ML19257D718

Evacuation Time Estimates for Areas Near Facility

ML19257D718
Person / Time
Site:	Rancho Seco
Issue date:	01/30/1980
From:	Byrd C, Gerber R, Hubenett R CENTER FOR PLANNING & RESEARCH, INC.
To:
Shared Package
ML19257D706	List: ML19257D705 ML19257D718
References
NUDOCS 8002050549
Download: ML19257D718 (56)

Text

,-

rs l

1 3

Final Report F.VACUATION TIME ESTIMATES FOR AREAS NEAR RANCHO SECO POWER PLANT 1

1 Prepared for:

SMUD Socramento Municipal Utility District Socramento, CA 95813 SMUD Contract No. 7503 I

I January 30, 1980 1876 145 I (

l 1

I I

Cen:er'or ?anning anc Researc1, nc.

I ec o m o m 1

I B

Final Report EVACUATION TIME ESTIMATES FOR AREAS NEAR RANCIIO SECO POWER PLANT I

by:

Robert Ilubenctte I

Cecil Byrd, Jr.

Robert Gerber Gerald Kopp Charles T. Raincy I

for:

SMUD I

Sacritmento Municipal Utility District Sacramento, CA 95813 I

SMUD Contract No. 7503 I

January 30, 1980 Center for Planning and Research, Inc.

2483 East 13ayshore Road Palo Alto, CA 94303 i

((415) 858-0252 I

l

[876 14'6 I

CONTENTS Pace ACKNOWLEDG M ENTS v

REVIEW NOTICE vi LIST OF FIGURES vii LI ST O F TA B LES.........................

viii I.

INT R O D U C TIO N..........................

I-1 Purpose and Scope 1-1 I

Setting...............................

1-2 Approach and Methodology....................

1-4 Local O f ficials' Co m m en ts....................

I-5 II. POPULATION DISTRIBUTION II-1 I

B o u n d ar i es.............................

II-1 Residential Populations......................

II-1 Work er Popula tion.........................

II-1 School Populations...................

II-4 I

Recreational Visitors II-5 Others II-5 Design Popula tion.........................

11 - 5 I

III. SPECIAL EVACU ATION CONSIDERATIONS HI-I I

Special Facilities..........................

III-1 Transportation Constraints....................

III-6 Adverse Weather Impact.....................

III-7 IV.

EV A C U ATIO N R O U TES......................

IV-1 L oca l P l a ns.............................

IV-1 Road Network..................

IV-1 R ou t es by S eg m e n t........................

IV-1 Capac it y Ana lysis.........................

IV-1 Adverse Weather Impact.....................

IV-3 V.

MOVEM ENT A N A LYSIS......................

V-1 Sec t or Selec tion..........................

V-1 Analysis Met hodology.......................

V-4 Notification of Government Officials V-4 Alerting of General Population..................

V-5 Movement Preparation Time...................

V-8 Travel Time V-9 Adverse Weather Impacts on Evacuation Time......... V-11 1876,147 g

I

VI EV A C U ATIO N VE RIFIC ATIO N..................

VI-1 Verification Time Using Marker Technique VI-1 Other Verification Techniques..................

VI-1 VII EFFECTIVENESS OF ALTERNATIVE PROTECTION ACTIONS VII-1 Protective Actions Prior to Cloud Arrival VII-I Protective Actions During the Cloud Passage VII-4 Protective Actions After the Cloud Passage VII-8 A PPENDIC ES A-Letter to All Power Reactor Licensees, Request for Evacuation Times (After Notification) for Areas Near Nuclear Power Plants, Nuclear Regulatory Commission, Nov. 28,1979.

B-Comments by Reviewers I

I I

I I

I I

1876 148 I

I

I v

I ACK NOWLEDGM ENTS I

The authors were greatly assisted by Donald W. Martin, Environmental Specialist, SMUD, who monitored the contract; IIal White, Emergency Operations Coordinator, I

Sacramento County; and C.A. Terhune, Superintendent, Preston School of Industry.

Substantial assistance in the data collection was provided by the Planning Departments of Sacramento, A m ador, and San Joaquin Counties as well as by the California Department of Transportation. The timely cooperation of these representatives and agencies is gratefully acknowledged; without their aid, the work reported could not have been done.

I I

I I

I I

I I

I I

I l

1876 149 I

vi REVIEW NOTICE The following representatives of responsible agencies have reviewed the report:

I I

IIal White George Nelson Emergency Operations Coordinator Assistant Director County of Sacramento City of Stockton I

Office of Emergency Services I

I Colleen Robertson Jack Kearns Emergency Coordinator Assistant Director Calaveras/Amador Counties State of California Office of Emergency Services I

I Cleo Janiw Emergency Coordinator San Joaquin County I

Appendix B presents the written comments of those reviewers who wished to go on record.

i876 150 I

vii FIGURES Page I-1 General Area Map Rancho Seco Facility I-3 11-1 Initial Segments Used in Analysis Il-2 Il-2 Population Distribution Estimate - 1978 1-10 Miles II-3 III-1 A Special Facilities - General Locations III-2 III-1 B Special Facilities - Ione Area III-3 III-2 100 Year Flood Zone III-8 IV-1 Principal Paved Roads in Study Area IV-2 V-1 Estimated Residential Population by Quadrant V-2 I

V-2 Correspondence Between 22-1/2 Sectors and 180 /90 Sectors V-3 V-3 General Shape of Alert Verification Distribution V-7 I

V-4 Distribution of Starting Times with Radio Warning V-10 V-5 Example Cumulative Starting Time Curve V-11 VII-1 Hypothetical Dose Rate IIistory Curve VII-2 VII-2 Conditional Probability of Execeding Whole Body Dose Versus Distance VII-5 VII-3 Conditional Probability of Exceeding Lung Doses Versus Distance VII-6 VII-4 Conditional Probability of Exceeding Thyroid Doses Versus Distance VII-7 VII-5 Three Sigma Half-widths of Gaussian Shaped Plumes vs I

Downwind Distance VII-10 I

I I

I I

i876 151 I

vili TABLES Pace II-1 Comparison of Alternative Design Populations H-6 V-1 Evacuation Time for Rancho Seco V-13 V-2 Time to Evacuate 50% of Residents Around Rancho Seco V-14 f

V-3 Estimated Population Isolated by 100 Year Flood V-15 V-4 Evacuation Time for Rancho Seco for Heavy Rain V-16 I

VI-1 Time to Verify Extent of Evacuation VI-2 VII-1 A Cloud Arrival Times VII-3 VII-1B Summary of Evacuation Times VII-3 I

i I

I I

I I

I.

INTRODUCTION Purpose and Scope This report presents the results of a study of the time necessary to evacuate the area around Rancho Seco Nuclear Power Plant. The study was conducted by the Center for Planning and Research, Inc. (CPR) for the Sacramemo Municipal Utility District (SMUD) under SMUD Contract No. 7503.

The overall purpose of the project is ta assist SMUD in responding to the Nuclear Regulatory Commission's letter dated November 28,1979 (Appendix A), requesting time I

estimates to evacuate the area around Rancho Seco. The scope of the work, which is discussed in detail in applicable chapters, is as follows:

o Acquire up to date information on population densities, trafic flow data, special facilities (e.g., schools, nursing homes, and other medical care institu-tions, correctional institutions, etc.) for the 2, 5 and 10 mile radius around SMUD's Rancho Seco Nuclear Generating Station Unit No. I site.

o Provide an evaluation of the "best" and " adverse" weather conditions as they relate to traffic flow on major roads which would be used during evacuation of the general public from the 2, 5, and 10 mile radius around Rancho Seco.

o Provide an estimate of the time it would take to notify and con 6uct a general evacuation for best and adverse weather conditions for the following:

Radius I

Distance Area 2 miles Two 180 Sectors 5 miles Four 90 Sectors 10 miles Four 90 Sectors o Provide an estimate of the time it would take to conduct evacuation of each of the special facilities identified within and immediately adjacent to the 10 mile radius.

Estimates should address both best and adverse weather conditions.

I I-2 o Determine the time it would take to confirm that evacuation of the general public has occurred.

Discuss the effectiveness of various verification techniques, o Assess the effectiveness of alternative protective actions, such as sheltering, and determine if greater reduction in exposures can be anticipated.

I Setting The Raneno Seco Nuclear Generating Station is a nuclear power plant operated by the Sacramento County Municipal Utility District (SMUD). The facility is located in the southeastern part of Sacramento County, approximately 26 miles north-northeast of Stockton and 25 miles southeast of the City of Sacramento. (See map at Figure I-1.)

The clmest city is Ione,10 miles due east, which has a population of 2,370.

At an elevation of approximately 160 feet, the terrain surrounding the facility is slightly rolling, and at a distance of about seven miles east the terrain rises slightly as the land approaches the Sierra Nevada foothills.

The land surrounding the site within a ten miie radius is used primarily for agricultural and grazing purposes.

The estimated 1978 residential population within a 10 mile radius of Rancho I

Seco, including the City of Ione, is 8,183 persons. The population within a 2 mile radius of the facility is 101; at 5 miles it increases to 697, with the largest concentrations being between the 5 and 10 mile radius, particularly in Ione and to the northwest.

Within the 10 mile radius from the facility there are 17 special facilities, and these are within Amador and Sacramento Counties. There are five schools; a day care center; a treatment center for TB and alcoholic patients; six residential care homes; an adult training center for developmentally disabled; a Fire Academy; the Preston School of Industry; and a nudist ranch.

State Route 104 runs just north of the facility in an cast-west direction and I

connects with U.S. Routes 50 and 99 to the west, and with State Route 49 to the cast. Rail access to the facility is available by a spur line from the Southern Pacific Railroad.

The only recreational facility wholly within the area is the Rancho Seco Recreational Park, located at the Reservoir Area approximately 2 miles from the plant.

This is a well-used facility with the largest single day attendance in 1979 being recorded at 3,000 persons.

A small tip of Comanche Reservoir extends into the study area.

1876 154

3, p v:

.), _ v anng q pl I-3

1 I

I I

=*

1.

um f-g___,,,

I

(

[ %u.m

(

.a. sa S

_ Lh.

g ";"a l

"\\

g; y

" u..== o

/

mits,

I

(

I ek._

(_ 4-,

V' m.

a

'l

• ~ ~ L-I Figure 1-1 General Arca Map Rancho Seco Facility i

I

-l 1BT6 155 I

I-4 Approach & Methodology I

Data Collection With the exception of the data on special facilities, all data for the analysis (population, roadway characteristics, employment, meteorology, flood zone, etc.) were obtained from secondary sources - generally governmental agencies. Direct contact was made with all of the special facilities within 10 miles of Rancho Seco. The area was flown over by the study staff for orientation purposes and for observation of all the roads and the general distribution of structures. In addition, a staff member drove the principal roads in the potential evacuation area to identify any capacity problems I

that might be obvious.

The population data obtained from the respective county planning departments is for 1978 and was compared with 1975 aerial photographs in order to verify the distribution of population by segment.

Employment data was obtained directly for the special facilities in the area including the Rancho Seco plant.

Employment not associated with a special facility was estimated by CPR on the basis of payroll statistics by employment sector. This data, along with most of the other data sets, was analyzed on a 22-1/2 degree sector ba sis,.

I Limited mobility population (those without normal access to private automobile) in the area were identified only if they were institutionalized. The distribution within the general population of handicapped people with mobility problems is estimated to be between 5 and 6%, of which half would be moderately handicapped and half severely handicapped.

Generally these people do not reside alone and they derive their transportation support from other members of the household. This pattern is assumed to prevail for the evacuation situation. Typically, approximately 1% of the population requires transportation assistance from public agencies under normal circumstances, and this requirement will probably continue in the event of evacuation. Because of the small number of these severely handicapped individuals in any evacuation sector, they will not change the evacuation times estimated in this report.

Movement Analysis The area surrounding the site was divided into 16 sectors of 22-1/2 degrees each and was further subdivided by radii at 2, 5, and 10 miles from the reactor site.

Movement routes were selected for each segment (22-1/2, 2-5-10 mile radii) using USGS 7-1/2' quadrangle maps.

Specific 2spacity calculations were done for SR 1876 156 0

I-5 104 and 88.

The other two-lane surfaced routes that were expected to carry more than a few hundred vehicles were assumed to have a capacity of 800 vehicles per hour. The assumptions for traffic volumes on the evacuation routes used a standard I-occupancy factor for type of traveler (i.e., workers and families).

It was further assumed that the traffic mix wou:d approximate the vehicle registrations within the county.

Adverse weather conditions were pmtulated and then applied as constraints to the alerting times and road capacities.

Surprisingly, the adverse weather conditions probable in th7 area had much less impact than anticipated because of compensating factors. These compensating factors are discussed in Chapters IV and V.

The movement analysis was conducted by analyzing each discrete element in the process of notification, alerting, and traveling. Where possible, distributed functions were used to describe the occurrence in probability terms. Since the various parameters considered did not result in significant sensitivities to total time to evacuate, the analysis also considered time to move 50% of the population.

Verification of Evacuation Time estimates for a windshield survey of dwelling units were based upon road distances. A number of other verification techniques are discussed but no time estimates have been made for them.

I Alternative Protective Actions The only facility in the area where evacuation was not the primary counter-I measure was the Preston School of Industry. The suitability of using on-site shelter for the Preston inmates was examined, and the shelter was found to be adequate.

Only one or two other facilities in the area were considered suitable for sheltering, but this subject was not pursued further since evacuation provided the most suitable solution.

Also studied were the protective actions to be taken before, during, and after cloud passage, for both the evacuation and shelter options.

Local Officials' Comments The analysis of evacuation times was reviewed with appropriate local government officials. Review meetings were held at the beginning of the study in order to advise these officials of the ;tudy purpose and to enlist their support in providing data.

Another meeting was Leld approximately one-third of the way through the study to i876 157 i

I-6 indicate the methodology being employed in the study and to obtain comments on the approach and assumptions. A final meeting was held at the completion of the study to review the draft report. Specific recommendations made by the local officials were incorporated into the final report.

Any comments from these officials received by 1

SMUD will be attached to this report as Appendix B.

t l

I J

I l

l I

E 1

e I

e I

i876 158 g

II. POPULATION DISTRIBUTION Boundaries The study area was subdivided using the convention established by the NRC of dividing the area into 22-1/2 sectors which are further subdivided by 2, 5, and 10 mile radii.

Since the City of Ione is intersected by the 10 mile radius, all of the population of lone, including special facilities, was considered to be inside the radii.

I Figure 11-1 indicates the initial segments used in the analysis.

Residential Population Estimates of the 1978 residential population were provided by Amador, Sacramento, and San Joaquin County planning departments. The planning department and emergency preparedness staff assisted in providing a preliminary distribution of the populations within the segments. These distributions were adjusted on the basis of CPR's aerial survey of the site and comparison with U.S.G.S. quadrangle maps. The final population distribution is shown in Figure II-2.

The total estimated residential I

population is 8,183.

This residential population would be typical for nighttime and weekend conditions.

Duri* g the work day, however, a substantial portion of the population commutes to n

work outside the study area. Based upon data from the Sacramento County Planning District 19 (Rural Sacramento-south of Sacramento City), it was estimated that only one-third of the residents work outside their homes, and 70% of these workers commute out of the area. That is, during the work day only 67% of residents are "at home",

another 10% work in the area, and 23% work outside the area.

I Worker Population The largest concentration of workers in the area is at the Rancho Seco facility.

I On an average work day, Rancho Seco has approximately 200 workers on site. This number is reduced to a small force of 20-25 (workers and security personnel) at night.

During periods when modification and construction activities are underway, another 200 workers could be added. During refueling operations another 300 workers would be added. Consequently, on occasion a total of 700 workers could be at the facility.

SMUD has indicated that during the next two years, the facility will be undergoing extensive modification and that an average of 350 workers can be expected to be on 1876 159

II-2 Figure II-l INITIAL SEGMENTS USED IN ANALYSIS

'.' V l}

• l

'~' "~ 3 s

')

,(o??,f{

f

^#

'/

ht

T D+

• .S.acrame.nto I

n/-

4 s

\\

..w-as

-4.p g) ys

/.' b '1

{

')

\\

l Q:c( gg'#f~_. E

• b T

~

N ts s

g k6

/;p., * d I

[

\\

o 3,

~g,

=

l' j"

//

o's q~$

\\

I a

a is g n{

{

' t,i s,... P ?,0--

, p.

.!p-l .,,o3,

.-10 h 42 l

A g

L.,;w-i s_.

a, e

l$[C D

11,l? "

h, I

)

E

,j fx I f

'.J

N

/f

,---g---

2( -

S.

I. e s

/..

\\

is..nu - a

~

I

,,f

\\1

[ g/.,..'

• (g

,v

/'a g

/s 23 70 #

I I

2 j

'N,,,",g[a

,'d * " -=

\\

g-

/

\\..x D

3

<,o s.

!.1L u g

- w %,,.

a Pj

\\ *.

/'

.. tl

~g

.~s;

'g"7

% h P.~. Q, (, **..,,

,'. ;i,2 [ /..

o d,,.._d

,.//

j l

,,.., y -

,\\

• • • ,.L.

.s4 ![ & '"

A 1

2

/l~ ~.....;'T T ~....r

.:x:i -

p l

=; l,, 1)~~7 '

7, s s j'

'Pfc

• .......N.,.. o uJ """

. j..... %....

pH s

i..

'l*j -

4..

p' i

".':'*... R ANCHO S E CO d

i}

- {-

1 e

i

>N, 1

t,.

I

,, j s \\.

'8 f.,.

(is

,., I i.w "g

y e u s'

/

s si

.. li./

\\

x Il t'.p'l

,.4

/'r' p

4 4

,,e 1-, 4{'p-l

\\

,f 7

i! Ij

il \\'*

' 1\\-

,*. 11 lg;~, \\

r i,

['.f :,

a a

KG 8

s s.~

a L.... ni e I

~, ". -

b ie11 sans,o ' / -

a s.,,

L Li jj.{

• j f.-2.

- i n, s, a 1

,.o M' gi ' h 0 N ']

q..> r.i il,,a ;M+N v* '"... w.: U b I

,l l

I

'r' r'

l --.' ' ' W., ~#o*'

B

~

cE

- yW.

3,,)I k.

E i

p1 i

4 s

'l { f "'" M I

i 8

\\...

Y-'

a,

't< < f l l'~& si >s >

M d

.64 ek I

E 45 i

l j3 J

• )

i A.\\e

'N I

• 2 ' -

a

"."h,, t.

',1, g

I

... s 1

y r

.g o

"g* -

~

1 O

'N 7,', \\' g f'y"j:

47/ w.v 8

q.

s 7

I N

. a. r q f..; &

4..

u w

a..

J ug /...I

. j*

s,

• s q. w= _ /

,.l-l

.g.-.le'.l w.

3 ia i

,,i j

I F

3 4

o SCA.LE.

-1876 160 l

-e nm; c

!rt g

11-3 Figure II-2 POPULATION DISTRIBUTION 1978 ESTIMATE O-10 MILES A

.)

10 0.

s I

R (0)

(247) 12 220 0

O (1,121) 16 1,031 5 -n.

I 0

27 0

50 p

14 90 1,239 2.,

0 65 0

0 0

0 0

E (388) 236 123 29 O

O 2,370 0

H 29 9

U 113 0

0 0

30 (41)

I 690 (832)

M 17 22 16 5

418 I

(16) g (534)

L 709 70 390 (726)

(92)

I (406)

J R ADIUS IN MILES 0-2 0-5 0 -10 ACCUMUL ATIVE TOTAL 101 697 8,183

( ) TOTAL POPULATION FROM O-10 MILES (lH SECTOR)

SOURCE: COUNTY PLANNING DEPARTMEN TS

(,i I

1876 161 I

II-4 site on any given day.

Consequently, the figure of 350 will be used in this report.

(Note:

SMUD indicated that, with the exception of a few people, the entire work force resides outside the study area.)

Ninety percent of the study area consists of farm land and pastureland. Through comparing an estimate of the population in the area against a 1970 U.S. Census estimate that 24.8% of the entire labor force in the Rural Sacramento Community Area are employed as farm laborers, it is estimated that approximately 680 persons are in this I

category. It is also estimated that the number of farm workers commuting into the area is equal to the number commuting out. Therefore, this factor is not continued in the analysis.

The following applies to the labor force within the City of Ione:

o The California Youth Authority's Preston School of Industry has an average of 300 staff members, with approximetely 200 commuting into the area.

I During nights 22 persons are on duty and on weekends approximately 100 persons are on duty. Typical weekday employment is 200.

o The California Department of Forestry's Fire Academy has 32 staff members, I

with 26 of these commuting into the area.

o General retail trade, services, and finance employment in the city is estimated I

to be 150 based upon per capita employment in rural Sacramento County.

Of these it is estimated that 50 are residents of the area and the remainder are from outside the study area. These are exclusive of employees in minerals and manufacturing who would be located south of the city outside the study area.

o AmCal Adult Training Center (AmCal) has seven employees.

The only other workers identified were 19 employees at rest homes throughout 3

the area, and approximately 70 employees at the schools in the area.

I School Populations Exclusive of school employees, who were identified aboee, the five schools (which are discussed in more detail in Chapter III) have a combined enrollment of 1,359 students.

Data provided by each school indicates that 1,230 of the students are residents of the study area and the remaining 129 commute into the area.

I Information from Mr. Raasch, Manager, Generation Engineering, SMUD.

I i876 f62 I

II-5 Recreational Visitors The only recreational facility within the area is the Rancho Seco Recreational Park, located at the Reservoir Area within the site boundaries.

During 1979, the I

Sacramento County Department of Parks and Recreation recorded 170,164 visitors at the facility, with the largest single day attendance being 3,000.

Ilowever, because the facility has only 403 parking spaces, it can accommodate (based on 3 persons per car) approximately 1,209 persons at one time.

In the southeast quadrant, a small tip of Comanche Reservoir extends into the study area. The portion that is in the study area does not contain significant recreational facilities, so that no population is attributable to the evacuation. However, it would be prudent to advise recreational visitors to leave the area in the event of an accident that resulted in a plume in that direction.

I Approximately 8 miles northeast of the site is the Rawhide Nudist Ranch, which is operated seasonally, and normally provides accommodations for 150 guests.

The California Department of Forestry's Fire Academy located in Ione conducts courses from October through May, with an average of 60 students attending each week.

Design Population In analyzing evacuation times, it is necessary to specify the design population or populations to be considered. Since there is no obvious correlation between these I

design populations and accident probability, the accident consideration is eliminated in selecting design populations.

The three most likely design population candidates are typical weekday, peak summer weekend, and nighttime. These estimates are presented in Table 11-1.

From the standpoint of population alone, it is clear that the peak summer weekend has the most people to be evacuated and therefore would be a more likely candidate than the other two. Under adverse weather conditions, the nighttime population provides the worst case.

In the continuing analysis, the peak summer weekend population will be used for the "best" weather estimates, and the nighttime population for " adverse" weather,3

)b7b lb estimates.

• Occasionally Regattas are held at the Reservoir at which time there is substantial overflow parking and peak populations may increase to between 2,000 and 3,000.

Since a large portion of this peak population probably consists of local residents, the maximum parking capacity of 403 vehicles was used as the worst case population to be added to resident population.

II-6 Table 11-1 COMPARISON OF ALTERNATIVE DESIGN POPULATIONS I

Annulus (miles)

Typical Weekday.

0-2 2-5 5-10 Total Residents 67 399 5,015 Non-Residents 134 I

Preston 86 Fire Academy Ione Co=mercial 100 Amcal 7

I Teachers 50 Students 129 SMUD 340 Totals 407 399 5,521 6,327 I

Summer Weekend.

Residents 101 596 7,486 Non-Residents C7 Preston Fire Academy 3

I Ione Commercial 100 SMUD 25 g

Rec. Park.

1,209*

5 Totals 1,335 596 7,656 9,587 I

Nighttime.

I Residents 101 596 7,486 Non-Residents 22 Preston 60 Fire Academy I

SMUD 25 Totals 126 596 7,568 8,290 I

• A substantial percentage of these visitors may come from residential population in the study area.

I III. SPECIAL EVACUATION CONSIDERATIONS Special Facilities There are 17 special facilities in Amador and Sacramento Counties within a 10 mile radius of Rancho Seco. They consist of four public schools (one high school and three elementary); one private elementary school; one pre-school; one treatment center for TB and alcoholic patients; six residential care homes; an adult training center for developmentally disabled; a Fire Academy; the Preston School of Industry; and a nudist ranch.

These facilities, which are listed below with numbers keyed to the map in Figure 11I-1 (A and B), may warrant special consideration in the evacuation analysis.

Schools 1.

Dillard Elementary (public) is located approximately 8 miles northwest of Rancho Seco. it has an enrollment of 327 students from Kindergarten to the 6th grade. It is in the Elk Grove Unified School District District, with student transportation being provided by the District's Garage in Elk Grove.

Buses make runs for Kindergarten in the morning as well as to Dillard School. According to the School District Transportation Manager, if evac-I untion is required during school hours, school buses could leave the District's Garage and evacuate the school i1 20 minutes under the best conditions, and 30 minutes under the worst conditions. In a worst case situation, buses could be making runs and only two would be available to move Dillard School students.

Ilowever, buses are equipped with two-way radios, so that if evacuation were ordered, certain runs could be cancelled and priority given to Dillard School 2.

Arcohe Elementary (public) is located 7-1/2 miles southwest of Rancho Seco and has an enrollment of 278 students from Kindergarten to the 8th grade.

The school, which is in the Arcohe Union Elementary District, has its own buses, located on the premises, and would have the capability to evacuate M

without any outside assistance.

3.

Wilton Christian (private) is located 8 miles northwest of Rancho Seco and has an enrollment of 100 students from pre-school to the 12th grade. The school has an adequate number of buses on the premises to meet its own needs.

1876 165 g

I

III-2 I

Ficure III-1A SPECIA1. FACTI.ITIES - GENERAI,1 OCATTONS k '*

4r'

%?

=

y l

!Y

-iJ 3

Q rl Vf, li.

ee I

m p

F N

J 1

J s

g

+

I l

bI U

I f,,,

w j

O w.,....

s hp '

q' e f.-

(

y e

l

/v

/

.s ;=R..! " y}

5

+

i e

m:

\\.

I

..l._.-.-.

__. $?.891'f L. _.

.u..._._......_

/

's

(,'

l

• "r I

f i

t ec 1

e)

\\

l e

I g

s=u.

g

_y...~.l.. - n )

9 u.

/

1 j.._ _

\\.y.4:. -.

I

r..

i:

s w

a

\\.,

!y j; -

.. x, 3

i

~

s

.' E

)'.

,,, } _ _ o/

\\.,

3 I

4 h

' \\

/'

[

g, a

./

5

. f p

};

c a

T l

~

)

I i 93 _ i..-....

g I

s o

i e

I

/

'. i T

O O

j I

s

, 'i f

]

N

=-

. f f

l q'.

o' :~~ '

l C;

y T

/

i l i '.

G/

M i

g

)

N =

"9 i M/[d I

i N

/

3 i

J

\\

g I

5 N

s a

I

~

Il j'

"*'"** \\

7 ',,,

\\ l !)'

}

(<

it

.i f t r.n i r., q / 'i, f,

O

. (,.-n d, lii

! !i' !

' i 1

t-u 3*.

3 l

u,,

L.,

a m-

/\\

r,n I

1876 166

III-3 I

Figure III-1B SPECIAL FACILITIES - IONE AREA B

N"

/ i/

l

.e,

l/

P#flT0se SCo* COL

'f 0F issou$fev a.-- ]

I

)

oe3l o

o (fY

f N>

-~

I l

ra e

ff Y

ll k'

l v

l

~

s 5

)

&\\

a'

.a n\\ lone

\\

s'

\\.,,c;!!'

-L~

a N. t,1 91

+f i

,\\

'{ St = YQ..

b,m*

Q J/' #ff ig P ^

\\[

i I

N**,' N ik cd'h " ' o

[.3!

d l

om

$,3

-g/

' \\_./ A-j,i

'k

.. )......q

,j :

\\

i....,,

...c..

),f I

j g;

\\

L1

...m i

il s

.T 1

\\

//

x--

a I

....._ a' i."

Source: Center for Planning and Research, Inc.

18~/6 167

III-4 4.

Ione Elementary (public) is located approximately 10. /2 miles cast of Rancho Seco in the City of Ione and has an enrollment of 447 students from Kindergarten to the 8th grade.

It is in the Ione Unified Schcol District with transportation being provided by the district.

5.

Ione Iligh (public) is also located in Ione in the Ione Unified School District I

with transportation being provided by the district. It has an enrollment of 207 students from the 9th to the 12th grades.

The Ione School District's buses are located at the fligh School when not in use. Because of the lack of a sufficient number of buses, the District will not be able to evacuate all students simultaneously. Ilowever, since virtually all of the students are from Ione and the schools are on the periphery of the risk area, it is assumed that most students would be picked ep by parents i

evacuating the area. Only the remainder would require bus evacuation. No special considerations are necessary in this case.

6 M&M Pre-School and Day Care Center is located at lone and has an enrollment I

of 20 children with transportation being provided by parents.

This could present a problem for working parents. Since neither public transportation or nearby shelter is available, expedient means must be used to evacuate.

Two alternatives were considered: (a) the small number of children could be evacuated i, private autos of the staff or nearby residents; and (b) an arrangement could be made with the school district to take the children by bus.

Rest llomes The following facilities provide residential care to persons who are either eldtrly, retarded, or mentally disturbed.

All six facilities have the transportation capability I

to move persors, in one trip, at any time. Listed are their names, number of patients and staff, and number of miles and direction from Rancho Seco.

7.

Edna Mullin's Twin Oaks Rest ilome: six patients and three staff residents, and located 8 miles northwest.

8.

Bufkin Residential Care Ilome: six patients and four staff members, and located 7 miles northwest.

l i8~/6;168 I

III-5 9.

James Board and Care Ilome:

12 patients and three staff members, and located 9 miles southwest.

10.

Dorothy Keesce Rest IIome: 15 patients and two resident staff members, ar>1 located 6 miles southwest.

11. Johnson Guest Rancl.:

15 patients and five resident staff members, and located 10 miles southwest.

12.

Leals Guest Ilome:

15 patients and one resident and one daytime staff member, and located 6 miles southwest.

Others 13.

Preston School of Industry: This facility is operated by the Califernia Youth Authority as a medium security facility, handling male inmates in the 17-24 age group. Of the inmate population of 500, approximately 275 are considered to be hard core, referred to Preston from other facilities. The facility has a daytime, weekday (Monday through Friday) staff of 200, with the daytime, weekend staff being reduced to approximately 100. Staff during night hours are further reduced to 22.

The facility's transportation consists of four passenger cars and vans and one 50 passenger (not caged) bus. It would not be feasible to evacuate the inmates and required staff personnel, because of the extcnsive security required and because of the lack of adequate transportation. (Note: Approximately 10 caged buses would be required for movement.) Consequently, since buildings on the facility afford an acceptable degree of protection against radioactive 'allout, the option would be to have the inmates and required security personnel seek shelter in the dining hall which has 825 spaces with a protection factor in excess of 40. (Shelter countermeasures are discussed in Chapter VII.)

14.

California Department of Forestry Fire Academy is located on the western edge of Ione.

It conducts courses from October to May with an average of 60 students attending each week and has a staff of 32 persons. Students provide their own transportation and pose no poblem in the evacuation.

.The protection factors noted ra the tex. ef >

the value computed for structures surveyed under the National Shelter Sin, ay,, j am conducted by the U.S. Federal Emergency Management Agency. The ned shieldig or attenuation factor that these structures provide against reactor release of fission pecducts would probably be 10 to 20% lower.

I 1876 169 I

s III-6 15.

Altua Village is a treatment center for TB cod alcoholic patients and is locatM 6 miles southwest of Rancho Seco in the unincorporated community of flerald.

It provias treatment for 55 resident patients, has a resident staff of nine employees, and has a sufficient number of vehicles and drivers to evacuate all persons in one trip.

I Rawhide Nudist Ranch is a seasonal facility (mostly summer months) located 16.

8 miles northwest of Rancho Seco.

It has a capacity of 150 guests, who are accommodated in 90 mobile homes, and a staff of five persons. Because guests normally arrive by private vehicles, it can be assumed that ther will be no transportation problem if evacuation is ordered.

17.

AmCal Adult Training Center is located in Ione and provides services to 36 developmentally disabled adults from Amador, Calaveras, and San Joaquin Counties. It is a day-care only facility staffed by seven persons. Trans-g portation is provided by Amador County Transit. Since these buses are not d

generally available in the area, the most effective solution would be an arrangement with Preston School of Industry to use their bus to evacuate the Training Center population.

Transportation Constraints Most of the special facilities have an adequate transportation capability, with the exception of the Preston School of Industry.

Staff transportation would present no problem, but the movement of 500 inmates (particularly the 275 hard core) would present a major problem. If the inmates and staff members were required to evacuhte, then at least ten 50-passenger, caged buses would be required. The only other option to movement would be to require everyone to remain "in-place".

This latter option will be discussed in Chapter VII (Effectiveness of Alternative Protective Actions).

The only special facility in the area that had evacuation problems that would require more time to evacuate than the general public was the Preston School of Industry. The finding fer Preston was that shciter was a more appropriate counter-measure than evacuation.

The only other special facilities that presented special problems were AmCal and H&H Pre-School & Day Care Center.

In both caes the problem was lack of available transportation vehicles, and there were expedient measures that could be employed to obtain the necessary vehicles. Since these special facilities are or should be on telephone alert lists, they will probably begin their evacuation before most of the general public are r. m alerted. Therefore, it can be concluded that all special facilities (other than Pres n) can be evacuated in less time than it takes to evacuate the general public.

1 876

,O I/

III-7 Adverse Weather Impact Two types of adverse weather, radiational fog and heavy rainfall, prevail in the evacuation area. The more common of the two is radiational fog, referred to locally as "Tule Fog". These fogs occur in the area between Novemoer and March. Historically the number of hours per year that the fog condition obtains varies from 150 hours0.00174 days <br />0.0417 hours <br />2.480159e-4 weeks <br />5.7075e-5 months <br /> to 7

1100 hours0.0127 days <br />0.306 hours <br />0.00182 weeks <br />4.1855e-4 months <br />. The fog is very heavy and hugs the ground with a tenacity that often reduces visibility to a few feet. It is particularly hazardous near rivers, swamps, and low lying areas which are prevalent in the Rancho Seco area. Should a reactor accident requiring evacuation occur during a period of heavy fog, it could affect the procedures for alerting the evacuees, reduce the capacity of the local roads, and increase the probability of traffic accidents.

Rainfall in the vicinity of the reactor site averages 14.5 inches per year. The maximum precipitation in any one month was 12.64 inches In December 1955. Maximum 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> rainfall was 5.59 inches in October 1962. Heavy rainfall during the evacuation could have the impact of reducing both travel speed and capacity; it also increases the probability of traffic accidents. Heavy rainfall and/or high winds can also preclude use of helicyters for alerting residents of the need to evacuate.

The most severe impact associated with heavy rainfall is from flooding. During the period of this study the Rancho Seco area experienced nine consecutive days of rein which produced isolated instances of flooding on local roads in the area. This rainfall period was the worst in terms of runoff since the major California flood of 1964/65. The water on the roadways was not deep enough to close the roads, but did restrict average speeds. The reactor site is not in a flood zone, but much of SR 104, I

the major highway in the area, and many of the lower volume roads in the area are in the 100 year flood zone, as indicated in Figure III-2.

Should a reactor accident requiring evacuation occur during a 100 year storm, most of the evacuation routes would be flooded and impassable.

It should be noted, however, that if a 100 year flood werc to occur in the Rancho Seco area, much of the populated area around the reactor site would already be evacuated because of rising flood waters and therefore would not be in further jeopardy from the potential reactor accident.

Of the two adverse weather situations postulated, it is our opinion that the fog condition would have the most severe impact on route capacity; however, the impact on evacuation would be the same for heavy rain and fog.

i8/6;171

III-8 Figure III-2 y m@m) [G ; c a ;ql:. ], i q q.

pp

, ' U

.Q '

100 Year Flood Zone :p,

J wh

'r 14 m

,3

-v g,.

..g

' 's " lf'A h

A=' ~ :i I

,.... g, J $p, w

N e*

I q

r

~"

d$i, 4,,"c'""' A. N G --

g 8

x

. 'm

=

g l\\'

A,

/

J. J I

'\\

A L; h..

,, e,

=

x

=c i

g

. ~,,,'5~ i'%

.1 I...\\ ';-.'.

, p~ 4, l

o 2

,y

, y

,~

o 3'1"4

).x,. /

• I

~

a.' dss?..,, 7" ' s s

3 g.:

I ni. _, ::=~

a-a

', / 2

.' :.M

' ' ~

,s,,...,.

3 i

1, e ~... ;

. s I#

l, q,

~":

g l

6 I

, L-

,2

-I

.p-

.y!:;:p 3

_. i.

a I

i-

\\'.

j

':::Y.*

.. 5 i

.s

,., (

W' w

J:l C=,_y>

3

,ican '-gy

_ j.

v t..

y i

g,

.f "2

anscso seco fg{,,

4 ~ ;w Sh

.. j s.

s 4

2,j, 'f{.., g 67;M

-.g gE O

[ $,,F"

~ ~ @M

~

'~'

3

\\

g..f g

i g

[f

'M M ' 7~

.,h~ [, [

\\, y /

W

~

',j

• s n\\

?

n

~

-a

%$4. e

\\

.a j

fp -..'s.

. '. no

,. t y

p y

3

.-g

s.. 11a s

g 3

e g

g-

-T..,.

\\s.

u y.

D-w

.1 p-

. jl.N 4i _ g.,L,-L__ A,

l., T< t, I

.,g',

i h

"w a

t

. c,

.,e.

3 m

s

. -- 3 g

i E

-]

,[ l' u

=

=

s,,,,,,

~

, ':. T.

pf ~._

%,=

~

~ ~ ~

m~ /.J s.

.p -

, s\\

. I '._ _ '._%,..,p ul /-

s,i I

,.i..

. "f2

= -. ;,.,....,....

--

=~

l Basic map reproduced by permission of the California State Automobile Association, copyright owner.

Source: Center for Planning and Research, Inc., from material supplied by the National Flood Insurance Program, U.S. Housing and Urban Development Department.

g 1876 172

IV. EVACUATION ROUTES Local Plans The Sacramento County Emergency Response plan identifies evacuation routes for all of the sectors within 5 miles (low population density zone) of Rancho Seco.

These routes were used to calculate travel tirr.es for the evacuation. Beyond the five mile radius CPR selected the evacuation routes based upon the shortest distance to exit the impacted area. Planning is underway in the counties for the 10 mile radius case.

I Road Network Figure IV-1 indicates the principal paved roads in the study area.

Numerous unimproved dirt roads exist throughout the area, particularly in the southeast quadrant.

Many of the isolated farm houses in that quadrant would be expected to use these dirt roads to exit the area since travel on paved roads is substantially farther and takes the evacuee back toward the reactor site. The most complete inventory of all Ivads was obtained from the U.S.G.S. quadrangle maps which were updated in 1975.

I Routes by Segment For each segment shown in Figure II-1 the centroid of housing was established and the outbound routes were located. These routes were then measured to determine the distance to the boundary of the risk area. Measurements began in the innermost radius and were cumulated for the next two radii to reach the 10 mile boundary.

Distances were calculated to the radius boundary except in the case of Ione. Since Ione straddles the 10 mile radius, the distance in that segment was calculated to one mile cast of the outermost city boundary.

Analysis of the remaining movement to final reception aceas was not within the scope of this study.

Capacity Analysis Data for the capacity analysis was obtained from the California Department of Transportation (CALTR ANS) "1975 California State Highway Log". All of the improved roads in the area are two-lane sections with traveled way widths that vary from 8 to 16 feet. Major portions of all roads are without treated outside shoulders.

The capacity calculations used the techniques contained in the " Highway Capacity Manual" published in 1965 by the Highway Research Board as Special Report 87.

Capacity is the maximum traffic flow under prevailing conditions. Traffic flow, or 1876 U3 I

IV-2

)

/5' Figure IV-1 tr'i.,

!is; -

f":p&,(- 9, m, <Tj e

q-

^

pr I

- i)a

. \\

PRINCIPAL PAVED ROADS IN STUDY AREA

" M lt'!b 0 b #Mt 2

\\

5

.=*r 2s # '

e s 9)}

E1

• yQ. '

I,r f 4 1.

n g, "",* L 1==

i-.,

~t f#

pd

,, T - i 75 4

""_ t,'

X

+

e g

I

,~

I d i1. u! \\',

g

~, L.d r~g-

[ f N$EY N

i a

ri i-5=e==eN

/

7 6, 8c !.

I 2i,,,,'\\D ! N'g 5

j 2

e I?

t a.

N p

N '-.

I h k;\\

i# "

3 f

}

A f

r '-

^ '. ^

\\

m4 y

I

=

so.. [.

I 4k e

• 2

.- =r

...n, 5 hrove f

, _j.

t2 i

2 E

L' ~l'

"'lE~ T

\\

i

\\

g A

• ,e,-

i

/

Ne..s Its g

I N

4,.

ra.P 3

I?

j'

,.g

. -. = '

=

s s

1

,~

Nj, 3 7,,

5

~=

a l

\\

. lone

/

{

-5 5

J t,g i

m.-

2, g '

u tr

../

.f'j

\\ '

=k RANCHO SECO 3 g*

C_

3 s

>s S

,s 1,

,>=m.

e"<

,s, 1, 3

'b,,

y.Q Az

,. e 9.\\

.A l

m,

x. v n)

[

9

,e.

l I,J' 8

.s =

, -, u _. i

. 6 s

/_ff

/

53 r.,,d, !7-

? i i,

-~

= \\ \\ r h,1. __ F J

\\ g,f * *' ( #

ie sa 1 N

p

,yc."</j/V&y' a

- a

,, ~

n w.

s

,gE x t.;.m,

' e. ~ g,. r,,

g m

f I

,~: _=",.; Y -

g[ s

-l

"".?' 9' r

>~C^-

h:p2T i.

,,; /

f k 'I '" \\

,V

~,

j 2

ss

.. x e

7.

t F '- ~

t _.__J t 'i smW

\\

...w,

• 5 d

a

.'7 u ~

s g'

!h!

,\\. l.'n - ! i

. g. 2 c,.';'". ", C -

i...

L o

[v-v 6

e

,aa.,,

i nt y\\

n i

i,

).

--3

,s-s, '

~:

\\

'.'.p )<;, q.i.,lI i

is.,,

gl,,L

}, -

!(

3 pm s

I g

.g_

g

-v i,

i

w. eno, i

2s g

n.

.,s

=

=

s, ' g..im -

3 n

-[

p I

s.s,

.s SCALE.

,7- ; T=;- ~-. T--f,--, ',5 _,,,,

m.-.

%a Basic map reproduced by permission of the California State Automobile Association, copyright owner.

IV-3 the volume of automobiles that can pass a point in an increment of time, is a function of speed and density.

The maximum flow is normally considered to be 2,000 vehicles / hour /lanc (vph/ lane). This generally occurs at a speed of 30 mph and a density of 66 vehicles / mile.

Variations from either the speed or the density result in a

[

reduction in flow rate. On a two-lane road the maximum flow rate is 2,000 vehicles / hour a

in both directions. If there are few opposing vehicles, the traffic can maintain optimum density by immediately passing to fill gaps that occur, and the 2,000 vph/ lane is reached I

in one direction.

However, as opposing traffic increases, the opportunity to pass is decreased; the traffic stecam breaks up and the gaps that occur are not immediately filled, which reduces density. Experience has shown that the 2,000 vph maximum flow rate for a two-lane road cannot be increased regardless of the distribution between lanes.

In the case of Rancho Seco the capacity was calculated for SR 104 and 88 based upon clear weather and the vehicle mix that would be anticipated during the evacuation. The capacity of SR 104 would be 1,300 vehicles per hour. The capacity of SR 88 would be 1,700 vehicles per hour. The number of vehicles in the evacuation I

usumed 0.64 vehicles per person for residential population; 0.33 vehic%s per person for recreational visitors; and one vehicle per person for workers. The vehicle mix was assuined to be the same as the mix of vehicles registered in Sacramento County. The maximum number of vehicles that would use SR 88 in the evacuation is 60.

The maximum number of vehicles that would use SR 104 is 900. In neither case will the capacity of the roads be approached.

The only other road with a substantial volume of traffic is Dillard Road traveling north from the site.

This road would carry approximately 850 vehicles.

Under the clear weather assumptions, these people would be evacuating over a period of at least an hour. This volume of traffic, if it occurs in one hour, would approach the capacity of the road. However, since there are numerous alternative routes in the vicinity of I

Dillard, it is unlikely that capacity would actually constrain the movement of evacuees from that quadrant.

Adverse Weather Impact There is limited prior research on the impact of adverse weather on capacity except in the case of snow and ice, neither of which prevails in the area. A study in Texas, based upon obscured conditions, indicates that capacity is reduced to between l

1876 175

IV-4 81 and 86% of normal capacity.

A theoretical analysis of environmental hazard (weather) suggests that severe fog may reduce capacity to 60% of normal."

The study also suggests that average speed would be reduced to 40% of normal.

The impact of heavy fog in auto streams is multi-faceted.

The decreased visibility results in reduced speeds often as low as 10 or 15 mph. The distance between vehicle pairs is quite small, generally on the order of 1 or 2 car lengths, but the gaps that develop as the vehicle stream breaks up into platoons tend to increase which reduces average density below the optimum of 66 vehicles per mile for maximum flow.

The increase in size and frequency of these gaps is fostered by the inability to pass in a fog where passing sight distance is only a few car lengths. The average density that prevails on a two-lane road operating at maximum capacity and evenly loaded in both directions is 30-35 vehicles per mile.

Under heavy fog conditions this lower density would probably prevail because the opportunity to pass is reduced to almost zero. A density of 30 vehicles per mile and a speed of 15 mph results in a maximum flow rate of 450 vph.

While heavy fog conditions could reduce the capacity of the roads from 1200 or 1300 vph to 450 vph, there are compensating factors that must also be considered.

The primary compensating factor is that the Tule fog is produced under calm wind conditions.

The plume that would be the cause for the evacuation would therefore move very slowly. These low wind conditions would result in a longer time to conduct the evacuation and/or would limit the area requiring evacuation. An analysis was done of the impact of fog on movement out of the five mile zone around the site. The maximum volume of traffic in this case is 200 vehicles on SR 104 westbound. This is well within the capacity of the road.

If it were necessary to evacuate out to the 10 mile distance, fog conditions I

would be a capacity constraint. The roads affected would be Clay Station /Elliot, SR 104, Dillard Road and Wilton.

These roads will be discussed further in Chapter V (Movement Analysis).

s e

i E. Ray Jones, et al., " Environmental Influence of Rain on Freeway Capacity", liighway Research Record 321, 1970.

William P. Walker, " Speed and Travel Time Measurement in Urban Areas", Ifighway Research Board Bulletin 156, 1957.

1876 176

I V.

MOVEMENT ANALYSIS I

Sector Sc1cetion Until this point in the study, all of the data analysis has been conducted using sixteen 22-1/2 degree sectors that are identified by letters A through R (excluding the letters I and O), as illustrated in Figure II-1. The NRC letter requires the evacuation I

times to be estimated for 180 and 90 sectors. The methodology for selecting these sectors is described in this section.

The sectors for the evacuation analysis can be selected using different criteria.

One criterion is wind direction. If the desire is to identify the sector combinations with the highest wind probability from the power plant, then a line drawn between sectors E and F provides a scmicircle with 69% wind probability. If the desire is to

~

provide equal wind probabilit!, a line between sectors B and C provides a 51% and 49% wind probability.

The same analysis indicates that the maximium wind probability in any one quadrant is 35%.

A line d." an between the following sections produces a quadrant that has a 35% probability of having wind blowing in that direction across the reactor site: AB, BC, EF, FG.

Another criterion is impact on the greatest number of people. In that case a line between sectors G and II or C and D provides the highest number of people in any one quadrant - 3,674 persons.

A third criterion is the road capacity for evacuation. The road that potentially has the highest volume-to-capacity ratio is State Route 104 east of Rancho Seco. The number of evacuees assigned to this route is maximum when a line is drawn between sectors B-C or F-G.

Since the most critical factor in the evacuation is the number of people to be I

evacuated, the primary dividing line for establishing the semicircle and quadrants for the analysis should be between either G and 11 or C and D.

Since these lines do not meet the wind or traffic capacity criteria, the decision between them will be,.ade on the basis of population alone. The G/II line provides slightly higher populations close in to the site; therefore it is used in the evacuation time analysis. Figure V-1 shows the quadrants used in the movement time analysis and the estimated residential population within each segment.

Figure V-2 shows the correspondence between the 180 and 90 sectors and the original 22-1/2 degree sectors.

1876 T77 I

I

V-2 i

3 POPULATION BY QUADRANT

,\\

.,=.'

I.%IIIII8.!l../

.. = _

t 1

as / g I

i

• i s

3 N

s e.,

es a.1 4 1. --.

I I

'g

%gg~ h... ) M't /, \\

p

!; Dw

...... K y-(, \\f a....s,-

I v,\\.

N c

q,

,=

!\\,

4

/

k

.s

[,

,\\

\\s [.. N' g

S 7

,,,,,r==

y ir% 8 o

>'_s s \\>. umr.r...;- -

g \\s,,

l s

\\

8 t

48

.s g

?

9

.s I

4

,f n

a s

j s

w

% -- br+':'""'&

e' n

D

• • \\,g[i

'd.."

I i

g *',

n

-.\\8,\\

/,14 9. ".

41 0

\\-

iis Ig

.s i

n,.

j 2

i B

x.

/

I

\\

.N

/

5 \\!.s 7 I

.h 3 O

E 14-i i<

391 r

I F.t

_ af l

i

.u, ~ 1 7

4. Q-6 3

i l

r 3

n; q

i,

6

[

I j,"""3196 mANCH ECO I

S 22 7

is a 6'

-l b=h

't ri is t_

o 6

7 y x

s g

e

. h Ki f

,,s I'iI 1587

- ['

Sq

-- t i

_._.,m

~

. q g"$,,,,,,,

,, p c

ga:/-

e_

l w-a e

w.

,a.

m 1 "_ (',nllI g.1]_%.,, W9 "rt.',,,, -

~

l 1'

i i

Il. i s.

>s

)

3

..,,a., J s

i m.=

3

. s==

4l

], '. \\

97/...,'('. If-l

' ' *,, ~

e 1.,

i I

\\'

s\\.

c.(

3 a. n e ;

,, /. _.:

\\,,

g, w-,

lM'-.I i

,,i

.I I

SCALE.

_a 1 Quadrant Number

.E.9 Population ISource:

n,...c,

.. =

Center for Planning and Research, Inc.

Basic map reproduced by permission of the California State Automobile Association, I

opyright owner.

1876 178 C

V-3 FIGURE V-2 CORRESPONDENCE BETWEEN 22 SECTORS AhT) 180 /90 SECTORS Rancho Seco 10 mile Zone I

I A

R 10 mi B

I

\\

l I

/

f O

/

/

C g

4 I

i

/

/

\\

/

/

s

's

Smi,

/

I s

N D

P s

\\

l

/

2 s

s.

i I

i I

/

s 9

'g2mi

~, -

s 7

's

,1-

~ _

y

_ 1 igg g

I

'. - - g/, c, - _3 ' '

. ),

~

,5 e

s

~_ -

I s

\\

N l

l

\\

s S/

s M

f

\\

s p

~

/'

\\

I N

/

6

/

I

\\

/

I I

/

/

\\

10 /

g I

\\

K H

J I

I Source: Center for Planning and Research Inc.

I I

1876 179 I

V-4 Analysis Methodology Determining the time to evacuate an area requires the consideration of at least four parameters or time frames: notification, alert verification, preparation, and travel

time, in order to include all of these parameters in an analysis of the time to evacuate, an estimate must be developed for each parameter. The following sectica I

provides a description of the basis for the estimates used in this study.

Notification of Government Officials The first parameter or time frame begins with the notification of the appropriate warning center by the on-site personnel operating the reactor that an accident is imminent or has occurred and ends with the initial dispatch of warning vehicles. Since the requirement to consider evacuation out ta 10 miles from the site is recent, there are currently no multi-county alerting systems around Rancho Seco for alerting the general public in the case of a reactor accident.

For Sacramento County the local government warning center, in the case of Rancho Seco, is the County Communications Center.

The center in turn notifies the Emergency Alert Officer, who advises the Communications Center to continue alerting the remainder of agencies on the standard warning list. When alerted the Law Enforcement Service then dispatches patrol units and helicopters to the scene to man road blocks and begin the notification of people in the area to be evacuated.

The notification from the reactor site to the communications Center, Alert Officer, and Law Enforcement Service (Sheriff's Department) can be accomplished in a matter of minutes.

The notification process can be separated into the time for physical actions and the decision making time.

The physical actions necessary to propagate the notification are controlled by the time necessary to establish communi-I cations.

Whether by telephone or radio, the time for contacting the appropriate officials would probably be less than a minute for each communication. The actual decision to evacuate, howaver, has serious consequences to both government and population in the area and is not likely to be an automatic response.

The Sacramento County plan states that the Emergency Alert Officer will advise the Communications Center dispatcher on how to proceed with the notification process.

This decision to evacuate may be delayed in order to obtain additional information on the reactor incident or to contact other officials who may have a role in the evacuation decision. For example, CPR staff recently visited the site of the evacuation of 200,000 people from a chlorine spill in Mississauga, Ontario, Canada.

In that case a train l

1876 180 I

V-5 accident occurred just before nidnight on Saturday of 1 long weekend. The notification of officials and the time necessary for those officials to decide to initiate evacuation actions took approximately one hour. This initial decision affected only a few thousand I

people that were in the immediate vicinity of the ruptured chlorine tank car. As the gas dispersed, the evacuation area was expanded until it covered an area that contained 250,000 people.

Historical data for evacuation from natural disasters and false alerts in the 1950's were reviewed and the data indicated that notification and decision times varied from an hour or two to many hours.

In the case of Rancho Seco, the awareness of the possibility of an accident requiring an evacuation is quite high among local public officials. In addition a great deal of planning has taken place that should result in an expeditious decision process leading to the notification to evacuate.

For lack of a I

more analytical basis (such as exercises), it is assumed that the time from notification by on-site operating staff to notification of sheriff's dispatcher to begin the evacuation alert would be on the order of 30 minutes.

A theoretical or ideal time for notification can be postulated based upon the communications time, assuming that no delay for the actual decision process is necessary.

This ideal time would be in the order of five rainutes.

The shortest time that can actually be expected would be dictated by the conditions in each of the three counties involved and possibly the role played by the State. It would probably vary between the 5 minutes cited above and the 30 minutes assumed for alerting under today's I

conditions.

I Alerting of General Population The second parameter is the alerting of the general population to the need to evacuate. The current local government response plans do not provide for assignment of alert functions in a manner specific enough that they can be analyzed to determine their effectiveness. There is no specific reference to using patrol cars with loudspeakers for alerting.

There is a reference to the use of the helicopter with PA system but no indication as to its priority assignment to this function. CPR staff were informed that some tests or exercises have been conducted with the helicopter for alerting, but it was not possible to verify this informati= No sirens or other rapid alerting systems I

exist in the area.

Regardless of how the alerting takes place, the people being warned must go through the following process:

1876 181

4%

V-6 1.

Hear signal i

2.

Recognize signal 3.

Seek confirmation of signal meaning and validity 4.

Find confirmation of signal meaning and validity 5.

Relate signal meaning to self 6.

Decide to act.

The alert and verification phase requires different times for each element of the general population. The time distribution for this function has been studied over the years as part of the federal civil defense program.

The distribution of time can 2

-t where f(t) is a probability be described by a relationship of the form: f(t) = a te a density distributing function. The general shape of this distribution is shown in Figure V-3.

The t is the time since the alert process began and a is a constant that determines the shape of the distribution.

The constant, a, has been calculated for a number of different alerting systems.

The most efficient alerting system is one that can be activated remotely from a central point with all points being alerted simultaneously.

An example would be a radio controlled device that turns on the radio or TV and tells the listener what has happened and what to do.

The value for a in this case would be 1.

A no alert system case would be one in which the population receives news of the disaster by word of mouth or from radio or TV which happened to be on at the time of the Based upon research in natural disasters such as floods or toxic gas releases, disaster.

for such conditions during the night has been calculated as 0.0275. A siren the a of 0.17 and an alert only radio that does not also provide alerting system has an a

a message has an a of 0.5.

The lower the value the longer it takes to complete the alert verification phase.

Since there is no formal alerting system in the Rancho Seco area, the use of patrol cars or a helicopter would be the most likely method to alert the population.

g m

The estimate of the time necessary to accomplish this alerting was based upon the road networks that would have to be driven or flown over to alert the population.

The assumptions used in this analysis were that the patrol cars would travel at an average speed of 12 mph while alerting the population.

In addition, a penalty time to reach the area to be alerted was applied. These were 15 minutes for a patrol car

.Robert A. Harker, Richard L. Goen, and Kendall D. Moll, "A Method for Evaluating Local Civil Defense Effectiveness", Stanford Research Institute for the Institute for I

Defense Analysis, Oct. 19G4.

A. E. Moon, " Population in Shelter, A Method of Measuring Effectiveness of Radio Warning", Stanford Research Institute, Nov.1965.

1876 182 l

t e

V-7 Figure V-3 GENERAL SHAPE OF ALERT VERIFICATION DISTRIBUTION I

l

.5-

.4-ALERT VERIFICATION TIME I

DISTRIBUTION ( = = 1.0)

.3_

l

-6 Yg 2_

E O

I i

E

.1 0;

~

I i:

I Bo 0

1 i

i w

1 I

I O

2 4

6 8

10 12 I

TIME - minutes Source: Center for Planning and Research, Inc.

I 1876.183 I

I

V-8 to reach the 2 mile radius,10 minutes to reach the 5 mile radius, and no time to reach the 10 mile radius. For the helicopter the assumption was that it would travel at 30 mph once it reached the evacuation area. It would take 15 minutes to reach the 2 mile radius and 10 minutes to reach any other point inside the 10 mile radius.

It was further assumed that only one helicopter would be used for alerting. The assumptions for oatrol car alerting were: only one car would be required within the 2 mile radius, twv cars within the 5 mile radius, and six cars within the 10 mile radius.

These values are cumulative - that is, if an entire sector out to 10 miles were evacuated, the maximum number of patrol cars used would be six.

In calculating the time for alerting, it was assumed that the patrol cars would travel approximately twice the road distance in the sector. This assumption compensates for the doubling back that would be required because of the large number of houses that are at the end of roads which do not provide for through circulation.

In the case of the helicopter, a factor of 1.5 times the road mileage was used.

For purposes of these calculations, this alerting process was assumed to have the same type of distribution as the alerting systems discussed earlier.

This is an approximation since those systems had the capability of simultaneously alerting, or at least alerting from multiple sources in the case of word of mouth alerting in the natural disaster experience.

The a developed by CPR for this study ranged from 0.27 to 0.0275 for alerting i

under today's conditions.

The values for a were calculated based upon the number and type of vehicles assigned to alerting as noted above, and the distances the alerting units must travel.

Movement Preparation Time The third parameter is movement preparation time.

This will include all of the actions the person must take between the time that he decides to act and the time he departs his location and begins to travel out of the area. These preparations include such actions as shutting off utilities, packing bags, deciding on the destination I

These assumptions were based upon existing plans in Sacramento County for alerting out to five miles.

If additional vehicles were assigned to the alert functions, they could be carried out in a shorter time frame.

se used in the analysis was 0.275 which corresponds to a word-of-mouth The smallest a alerting.

I i876 184

V-9 and routes to be traveled in evacuating, taking care of livestock, collecting other family members, loading the automobile, etc.

The movement preparation time is also a distributed event which varies from household to household. Research conducted by Operations Research, Inc. on nighttime preparation time indicates that the distribution is similar to that for alerting time.

~6 The formula is:

f(t) = 6 te The value for s has been computed previously as 0.41.

The assumptions relative to that value are considered appropriate for I

nighttime conditions in this study.

The movement preparation time, when combined with the alerting time, results in a distribution of time that elements of the population would be starting to travel out of the area.

It should be noted that these two distributions cannot be simply added together but must be combined as probabilities which require rather sophisticated mathematical treatment or a laborious numerical analysis.

Figure V-4 illustrates the two distributions (alerting and preparation time) and the combined starting time distribution for radio warning at night. Figure V-5 shows the cumulative time curve resulting from these two distributions.

I The NRC letter requesting this analysis indicated that an ideal " notification" time (in this case the NRC's use of notification time corresponds to our alert /ver-nication/ preparation time) would be 15 minutes. Our analysis suggests that an ideal or best possible alerting / verification / preparation time would be a distribution that has a maximum time of 20 minutes.

Travel Time The fourth parameter is movement time. Movement time is a function of the road distance to the boundary of the evacuation area, vehicle used for evacuation, and auto tra ffic conditions (traffic volumes, road capacity, weather conditions).

The travel times for the evacuation were calculated on the basis of the population I

distribution within each segment and the shortest road distance to the boundary of the evacuation area. The centroid of population was estimated using U.S.G.S. maps which indicated the location of each structure in the area. A " map wheel" was used to calculate the road distance. Judgment was exercised in a few cases to decide whether the evacuation route would be a shorter dirt road or a larger paved road.

l

.Whenever the calculated values for Alerting / Preparation time resulted in values greater than 200 minutes, a 200 minute default value v as substituted.

I

V-10 I

I Figure V-4 DISTRIBUTION OF STARTING TIMES WITH RADIO WARNING I

.5 4

I ALERT VERIFICATION TIME

1. y DISTRIBUTION (v = 1.0 )

I 1

y I

e.

UZ 3 ~2 MOVEMENT PREPARATION' I

' TIME DISTRIBUTION (.8=.41 )

5 STARTING TIME y

DISTRIBUTION a.i I

~

y s

x NN a

xx I

Z l

f 0

0 2

4 6

8 10 12 TIME - minutes I

A.E. Moon, " Population in Shelter, A Method of Measuring the Source:

Effectiveness of Radio Warning," Stanford Research Institute, Nov. 1965.

• I 1876 186 I

I

A-TI I

I I

m 5-H

~

Oz T

H n

y v

~w 0

H O

h

~

E H n "G~

~

Y W

3 C

m o

- E, W

W W

C W

C m

O M

3 m

w e

C m

s O

U H

N p

2 v

e r 7

O D

D o

4 W

U oc

~

g y

B ca

$9 H

M y

O v

m m

I 3

0(

O m W cn u

M C

O m

l I

1 1

1

=

=

s I

aerA ksiR fo tuO levarT ot gnitratS noitalupoP fo tnecreP I

l9/9 181 I

I

V-12 It was assumed that the average speed for vehicles leaving the area varied from 20 mph to 40 mph. While the traffic volumes in the evacuation area are generally low, the traffic is traveling on secondary roads, access roads, and in a few cases, dirt roads. It is expected that average speeds on SR 104 and 88 will be on the order of 40 to 50 mph.

The scope of work for this project required that evacuation times be calculated for complete evacuation of the area. Since the time required to get 100% of the population out of the risk area is always controlled by the last few vehicles, the various travel times do not appear to be sensitive to the variables considered in the study.

If, however, one were to examine the distribution of people departing the area, a greater sensitivity is shown. To illustrate this, two conditions were calculated.

One was the use of an ideal radio alerting system. The other was the time necessary for 50% of the population to evacuate the area.

Table V-1 indicates the estimated evacuation times for today's conditions. Table I

V-2 provides similar information for the time necessary to evacuate 50% of the population. A method of comparing these times with a theoretical ideal evacuation time is given in the footnotes to the table.

Adverse Weather Impacts on Evacuation Time As was discussed in Chapters III and IV, adverse weather will impact the evacuation times.

In the most severe case, a hundred year flood would interdict most of the major evacuation routes. While almost one-third of the population would be displaced by flood waters, the remaining two-thirds would have problems in I

evacuating from a reactor accident. Table V-3 indicates the approximate percentage of population in each segment that could be isolated or trapped in such a condition if they had not already evacuated before the flood waters closed the vulnerable roads.

In the case of heavy rainfall the reduced visibility and more hazardous driving conditions would reduce road capacities to approximately 80% of normal. This capacity constraint would not result in any of the traffic flows exceeding capacity except on Dillard Road. In this case the traffic assigned to Dillard would be 32% greater than its capacity.

As was mentioned earlier, most residents of the area would seek the available alternate routes for evacuation. The impact on average speed could tend to reduce normal speeds by as much as 50% based upon the earlier cited adverse weather studies. IIcavy rain could also impact alerting times. Under adverse weather I

conditions patrol cars would have to be used for alerting rather than the more efficient helicopters. The combined impact of these constraints on evacuation time is shown gJ{

}88 on Table V-4.

I

V-13 Table V-1 EVACUATION TIME FOR RANCHO SECO I

(in minutes)

Alerting &

Notification /

Preparation Travel Totalc Segments Evacuated Decision Time Time Timeb Time I

1 30 35 3

68 1-2 30 35 9

74 1-2-4 30 155 20 205 1-3 30 35 9

74 1-3-5 30 200 20 250 6

30 40 3

73 I

6-7 30 70 9

109 6-7-9 30 200 20 250 6-8 30 55 11 88 6-8-10 30 200 20 250

1. Assumes helicopter alerting of residents during nighttime conditions.

The maximum value for this time phase is 200 minutes which corresponds to word-of-mouth alerting.

I Assumes 20 mph average travel speed for the first 2 miles, 30 mph for 2 to 5 miles, and 40 mph for distances beyond 5 miles.

' A theoretical ideal evacuation time can be estimated by adding an ideal notification and alert time to the travel time shown in the Table.

Assume radio alert system that results in 20 minutes maximum time to alert, verify and prepare to move.

Also assume a notification system that allows the decision to evacuate to be made in 5 minutes.

Source:

Center for Planning and Research, Inc.

I

V-14 I

Table V-2 I

TIME TO EVACUATE 50% OF RESIDENTS AROUND RANCHO SECO (in minutes)

I Alerting 6 a b

C Segments Notification /

Preparation Travel Total Evacuated Decision Time Time Time Time 1

30 11 3

44 1-2 30 12 8

50 1-2-4 30 44 18 92 1-3 30 12 6

48 1-3-5 30 70 10 110 6

30 13 3

46 6-7 30 21 6

57

.6-7-9 30 70 9

109 6-8 30 17 8

55 6-8-10 30 70 6

106 I

" Assunes helicopter alerting of residents during nighttime conditions.

Assumes 20 mph average travel speed for the first 2 miles, 30 mph for 2 to 5 miles, and 40 mph for distances beyond 5 miles.

' A theoretical ideal time for evacuation of 50% of population can be estimated by adding an ideal notification and alert time to the I

travel time.

Assume a radio alert system that results in 6 minutes for 50% of evacuees to alert, verify, and prepare to move.

Also assume notification system that allows the decision to evacuate to I

be made in 5 minutes.

I Source:

Center for Planning & Research, Inc.

I

V-15 Table V-3 ESTIMATED POPULATION WITH POTENTIAL FOR ISOLATION BY 100 YEAR FLOOD I

Percent Segment Population Isolated 1

14 0

1-2 55 3C 1-2-4 303 80 1-3 36 40 1

1-3-5 2,491 20 6

87 70 6-7 478 50 6-7-9 3,674 80 6-8 229 70 6-8-10 1,816 10 Source: Center for Planning and Research, Inc.

I I

I I

.a I

I 1876 191 g

I

V-16 I

Table V-4 EVACUATION TIME FOR RANCH 0 SECO FOR HEAVY RAIN I

(in minutes) a Alerting L Notification /

Preparation Travel Total Segments Evacuated Decision Time Time Ticie _

Time 1

30 40 6

76 1-2 30 40 24 94 1-2-4 30 200 66 296 1-3 30 40 24 82 1-3-5 30 200 66 296 6

30 45 6

81 6-7 30 90 12 132 6-7-9 30 200 64 294 6-8 30 70 28 128 6-8-10 30 200 64 294 I

" Assumes patrol car alerting to residents during nighttime conditions.

Assumes 10 mph average travel speed for first 2 miles, 15 mph for 2 to 5 miles and 20 mph over 5 miles.

I Source:

Center for Planning & Research, Inc.

M76 "

I I

V-17 The most probable adverse weather case is "Tule" fog which has an impact I

similar to heavy rain on average vehicle speeds but a greater impact on highway capacity.

Ilowever, because the severe fog condition only occurs when there are very calm winds, the area that would require evacuation would be limited to within less than five miles of the site. The more severe capacity constraints do not become a factor in this case since traffic volumes within five miles of the site are all less than the resulting road capacities. Therefore, the impact of severe fog on evacuation times would be the same as for heavy rain shown on Table V-4.

F l

I I

I I

I I

i876 193

I VI. EVACUATION VERIFICATION It is extremely difficult to verify that an area has been evacuated. The letter from NRC identified a technique consisting of the placing of a marker outside a house I

by the occupants to indicate they had left. Techniques such as this have a number of inherent weaknesses. One is that residents are likely to forget to place the marker in their haste to evacuate. Some residents, fearful of looting, may not use the marker even if they remember it, in virtually every evacuation, some residents choose to stay behind. The percentage remaining in the evacuated area may be substantial. In flood and hurricane evacuations it is not unusual for 10 or 20% of the residents to stay put.

It would also be reasonable for residents who wished to stay put to place the evacuated marker outside so that they would not be bothered.

Verification Time Using Marker Technique If the marker technique were to be employed, someone would have to enter the hazardous area and sight the houses marked. Homes without markers would have to be inspected to determine if anyone were present. A possible variation of this technique would be to place a large marker, such as a weighted down bedsheet, in front of the house, so that it could be sighted from the air. Table VI-1 indicates estimated time to dr.ive over the area to sight all houses.

The estimates are based upon measured road distances and average driving speed of 15 mph.

I Other Verification Techniques Other pmsible verification techniques include the polling of residences by tele-phone.

The telephone company's cooperation would be required to obtain access to unlisted numbers.

The time to conduct such a poll would be on the order of 1/2 minute per residence provided that the telephone numbers were organized in a sequential fashion.

It may also be possible to have the telephone company monitor telephone calls in the impacted area.

The placing or receiving of a telephone call would be prima facie evidence of occupancy. It would also be likely that anyone staying behind would use the telephone in order to be in contact with the outside world.

I 1876 194

VI-2 I

Table VI-l TIME TO VERIFY EXTENT OF EVACUATION Exterior Marker Technicue I

Total Vehicle Minutes Estimated Number I

Segments To Complete of Houses a

b Evacuated Windshield Survev Per Segment I

1 16 6

1-2 30 24 1-2-4 270 134 1-3 27 16 1-3-5 343 1102 6

22 38 i

6-7 98 211 6-7-9 652 1625 6-8 132 101 6-8-10 760 803 I

I

" No allowance made for leaving vehicle to check unmarked houses.

Times can be reduced by assigning multiple vehicles to verification activities.

Use of a helicopter for verification would reduce these I

times by two-thirds.

b Estimated total number of dwelling units within 10 miles of Rancho Seco is 3,620.

I i876 195 Source: Center for Planning & Research, Inc.

I

I VII. ALTERNATIVE PROTECTIVE ACTIONS To assess the potential effectiveness of alternative protective measures, it is necessary to cor sider the possible dynamics of a rediological emergency and identify the protective measures that might be appropriate for the various stages of the emergency.

I Figure VII-1 presents a hypothetical dose rate curve for a location somewhere downwind of the site. The curve may be interpreted as follows:

I o

t is the time that the notification process begins.

N o

t is the time that an accidental release begins to occur at Rancho Seco, R

assumed to be at least 30 minutes after t A e rding to NUREG 0396, N.

the time between the initiating event and the start of atmospheric release may range from 0.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> to one day.

o aT is the period of time between t and the time that the leading I

1 N

edge of the cloud reaches and begins to affect the location of interest.

o AT is the time required for the cloud to pass the location. During this 2

period the dose rate would increase from background, pass through a peak, and then decrease.

o AT is the period of time that begins when the trailing edge of the 3

cloud passes the lecation and ends when residual hazard from radioactive materials deposited in the location is no longer significant.

Protective Actions Prior to Cloud Arrival During the period A T, the central question will be whether it is possible to 1

evacuate the population from downwind locations, prior to their being affected - i.e.,

prior to the arrival of the cloud of radioactive material. Table VII-1 A indicates cloud arrival times for different effective wind speeds.

For comparison purposes, Table VII-1B summarizes the estimated evacuation times previously presented in Table V-1.

By comparing the data in Tables VII-1 A and IB, it is evident that in order to complete 1876 196

I VII-2 I

Figure VII-l I

IIYPOTilETICAL DOSE RATE IIISTORY CURVE I

E I

I I

I i

c>

i g

f'-

l i

tN........tR AT AT AT3 I

I Source: Center for Planning and Research Inc.

I I

1876 197 I

VII-3 Table VII-1A CLOUD ARRIVAL TIMES *

(minutes including notification time)

Effective Locations Locations Locations I

Wind 1.0 miles 3.5 miles 7.5 miles Speed Downwind Downwind Downwind 2 m pli 60 135 255 5 mph 42 72 120 10 mph 36 51 75 20 mph 33 41 53 I

• Assumes release starts 30 minutes after decision to start notification.

Locations chosen were the mid points of each segment.

Table VII-1B SIBNARY OF EVACUATION TIMES *

(minutes including notificaticn time)

Locations Locations Locations I

0-2 miles 0-5 miles 5-10 miles Downwind Downwind Downwind I

Existing Conditions 68-73 74-109 205-250 Ideal Conditions 28 34-36 45 Heavy Rain Conditions 76-81 82-132 294-296

• For greater detail see Tabics V-1 and V-4.

Source: Center for Planning and Research, Inc.

1876 198

VII-4 evacuation of the general population prior to cloud arrival under existing conditions, it would be necessary to start the notification process prior to the actual release.

Developing the idealized notification and warning system would effectively increase I

the time available for evacuation. As discussed previously, evacuating the inmates of Preston does not appear to be feasible, because of security reasons, prior to cloud arrival.

Protective Actions During the Cloud Passage During the period that the cloud passes over a location,t> T, the population at 2

that location would be subjected to three expmure pathways: external, contact, and inhalation.

Calculating probable exposures is beyond the scope of this contract.

flowever, to give a perspective of the conditional probabilities of whole body, lung, g

E and thyroid exposures at varying distances from a core melidown accident (this we assume to be the worst case), we have included the following three figures taken from I

NUREG 0396 (Figures VII-2 through VII-4).

If the general population cannot be evacuated prior to cloud arrival, one alternative would be to instruct them to go indoors to the best available shelter. Most of the facilities in the area are of wood frame construction without basements. Such f acilities, if closed up, would offer some protection against inhalation and contact expmures, and would afford an estimated shielding factor of about 0.75 to 0.33 against external exposure (assuming the mixture of radioisotopes has radiation energies similar i

to mixed fission products).

This does not appear to be a good alternative.

The preferred alternative appears to be to continue or start evacuation, in which case population groups would be subject to external, contact, and inhalation pathways, if they drove through the advancing cloud.

However, the inhalation and contact exposures could be materially reduced by rolling up vehicle windows. The mass of the vehicle would provide some shielding against external exposure.

The short exposure period measured in minutes, coupled with some protection en route would generally result in less exposure than if the population group waited in marginal shelters until the limits of the hazardous area were determined and remedial movement plans put into effect.

." Planning Basis for the Development of State and Local Government Radiological Emergency Response Plans in Support of Light Water Nuclear Power Plants", A Report I

Prepared by a U.S. Nuclear Regulatory Commission and U.S. Environmental Protection Agency Task Force on Emergency Plmining, December 1978.

.." Examination of Offsite Emergency Protective Mensures for Core Melt Accidents",

L Aldrich et al., Sandia Laboratories and Rasmussen, Department of Nuclear Engineering, 1876 194 g

VII-5 Figure vil-2 CONDITIONAL PitODABILITY OF EXCEEDING WilOLE BODY DOSE VEllSUS DISTANCE Probabilities are Conditional on a Core Melt Accident (5 x 10~ )

U 1

l I

0 8

I 1 REM O

I

>O 0.1 5 REM g$

wE I

80 jU 2

eb I

2m 5 :s EE 50 REM U O I

yu m<

&>5 o

0.01 I

2***

e

~

g 200 REM E

I

l I-0.001 1

10 100 1000 DISTANCE (MILES)

I I

Whole body dose calculated includes: external dose to the whole body due to the passing cloud, exposure to radionuclides on ground, and the dose to the whole body from inhaled radionuclides.

Dose calculations assumed no protective actions taken, and straight line plume trajectory.

Source:

NUPIG 0396 I

1876 200 I

VII-6 Figure VII-3 CONDITION AL PROBABILITY OF EXCEEDING LUNG DOSES VERSUS DISTANCE Probabilities are Conditional on a Core Melt Accident (5 x 10 -5) 1 i

i e i s i s i.

eisig i

i e i sisig i

i e i I

[

YA SREM I

op!!

I o_J u 25 REM u

z<

ob

$w I

U -:

Xw wc

u. O

~

~

OU E

>w=

4 ta h h 0.01 _-

7 IO x

c 300 R EM I

3000 REM

'I

I 0.001 1

10 100 1000 DISTANCE (MILES) r I

Lung dose calculated includes: external dose to the lung due to the passing cloud, exposure to radionuclides on ground, and the dose to the lung from inhaled radionuclides within 1 year.

Dose calculations assumed no protective actions taken, and straight line trajectory.

}hfb Source:

NUREG 0396 I

VII-7 Figure VII-4 CONDITIONAL PROBABILITY OF EXCEEDING TilYROID DOSES VERSUS DISTANCE Probabilities are Conditional on a Core Melt Accident (5 x 10-5)

I g

I

~

~

5 REM O

O O p.

o2 0.1 7

mw

>- c PU 25 REM I

I U

04 2 >-

5$ :g s

l WE

• 8 300 REM Oq b'

d 2 0.01 me 4

l

,O

~

e I

0.001 1

10 100 1000 I

DISTANCE (MILES) s I

Thyroid dose calculated includes: external dose to the thyroid due to the passing cloud, exposure to radionuclides on ground, and the dose to the thyroid from inhaled radionuclides.

Dose calculations assumed no protective actions taken, and straight line trajectory, g

Source:

NUREG 0396 1876 202 I

VII-8 The National Fallout Shelter Survey identified three fallout shelters in the 10 mile radius study area. Two of these are mines near Ione, which are near the 10 mile boundary.

Moving part of the population to shelter in these mines does not appear any more effective than evacuating, because the movement to the mines would take as much time as moving out of the area.

I A third group of fallout shelters are located at the Preston School of Industry.

Fallout shelter for 981 persons is located in buildings at Preston (133 in the Barracks; I

23 in the Chapel; and 825 in the Dining IIall).

Consequently, the 500 inmates (who offer a unique security problem) and approximately 150 staff personnel (mostly security) might remain at the facility as an alternative to evacuating.

If the buildings were closed, they would also provide protection against inhalation and contact exposures.

Consideration should be given to surveying the Preston Facility barracks to determine the shielding afforded against radiation associated with a reactor release.

The ability to shelter inmates in tneir barracks would probably present less security problems than sheltering them in the Dining liall.

Protective Actions After the Cloud Passage After the cloud has passed a location, ground shine from deposited radioactive material may pose a continuing risk to those in the affected area (presumably indoors, or in the case of Preston, in-shelter).

If this is the case, relocation or remedial movement to radiation free areas would be the appropriate countermeasure.

The general process would be to determine the extent of the contaminated area by monitoring estual dose rates, and then instruct the population remaining in the area I

to move out (generally a short distance) along routes designated at the time. Using helicopters for warning appears to be more effective than using surface vehicles.

Because the general population does not have cecess to shelters affording substantial protection, there could be some urgency in executing remedial movement for those in the center of the fallout pattern, depending on the dose already received during AT2 (the cloud passage) and the intensity of residual radiation (ground shine).

O Ilowever, under nearly all scenarios, it would be necessary to move only a short distance eThe identified shelter spaces offerred a protection factor of at least 40 (i.e., shielding Q

factor 0.025) for fallout radiation. Because the energy level of the gamma photons E

can be slightly higher for the fission products of a reactor accident than those used for shielding calculations for nuclear weapons, the PF's from the National Shelter Survey slightly overstate the protection for these shelters in the case of'Rgpg,hg Seco.

I 1876 zu)

I

VII-9 (from 500 to 2,000 meters at 5 miles from the site - see Figure VII-5) to radiation free areas.

Again, the inmates at the Preston School of Industry present a unique problem.

In the unlikely event that the school were to be in the center of the fallout pattern, it might be necessary to relocate the inmates to another facility which can provide I

the necessary security.

A considerable period of time would be available to marshal the transportation (caged buses) needed to transfer the inmates from Preston to another secure facility. An alternative to remedial movement would be decontaminaton of the Preston facility. Prior research has demonstrated the feasibility of obtaining decon-tamination factors of 10:1 using normally available equipment (firchosing, sweeping, etc.).

1 i

Franklin I3 rooks et al., Radiological Defense", Tech. Ops. Inc. Report T01 58-26, July 31, 1958 for the Office of Civil And Defense Mobilization. This research has considered the length of shelter stay and the optimum time for remedial movement from fallout shelters postattack. For weapon caused fallout radiation, the otimum time to leave shelter and move to a radiation free area is equal to 3/5 of the shelter protection factor (PF) times the number of hours required to reach a radiation free area.

I Asspming a minimum PF of 40 in Preston shelters and one hour transit time, the optimum time to start remedial movement would be 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after release, for the case where the decay curve approximated that of mixed fission products.

I

..Miller, C.F. Fallout and Radiological Countermeasures Volumes I and II, Stanford Research Institute, May 1967.

l I

I I

I I

i876 204 I

VII-10 Figure VII-5 TIIREE SIGMA IIALF-WIDTils OF GAUSSIAN SIIAPED PLUMES H DOWNWIND DISTANCE For Pasquill Stability Classes B, D and F Also shown are travel times of plume fronts for wind speeds of 1, 2 and 5 m/sec, and crosswind travel times for travel speeds of 2, 5 and 10 mph.

CROS$ WIND TR AVEL TIME IMIN )

2 5

10 mph r

- 120 1

10,000

-iso s

- 60

- 30 6

120 I

6 4

- 30

-15 g

3

~

-10 cp

/

l T

U

- 30 S

~

w 10 5

[

s'*

Z

/

2_

1000 20 o"

w *c:

l 1s c[s O

3 o

22 6

10

-2 qw 3

[

=.

6-U

!G w OE 3

-5

-2 o3 z'

32 DOWNWIND - PLUME F RONT TRAVEL TIME (MIN )

u)

OS 100

-2 a-U 2

U

• 5mhec I,0 2,0 30 60 90 120 180 300 s

5 10 2J 30 60 90 120 180 300 600 G = 2m/wc t

t f

I t

t t

i s

i 3

10 20 30 60 90 120 180 300 600 1200 u = 1mhec i e

i t

I t i

i i

i f

i n

K ff 'l ' illif fill '

ris I

// /

a 7

I I

I I

I I

I I

I 10 1

2 3

5 10 20 30 50 MILES l

l lKM 01 1

10 DOWNWIND DISTANCE Source: NUREG 0396 I

I I

APPENDIX A I

Letter to All Power Reactor Licensees I

I I

I I

I I

I 1876 206 I

_REOUEST FOR EVACUATION TIME ESTIMATES (AFTER NOTIFICATION)

TOR AREAS NEAR NUCLEAR POWER PLANTS

Background

Prior to recent NRC requests that means for prompt notification to the public be installed around each nuclear power plant site, a significant component of evacuation time estimates was the time required to notify the public of a I

neec evacuation.

Studies of actual evacuations that have taken place generally do not distinguish between the time required for notification, the time required to implement the evacuation, and the time required to confirm that an evacuation has taken place.1 The estimates for time required for evacuations now requested relate primarily to the time to implement an evacuation as opposed to the time required for notification.

These estimates I

may be based on previous loca t experiences (e.g. chemical spills or floods) or may be based on studies related to population density, local geography and road capacities.

No standard method for making such estimates is identified for use at this time.

The basis for the method chosen should be described in I

the response.

As an independent check on the evacuation time estimates, agreement with or comments on the time estimates made should be obtained from the principal local officials responsible for carrying out such evacuations.

Such agreement should be documented or the areas of disagreement indicated in the submittal.

The format given below is appropriate for reporting to the NRC estimates of the time required to implement evacuation of areas near nuclear power plants.

These estimates, are to be made for the primary purpose of making available, to those officials who would make evacuation decisions in an emergency situation, knowledge of the time required to complete one of the protective action options (evacuation) available for a particular potentially affected segment of the population.

A second purpose of these estimates is to identify to all concerned those instances in which unusual evacuation constraints exist and that special planning measures should be considered.

In some cases of extreme difficulty where a large population is at risk, special facility modifications may also be considered.

Given a decision to evacuate rather than shelter in an actual event, fewer I

might be evacuated should this be the chosen protective action.

or more sectors or different distances than given in the reporting format For example, three 22-sectors might be initially evacuated in a downwind direction (the sector containing the plume and an adjacent sector on each side), followed by the evacuation of other sectors as a precautionary measure.

I1 Hans, J.M.,

Jr. and T.C. Sell, 1974 Evacuation Risks - An Evaluation, U.S. Environmental Protection Agency, National Environmental Research Center, Las Vegas, EPA-520/6-74-002.

1876 207 Format for Reporting Information The areas for which evacuation estimates are required must encompass the entire area within a circle of about 10 miles radius, and have outer boundaries corresponding to the plume exposure EPZ.

These areas are as follows:

Distance Area 2 miles two 180' sectors 5 miles four 90* sectors I

about 10 miles four 90' sectors I

Estimates for the outer sectors should assume that the inner adjacent sectors are being evacuated simultaneously.

To the extent practical, the sector boundaries should not divide densely populated areas. Where a direction corresponding to the edges of areas for which estimates have been made is I

thought not to be adequately represented by the time estimates for adjacent areas, an additonal area should be defined and a separate estimate made for this case.

The format for submittal should include both a table and a figure (overlaid on a map) which each give the information requested in items 1 and 2 below. Additional material may be provided in associated text.

Required Information 1.

Two estimates are requested in each of the areas defined in item 1 for a general evacuation of the population (not including special facilities).

I A best estimate is required and an adverse weather estimate is required for movement of the population.

2.

The total time required to evacuate special facilities (e.g. hospitals) within each area must be specified (best estimate and adverse weather).

~

3.

The time required for confirmation of evacuation should be indicated.

Confirmation times may consider special instructions to the public (e.g. tying a handkerchief to a door or gate to indicate the occupant has left the premises).

4.

Where plans and prompt notification systems have not been put in place for areas out to about 10 miles, estimates of the times reqtired to evacuate I

until such measures are in place for the plume exposure emergency planning zone (EPZ) should also be given.

Notification times greater than 15 minutes should be included in the evacuation times and footnoted to indicate the notification time.

I i876 208 I

I I

5.

Where special evacuation problems are identified (e.g. in high I

population density areas), specify alternative protective actions, such as sheltering, which would reduce exposures and the effectiveness of these measures.

6.

A short background document should be submitted giving the methods used to make the estimates and the assumptions made including the routes and methods of transportation used.

This document should also note the agreement or areas of disagreement with principal local officials regarding these estimates.

I I

I I

I I

I 1876 209

I I

I I

I I

I APPENDIX B Comments by Reviewers I

I I

I I

I I

e I

I 1876 210 I