ML12355A751

Evacuation Time Estimate Study; Section 5 Through 12

ML12355A751
Person / Time
Site:	Palo Verde
Issue date:	12/14/2012
From:	KLD Engineering, PC
To:	Office of Nuclear Material Safety and Safeguards, Arizona Public Service Co
References
Download: ML12355A751 (85)

Text

5 ESTIMATION OF TRIP GENERATION TIME Federal Government guidelines (see NUREG CR-7002) specify that the planner estimate the distributions of elapsed times associated with mobilization activities undertaken by the public to prepare for the evacuation trip. The elapsed time associated with each activity is represented as a statistical distribution reflecting differences between members of the public.

The Aluantification of these activity-based distributions relies largely on the results of the telephone survey. We define the sum of these distributions of elapsed times as the Trip Generation Time Distribution.

5.1 Background In general, an accident at a nuclear power plant is characterized by the following Emergency Classification Levels (see Appendix 1 of NUREG 0654 for details):

1. Unusual Event

2. Alert

3. Site Area Emergency

4. General Emergency At each level, the Federal guidelines specify a set of Actions to be undertaken by the Licensee, and by State and Local offsite authorities. As a Planning Basis. we will adopt a conservative posture, in accordance with Section 1.2 of NUREG/CR-7002, that a rapidly escalating accident will be considered in calculating the Trip Generation Time. We will assume:

1. The Advisory to Evacuate will be announced coincident with the siren notification.

2. Mobilization of the general population will commence within 15 minutes after the siren notification.

3. ETE are measured relative to the Advisory to Evacuate.

We emphasize that the adoption of this planning basis is not a representation that these events will occur within the indicated time frame. Rather, these assumptions are necessary in order to:

1. Establish a temporal framework for estimating the Trip Generation distribution in the format recommended in Section 2.13 of NUREG/CR-6863.

2. Identify temporal points of reference that uniquely define "Clear Time" and ETE.

It is likely that a longer time will elapse between the various classes of an emergency.

For example, suppose one hour elapses from the siren alert to the Advisory to Evacuate. In this case, it is reasonable to expect some degree of spontaneous evacuation by the public during this one-hour period. As a result, the population within the EPZ will be lower when the Advisory to Evacuate is announced, than at the time of the siren alert. In addition, many will engage in preparation activities to evacuate, in anticipation that an Advisory will be broadcast.

Thus, the time needed to complete the mobilization activities and the number of people remaining to evacuate the EPZ after the Advisory to Evacuate, will both be somewhat less than Palo Verde 5-1 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

the estimates presented in this report. Consequently, the ETE presented in this report are higher than the actual evacuation time, if this hypothetical situation were to take place.

The notification process consists of two events:

1. Transmitting information using the alert notification systems available within the EPZ (e.g. sirens, tone alerts, EAS broadcasts, loud speakers).

2. Receiving and correctly interpreting the information that is transmitted.

The population within the EPZ is dispersed over an area of approximately 315 square miles~and is engaged in a wide variety of activities. It must be anticipated that some time will elapse between the transmission and receipt of the information advising the public of an accident.

The amount of elapsed time will vary from one individual to the next depending on where that person is, what that person is doing, and related factors. Furthermore, some persons who will be directly involved with the evacuation process may be outside the EPZ at the time the emergency is declared. These people may be commuters, shoppers and other travelers who reside within the EPZ and who will return to join the other household members upon receiving notification of an emergency.

As indicated in Section 2.13 of NUREG/CR-6863, the estimated elapsed times for the receipt of notification can be expressed as a distribution reflecting the different notification times for different people within, and outside, the EPZ. By using time distributions, it is also possible to distinguish between different population groups and different day-of-week and time-of-day scenarios, so that accurate ETE may be computed.

For example, people at home or at work within the EPZ will be notified by siren, and/or tone alert and/or radio (if available). Those well outside the EPZ will be notified by telephone, radio, TV and word-of-mouth, with potentially longer time lags. Furthermore, the spatial distribution of the EPZ population will differ with time of day - families will be united in the evenings, but

-dispersed during the day. In this respect, weekends will differ from weekdays.

As indicated in Section 4.1 of NUREG/CR-7002, the information required to compute trip generation times is typically obtained from a telephone survey of EPZ residents. Such a survey was conducted in support of this ETE study. Appendix F presents the survey sampling plan, survey instrument, and raw survey results. It is important to note that the shape and duration of the evacuation trip mobilization distribution is important at sites where traffic congestion is not expected to cause the evacuation time estimate to extend in time well beyond the trip generation period. The remaining discussion will focus on the application of the trip generation data obtained from the telephone survey to the development of the ETE documented in this report.

Palo Verde 5-2 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

5.2 Fundamental Considerations The environment leading up to the time that people begin their evacuation trips consists of a sequence of events and activities. Each event (other than the first) occurs at an instant in time and is the outcome of an activity.

Activities are undertaken over a period of time. Activities may be in "series" (i.e. to undertake an activity implies the completion of all preceding events) or may be in parallel (two or more activities may" take place over the same period of time). Activities conducted in series are functionally dependent on the completion of prior activities; activities conducted in parallel are functionally independent of one another. The relevant events associated with the public's preparation for evacuation are:

Event Number Event Description 1 Notification 2 Awareness of Situation 3 Depart Work 4 Arrive Home 5 Depart on Evacuation Trip Associated with each sequence of events are one or more activities, as outlined below:

Table 5-1. Event Sequence for Evacuation Activities 1 -4 2 - Receive Notification 1 2--3 Prepare to Leave Work 2 2,3 -) 4 Travel Home 3 2,4 -> 5 Prepare to Leave to Evacuate 4 These relationships are shown graphically in Figure 5-1.

" An Event is a 'state' that exists at a point in time (e.g., depart work, arrive home)

" An Activity is a 'process' that takes place over some elapsed time (e.g., prepare to leave work, travel home)

As such, a completed Activity changes the 'state' of an individual (e.g. the activity, 'travel home' changes the state from 'depart work' to 'arrive home'). Therefore, an Activity can be described as an 'Event Sequence'; the elapsed times to perform an event sequence vary from one person to the next and are described as statistical distributions on the following pages.

An employee who lives outside the EPZ will follow sequence (c) of Figure 5-1. A household within the EPZ that has one or more commuters at work, and will await their return before Palo Verde 5-3 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

beginning the evacuation trip will follow the first sequence of Figure 5-1(a). A household within the EPZ that has no commuters at work, or that will not await the return of any commuters, will follow the second sequence of Figure 5-1(a), regardless of day of week or time of day.

Households with no commuters on weekends or in the evening/night-time, will follow the applicable sequence in Figure 5-1(b). Transients will always follow one of the sequences of Figure 5-1(b). Some transients away from their residence could elect to evacuate immediately without returning to the residence, as indicated in the second sequence.

It is seen from Figure 5-1, that the Trip Generation time (i.e. the total elapsed time from Event 1 to Event 5) depends on the scenario and will vary from one' household to the next.

Furthermore, Event 5 depends, in a complicated way, on the time distributions of all activities preceding that event. That is, to estimate the time distribution of Event 5, we must obtain estimates of the time distributions of all preceding events. For this study, we adopt the conservative posture that all activities will occur in sequence.

In some cases, assuming certain events occur strictly sequential (for instance, commuter returning home before beginning preparation to leave) can result in rather conservative (that is, longer) estimates of mobilization times. It is reasonable to expect that at least some parts of these events will overlap for many households, but that assumption is not made in this study.

Palo Verde 5-4 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

1 2 3 4 5 As _Aa Af Residents W IW W 4

I Households wait for Commuters' Households without 1 2 5 Commuters and Residents households who do not wait for Commuters Residents, Transients 1 2 4 5 Return to residence, away from then evacuate Residence Residents, 1 2 5 Residents at home; Transients at transients evacuate directly Residence 1

e--42 --0 3,5 ACTIVITIES EVENTS 1

• 2 Receive Notification 1. Notification

2. 3 Prepare to Leave Work 2. Aware of situation 2, 3 -. 0. 4 Travel Home 3. Depart work 2, 4 0 5 Prepare to Leave to Evacuate 4. Arrive home

5. Depart on evacuation trip A

Activities Consume Time 1 Applies for evening and weekends also if commuters are at work.

2 Applies throughout the year for transients.

Figure 5-1. Events and Activities Preceding the Evacuation Trip Palo Verde 5-5 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

5.3 Estimated Time Distributions of Activities Preceding Event 5 The time distribution of an event is obtained by "summing" the time distributions of all prior contributing activities. (This "summing" process is quite different than an algebraic sum since it is performed on distributions - not scalar numbers).

Time Distribution No. 1, Notification Process: Activity 1 -- 2 It is assumed (based on the presence of sirens within the EPZ) that 87 percent of those within the EPZ will be aware of the accident within 30 minutes with the remainder notified within the following 15 minutes. The notification distribution is given below:

Table 5-2. Time Distribution for Notifying the Public Elpe Tim Pecnto (Minutes Pouato Noife 0 0.0%

5 7.1%

10 13.3%

15 26.5%

20 46.9%

25 66.3%

30 86.7%

35 91.8%

40 96.9%

45 100.0%

Palo Verde 5-6 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Distribution No. 2, Prepare to Leave Work: Activity 2 -- 3 It is reasonable to expect that the vast majority of business enterprises within the EPZ will elect to shut down following notification and most employees would leave work quickly. Commuters, who work outside the EPZ could, in all probability, also leave quickly since facilities outside the EPZ would remain open and other personnel would remain. Personnel or farmers responsible for equipment/livestock would require additional time to secure their facility. The distribution of Activity 2 --> 3 shown in Table 5-3 reflects data obtained by the telephone survey. This distribution is plotted in Figure 5-2.

Table 5-3. Time Distribution for Employees to Prepare to Leave Work 0 0.0% 40 88.6%

5 35.2% 45 91.5%

10 57.0% 50 91.5%

15 70.3% 55 91.5%

20 76.6% 60 98.3%

25 77.1% 75 99.3%

30 87.3% 90 100.0%

35 87.6%

NOTE: The survey data was normalized to distribute the "Don't know" response. That is, the sample was reduced in size to include only those households who responded to this question. The underlying assumption is that the distribution of this activity for the "Don't know" responders, if the event takes place, would be the same as those responders who provided estimates.

Palo Verde 5-7 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Distribution No. 3. Travel Home: Activity 3 -* 4 These data are provided directly by those households which responded to the telephone survey. This distribution is plotted in Figure 5-2 and listed in Table 5-4.

Table 5-4. Time Distribution for Commuters to Travel Home 0 0.0% 40 75.9%

5 9.6% I 45 85.9%

10 15.3% 50 87.6%

15 27.3% 55 87.6%

20 41.6% 60 96.9%

25 46.9% 75 98.2%

30 66.1% 90 100.0%

NOTE: The survey data was normalized to distribute the "Don't know" response Palo Verde 5-8 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Distribution No. 4, Prepare to Leave Home: Activity 2,4 -> 5 These data are provided directly by those households which responded to the telephone survey. This distribution is plotted in Figure 5-2 and listed in Table 5-5.

Table 5-5. Time Distribution for Population to Prepare to Evacuate 0 0.0% 105 87.1%

15 16.0% 120 94.5%

30 55.9% 135 96.9%

45 62.3% 150 97.1%

60 78.9% 165 97.4%

75 84.4% 180 98.7%

90 86.6% 195 100.0%

NOTE: The survey data was normalized to distribute the "Don't know" response Palo Verde 5-9 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Mobilization Activities 100% y C

.2 80%

(j:f

.N 0

60%

/V/ .

4.

- Notification E

C 40% IT -- Prepare to Leave Work

-Travel Home 0 -Prepare Home I 20%

C.

0%

0 30 60 90 120 150 180 210 Elapsed Time from Start of Mobilization Activity (min)

Figure 5-2. Evacuation Mobilization Activities KLD Engineering, P.C.

Palo Verde Palo Verde 5-10 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

5.4 Calculation of Trip Generation Time Distribution The time distributions for each of the mobilization activities presented herein must be combined to form the appropriate Trip Generation Distributions. As discussed above, this study assumes that the stated events take place in sequence such that all preceding events must be completed before the current event can occur. For example, if a household awaits the return of a commuter, the work-to-home trip (Activity 3,-+ 4) must precede Activity 4 -+ 5.

To calculate the time distribution of an event that is dependent on two sequential activities, it is necessary to "sum" the distributions associated with these prior activities. The distribution summing algorithm is applied repeatedly as shown to form the required distribution. As an outcome of this procedure, new time distributions are formed; we assign "letter" designations to these intermediate distributions to describe the procedure. Table 5-6 presents the summing procedure to arrive at each designated distribution.

Table 5-6. Mapping Distributions to Events Apl "Sm ingAloih To Ditibto Obaie EvntDeine Distributions 1 and 2 Distribution A I Event 3 Distributions Aand 3 Distribution B T Event 4 Distributions Band 4 Distribution C Event 5 Distributionsl1and 4 Distribution D Event 5 Table 5-7 presents a description of each of the final trip generation distributions achieved after the summing process is completed.

Palo Verde 5-11 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 5-7. Description of the Distributions F- -SI Time distribution of commuters departing place of work (Event 3). Also applies A to employees who work within the EPZ who live outside, and to Transients within the EPZ.

B Time distribution of commuters arriving home (Event 4).

home, leaving home Time distribution of residents with commuters who return to begin the evacuation trip (Event 5).

D Time distribution of residents without commuters returning home, leaving home to begin the evacuation trip (Event 5).

5.4.1 Statistical Outliers As already mentioned, some portion of the survey respondents answer "don't know" to some questions or choose to not respond to a question. The mobilization activity distributions are based upon actual responses. But, it is the nature of surveys that a few numeric responses are inconsistent with the overall pattern of results. An example would be a case in which for 500 responses, almost all of them estimate less than two hours for a given answer, but 3 say "four hours" and 4 say "six or more hours".

These "outliers" must be considered: are they valid responses, or so atypical that they should be dropped from the sample?

In assessing outliers, there are three alternates to consider:

1) Some responses with very long times may be valid, but reflect the reality that the respondent really needs to be classified in a different population subgroup, based upon special needs;

2) Other responses may be unrealistic (6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> to return home from commuting distance, or 2 days to prepare the home for departure);

3) Some high values are representative and plausible, and one must not cut them as part of the consideration of outliers.

The issue of course is how to make the decision that a given response or set of responses are to be considered "outliers" for the component mobilization activities, using a method that objectively quantifies the process.

There is considerable statistical literature on the identification and treatment of outliers singly or in groups, much of which assumes the data is normally distributed and some of which uses non-parametric methods to avoid that assumption. The literature cites that limited work has been done directly on outliers in sample survey responses.

In establishing the overall mobilization time/trip generation distributions, the following principles Palo Verde 5-12 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

are used:

1) It is recognized that the overall trip generation distributions are conservative estimates, because they assume a household will do the mobilization activities sequentially, with no overlap of activities;

2) The individual mobilization activities (prepare to leave work, travel home, prepare home) are reviewed for outliers, and then the overall trip generation distributions are created (see Figure 5-1, Table 5-6, Table 5-7);

3) Outliers can be eliminated either because the response reflects a special population (e.g.

special needs, transit dependent) or lack of realism, because the purpose is to estimate trip generation patterns for personal vehicles;

4) To eliminate outliers, a) the mean and standard deviation of the specific activity are estimated from the responses, b) the median of the same data is estimated, with its position relative to the mean noted, c) the histogram of the data is inspected, and d) all values greater than 3.5 standard deviations are flagged for attention, taking special note of whether there are gaps (categories with zero entries) in the histogram display.

In general, only flagged values more than 4 standard deviations from the mean are allowed to be considered outliers, with gaps in the histogram expected.

When flagged values are classified as outliers and dropped, steps "a" to "d" are repeated.

Palo Verde 5-13 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

5) As a practical matter, even with outliers eliminated by the above, the resultant histogram, viewed as a cumulative distribution, is not a normal distribution. A typical situation that results is shown below in Figure 5-3.

100.0% ,

90.0%

80.0%

' 70.0%

4' f 60.0%

U 50.0%

> 40.0%

" 30.0%

E 20.0%

10.0%

0.0%

r , r (N4. r4J r- r4 N. r4J r, r4 r, N- CN r, (4

,4 (N (N m m 4 tr"L LLn %D O oC -

Center of Interval (minutes)

- Cumulative Data - - Cumulative Normal Figure 5-3. Comparison of Data Distribution and Normal Distribution

6) In particular, the cumulative distribution differs from the normal distribution in two key aspects, both very important in loading a network to estimate evacuation times:

> Most of the real data is to the left of the "normal" curve above, indicating that the network loads faster for the first 80-85% of the vehicles, potentially causing more (and earlier) congestion than otherwise modeled;

> The last 10-15% of the real data "tails off" slower than the comparable "normal" curve, indicating that there is significant traffic still loading at later times.

Because these two features are important to preserve, it is the histogram of the data that is used to describe the mobilization activities, not a "normal" curve fit to the data. One could consider other distributions, but using the shape of the actual data curve is unambiguous and preserves these important features;

7) With the mobilization activities each modeled according to Steps 1-6, including preserving the features cited in Step 6, the overall (or total) mobilization times are constructed.

This is done by using the data sets and distributions under different scenarios (e.g. commuter returning, no commuter returning). In general, these are additive, using weighting based upon the Palo Verde 5-14 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

probability distributions of each element; Figure 5-4 presents the combined trip generation distributions designated A, C, and D. These distributions are presented on the same time scale.

(As discussed earlier, the use of strictly additive activities is a conservative approach, because it makes all activities sequential - travel home from work follows preparation to leave work, preparation for departure follows the return of the commuter, and so forth. In practice, it is reasonable that some of these activities are done in parallel, at least to some extent - for instance, preparation to depart begins by a household member at home while the commuter is still on the road.)

The mobilization distributions that result are used in their tabular/graphical form as direct inputs to later computations that lead to the ETE.

The DYNEV II simulation model is designed to accept varying rates of vehicle trip generation for each origin centroid, expressed in the form of histograms. These histograms, which represent Distributions A, C, and D, properly displaced with respect to one another, are tabulated in Table 5-8 (Distribution B, Arrive Home, omitted for clarity).

The final time period (13) is 600 minutes long. This time period is added to allow the analysis network to clear, in the event congestion persists beyond the trip generation period. Note that there are no trips generated during this final time period.

5.4.2 Staged Evacuation Trip Generation As defined in NUREG/CR-7002, staged evacuation consists of the following:

1. Sectors comprising the 2 mile region are advised to evacuate immediately

2. Sectors comprising regions extending from 2 to 5 miles downwind are advised to shelter in-place while the 2 mile region is cleared

3. As vehicles evacuate the 2 mile region, sheltered people from 2 to 5 miles downwind continue preparation for evacuation

4. The population sheltering in the 2 to 5 mile region are advised to begin evacuating when approximately 90% of those originally within the 2 mile region evacuate across the 2 mile region boundary

5. Non-compliance with the shelter recommendation is the same as the shadow evacuation percentage of 20%

Assumptions

1. The EPZ population in sectors beyond 5 miles will react as does the population in the 2 to 5 mile region; that is they will first shelter, then evacuate after the 90th percentile ETE for the 2 mile region

2. The population in the shadow region beyond the EPZ boundary, extending to 15 miles radially from the plant, will react as they do for all non-staged evacuation scenarios.

That is 20% of these households will elect to evacuate with no shelter delay.

Palo Verde 5-15 KLD Engineering, P.C.

Evacuation Time Estimate Rev. I

3. The transient population will not be expected to stage their evacuation because of the limited sheltering options available to people who may be at parks, on a beach, or at other venues. Also, notifying the transient population of a staged evacuation would prove difficult.

4. Employees will also be assumed to evacuate without first sheltering.

Procedure

1. Trip generation for population groups in the 2 mile region will be as computed based upon the results of the telephone survey and analysis.

2. Trip generation for the population subject to staged evacuation will be formulated as follows:

a. Identify the 90th percentile evacuation time for the sectors comprising the two mile region. This value, Tscen*, is obtained from simulation results. It will become the time at which the region being sheltered will be told to evacuate for each scenario.

b. The resultant trip generation curves for staging are then formed as follows:

i. The non-shelter trip generation curve is followed until a maximum of 20%

of the total trips are generated (to account for shelter non-compliance).

ii. No additional trips are generated until time Tscen*

iii. Following time Tscen , the balance of trips are generated:

1. by stepping up and then following the non-shelter trip generation curve (if Tscen* is < max trip generation time) or

2. by stepping up to 100% (if Tscen* is > max trip generation time)

c. Note: This procedure implies that there may be different staged trip generation distributions for different scenarios. NUREG/CR-7002 uses the statement "approximately 9 0 th percentile" as the time to end staging and begin evacuating.

The value of Tscen* is about 90 minutes, on average.

3. Staged trip generation distributions are created for the following population groups:

a. Residents with returning commuters

b. Residents without returning commuters Figure 5-5 presents the staged trip generation distributions for both residents with and without returning commuters; the 90th percentile two-mile evacuation time is 90 minutes for good weather. At the 9 0 th percentile evacuation time, 20% of the population (who normally would have completed their mobilization activities for an un-staged evacuation) advised to shelter has nevertheless departed the area. These people do not comply with the shelter advisory. Also included on the plot are the trip generation distributions for these groups as applied to the regions advised to evacuate immediately.

Since the 9 0 th percentile evacuation time occurs before the end of the trip generation time, after the sheltered region is advised to evacuate, the shelter trip generation distribution rises to meet the balance of the non-staged trip generation distribution. Following time Tscen*, the balance of staged evacuation trips that are ready to depart are released within 15 minutes. After Palo Verde 5-16 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Tscen*+15, the remainder of evacuation trips are generated in accordance with the un-staged trip generation distribution.

Table 5-9 provides the trip generation histograms for staged evacuation.

5.4.3 Trip Generation Time for Recreational Areas As indicated in Table 5-2, this study assumes 100% notification in 45 minutes. Table 5-9 indicates that all transients will have mobilized within 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />. It is assumed that this 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> timeframe is sufficient time for campers and other transients to return to their vehicles and begin their evacuation trip.

Palo Verde 5-17 KLD Engineering, P.C.

Evacuation Time Estimate Rev. I

Table 5-8. Trip Generation Histograms for the EPZ Population for Un-staged Evacuation 1 15 6% 6% 0% 1%

2 15 32% 32% 0% 9%

3 15 35% 35% 2% 24%

4 15 14% 14% 6% 23%

5 15 6% 6% 12% 13%

6 15 5% 5% 15% 10%

7 30 2% 2% 28% 7%

8 30 0% 0% 17% 8%

9 30 0% 0% 10% 2%

10 30 0% 0% 5% 2%

11 30 0% 0% 3% 1%

12 60 0% 0% 2% 0%

13 600 0% 0% 0% 0%

NOTE:

Shadow vehicles are loaded onto the analysis network (Figure 1-2) using Distributions C for good weather.

Special event vehicles are loaded using Distribution A.

Palo Verde 5-18 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Trip Generation Distributions

-Employees/Transients - Residents with Commuters - Residents with no Commuters I-U C

0 80 80 '00'ý 40 w 2 a.26 1A 00 60 120 180 240 300 Elapsed Time from Evacuation Advisory (min)

Figure 5-4. Comparison of Trip Generation Distributions Palo Verde 5-19 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 5-9. Trip Generation Histograms for the EPZ Population for Staged Evacuation Percent of Total Trips Generated Within Indicated Time Period' Time Duration Residents with Commuters Residents Without Period (Min) (Distribution Q Commuters (Distribution D) 1 15 0% 0%

2 15 0% 2%

3 15 0% 5%

4 15 2% 4%

5 15 2% 3%

6 15 3% 2%

7 30 56% 71%

8 30 17% 8%

9 30 10% 2%

10 30 5% 2%

11 30 3% 1%

12 60 2% 0%

13 600 0% 0%

• Trip Generation for Employees and Transients (see Table 5-8) is the same for Un-staged and Staged Evacuation.

Palo Verde 5-20 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Staged and Un-staged Evacuation Trip Generation

-Employees / Transients - Residents with Commuters

- Residents with no Commuters - Staged Residents with Commuters

-Staged Residents with no Commuters 100 C.

I-0 to 80 U

fa w

to 60 C

0 7.

0 40 jleoJ CL 20 C

0 0 30 60 90 120 150 180 210 240 270 300 Elapsed Time from Evacuation Advisory (min)

Figure 5-5. Comparison of Staged and Un-staged Trip Generation Distributions in the 2 to 5 Mile Region P.C.1 KLD Engineering,Rev.

Verde Palo Verde 5-21 KLD Engineering, P.C.

Evacuation Time Estimate Time Estimate Rev. 1

6 DEMAND ESTIMATION FOR EVACUATION SCENARIOS An evacuation "case" defines a combination of Evacuation Region and Evacuation Scenario.

The definitions of "Region" and "Scenario" are as follows:

Region A grouping of contiguous evacuating sectors that forms either a "keyhole" sector-based area, or a circular area within the EPZ, that must be evacuated in response to a radiological emergency.

Scenario A combination of circumstances, including time of day, day of week, season, and weather conditions. Scenarios define the number of people in each of the affected population groups and their respective mobilization time distributions.

A total of 52 Regions were defined which encompass all the groupings of sectors considered.

These Regions are defined in Table 6-1. The sector configurations are identified in Figure 6-1.

Each keyhole sector-based area consists of a central circle centered at the power plant, and three adjoining sectors, each with a central angle of 22.5 degrees, as per NUREG/CR-7002 guidance. The central sector coincides with the wind direction. These sectors extend to 5 miles from the plant (Regions R04 through Region R19) or to the EPZ boundary (Regions R20 through R35). Regions R01, R02 and R03 represent evacuations of circular areas with radii of 2, 5 and 10 miles, respectively. Regions R36 through R52 are identical to Regions R02, R04 through R19, respectively; however, those sectors between 2 miles and 5 miles are staged until 90% of the 2-mile region (Region R01) has evacuated. When the wind is blowing from the east, east-southeast or southeast, two sectors on either side of the central sector (5 sectors total) are considered due to the swirling effect of the wind in the mountain range to the west of the plant.

A total of 12 Scenarios were evaluated for all Regions. Thus, there are a total of 52x12=624 evacuation cases. Table 6-2 is a description of all Scenarios.

Each combination of region and scenario implies a specific population to be evacuated. Table 6-3 presents the percentage of each population group estimated to evacuate for each scenario.

Table 6-4 presents the vehicle counts for each scenario for an evacuation of Region R03 - the entire EPZ.

The vehicle estimates presented in Section 3 are peak values. These peak values are adjusted depending on the scenario and region being considered, using scenario and region specific percentages; the scenario percentages are presented in Table 6-3, while the regional percentages are provided in Table H-1. The percentages presented in Table 6-3 were determined as follows:

The number of residents with commuters during the week (when workforce is at its peak) is equal to the product of 61% (the number of households with at least one commuter) and 42%

(the number of households with a commuter that would await the return of the commuter prior to evacuating). See assumption 3 in Section 2.3. It is estimated for weekend and evening Palo Verde 6-1 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

scenarios that 10% of households with returning commuters will have a commuter at work during those times.

Employment is assumed to be at its peak during the winter, midweek, midday scenarios.

Employment is reduced slightly (96%) for summer, midweek, midday scenarios. This is based on the estimation that 50% of the employees commuting into the EPZ will be on vacation for a week during the approximate 12 weeks of summer. It is further estimated that those taking vacation will be uniformly dispersed throughout the summer with approximately 4% of employees vacationing each week. It is further estimated that only 10% of the employees are working in the evenings and during the weekends.

Based on discussions with emergency planning personnel from APS who are familiar with the transient attractions in the EPZ, transient activity is estimated to be at its peak (100%) during winter evenings and less (75%) during winter days, 40% during summer evenings and 30%

during summer days. During the outage (Scenario 11), most of the transient attractions are occupied by outage workers who are already counted as special event population. Thus, the transient percentage is significantly less - 10% - for this scenario.

As noted in the shadow footnote to Table 6-3, the shadow percentages are computed using a base of 20% (see assumption 5 in Section 2.2); to include the employees within the shadow region who may choose to evacuate, the voluntary evacuation is multiplied by a scenario-specific proportion of employees to permanent residents in the shadow region. For example, using the values provided in Table 6-4 for Scenario 1, the shadow percentage is computed as follows:

20% x + 1,543 2 ' x 1,634 + 4,655)

One special event - an outage at the plant - was considered as Scenario 11. Thus, the special event traffic is 100% evacuated for Scenario 11, and 0% for all other scenarios. As discussed above, since most of the additional outage workers stay at the two transient facilities in the EPZ, the percentage of transients was reduced to 10% for this scenario.

It is estimated that summer school enrollment is approximately 10% of enrollment during the regular school year for summer, midweek, midday scenarios. School is not in session during weekends and evenings, thus no buses for school children are needed under those circumstances. As discussed in Section 7, schools are in session during the winter season, midweek, midday and 100% of buses will be needed under those circumstances. Transit buses for the transit-dependent population are set to 100% for all scenarios as it is assumed that the transit-dependent population is present in the EPZ for all scenarios.

External traffic is estimated to be reduced by 60% during evening scenarios and is 100% for all other scenarios.

Palo Verde 6-2 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 6-1. Description of Evacuation Regions I sector 10 C I D E IFI G H ILJIK IL LM N PIQ IR]

Reg~io RZ 2-Wind Direction D E I F G J K L M N I P a R Region From: 5- 2-SI-102-SS-112-IS-012!5__ 02-5SI-10I2-SI54(412-515-Iflt2-5IS-101 5-Sr1012-55-1012-5SI -1012-SIS-10[2-5frS-044 S l ROE SSW ROE SW F&7 WSW ROE W R09 WNW R10 NW RuI NNW R22 N R13 NNE R14 NE RUS ENE RIG E R17 ESE R18 SE R19 SSE

-Me RIus an Down 'es Wind ____ ___ ____ ____ ___ ____ ___ Sector Direction H I J K L I M P I C Region I From: 7S.c 5-10 2-5 S-10 2-51 S-101 2-5 5-10 2- S-101 2- S-10 2--5510 S m SSW I I I I I I I I I I RUl SW I I I l l l l l lI i R23 WSW W

r:T=!

1 I I I

I I

I I

I I

RZIB WNW I 1 R25 NW I I NNW N

R26 NNE R27 NE ENE I I R30 E I I i I R31 ESE I I I I SE I I I I I I I SSE . . . I I I I I I I I I I I I I~~~ I I IIrII ~~ I I I III.. . . . h I.. . Sector(s) Shelter-,-,Pace Palo Verde 6-3 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Staged Evacuation Mile Radius Evacuates, then Evacuate Downwind to S Miles Wind Sector Direction a C D E F G H J K I- M N P n 1 R 0 F G N L N n I R Region From: '-F05-IL'0 5-Ifl 51 5 S0 -102- S-01-SL~9IS-10 -SS-0 -S5 C5402-

-02.US 1e2SS S-1 S-14 R36 5-Mile Radius R37 S Fi SSW R39 SW MO- WSW R41 W R42 WNW R43 NW R44 NNW N

NNE NE ENE E

ESE SE mm Sector________s)f SheltihihhInIIlacl Palo Verde KLD Engineering, P.C.

6-4 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Figure 6-1. Palo Verde EPZ Sectors Palo Verde KLD Engineering, P.C.

6-5 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 6-2. Evacuation Scenario Definitions e na Ss D

• 1 Summer Midweek Midday Good None 2 Summer Midweek Midday Rain None 3 Summer Weekend Midday Good None 4 Summer Weekend Midday Rain None 5 Summer Midweek, Evening Good None Weekend 6 Winter Midweek Midday Good None 7 Winter Midweek Midday Rain None 8 Winter Weekend Midday Good None 9 Winter Weekend Midday Rain None 10 Winter Midweek, Evening Good None Weekend 11 Winter Midweek Midday Good Outage at PVNGS Roadway Impact 12 Summer Midweek Midday Good - Lane Closure on 1-10 Eastbound Palo Verde 6-6 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 6-3. Percent of Population Groups Evacuating for Various Scenarios Hoshod Houshols - S 1 26% 74% 96% 30% 25% 0% 10% 100% 100%

2 26% 74% 96% 30% 25% 0% 10% 100% 100%

3 2.6% 97.4% 10% 30% 21% 0% 0% 100% 100%

4 2.6% 97.4% 10% 30% 21% 0% 0% 100% 100%

5 2.6% 97.4% 10% 40% 21% 0% 0% 100% 40%

6 26% 74% 100% 75% 25% 0% 100% 100% 100%

7 26% 74% 100% 75% 25% 0% 100% 100% 100%

8 2.6% 97.4% 10% 75% 21% 0% 0% 100% 100%

9 2.6% 97.4% 10% 75% 21% 0% 0% 100% 100%

10 2.6% 97.4% 10% 100% 21% 0% 0% 100% 40%

11 26% 74% 100% 10% 25% 100% 100% 100% 100%

12 26% 74% 96% 30% 25% 0% 10% 100% 100%

Resident Households with Commuters ....... Households of EPZ residents who await the return of commuters prior to beginning the evacuation trip.

Resident Households with No Commuters ..Households of EPZ residents who do not have commuters or will not await the return of commuters prior to beginning the evacuation trip.

Employees ................................................. EPZ employees who live outside the EPZ Transients .................................................. People who are In the EPZ at the time of an accident for recreational or other (non-employment) purposes.

Shadow ...................................................... Residents and employees In the shadow region (outside of the EPZ) who will spontaneously decide to relocate during the evacuation. The basis for the values shown Is a 20% relocation of shadow residents along with a proportional percentage of shadow employees.

Special Events ............................................ Additional vehides in the EPZ due to the Identified special event.

School and Transit Buses ............................ Vehicle-equivalents present on the road during evacuation servidng schools and transit-dependent people (1 bus is equivalent to 2 passenger vehicles).

External Through Traffic ............................. Traffic on Interstates/freeways and major arterial roads at the start of the evacuation. This traffic is stopped by access control approximately 45 minutes after the evacuation begins.

Palo Verde 6-7 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 6-4. Vehicle Estimates by Scenario Household Hoshod With ithou Tota Re u nn Re u n n Sp ca Sc oo Tr ni0 tcrnaIS e ai 1 1,634 4,655 1,543 212 1,206 5 32 2,288 11,575 2 1,634 4,655 1,543 212 1,206 - 5 32 2,288 11,575 3 163 6,126 161 212 993 - - 32 2,288 9,975 4 163 6,126 161 212 993 - 32 2,288 9,975 5 163 6,126 161 283 993 - - 32 915 8,673 6 1,634 4,655 1,607 531 1,216 - 50 32 2,288 12,013 7 1,634 4,655 1,607 531 1,216 - 50 32 2,288 12,013 8 163 6,126 161 531 993 -- 32 2,288 10,294 9 163 6,126 161 531 993 32 2,288 10,294 10 163 6,126 161 708 993 - - 32 915 9,098 11 1,634 4,655 1,607 71 1,216 1,444 50 32 2,288 12,997 12 1,634 4,655 1,543 212 1,206 - 5 32 2,288 11,575 Note: Vehicle estimates are for an evacuation of the entire EPZ (Region R03)

KLD Engineering, P.C.

Palo Verde 6-8 6-8 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

7 GENERAL POPULATION EVACUATION TIME ESTIMATES (ETE)

This section presents the ETE results of the computer analyses using the DYNEV II System described in Appendices B, C and D. These results cover 52 regions within the Palo Verde EPZ and the 12 Evacuation Scenarios discussed in Section 6.

The ETE for all Evacuation Cases are presented in Table 7-1 and Table 7-2. These tables present the estimated times to clear the indicated population percentages from the Evacuation Regions for all Evacuation Scenarios. The ETE of the 2-mile region in both staged and un-staged regions are presented in Table 7-3 and Table 7-4. Table 7-5 defines the Evacuation Regions considered.

The tabulated values of ETE are obtained from the DYNEV II System outputs which are generated at 5-minute intervals.

7.1 Voluntary Evacuation and Shadow Evacuation "Voluntary evacuees" are people within the EPZ in Sectors for which an Advisory to Evacuate has not been issued, yet who elect to evacuate. "Shadow evacuation" is the voluntary outward movement of some people from the Shadow Region (outside the EPZ) for whom no protective action recommendation has been issued. Both voluntary and shadow evacuations are assumed to take place over the same time frame as the evacuation from within the impacted Evacuation Region.

The ETE for the Palo Verde EPZ addresses the issue of voluntary evacuees in the manner shown in Figure 7-1. Within the EPZ, 20 percent of people located in Sectors outside of the evacuation region who are not advised to evacuate, are assumed to elect to evacuate. Similarly, it is assumed that 20 percent of those people in the Shadow Region will choose to leave the area.

Figure 7-2 presents the area identified as the Shadow Region. This region extends radially from the plant to cover a region between the EPZ boundary and approximately 15 miles. The population and number of evacuating vehicles in the Shadow Region were estimated using the same methodology that was used for permanent residents within the EPZ (see Section 3.1). As discussed in Section 3.2, it is estimated that a total of 9,546 people reside in the Shadow Region; 20 percent of them would evacuate. See Table 6-4 for the number of evacuating vehicles from the Shadow Region.

Traffic generated within this Shadow Region, traveling away from the PVNGS location, has the potential for impeding evacuating vehicles from within the Evacuation Region. All ETE calculations include this shadow traffic movement.

7.2 Staged Evacuation As defined in NUREG/CR-7002, staged evacuation consists of the following:

1. Sectors comprising the 2 mile region are advised to evacuate immediately.

2. Sectors comprising regions extending from 2 to 5 miles downwind are advised to shelter in-place while the two mile region is cleared.

Palo Verde 7-1 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

3. As vehicles evacuate the 2 mile region, people from 2 to 5 miles downwind continue preparation for evacuation while they shelter.

4. The population sheltering in the 2 to 5 mile region is advised to evacuate when approximately 90% of the 2 mile region evacuating traffic crosses the 2 mile region boundary.

5. Non-compliance with the shelter recommendation is the same as the shadow evacuation pertentage of 20%.

See Section 5.4.2 for additional information on staged evacuation.

7.3 Patterns of Traffic Congestion during Evacuation Figure 7-3 through Figure 7-6 illustrate the patterns of traffic congestion that arise for the case when the entire EPZ (Region R03) is advised to evacuate during the summer, midweek, midday period under good weather conditions (Scenario 1).

Traffic congestion, as the term is used here, is defined as Level of Service (LOS) F. LOS F is defined as follows (HCM 2010, page 5-5):

The HCM uses LOS F to define operations that have either broken down (i.e., demand exceeds capacity) or have exceeded a specified service measure value, or combination of service measure values, that most users would consider unsatisfactory. However, particularly for planning applications where different alternatives may be compared, analysts may be interested in knowing just how bad the LOS F condition is. Several measures are available to describe individually, or in combination, the severity of a LOS F condition:

• Demand-to-capacity ratios describe the extent to which capacity is exceeded during the analysis period (e.g., by 1%, 15%, etc.);

• Duration of LOS F describes how long the condition persists (e.g., 15 min, 1 h, 3 h); and

• Spatial extent measures describe the areas affected by LOS F conditions. These include measures such as the back of queue, and the identification of the specific intersection approaches or system elements experiencing LOS F conditions.

All highway "links" which experience LOS F are delineated in these Figures by a thick red line; all others are lightly indicated. At 30 minutes after the ATE, Figure 7-3 displays congestion on the plant access road caused by employee vehicles evacuating from the plant. Moderate traffic congestion (LOS C) exists on 379th Ave northbound as plant employees are accessing 1-10.

At 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> after the ATE, Figure 7-4 displays moderate levels of traffic congestion (LOS B, C and D) within the EPZ. The plant access road still exhibits LOS F, as approximately 90% of employees have mobilized. State Highway 85 northbound is two lanes, but narrows to a single lane to access 1-10 Eastbound. Congestion begins to build at this bottleneck as shadow evacuees traveling along State Highway 85 merge with those EPZ evacuees traveling along 1-10 Palo Verde 7-2 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Eastbound.

At 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and 10 minutes after the ATE, congestion persists at the plant access road. The on ramp to 1-10 eastbound from State Highway 85 northbound is now experiencing LOS F conditions.

At 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />, 40 minutes, Figure 7-6 displays an EPZ that is essentially clear of traffic congestion, which is well before the completion of the trip-generation (mobilization) time. The lone remnant of congestion is in the Shadow Region to the east, on Interstate 10 after the junction with State Highway 85.

7.4 Evacuation Rates Evacuation is a continuous process, as implied by Figure 7-7 through Figure 7-18. These figures indicate the rate at which traffic flows out of the indicated areas for the case of an evacuation of the full EPZ (Region R03) under the indicated conditions. One figure is presented for each scenario considered.

As indicated in Figure 7-7, there is typically a long "tail" to these distributions. Vehicles begin to evacuate an area slowly at first, as people respond to the ATE at different rates. Then traffic demand builds rapidly (slopes of curves increase). When the system becomes congested, traffic exits the EPZ at rates somewhat below capacity until some evacuation routes have cleared. As more routes clear, the aggregate rate of egress slows since many vehicles have already left the EPZ. Towards the end of the process, relatively few evacuation routes service the remaining demand.

This decline in aggregate flow rate, towards the end of the process, is characterized by these curves flattening and gradually becoming horizontal. Ideally, it would be desirable to fully saturate all evacuation routes equally so that all will service traffic near capacity levels and all will clear at the same time. For this ideal situation, all curves would-retain the same slope until the end - thus minimizing evacuation time. In reality, this ideal is generally unattainable reflecting the spatial variation in population density, mobilization rates and in highway capacity over the EPZ.

Palo Verde 7-3 KLD Engineering. P.C.

Evacuation Time Estimate Rev. 1

7.5 Evacuation Time Estimate (ETE) Results Table 7-1 through Table 7-2 present the ETE values for all 52 Evacuation Regions and all 12 Evacuation Scenarios. Table 7-3 through Table 7-4 present the ETE values for the 2-Mile region for both staged and un-staged keyhole regions downwind to 5 miles. They are organized as follows:

Table ontent ETE represents the elapsed time required for 90 percent of the 7-1 population within a Region, to evacuate from that Region. All Scenarios are considered, as well as Staged Evacuation scenarios.

ETE represents the elapsed time required for 100 percent of the 7-2 population within a Region, to evacuate from that Region. All Scenarios are considered, as well as Staged Evacuation scenarios.

ETE represents the elapsed time required for 90 percent of the 7-3 population within the 2-mile Region, to evacuate from that Region with both Concurrent and Staged Evacuations.

ETE represents the elapsed time required for 100 percent of the 7-4 population within the 2-mile Region, to evacuate from that Region with both Concurrent and Staged Evacuations.

The animation snapshots described above reflect the ETE statistics for the concurrent (un-staged) evacuation scenarios and regions, which are displayed in Figure 7-3 through Figure 7-6.

There is minimal traffic congestion with the EPZ, which results in ETE values which parallel mobilization time; this is reflected in the ETE statistics:

" The 9 0 th percentile ETE for Region R01 (2-mile area) is approximately 45 minutes less (on average) than for Region R02 and R03, primarily because almost all evacuees from Region R01 are employees at PVNGS, who mobilize quickly. As shown in Figure 5-4, 90 percent of employees mobilize in about 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and 10 minutes. The ETE for R01 is about I hour and 25 minutes, on average (not including the special event). The additional 15 minutes (1:25 - 1:10) is the result of the traffic congestion along the plant access road.

" Because of the large number of transients within the 5 to 10-mile region, the 9 0 th percentile ETE for Region R02 (5-mile region) is greater than R03 (entire EPZ) for some scenarios. Transients mobilize more quickly than permanent residents (see Figure 5-4),

enabling R03 to reach the 901h percentile more quickly than R02.

The 1 0 0 th percentile ETE for all Regions and for all Scenarios parallel mobilization time, as well.

This fact implies that the congestion within the EPZ dissipates prior to the end of mobilization, as is displayed in Figure 7-6.

Comparison of Scenarios 6 and 11 in Table 7-1 indicates that the Special Event - an outage at the plant - has a significant impact on the ETE for the 9 0 th percentile. As discussed in Section Palo Verde 7-4 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

3.6, an additional 1,444 vehicles are present at the plant. The additional vehicles increase congestion on the plant access road, extending the ETE of Region R01 by 50 minutes. For those keyhole regions with wind from the north, the ETE is largely dictated by plant employees; therefore the 9 0 th percentile ETE for these regions is also significantly impacted by the additional outage employees with increases of up to 55 minutes. Even with the additional vehicles present for the special event, traffic congestion within the EPZ still clears well before the mobilization time. As a result, the 1 0 0 th percentile ETE are not impacted by the special event.

Comparison of Scenarios 1 and 12 in Table 7-1 indicates that the roadway closure - one lane eastbound on 1-10 from the interchange with S Wintersburg Rd (Exit 98) to the interchange with State Highway 85 (Exit 112) - has no material impact on 9 0 th or 1 00th percentile ETE. The interstate never experiences sustained traffic congestion (LOS F), which means it has excess capacity to service the evacuating traffic demand. The ramps to access the interstate are bottlenecks, not the mainline of the interstate.

7.6 Staged Evacuation Results Table 7-3 and Table 7-4 present a comparison of the ETE compiled for the concurrent (un-staged) and staged evacuation studies. Note that Regions R36 through R52 are the same geographic areas as Regions R02, R04 through R19, respectively.

To determine whether the staged evacuation strategy is worthy of consideration, one must show that the ETE for the 2 Mile region can be reduced without significantly affecting the region between 2 miles and 5 miles. In all cases, as shown in these tables, the ETE for the 2 mile region is unchanged when a staged evacuation is implemented. The reason for this is that no congestion exists beyond 2 miles of PVNGS.

While failing to provide assistance to evacuees from within 2 miles of the PVNGS, staging produces a negative impact on the ETE for those evacuating from within the 5-mile region. A comparison of ETE between Regions R02, R04 through R19 and R36 through R52; reveals that staging retards the 9 0 th percentile evacuation time for those in the 2 to 5-mile area by up to 20 minutes (see Table 7-1). This extending of ETE is due to the delay in beginning the evacuation trip experienced by those who shelter.

In summary, the staged evacuation option provides no benefits to evacuees from within 2 miles and adversely impacts many evacuees located beyond 2 miles from the PVNGS.

Palo Verde 7-5 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

7.7 Guidance on Using ETE Tables The user first determines the percentile of population for which the ETE is sought (The NRC guidance calls for the 9 0 th percentile). The applicable value of ETE within the chosen table may then be identified using the following procedure:

1. Identify the applicable Scenario:

• Season

" Summer

" Winter (also Autumn and Spring)

" Day of Week

" Midweek

• Weekend

" Time of Day

" Midday

" Evening

• Weather Condition

" Good Weather

" Rain

• Special Event

" An Outage at PVNGS

" Road Closure (A lane on 1-10 EB is closed)

• Evacuation Staging

" No, Staged Evacuation is not considered

" Yes, Staged Evacuation is considered While these Scenarios are designed, in aggregate, to represent conditions throughout the year, some further clarification is warranted:

• The conditions of a summer evening (either midweek or weekend) and rain are not explicitly identified in the Tables. For these conditions, Scenarios (2) and (4) apply.

• The conditions of a winter evening (either midweek or weekend) and rain are not explicitly identified in the Tables. For these conditions, Scenarios (7) and (9) for rain apply.

• The seasons are defined as follows:

" Summer assumes that public schools are not in session.

" Winter (includes Spring and Autumn) considers that public schools are in session.

• Time of Day: Midday implies the time over which most commuters are at work or are travelling to/from work.

2. With the desired percentile ETE and Scenario identified, now identify the Evacuation Region:

• Determine the projected azimuth direction of the plume (coincident with the wind direction). This direction is expressed in terms of compass orientation: from S, SSW, SW, ...

• Determine the distance that the Evacuation Region will extend from the nuclear Palo Verde 7-6 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

power plant. The applicable distances and their associated candidate Regions are given below:

0 2 Miles (Region R01)

E To 5 Miles (Region R02, R04 through R19) 0 To 10 Miles (Regions R03, R20 through R35)

• Enter Table 7-5 and identify the applicable group of candidate Regions based on the distance that the selected Region extends from the PVNGS. Select the Evacuation Region identifier in that row, based on the azimuth direction of the plume, from the first column of the Table.

3. Determine the ETE Table based on the percentile selected. Then, for the Scenario identified in Step 1 and the Region identified in Step 2, proceed as follows:

• The columns of Table 7-1 are labeled with the Scenario numbers. Identify the proper column in the selected Table using the Scenario number defined in Step 1.

• Identify the row in this table that provides ETE values for the Region identified in Step 2.

• The unique data cell defined by the column and row so determined contains the desired value of ETE expressed in Hours:Minutes.

Example It is desired to identify the ETE for the following conditions:

• Sunday, August 10th at 4:00 AM.

• It is raining.

• Wind direction is from the northeast (NE).

• Wind speed is such that the distance to be evacuated is judged to be a 2-mile radius and downwind to 10 miles (to EPZ boundary).

• The desired ETE is that value needed to evacuate 90 percent of the population from within the impacted Region.

• A staged evacuation is not desired.

Table 7-1 is applicable because the 90th percentile ETE is desired. Proceed as follows:

1. Identify the Scenario as summer, weekend, evening and raining. Entering Table 7-1, it is seen that there is no match for these descriptors. However, the clarification given above assigns this combination of circumstances to Scenario 4.

2. Enter Table 7-5 and locate the Region described as "Evacuate 2-Mile Radius and Downwind to 10 miles" for wind direction from the NE and read Region R30 in the first column of that row.

3. Enter Table 7-1 to locate the data cell containing the value of ETE for Scenario 4 and Region R30. This data cell is in column (4) and in the row for Region R30; it contains the ETE value of 1:45.

Palo Verde 7-7 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 7-1. Time to Clear the Indicated Area of 90 Percent of the Affected Population Summer Summer Summer Winter Winter Winter Winter Summer Midweek Weekend Mdek Midweek Weekend MdekMidweek Midweek Weekend Weekend Midday Region GoodMiddayF. Evening God°odI'nGoodl Midday Midday Evening Good Midday Midday-F Specal Rodway Reion Weather Ran Weather Weather Weather Ran Weather Ran Weather Event Impact Entire 2-Mile Region, 5-Mile Region, and EPZ Scnrio: (1)5 1:55 2:10 2:10 2:10 1:55 1:55 2:10 2:10 2:10) 2:15 (1:55 Scnrio:

R02 2:0 21 2:15 l2:15 2:15 2, 2:10 2:15 2:1 2:15 2:20 2:10 I R02 2-Mile Region and Keyhole to S Miles R04 1:55 1:55 2:10 2:10 2:10 1:55 1:55 2:10 2:10 2:10 2:15 1:55 R04 R06 1:55 1:5 2:10 2:10 2:10 1:55 1:55 2:10 2:10 2:10 2:15 1:55 R06 ROB 1:50 1:50 2:10 2:10 2:10 1:50 1:50 2:10 2:10 2:10 2:15 1:50 ROB RIO 1:30 1:30 2:00 2:00 2:00 1:30 1:30 2:00 2:00 2:00 2:15 1:30 RIO R12 1:20 1:25 1:35 1:35 1:35 1:25 1:25 1:35 1:35 1:35 2:15 1:20 R12 R13 1:20 1:20 1:35 1:4 1:35 1:20 1:20 1:35 1:40 1:35 2:10 1:20 R13 R14 1:20 1:20 1:35 1:40 1:35 1:20 1:20 1:35 1:40 1:35 2:10 1:20 R14 R16 1:45 1:45 2:05 2:05 2:05 1:45 1:45 2:05 2:05 2:35 2:00 1:20 R16 R18 1:45 1:4 2:05 2:05 2:05 1:45 1:45 2:05 2:05 2:05 2:15 1:45 R18 Palo Verde 7-8 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Summer Summer Summer Winter Winter Winter Winter Summer Midweek Midweek Midweek Weekend Weekend Weekend Midweek Weekend Weekend Weekend Midweek Midweek Midday Midday Evening Midday Midday Evening Midday Midday Rain SRood Good Rain Good GoodI Good Rain Good Special Roadway Region Weather Weather Rain Weather Weather Rain Weather I Weather Event Impact 2-Mile Region and Keyhole to EPZ Boundary R20 1:50 1:50 1:40 1:40 2:00 1:45 1:45 1:35 1:35 1:55 2:15 1:50 R20 R21 1:55 1:55 1:40 1:40 2:05 1:50 1:50 1:40 1:40 2:05 2:15 1:55 R21 11- 1:55 1:5 1:45 1:45 2:05 1:55 1:55 1:45 1:45 2:05 2:15 2:00 R22 R23 2:00 2:00 1:55 1:55 2:10 2:00 2:00 1:55 1:55 2:10 2:15 2:05 R23 R24 2:05 2:05 2:00 2:00 2:10 2:00 2:00 2:00 2:00 2:10 2:20 2:05 R24 R25 2:00 2:00 2:10 2:15 2:10 2:00 2:00 2:10 2:15 2:10 2:20 2:00 R25 R26 1:50 1:50 2:10 2:10 2:10 1:45 1:45 2:10 2:10 2:10 2:20 1:50 R26 R27 1:30 1:35 2:05 2:05 2:05 1:30 1:35 2:05 2:05 2:05 2:20 1:30 R27 R28 1:25 1:25 1:50 1:50 1:50 1:25 1:25 1:50 1:50 1:50 2:15 1:25 R2 R29 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:15 1:20 R29 R30 1:20 1:20 1:45 1:45 1:45 1:20 1:20 1:45 1:45 1:45 2:00 1:20 R30 R31 1:20 1:20 1:45 1:45 1:45 1:20 1:25 1:45 1:45 1:45 2:10 1:20 R31 R32 1:35 1:40 2:05 2:05 2:05 1:35 1:35 2:00 2:00 2:00 2:15 1:35 R32 R33 1:30 1:30 1:25 1:25 1:45 1:30 1:30 1:25 1:25 1:40 2:10 1:30 R33 R34 1:45 145 1:35 1:35 1:55 1:40 1:40 1:35 1:35 1:50 2:10 1:4 R34 R3S 1:45 1:45 1:35 1:35 1:55 1:45 1:45 1:35 1:35 1:50 2:10 1:45 R35 Palo Verde 7-9 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Summer Summer Summer Winter Winter Winter Winter Summer Midweek Weekend Midweek Midweek Weekend Midweek Midweek Midweek Sc na io (i] [(2)

(3) (4 Weekend (5 (6)] (7) (8 (9)

Weekend (1 )(1) -(12) Scen rio Midday Midday Evening Midday Midday Evening Midday Midday Region G Rain Good Rain Good Good I Rain Good Rain Good Special Roadway Region Weather Weather I Weather Weather Weather Weather Event Impact Staged Evacuation Mile Region and Keyhole to 5 Miles R37 2:05 2:05 2:10 2:10 2:10 2:05 2:05 2:10 2:10 2:10 2:15 2:05 R37 R38 2:05 2:05 2:10 2:15 2:10 2:05 2:05 2:10 2:15 2:10 2:20 2:05 R38 R39 2:00 2:05 2:10 2:10 2:10 2:00 2:00 2:10 2:10 2:10 2:15 2:00 R39 R41 1:55 1:55 2:10 2:10 2:10 1:55 1:55 2:10 2:10 2:10 2:15 1:55 R41 R43 1:50 1:50 2:05 2:05 2:05 1:50 1:50 2:05 2:05 2:05 2:15 1:50 R43 R45 1:25 1:25 1:55 1:55 1:55 1:25 1:25 1:55 1:55 1:55 2:15 1:25 R45 R47 1:20 1:20 1:55 1:55 1:55 1:20 1:25 1:55 1:55 1:55 2:15 1:20 R47 R49 1:20 1:20 1:55 1:55 1:55 1:20 1:25 1:55 1:55 1:55 2:15 1:20 R49 R51 2:00 2:00 2:10 2:10 2:10 2:00 2:00 2:10 2:10 2:10 2:15 2:00 R51 R52~~ ~

2.) ~ :1 ~ ~

0 ~ 2:0 ~ ~ 20 ~ 2:00~20 ~ ii* :1 :0 2:0 21 20 5 Rev. 1 Palo Verde 7-10 KLD Engineering, P.C.

Evacuation Time Estimate Time Estimate Rev. 1

Table 7-2. Time to Clear the Indicated Area of 100 Percent of the Affected Population Summer Summer Summer Winter Winter Winter Winter Summer Midweek Weekend Mdek Midweek Weekend MdekMidweek Midweek Weekend Weekend Sce*nario (1) [(2] [(3] (4) (S) (6) (7) (8) (9) (10) (1) (1 ) ce aro Region -iooo0I Rain Midday Weather GoodL Ran Oood Miday Weather Evening Weather GoodMday Weather I Rain GoodMidday IWeather Ran oodSpeia Roadway RIn Evening Weather Midday Event Midday Impact Region Entire 2-Mile Region, 5-Mile Region, and EPZ R01 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 ~5:00 5:00 5:00 5:0 R11 R02 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R02 R03 5:10 5:101 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:SA 5:10 R0 2-Mile Region and Keyhole to 5 Miles I

R04 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R04 OSs 505 5:05 5:05 505 5:05 5:05 5:05 5:05 505 5:05 5:05 5:05 R06 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R06 R07 505 5:05 5:05 5:05 5:05 5:05 5:05 5:05 505 5:05 505 505 R07 RO0 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 RO9 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 505 5:05 5:05 5:05 RIO Rio 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R09 5:05 5:05 5:05 5:05 505 5:05 5:05 5:05 5:05 505 5:05 505 R12 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R12 R13 5:05 5:05 5:05 5:0 5:05 5:05 5:05 5:05 505 5:05 5:05 5:05 R13 R14 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R14 R15 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R15 R16 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R16 R17 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R17 R18 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R18 R19 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 Palo Verde 7-11 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Summer Summer Summer Winter Winter Winter Winter Summer Midweek Midweek Midweek Weekend Weekend Midweek Weekend Weekend Midweek Midweek Weekend Weekend Region GooMidday Rein ---

Weather d Rain GoodMiddayI Goo Weather Rain Rain Evening Good Weather Good Weather 1o 1~~

Midday I an Good Midday Weather I an ain Evening Good Weather Midday Special Event Midday Roadway mact Region 2-Mile Region and Keyhole to EPZ Boundary R21 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 R21 R21 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 R21 R22 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 R22 R23 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 R23 R24 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 R24 R25 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 R25 26:10 :10 5:10 5:10 5:10 :10 0 5:0 :1 R R32 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 R33 R28 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 R28 R35 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 5:10 R35 Palo Verde 7-12 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Summer Summer Summer Winter Winter Winter Winter Summer Midweek Midweek Midweek Weekend Weekend Midweek Weekend Weekend Midweek Midweek Midday Midday Evening Midday Midday Evening Midday Midday ood Raiion Rain ood Rain Good Good Good Good Special Roadway Region Weather Weather Weather Weather Rain Weather In Weather Event Impact Staged Evacuation Mile Region and Keyhole to 5 Miles R36 5:05 5:5 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R37 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R37 R37 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R38 R39 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R39 R40 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R40 R41 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R41 R42 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R42 R43 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R43 R45 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R44 R45 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R45 R46 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R46 R47 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R47 R48 505 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 505 R R49 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R49 RSO 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 USO R51 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 R51 R52 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 5:05 RS2 Palo Verde 7-13 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 7-3. Time to Clear 90 Percent of the 2-Mile Area within the Indicated Region Summer Summer Summer Winter Winter Winter Winter Summer Midweek Midweek Midweek Weekend wek Weekend Midweek Weekend Weekend Weekend Midweek Midweek

-cenario: -1 -2) -] (6--4) -( -7) -0enro Midday Midday Evening Midday Midday Evening Midday Midday Region Good [ Good Good Good in Good Good Special Roadwa Region Weather Weather ain Weather Weather Weather R Weather Event Impact _ _

Unstaged Evacuation Mile Region and Keyhole to S-Miles R02 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R02 R* 2 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R02 R04 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R04 R05 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R05 R06 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R06 R07 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R07 R1S 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R1I R15 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R15 R13 1:20 1:20 1:35 1:35 1:35 1:20 ..... 1:35 1:35 1:35 2:10 1:20 R13 R17 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R1N Palo Verde 7-14 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Summer Summer Summer Winter Winter Winter Winter Summer Midweek Midweek Midweek Weekend Weekend Weekend Midweek Weekend Weekend Weekend Midweek Midweek Midday Midday Evening Midday Midday Evening Midday Midday Region Goo i Good [ Good I GSindGoSpecial Roadway Region Weather Weather Weather Weather I I Weather a Weather Event Impact Staged Evacuation Mile Region and Keyhole to S-Miles R36 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R36 R37 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R37 R38 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R39 R39 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R39 R4O 1.20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R4 R41 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R41 R42 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R4 R43 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R43 R44 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R41 R45 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R45 R46 1:20 120 1:5 1:35 1:35 1:20 120 1:35 1:35 1:35 2:10 1:20 R46 R47 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R47 R49 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R48 R49 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R49 R50 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R51 R51 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35 1:35 1:35 2:10 1:20 R51 R52 1:20 1:20 1:35 1:35 1:35 1:20 1:20 1:35~ 1:35 1:35 2:10 1:20 BR2 KLD Engineering, P.C.

Palo Verde 7-15 7-15 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 7-4. Time to Clear 100 Percent of the 2-Mile Area within the Indicated Region Summer Summer Summer Winter Winter Winter Winter Summer Midweek Midweek Midweek Weekend Weekend Weekend Midweek Weekend Weekend Midweek WeekendII Midweek

-e-nario: - (5 ---

(6) (7) ( -3 -4) i()11 Sna Midday Midday Evening Midday Midday Evening Midday Midday Reaion R Rain Good Good n Good Good Special Roadway Region WahWetr Weather RiWeather Weather Weather Event Impact Unstaged Evacuation Mile Region and Keyhole to 5-Miles R02 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R02 R02 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 RO5 R07 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R07 R09 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R09 R06 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 500 5:00 5:00 R06 R13 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R13 ROB 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 ROB R17 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R17 RI9 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R19 Rev. 1 Palo Verde 7-16 KLD Engineering, P.C.

Time Estimate Evacuation Time Estimate Rev. 1

Summer Summer Summer Winter Winter Winter Winter Summer Midweek Midweek Midweek Weekend Weekend Weekend Midweek Weekend Weekend Weekend Midweek Midweek Jc--] (2- ([ -3) (7) -(- - [12) S -6)

Midday Midday Evening Midday Midday Evening Midday Midday Region Good Rain Good Too Good] Good Special Roadway Region weather Weather Riood GRain Goo I Rain Ran aI Weather Weather I Weather I Weather Event Impact Staged Evacuation Mile Region and Keyhole to S-Miles R36 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R36 R37 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R37 R38 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R38 R39 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R39 R40 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R40 R41 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R41 R42 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R42 R43 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R43 R44 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R44 R45 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 14S R46 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R46 R47 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R47 R48 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R48 R49 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R49 R50 5:00 5-00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 S RS1 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 R51 R52 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:00 5:0)0 5:00 _____

KLD Engineering, P.C.

Palo Verde 7-17 7-17 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 7-5. Description of Evacuation Regions Wind Direction I M N P Q R Region From: 2-515-WI2-515-WIZ-SIS-W!2-SIS-W!2-5 S-M R04 S ROS R06 ROT R011 SSW SW WSW E~~]IEEEEIE LLI - LEE WTI~LL LI - I RO9 W

R.OB WNW Rio NW Rll NNW R12 N R13 NNE R14 NE R15 ENE R16 E R17 ESE - - - -~

RIB R19 SE SSE ILL LI - EI~E1L1~L LI - t-Evacuate 2-Mile Radius and Downwind to 10 miles Wind _______ _______

_______ _______ sector _ __ __ _

Direction Region From:

S R20 SSW Rfl SW R23 WSW R24 W WNW NNW RUS NNW R29 N NNE a3 NE ENE E

ESE SE SSE

... . . . . . . . . . . ..M , ,I'heItIIP- ' . .' 1I.. . .. . ... . . I Palo Verde 7-18 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Staged Evacuation Mile Radius Evacuates, then Evacuate Downwind to 5 Miles I-Wind Sector Direction A B I C I D I E I F I G I H I J I K I L I M I N I P I Q I R From: PVNGS 2 Mile -101 2-Si S-101 2-SiS5-101 2-Si S-101 2-SIS12S 5-1012-S515-101 2-SI S-101 2-SIS-1012-5IlS-1012-51-01S 5-1012-Si S-101 2-SiS5-1012-51 S-101 2-51 5-1012-Si51&I MINES 5-Mile Radius R36 S SSW R39 SW R40 WSW R41 W R42 WNW RM3 NW R44 NNW RM5 N M46 NNE R47 NE R48 ENE R49 E ESE l m m m m m SE ii 112Th 1 IZIZLI IZIIIZIIIIZK 12- I tzm: -

m m Sector(s) Shelter-in-Place Rev. 1 Palo Verde 7-19 KLD Engineering, P.C.

Evacuation Time Estimate Evacuation Time Estimate Rev. 1

mu - - J I* PtW I*essbw*$0S Ws EvmuaW: IS wwfmC t fwO - -a :0*w 0 eu'-"

Figure 7-1. Voluntary Evacuation Methodology KLD Engineering, P.C.

Palo Verde 7-20 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Figure 7-2. Palo Verde Shadow Region 7-21 KLD Engineering, P.C.

Palo Verde 7-21 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Figure 7-3. Congestion Patterns at 30 Minutes after the Advisory to Evacuate Palo Verde 7-22 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Figure 7-4. Congestion Patterns at 1 Hour after the Advisory to Evacuate Palo Verde 7-23 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Figure 7-5. Congestion Patterns at 1 Hour 10 Minutes after the Advisory to Evacuate Palo Verde 7-24 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Figure 7-6. Congestion Patterns at I Hour 40 Minutes after the Advisory to Evacuate Palo Verde 7-25 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Evacuation Time Estimates Summer, Midweek, Midday, Good (Scenario 1)

Mile Region - S-Mile Region - Entire EPZ 0 90% 0 100%

14 12

= 10 UC 8 o!7 U-2 4 2

0 0 30 60 90 120 150 180 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-7. Evacuation Time Estimates - Scenario 1 for Region R03 Evacuation Time Estimates Summer, Midweek, Midday, Rain (Scenario 2)

- 2-Mile Region - 5-Mile Region - Entire EPZ 0 90% 0 100%

14 12

.C 10 4'..

M S8 LU-

.~ 4 2Ne 01 0 30 60 90 120 150 180 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-8. Evacuation Time Estimates - Scenario 2 for Region R03 Palo Verde 7-26 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Evacuation Time Estimates Summer, Weekend, Midday, Good (Scenario 3)

Mile Region - 5-Mile Region - Entire EPZ

• 90% 0 100%

14 12 ta

_4 10

~C 8 06 0 30 60 90 120 150 180 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-9. Evacuation Time Estimates - Scenario 3 for Region R03 Evacuation Time Estimates Summer, Weekend, Midday, Rain (Scenario 4)

- 2-Mile Region - 5-Mile Region - Entire EPZ

• 90%

• 100%

14 12

~ _10

~c 8 f 0 6 ai 44 2

0 0 30 60 90 120 150 180 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-10. Evacuation Time Estimates - Scenario 4 for Region R03 KLD Engineering, P.C.

Palo Verde Palo Verde 7-27 KLD Engineering, P.C.

Evacuation Time Estimate Rev. I

Evacuation Time Estimates Summer, Midweek, Weekend, Evening, Good (Scenario 5)

- 2-Mile Region - 5-Mile Region - Entire EPZ 0 90% 0 100%

14 12 10

~C 8 in 4 2

0 0 30 60 90 120 150 180 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-11. Evacuation Time Estimates - Scenario 5 for Region R03 Evacuation Time Estimates Winter, Midweek, Midday, Good (Scenario 6)

Mile Region Mile Region - Entire EPZ 0 90% 0 100%

14 12

.C 10

-0 8 to U'-4 2

0 0 30 60 90 120 150 180 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-12. Evacuation Time Estimates - Scenario 6 for Region R03 Palo Verde 7-28 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Evacuation Time Estimates Winter, Midweek, Midday, Rain (Scenario 7)

- 2-Mile Region - 5-Mile Region - Entire EPZ

• 90% 0 100%

14 12 to S 10

~C 8

~O6

• 4 2

0 0 30 60 90 120 150 180 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-13. Evacuation Time Estimates - Scenario 7 for Region R03 Evacuation Time Estimates Winter, Weekend, Midday, Good (Scenario 8)

Mile Region -- 5-Mile Region - Entire EPZ

• 90% 0 100%

14 12

,~10 4U 06

• 4 2

0 0 30 60 90 120 150 1S8 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-14. Evacuation Time Estimates - Scenario 8 for Region R03 Palo Verde 7-29 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Evacuation Time Estimates Winter, Weekend, Midday, Rain (Scenario 9)

- 2-Mile Region - 5-Mile Region - Entire EPZ 0 90% 0 100%

14 12 CA C 10 C 8 0 6 4

2 All 0

0 30 60 90 120 150 180 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-15. Evacuation Time Estimates - Scenario 9 for Region R03 Evacuation Time Estimates Winter, Midweek, Weekend, Evening, Good (Scenario 10)

Mile Region - 5-Mile Region - Entire EPZ 0 90% 0 100%

14 12 lO

,~10

~c 8 LU

• , 6 2 Z 0

0 30 60 90 120 150 180 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-16. Evacuation Time Estimates - Scenario 10 for Region R03 Palo Verde 7-30 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Evacuation Time Estimates Winter, Midweek, Midday, Good, Special Event (Scenario 11)

Mile Region - 5-Mile Region - Entire EPZ 0 90% 0 100%

14 12 S 10 C 8 U.-

i 4 2

0 0 30 60 90 120 150 180 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-17. Evacuation Time Estimates - Scenario 11 for Region R03 Evacuation Time Estimates Summer, Midweek, Midday, Good, Roadway Impact (Scenario 12)

- 2-Mile Region Mile Region - Entire EPZ

• 90% 0 100%

14 12 10 c8

'U V U1 6 Z~ 4 2 01 01 0 30 60 90 120 150 180 210 240 270 300 330 Elapsed Time After Evacuation Recommendation (min)

Figure 7-18. Evacuation Time Estimates - Scenario 12 for Region R03 Rev. 1 Palo Verde 7-31 KLD Engineering, P.C.

Evacuation Time Estimate Evacuation Time Estimate Rev. 1

8 TRANSIT-DEPENDENT AND SPECIAL FACILITY EVACUATION TIME ESTIMATES This section details the analyses applied and the results obtained in the form of evacuation time estimates for transit vehicles. The demand for transit service reflects the needs of three population groups: (1) residents with no vehicles available; (2) residents of special facilities such as schools; and (3) access and functional needs population.

These transit vehicles mix with the general evacuation traffic that is comprised mostly of "1passenger cars" (pc's). The presence of each transit vehicle in the evacuating traffic stream is represented within the modeling paradigm described in Appendix D as equivalent to two pc's.

This equivalence factor represents the longer size and more sluggish operating characteristics of a transit vehicle, relative to those of a pc.

Transit vehicles must be mobilized in preparation for their respective evacuation missions.

Specifically:

• Bus drivers must be alerted

• They must travel to the bus depot

• They must be briefed there and assigned to a route or facility These activities consume time. Based on discussion with the offsite agencies, it is estimated that bus mobilization time will average approximately 10 minutes extending from the Advisory to Evacuate as buses and drivers remain at the schools throughout the day.

During this mobilization period, other mobilization activities may be taking place. One of these is the action taken by parents, neighbors, relatives and friends to pick up children from school prior to the arrival of buses, so that they may join their families. Virtually all studies of evacuations have concluded that this "bonding" process of uniting families is universally prevalent during emergencies and should be anticipated in the planning process. The current public information disseminated to residents of the Palo Verde EPZ indicates that schoolchildren will be evacuated to reception and care centers (RCC) if an evacuation is ordered, and that parents should pick schoolchildren up at the designated RCC. As discussed in Section 2, this study assumes a fast breaking general emergency. Therefore, children are evacuated to a RCC. Picking up children at school could add to traffic congestion at the schools, delaying the departure of the buses evacuating schoolchildren, which may have to return in a subsequent "wave" to the EPZ to evacuate the transit-dependent population. This report provides estimates of buses under the assumption that no children will be picked up by their parents (in accordance with NUREG/CR-7002), to present an upper bound estimate of buses required.

The procedure for computing transit-dependent ETE is to:

• Estimate demand for transit service

• Estimate time to perform all transit functions

• Estimate route travel times to the EPZ boundary and to the RCC Palo Verde 8-1 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

8.1 Transit Dependent People Demand Estimate The telephone survey (see Appendix F) results were used to estimate the portion of the population requiring transit service:

• Those persons in households that do not have a vehicle available.

• Those persons in households that do have vehicle(s) that would not be available at the time the evacuation is advised.

In the latter group, the vehicle(s) may be used by a commuter(s) who does not return (or is not expected to return) home to evacuate the household.

Table 8-1 presents estimates of transit-dependent people. Note:

Estimates of persons requiring transit vehicles include schoolchildren. For those evacuation scenarios where children are at school when an evacuation is ordered, separate transportation is provided for the schoolchildren. The actual need for transit vehicles by residents is thereby less than the given estimates. However, estimates of transit vehicles are not reduced when schools are in session.

It is reasonable and appropriate to consider that many transit-dependent persons will evacuate by ride-sharing with neighbors, friends or family. For example, nearly 80 percent of those who evacuated from Mississauga, Ontario who did not use their own cars, shared a ride with neighbors or friends. Other documents report that approximately 70 percent of transit dependent persons were evacuated via ride sharing. We will adopt a conservative estimate that 50 percent of transit dependent persons will ride share, in accordance with NUREG/CR-7002.

The estimated number of bus trips needed to service transit-dependent persons is based on an estimate of average bus occupancy of 30 persons at the conclusion of the bus run. Transit vehicle seating capacities typically equal or exceed 60 children on average (roughly equivalent to 40 adults). If transit vehicle evacuees are two thirds adults and one third children, then the number of "adult seats" taken by 30 persons is 20 + (2/3 xlO) = 27. On this basis, the average load factor anticipated is (27/40) x 100 = 68 percent. Thus, if the actual demand for service exceeds the estimates of Table 8-1 by 50 percent, the demand for service can still be accommodated by the available bus seating capacity.

[20 + (2 x 10)] + 40 x 1.5= 1.00 Table 8-1 indicates that transportation must be provided for 455 people. Therefore, a total of 16 bus runs are required to transport this population to RCC.

Palo Verde 8-2 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

To illustrate this estimation procedure, we calculate the number of persons, P, requiring public transit orride-share, and the number of buses, B, required for the Palo Verde EPZ:

n P = No. of HH x -- (% HH with i vehicles) x [(Average HH Size) - i]} x AiCi i=O

Where, A = Percent of households with commuters C = Percent of households who will not await the return of a commuter P = 4,301 x [0.026 x 1.31 + 0.276 x (2.30 - 1) x 0.61 x 0.58 + 0.430 x (2.96 - 2) x (0.61 x 0.58)2] = 4,301 x 0.212 = 910 B = (0.5 x P) + 30 = 16 These calculations are explained as follows:

All members (1.31 avg.) of households (HH) with no vehicles (2.6%) will evacuate by public transit or ride-share. The term 4,301 (number of households) x 0.026 x 1.31, accounts for these people.

The members of HH with 1 vehicle away (27.6%), who are at home, equal (2.30-1).

The number of HH where the commuter will not return home is equal to (4,301 x 0.276 x 1.30 x 0.61 x 058), as 61% of EPZ households have a commuter, 58% of which would not return home in the event of an emergency. The number of persons who will evacuate by public transit or ride-share is equal to the product of these two terms.

The members of HH with 2 vehicles that are away (43.0%), who are at home, equal (2.96 - 2). The number of HH where neither commuter will return home is equal to 4,301 x 0.430 x 0.96 x (0.61 x 0.58)2. The number of persons who will evacuate by public transit or ride-share is equal to the product of these two terms (the last term is squared to represent the probability that neither commuter will return).

Households with 3 or more vehicles are assumed to have no need for transit vehicles.

The total number of persons requiring public transit is the sum of such people in HH with no vehicles, or with 1 or 2 vehicles that are away from home.

Palo Verde 8-3 KLD Engineering, P.C.

Evacuation Time Estimate Rev. I

8.2 School Population - Transit Demand Table 8-2 presents the school population and transportation requirements for the direct evacuation of all schools within the EPZ for the 2011-2012 school year. This information was provided by the Maricopa County DEM. The column in Table 8-2 entitled "Buses Required" specifies the number of buses required for each school under the following set of assumptions and estimates:

• No students will be picked up by their parents prior to the arrival of the buses.

• While many high school students commute to school using private automobiles (as discussed in Section 2.4 of NUREG/CR-7002), the estimate of buses required for school evacuation do not consider the use of these private vehicles.

• Bus capacity, expressed in students per bus, is set to 70 for primary schools and 50 for middle and high schools.

• Those staff members who do not accompany the students will evacuate in their private vehicles.

• No allowance is made for student absenteeism, typically 3 percent daily.

• Palo Verde Elementary School will be evacuated (even though it is in the Shadow Region) in the event of an emergency at the PVNGS according to the Maricopa County emergency plan.

The need for buses would be reduced by any high school students who have evacuated using private automobiles (if permitted by school authorities). Those buses originally allocated to evacuate schoolchildren that are not needed due to children being picked up by their parents, can be gainfully assigned to service other facilities or those persons who do not have access to private vehicles or to ride-sharing.

Table 8-3 presents a list of the RCC for each school in the EPZ. Figure 10-1 maps each of the RCC. Schools were routed to the closest RCC. Students will be transported to these centers where they will be subsequently retrieved by their respective families.

8.3 Evacuation Time Estimates for Transit Dependent People EPZ bus resources are assigned to evacuating schoolchildren (if school is in session at the time of the ATE) as the first priority in the event of an emergency. In the event that the allocation of buses dispatched from the depots to the various facilities and to the bus routes is somewhat "inefficient", or if there is a shortfall of available drivers, then there may be a need for some buses to return to the EPZ from the RCC after completing their first evacuation trip, to complete a "second wave" of providing transport service to evacuees. For this reason, the ETE for the transit-dependent population will be calculated for both a one wave transit evacuation and for two waves. Of course, if the impacted Evacuation Region is other than R03 (the entire EPZ),

then there will likely be ample transit resources relative to demand in the impacted Region and this discussion of a second wave would likely not apply.

When school evacuation needs are satisfied, subsequent assignments of buses to service the transit-dependent should be sensitive to their mobilization time. Clearly, the buses should be Palo Verde 8-4 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

dispatched after people have completed their mobilization activities and are in a position to board the buses when they arrive at the pick-up points.

Evacuation Time Estimates for transit trips were developed using both good weather and adverse weather conditions. Figure 8-1 presents the chronology of events relevant to transit operations. The elapsed time for each activity will now be discussed with reference to Figure 8-1.

Activity: Mobilize Drivers (A--'B--C)

Mobilization is the elapsed time from the Advisory to Evacuate until the time the buses arrive at the school to be evacuated. It is assumed that for a rapidly escalating radiological emergency with no observable indication before the fact, drivers would likely require 10 minutes to mobilize because the buses and drivers remain on-site during the school day. Mobilization time is slightly longer in adverse weather - 20 minutes when raining.

Activity: Board Passengers (C--1D)

Based on discussions with offsite agencies, a loading time of 15 minutes (20 minutes for rain) for school buses is used.

For multiple stops along a pick-up route (transit-dependent bus routes) estimation of travel time must allow for the delay associated with stopping and starting at each pick-up point. The time, t, required for a bus to decelerate at a rate, "a", expressed in ft/sec/sec, from a speed, "v", expressed in ft/sec, to a stop, is t = v/a. Assuming the same acceleration rate and final speed following the stop yields a total time, T, to service boarding passengers:

T=t+B+t=B+2t=B+--, 2v a

Where B = Dwell time to service passengers. The total distance, "s" in feet, travelled during the deceleration and acceleration activities is: s = v 2/a. If the bus had not stopped to service passengers, but had continued to travel at speed, v, then its travel time over the distance, s,-

would be: s/v = v/a. Then the total delay (i.e. pickup time, P) to service passengers is:

P = T- = B+-

a a Assigning reasonable estimates:

B = 50 seconds: a generous value for a single passenger, carrying personal items, to board per stop v = 25 mph = 37 ft/sec a = 4 ft/sec/sec, a moderate average rate Then, P = 1 minute per stop. Allowing 30 minutes pick-up time per bus run implies 30 stops per run, for good weather. It is assumed that bus acceleration and speed will be less in rain; total loading time is 40 minutes per bus in rain.

Palo Verde 8-5 KLD Engineering, P.C.

Evacuation Time Estimate Rev. I

Activity: Travel to EPZ Boundary (D-4E)

School Evacuation Transportation resources available were provided by the county emergency management agency and are summarized in Table 8-4. Also included in the table is the number of buses needed to evacuate schools, transit-dependent population, and access and functional needs (discussed below in Section 8.4). These numbers indicate that there are sufficient resources to evacuate all transit-dependent population in a single wave.

The buses servicing the schools are ready to begin their evacuation trips at 25 minutes after the advisory to evacuate - 10 minutes mobilization time plus 15 minutes loading time - in good weather. The UNITES software discussed in Section 1.3 was used to define bus routes along the most likely path from a school being evacuated to the EPZ boundary, traveling toward the appropriate RCC. This is done in UNITES by interactively selecting the series of nodes from the school to the EPZ boundary. Each bus route is given an identification number and is written to the DYNEV II input stream. DYNEV computes the route length and outputs the average speed for each 5 minute interval, for each bus route. The specified bus routes are documented in Table 8-5 (refer to the maps of the link-node analysis network in Appendix K for node locations). Data provided by DYNEV during the appropriate timeframe depending on the mobilization and loading times (i.e., 25 minutes after the advisory to evacuate for good weather) were used to compute the average speed for each route, as follows:

Average Speed (flu)

[. L=

length of link i (mi)

Ifl Delay on link i (min. ) +

oni.)xlhr__- length of link i (mi.)

currentspeed on link i M 60 min.

1h 60 min.

X-1 hr.

The average speed computed (using this methodology) for the buses servicing each of the schools in the EPZ is shown in Table 8-6 and Table 8-8Table 8-7 for school evacuation, and in Table 8-9 and Table 8-10 for the transit vehicles evacuating transit-dependent persons, which are discussed later. The travel time to the EPZ boundary was computed for each bus using the computed average speed and the distance to the EPZ boundary along the most likely route out of the EPZ. The travel time from the EPZ boundary to the RCC was computed assuming an average speed of 65 mph and 58.5 mph (10% less) for good weather and rain, respectively.

Speeds were reduced in Table 8-6, Table 8-7, Palo Verde 8-6 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 8-9 and Table 8-10 to 65 mph (58.5 mph for rain - 10% decrease) for those calculated bus speeds which exceed 65 mph, as the school bus speed limit for state routes in Arizona is 65 mph.

Table 8-6 (good weather) and Table 8-7 (rain) present the following evacuation time estimates (rounded up to the nearest 5 minutes) for schools in the EPZ: (1) The elapsed time from the Advisory to Evacuate until the bus exits the EPZ; and (2) The elapsed time until the bus reaches the School RCC. The evacuation time out of the EPZ can be computed as the sum of times associated with Activities A->B->C, C--)D, and D-)E (For example: 10 min. + 15 + 11 = 0:40 for Ruth Fisher Elementary School, with good weather). The evacuation time to the School RCC is determined by adding the time associated with Activity E->F (discussed below), to this EPZ evacuation time.

Evacuation of Transit-DependentPopulation The buses dispatched from the depots to service the transit-dependent evacuees will be scheduled so that they arrive at their respective routes after their passengers have completed their mobilization. As shown in Figure 5-4 (Residents with no Commuters), 85 percent of the evacuees will complete their mobilization when the buses will begin their routes, approximately 105 minutes after the Advisory to Evacuate. Two groups of buses will be dispatched. The start of service on these routes is separated by 20 minute headways, as shown in Table 8-9 and Table 8-10. The use of bus headways ensures that those people who take longer to mobilize will be picked up. Mobilization time is 10 minutes longer in rain to account for slower travel speeds and reduced roadway capacity.

Discussions with local OROs indicated that fixed bus routes would not be used for the evacuation of transit-dependent people within the PVNGS EPZ. Rather, people fill out an Assistance Form in advance or call the local ORO to request transportation assistance. Buses would then be dispatched to individual homes. We modeled this approach by defining a set of routes that would encompass the most populated areas of the EPZ. Buses servicing the transit-dependent evacuees will first travel along their pick-up routes, then proceed out of the EPZ.

Buses will travel along the major routes in the EPZ as described in Table 8-8 and shown graphically in Figure 8-2. These routes are only used in this study for the purpose of computing ETE. No pre-established transit dependent bus routes exist in the county emergency plans. On page ERO-5 of the State of Arizona - Maricopa County Offsite Emergency Response Plans, it states that Maricopa County Department of Emergency Management will respond to the requests for transportation from those residents of the Plume Exposure Pathway EPZ.

As previously discussed, a pickup time of 30 minutes (good weather) is estimated for 30 individual stops to pick up passengers, with an average of one minute of delay associated with each stop. A longer pickup time of 40 minutes was used for rain.

The travel distance along the respective pick-up routes within the EPZ is estimated using the UNITES software. Bus travel times within the EPZ are computed using average speeds computed by DYNEV, using the aforementioned methodology that was used for school evacuation. Table 8-9 and Table 8-10 present the transit-dependent population evacuation Palo Verde 8-7 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

time estimates for each bus route calculated using the above procedures for good weather and rain.

For example, the ETE for the Transit Dependent Bus Route 1 is computed as 105 + 12 + 30 =

2:30 for good weather (rounded up to nearest 5 minutes). Here, 12 minutes is the time to travel 11 miles at 54.6 mph, the average speed output by the model for this route at 105 minutes.

The ETE for a second wave (discussed below) is presented in the event there is a shortfall of available buses or bus drivers, as previously discussed.

Activity: Travel to RCCs (E-MF)

The distances from the EPZ boundary to the RCCs are measured using GIS software along the most likely route from the EPZ exit point to the RCC. For a one-wave evacuation, this travel time outside the EPZ does not contribute to the ETE. For a two-wave evacuation, the ETE for buses must be considered separately, since it could exceed the ETE for the general population.

Assumed bus speeds of 65 mph and 58.5 mph for good weather and rain, respectively, will be applied for this activity.

Activity: Passengers Leave Bus (F-)G)

A bus can empty within 5 minutes. The driver takes a 10 minute break.

Activity: Bus Returns to Route for Second Wave Evacuation (G-KC)

The buses assigned to return to the EPZ to perform a "second wave" evacuation of transit-dependent evacuees will be those that have already evacuated transit-dependent people who mobilized more quickly. The first wave of transit-dependent people depart the bus, and the bus then returns to the EPZ, travels to its route and proceeds to pick up more transit-dependent evacuees along the route. The travel time back to the EPZ is equal to the travel time to the RCC. Bus travel times within the EPZ are computed using average speeds computed by DYNEV, using the aforementioned methodology that was used for school evacuation.

The second-wave ETE for the Transit Dependent Bus Route 1 is computed as follows for good weather:

Bus arrives at RCC at 2:37 in good weather (2:30 to exit EPZ + 7 minute travel time to RCC).

Bus discharges passengers (5 minutes) and driver takes a 10-minute rest: 15 minutes.

Bus returns to EPZ and completes second route: 7 minutes (equal to travel time to RCC) + 12 minutes (11 miles @ 54.6 mph) + 11 mph (11 miles @ 65 mph) = 29 minutes

• Bus completes pick-ups along route: 30 minutes.

• Bus exits EPZ at time 2:30 + 0:07 + 0:15 + 0:29 + 0:30 = 3:55 (rounded up to nearest 5 minutes) after the Advisory to Evacuate.

The ETE for the completion of the second wave for all transit-dependent bus routes are provided in Palo Verde 8-8 KLD Engineering, P.C.

Evacuation Time Estimate Rev. I

Table 8-9 and Table 8-10. The average ETE for a two-wave evacuation of transit-dependent people exceeds the ETE for the general population at the 9 0 th percentile.

The relocation of transit-dependent evacuees from the RCCs to congregate care centers, if the county decides to do so, is not considered in this study.

8.4 Special Needs Population The county emergency management agency has a registration for transit-dependent and access and functional needs persons. Based on data provided by the county, there are an estimated 1,082 people who require transportation assistance to evacuate. 40 households who require a wheelchair capable vehicle have a population of 102. The remaining 980 people are ambulatory and include both the access and functional needs and transit dependent population (computed to be 455 in Section 8.1). Therefore, of the remaining 980 registered transit dependent people, 525 of them are ambulatory access and functional needs who require a bus.

ETE for Access and FunctionalNeeds Persons Table 8-11 summarizes the ETE for access and functional needs people. The table is categorized by type of vehicle required and then broken down by weather condition. The table takes into consideration the deployment of multiple vehicles to reduce the number of stops per vehicle.

It is conservatively assumed that ambulatory and wheelchair bound special needs households are spaced 3 miles apart and bedridden households are spaced 5 miles apart. Van and bus speeds approximate 40 mph between households in good weather (10% slower in rain). A mobilization time of 60 minutes is used (70 minutes for rain). The last HH is assumed to be 5 miles from the EPZ boundary, and the network-wide average speed, capped at 65 mph (58.5 mph for rain), after the last pickup is used to compute travel time. ETE is computed by summing mobilization time, loading time at first household, travel to subsequent households, loading time at subsequent households, and travel time to EPZ boundary. All ETE are rounded to the nearest 5 minutes.

For example, assuming no more than one special needs person per HH implies that 525 ambulatory households need to be serviced. While only 18 buses are needed from a capacity perspective, if 40 buses are deployed to service these special needs HH, then each would require about 14 stops. The following outlines the ETE calculations:

1. Assume 40 buses are deployed, each with about 14 stops, to service a total of 525 HH.

2. The ETE is calculated as follows:

a. Buses arrive at the first pickup location: 60 minutes

b. Load HH members at first pickup: 5 minutes

c. Travel to subsequent pickup locations: 13 @ 9 minutes = 117 minutes

d. Load HH members at subsequent pickup locations: 13 @ 5 minutes = 65 minutes

e. Travel to EPZ boundary at 4:10: 5 minutes (5 miles at 60 mph).

ETE: 60 + 5 + 117 + 65 + 5 = 4:15 rounded up to the nearest 5 minutes Palo Verde 8-9 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

(Subsequent Wave)

• Z C I Dii I E K3 0 Uil I Time A Advisory to Evacuate B Bus Dispatched from Depot C Bus Arrives at Facility/Pick-up Route D Bus Departs for Reception and Care Center E Bus Exits Region F Bus Arrives at Reception and Care Center G Bus Available for "Second Wave" Evacuation Service A-+B Driver Mobilization B->C Travel to Facility or to Pick-up Route C-*D Passengers Board the Bus D--E Bus Travels Towards Region Boundary E--F Bus Travels Towards Reception and Care Center Outside the EPZ F--G Passengers Leave Bus; Driver Takes a Break Figure 8-1. Chronology of Transit Evacuation Operations Palo Verde 8-10 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Figure 8-2. Transit-Dependent Bus Routes Palo Verde 8-11 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 8-1. Transit-Dependent Population Estimates Suve PercentS

• vrg Suve Siz 0H Suve Pecn Suve PecnTtlPepe S Pplto 0niae wit No 0f 06imte wit Iniae No of Pecn5wt o - epe Etmtd Reurn eurn 2011 I3L Z E~9 4,30 2.hcl% 27.6% 43.0%le 61%h Reurn 910iin 50%san 455li Publ%

Palo Verde 8-12 KILD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 8-2. School Population Demand Estimates 5

ý A-8 ýTonopah Valley High School 380 8 F-4 Arlington Elementary School 282 5____

Shadow Palo Verde Elementary School 450 I 7 Region Table 8-3. School Reception and Care Centers Scho Reeto an Car Cente Arlington Elementary School Buckeye Union High School Palo Verde Elementary School Ruth Fisher Elementary School Desert Edge High School Tonopah Valley High School Note: Schools were routed to the closest RCC.

Table 8-4. Summary of Transportation Resources Ruth Fisher Elementary/Tonopah Valley High 21 School Arlington Elementary School 7 Palo Verde Elementary School 8 Valley Metro - Phoenix 481 125 Valley Metro - Tempe 117 RPTA 780 68 Palo Verde 8-13 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 8-5. Bus Route Descriptions 6.4/,155, 1 U,56,J167, 156, 1 Arlington Elementary School 64, 155, 169, 56, 167, 156 2 Ruth Fisher Elementary School 86, 41, 8, 135, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 3 Tonopah Valley High School 86, 41, 8, 135, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 4 Transit Dependent Bus Route 1 25, 186, 185, 184, 116, 83, 44, 157, 45, 46, 47, 187, 151, 49, 63, 64, 155, 169, 56, 167, 156 25, 26, 189, 193, 27, 38, 30, 117, 136, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 Palo Verde 8-14 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 8-6. School Evacuation Time Estimates - Good Weather Table 8-7. School Evacuation Time Estimates - Rain Palo Verde 8-15 -, KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 848. Summary of Transit-Dependent Bus Routes No ofLnt Rot Bue ot ecrpin( .

Salome Hwy eastbound at the intersection of S 379th Ave to Old Hwy 80, then 11.0 6 8 eastbound out of EPZ Salome Hwy at the intersection of S 379th Ave to 411th Ave northbound to 1-10, 21.0 then westbound out of EPZ Table 8-9. Transit-Dependent Evacuation Time Estimates - Good Weather Palo Verde 8-16 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Table 8-10. Transit-Dependent Evacuation Time Estimates - Rain Table S-11. Access and Functional Needs Population Evacuation Time Estimates Palo Verde 8-17 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

9 TRAFFIC MANAGEMENT STRATEGY This section discusses the suggested traffic control and management strategy that is designed to expedite the movement of evacuating traffic. The resources required to implement this strategy include:

• Personnel with the capabilities of performing the planned control functions of traffic guides (preferably, not necessarily, law enforcement officers).

" Traffic Control Devices to assist these personnel in the performance of their tasks. These devices should comply with the guidance of the Manual of Uniform Traffic Control Devices (MUTCD) published by the Federal Highway Administration (FHWA) of the U.S.D.O.T. All state and most county transportation agencies have access to the MUTCD, which is available on-line: http://mutcd.fhwa.dot.gov which provides access to the official PDF version.

• A plan that defines all locations, provides necessary details and is documented in a format that is readily understood by those assigned to perform traffic control.

The functions to be performed in the field are:

1. Facilitate evacuating traffic movements that safely expedite travel out of the EPZ.

2. Discourage traffic movements that move evacuating vehicles in a direction which takes them significantly closer to the power plant, or which interferes with the efficient flow of other evacuees.

The terms "facilitate" and "discourage" are employed rather than "enforce" and "prohibit" to indicate the need for flexibility in performing the traffic control function. There are always legitimate reasons for a driver to prefer a direction other than that indicated. For example:

" A driver may be traveling home from work or from another location, to join other family members prior to evacuating.

• An evacuating driver may be travelling to pick up a relative, or other evacuees.

• The driver may be an emergency worker en route to perform an important activity.

The implementation of a plan must also be flexible enough for the application of sound judgment by the traffic guide.

The traffic management plan is the outcome of the following process:

1. The existing TCPs and ACPs identified by the offsite agencies in their existing emergency plans serve as the basis of the traffic management plan, as per NUREG/CR-7002.

2. The existing TCPs and ACPs and how they were applied in this study are discussed in Appendix G.

3. Computer analysis of the evacuation traffic flow environment (see Figures 7-3 through 7-6). As discussed in Section 7.3, congestion within the EPZ is clear by 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and 40 minutes after the ATE. The only intersection that experiences LOS F is the plant access road intersections with S 379th Ave (Wintersburg Rd). Based on the limited traffic congestion within the EPZ, no additional TCPs or ACPs are identified as a result of this Palo Verde 9-1 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

study. The existing traffic management plans are adequate.

The use of Intelligent Transportation Systems (ITS) technologies (if available) can reduce manpower and equipment needs, while still facilitating the evacuation process. Dynamic Message Signs (DMS) can be placed within the EPZ to provide information to travelers regarding traffic conditions, route selection, and reception center information. DMS can also be placed outside of the EPZ to warn motorists to avoid using routes that may conflict with the flow of evacuees away from the power plant. Highway Advisory Radio (HAR) can be used to

'broadcast information to evacuees en route through their vehicle stereo systems. Automated Traveler Information Systems (ATIS) can also be used to provide evacuees with information.

Internet websites can provide traffic and evacuation route information before the evacuee begins their trip, while on board navigation systems (GPS units), cell phones, and pagers can be used to provide information en route. These are only several examples of how ITS technologies can benefit the evacuation process. Consideration should be given that ITS technologies be used to facilitate the evacuation process, and any additional signage placed should consider evacuation needs.

The ETE analysis treated all controlled intersections that are existing TCP locations in the offsite agency plans as being controlled by actuated signals.

Chapters 2N and 5G, and Part 6 of the 2009 MUTCD are particularly relevant and should be reviewed during emergency response training.

The ETE calculations reflect the assumption that all "external-external" trips are interdicted and diverted after 45 minutes have elapsed from the ATE.

All transit vehicles and other responders entering the EPZ to support the evacuation are assumed to be unhindered by personnel manning ACPs and TCPs.

Study Assumptions 5 and 6 in Section 2.3 discuss ACP and TCP staffing schedules and operations.

Palo Verde 9-2 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

10 EVACUATION ROUTES Evacuation routes are comprised of two distinct components:

• Routing from sectors being evacuated tothe boundary of the Evacuation Region and thence out of the EPZ.

" Routing of transit-dependent evacuees from the EPZ boundary to reception and care.

centers.

Evacuees will select routes within the EPZ in such a way as to minimize their exposure to risk.

This expectation is met by the DYNEV II model routing traffic away from the location of the plant, to the extent practicable. The DTRAD model satisfies this behavior by routing traffic so as to balance traffic demand relative to the available highway capacity to the extent possible.

See Appendices B through D for further discussion.

The routing of transit-dependent evacuees from the EPZ boundary to reception and care centers is designed to minimize the amount of travel outside the EPZ, from the points where these routes cross the EPZ boundary.

Figure 10-1 presents a map showing the general population and school reception and care centers for evacuees. The major evacuation routes for the EPZ are presented in Figure 10-2.

It is assumed that all school evacuees will be taken to the appropriate reception and care center and subsequently picked up by parents or guardians. Transit-dependent evacuees are transported to the nearest reception and care center. This study does not consider the transport of evacuees from reception and care centers to congregate care centers, if the county does make the decision to relocate evacuees.

Palo Verde 10-1 10-1 KLD Engineering, P.c.

KLD Engineering, P.C.

Evacuation Time Estimate Rev. I

Figure 10-1. Reception and Care Centers KLD Engineering, P.C.

Palo Verde 10-2 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

Figure 10-2. Major Evacuation Routes Palo Verde 10-3 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

11 SURVEILLANCE OF EVACUATION OPERATIONS There is a need for surveillance of traffic operations during the evacuation. There is also a need to clear any blockage of roadways arising from accidents or vehicle disablement. Surveillance can take several forms.

1. Traffic control personnel, located at Traffic Control and Access Control points, provide fixed-point surveillance.

2. Ground patrols may be undertaken along well-defined paths to ensure coverage of those highways that serve as major evacuation routes.

3. Aerial surveillance of evacuation operations may also be cbnducted using helicopter or fixed-wing aircraft, if available.

4. Cellular phone calls (if cellular coverage exists) from motorists may also provide direct field reports of road blockages.

These concurrent surveillance procedures are designed to provide coverage of the entire EPZ as well as the area around its periphery. It is the responsibility of the County to support an emergency response system that can receive messages from the field and be in a position to respond to any reported problems in a timely manner. This coverage should quickly identify, and expedite the response to any blockage caused by a disabled vehicle.

Tow Vehicles In a low-speed traffic environment, any vehicle disablement is likely to arise due to a low-speed collision, mechanical failure or the exhaustion of its fuel supply. In any case, the disabled vehicle can be pushed onto the shoulder, thereby restoring traffic flow. Past experience in other emergencies indicates that evacuees who are leaving an area often perform activities such as pushing a disabled vehicle to the side of the road without prompting.

While the need for tow vehicles is expected to be low under the circumstances described above, it is still prudent to be prepared for such a need. Consideration should be given that tow trucks with a supply of gasoline be deployed at strategic locations within, or just outside, the EPZ. These locations should be selected so that:

" They permit access to key, heavily loaded, evacuation routes.

" Responding tow trucks would most likely travel counter-flow relative to evacuating traffic.

Consideration should also be given that the state and local emergency management agencies encourage gas stations to remain open during the evacuation.

Palo Verde 11-1 KLD Engineering, P.C.

Evacuation Time Estimate Rev. 1

12 CONFIRMATION TIME It is necessary to confirm that the evacuation process is effective in the sense that the public is complying with the Advisory to Evacuate. The EPZ county radiological emergency plan does not discuss a procedure for confirming evacuation. Should procedures not already exist, the following alternative or complementary approach is suggested.

The suggested procedure employs a stratified random sample and a telephone survey. The size of the sample is dependent on the expected number of households that do not comply with the Advisory to Evacuate. We believe it is reasonable to assume, for the purpose of estimating sample size that at least 80 percent of the population within the EPZ will comply with the Advisory to Evacuate. On this basis, an analysis could be undertaken (see Table 12-1) to yield an estimated sample size of approximately 300.

The confirmation process should start at about 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> after the Advisory to Evacuate, which is when approximately 90 percent of evacuees have completed their mobilization activities (see Table 5-9). At this time, virtually all evacuees will have departed on their respective trips and the local telephone system will be largely free of traffic.

As indicated in Table 12-1, approximately 7',2 person hours are needed to complete the telephone survey. If six people are assigned to this task, each dialing a different set of telephone exchanges (e.g., each person can be assigned a different set of sectors), then the confirmation process will extend over a timeframe of about 75 minutes. Thus, the confirmation should be completed before the evacuated area is cleared. Of course, fewer people would be needed for this survey if the Evacuation Region were only a portion of the EPZ. Use of modern automated computer controlled dialing equipment or other technologies (e.g., reverse 911 or equivalent if available) can significantly reduce the manpower requirements and the time required to undertake this type of confirmation survey.

If this method is indeed used by the offsite agencies, consideration should be given to maintain a list of telephone numbers within the EPZ in the EOC at all times. Such a list could be purchased from vendors and could be periodically updated. As indicated above, the confirmation process should not begin until 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> after the Advisory to Evacuate, to ensure that households have had enough time to mobilize. This 3-hour timeframe will enable telephone operators to arrive at their workplace, obtain a call list and prepare to make the necessary phone calls.

Should the number of telephone responses (i.e., people still at home) exceed 20 percent, then the telephone survey should be repeated after an hour's interval until the confirmation process is completed.

Other techniques could also be considered. After traffic volumes decline, the personnel manning TCPs can be redeployed to travel through residential areas to observe and to confirm evacuation activities.

Palo Verde 12-1 KLD Engineering, P.C.

Evacuation Time Estimate Rev. I

Table 12-1. Estimated Number of Telephone Calls Required for Confirmation of Evacuation Problem Definition Estimate number of phone calls, n, needed to ascertain the proportion, F of households that have not evacuated.

Reference:

Burstein, H., Attribute Sampling McGraw Hill, 1971 Given:

• No. of households plus other facilities, N, within the EPZ (est.) = 4,350

" Est. proportion, F, of households that will not evacuate = 0.20

• Allowable error margin, e: 0.05

• Confidence level, a: 0.95 (implies A = 1.96)

Applying Table 10 of cited reference, p=F+e=0.25; q = l-p=0.75 A 2 pq + e 3 n = e2 - 308 Finite population correction:

nN nF --n+N-1 N - 288 Thus, some 300 telephone calls will confirm that approximately 20 percent of the population has not evacuated. If only 10 percent of the population does not comply with the Advisory to Evacuate, then the required sample size, nF = 206.

Est. Person Hours to complete 300 telephone calls Assume:

" Time to dial using touch tone (random selection of listed numbers): 30 seconds

" Time for 6 rings (no answer): 36 seconds

" Time for 4 rings plus short conversation: 60 sec.

• Interval between calls: 20 sec.

Person Hours:

300[30 + 0.8(36) + 0.2(60) + 20] 7.6 3600 3600 Palo Verde 12-2 KLD Engineering, P.c.

Evacuation Time Estimate Rev. 1