboratory. Synposium on Atmospheric Diffusion and Air Pollution, Santa Barbara, California, 289-295.

Raynor, G.S., P. Michael, and S. SethuRamen,1979: Recommendations for Meteoro-logical Measurement Programs and Atmospheric Diffusion Prediction Methods for Use at Coastal Nuclear Reactor Sites. NUREG/CR-0936, USNRC - Brookhaven hational Laboratory.

Septoff, M., A.E. Mitchell, Jr. and L.H. Teuscher, October 1976: The Design I

of an Onshore Tracer Program at the San Onofre Nuclear Generating Station.

NUS-2002, Appendix A.

NUS Corporation, Rockville, MD.

Turner, D.B., 1970: Workbook of Atmospheric Dispersion Estimates. PHS No. 999-AP-26.

U.S. Department of Health, Education and Welfare.

USNRC, 1972: Onsite Meteorological Programs, Regulatory Guide 1.23 (Safety Guide 23).

U.S. Nuclear Regulatory Comission.

USNRC, 1975: Calculation of Reactor Accident Consequences. Reactor Safety Study. WASH-1400.

U.S. Nuclear Regulatory Commission.

USNRC, October 1977: Calculation of Annual Doses to Man from Routine Releases I

of Reactor Effluents for the Purpose of Evaluating Compliance With 10 CFR Part 50, Appendix I.

Regulatory Guide 1.109, Revision 1, U.S. Nuclear Regulatory i

Comission.

I 5-2 Revision 1

I

\\

l I

USNRC, August 1979: Atmopsheric Dispersion Models for Potential Accident Con-I sequence Assessments at Nuclear Power Plants. Regulatory Guide 1.145.

U.S.

Nuclear Regulatory Comission.

USNRC, September 1980a: Meteorological Programs in Suport of Nuclear power I

Plants. Proposed Rev. 1, Regulatory Guide 1.23.

U.S. Nuclear Regulatory Com-mission.

I USNRC, October 1980b: Criteria for Preparation and Evaluation of Radiological Emergency Response Plans and Preparedness in Support of Nuclear Power Plants.

NUREG-0654/ FEMA-REP-1, Rev 1.

(Endorsed by Regulatory Guide 1.101, Rev. 2, October 1981: Emergency Planning and Preparedness for Nuclear Power Reactors.)

Webb, J.S., 1974:

Surface Temperatures of Lake Erie, Water Resources Research, Vol. 10, No. 2, pp. 199-210.

I I

I I

I 5-3 Revision 1

6.0 CROSS-REFERENCE TO NRC COMMENTS I

6.1 Round 1 Questions This lists the sections of this document in which NRC Round 1 questions on the Perry Emergency Plan are addressed. Only questions directly relating to emergency offsite dose calculations are cross-referenced. References to Chapter 2 concern the hand-calculated method; Chapter 3 contains the automated methodology.

Question Location (s) Addressed Coments I.3 Sections 2.0, 2.4-2.10 Identification of the accident, and 8

Section 3.1 therefore, the source terms, is the responsibility of the operator. Once the accident is identified, the source terms and release magnitude can be

" fine-tuned" using results of effluent analyses.

I.4 Secticks-2.4, 2.5, 2.8 Section 3.2 includes an incorporation Section 3.1 of the lake breeze.

I I

I.6 Sections 2.9, 3.2, 3.1 If the ERIS computer is inoperable, use i

entire Chapter 2.

I.10 Sections 2.8, 2.10 Integrated doses are addressed in Sec-I Section 3.1 tions 2.10 and 3.1.

6.2 Round 2 Questions This lists the sections of this document in which NRC Round 2 questions on the I

Perry Emergency Plan are addressed. Only questions directly relating to emergency offsite dose calculations are cross-referenced. References to Chapter 2 concern the hand-calculated method; Chapter 3 contains the automated methodology.

I Question Location (s) Addressed Comments H.5 Sections 2.1, 2.2, 2.3 Sections 2.2 and 2.3 contain backup Section 3.1 methods for estimating dispersion I

parameters based on visual observa-tions.

6-1 Revision 1

H.6 Section 3.1 Actual dose calculation methods are in I

Section 3.1.

H.7 Sections 2.1, 2.2, 2.3 Sections in Chapter 2 contain backup Sections 3.1, 3.2 methods for obtaining dispersion esti-I mates.

I.3 Sections 2.4, 2.5 Backup methods are in Sections 2.4 and Section 3.1 2.5.

8 I.4 Chapter 2, Chapter 3 Chapter 2 addresses hand-calculated I

methods; Chapter 3 discussed automated methods.

I.5 Sections 2.9, 3.1, 3.2 If the ERIS computer is inoperable, use i

entire Chapter 2.

6.3 Contractor Evaluation Findings on Meteorology This lists the sections of this document in which the NRC-Contractor evaluation is addressed.

(Reference NRC letter August 22, 1984 (Youngblood to Edelman)).

Recomendation Location (s) Addressed 1

Sections 3.1, 3.2, 3.3 2

Sections 3.2, 3.4 l

l l

6-2 Revision 1

\\

+,-

I!

i NUS Corporation Domestic Offices intomational Offices 910 Clopper Road Aiken, SC Grazil Gaithersburg, Maryland 20078 Arlington, VA Germany 301/2584000 Aurora, CO Japan-TWX: 7108284540 Chicago, IL Saudi Arabia NUSWASH Clearwater, FL Columbia, SC

' Gaithersburg, MO Tucker, GA

. Houston, TX

' Menlo Park, CA Bedford, MA Edison, NJ Wayne, PA Pittsburgh, PA Portland, OR

. San Diego, CA E

' Sherman Oaks, CA g'

i i

l'5 -

e

,e.

c

>?

L

(

I 1

[

Il

_