ML20248F865
| ML20248F865 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 04/03/1989 |
| From: | Adler T MASSACHUSETTS, COMMONWEALTH OF, RESOURCE SYSTEMS GROUP |
| To: | |
| Shared Package | |
| ML20248F851 | List: |
| References | |
| OL, NUDOCS 8904130234 | |
| Download: ML20248F865 (33) | |
Text
_ _ _ _ _ . - _ _ _
qe k EfUllRQ LH4M @ ptA u 00CKEIED
, USNBC UNITED STATES OF AMERI'CA '89 m -5 pl 36 NUCLEAR REGULATORY COMMISSION-ATOMIC SAFETY AND LICENSING BOAkd[.
Before the Administrative Judges:
Ivan W. Smith, Chairman Dr. Richard F. Cole Kenneth A. McCollom ..
L
)
In the Matter of ). Docket Nos. 50-443-OL
) 50-444-OL PUBLIC SERVICE. COMPANY ) (Off-Site EP)
OF NEW HAMPSHIRE, EI AL. )
)
(Seabrook Station, Units 1 and 2) ) April 3, 1989
)
(
TESTIMONY OF DR. THOMAS J. ADLER ON BEHALF OF JAMES M. SHANNON, ATTORNEY GENERAL FOR THE COMMONWEALTH OF MASSACHUSETTS, CONCERNING CRMIENTIONS JI-1, JI-2 (ETEs) AND JI-3 (DATA COLLECIIQN1 Department of the Attorney General Commonwealth of Massachusetts One Ashburton Place Boston, Massachusetts 02108-1698 (617) 727-2200 8904130234 PDR 990403 ADOCK 05000443 PDR 1
t . .
o l
l- .,
UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION ATOMIC SAFETY AND LICENSING BOARD l
Before the Administrative Judges:
Ivan W. Smith, Chairman Dr. Richard F. Cole Kenneth A. McCollom
)
In the Matter of ) Docket Nos. 50-443-OL
) 50-444-OL PUBLIC SERVICE' COMPANY ) (Off-Site EP)
OF NEW HAMPSHIRE, EI AL. )
)
(Seabrook Station, Units 1 and 2) ) April 3, 1989
)
TESTIMONY OF DR. THOMAS J. ADLER ON BEHALF OF JAMES M. SHANNON, ATTORNEY GENERAL FOR THE COMMONWEALTH OF MASSACHUSETTS, CONCERNING CONTENTIONS JI-1, JI-2 (ETEs) AND JI-3 (DATA COLLECTION)
I.
SUMMARY
OF TESTIMONY In this testimony Dr. Adler presents his views in support of three contentions: JI-1, JI-2, and JI-3. As to JI-1, Dr. )
Adler agrees that there is no evacuation time study report ,
i which describes the derivation of the SPMC's ETEs, and he explains why Volume 6 of the NHRERP cannot suffice as such a study for the SPMC's ETEs. As to JI-2, which asserts that the SPMC's ETEs are too unrealistic to form the basis of adequate protective action decision-making, Dr. Adler sets forth the
reasons why, in his opinion, the SPMC's ETEs are unrealistic and are likely to be longer. He also explains why, in his l l
opinion, the lack of.special facility ETEs and an ETE for the 1 I
transit-dependent segment.of the population violate the
{
l provisions of NUREG-0654, Appendix 4. Finally, as to JI-3, Dr. j i
Adler sets forth the reasons why protective action decision-makers need to have the abilities to make real-time estimates of the beach vehicle population'and to translate those estimates into ETEs, and he explains a range of methodologies for accomplishing these results at a reasonable ,
I cost.
II. IDENTIFICATION OF WITNESS Q.l. What is you name and current occupation? !
A.l. My name is Thomas J. Adler, and I am President of Resource Systems Group of Norwich, Vermont.
Q.2. Are your professional qualifications as set forth in the testimony and attachments filed for this case on September 14, 1987, and summarized in testimony submitted on February 21, 19897 A.2. Yes, they are.
III. IESTIMONY Q.3. Are you familiar with Contention JI-l?
A.3. Yes. It asserts that no evacuation time study has been done to assess what the realistic evacuation times would be in i
l
-2 -
f-
t l
the Massachusetts portion of the EPZ in light of'the special difficulties, circumstances, and delays in conducting an evacuation in Massachusetts under the Seabrook Plan for Massachusetts Communities ("SPMC"). The contention further asserts that Volume 6 of the New Hampshire Radiological Emergency Response Plan ("NHRERP"), which was published in August 1986 and contains KLD's ETE assessment for the Seabrook Station EPZ, did not take into account the special circumstances and delays in conducting an evacuation in Massachusetts under the SPMC; so it cannot be relied upon solely as the ETE study for the SPMC.
Finally, the contention asserts that a new ETE study needs to be conducted to meet the requirements of the NRC's regulations and NUREG-0654.
Q.4. Do you agree with the contention?
A.4. Yes I do. I am very familiar with Volume 6, which describes the NHRERP's ETEs and how they were derived, and it makes a number of assumptions about conditions in the Seabrook EPZ which either will not or may not apply for an evacuation conducted in the Massachusetts portion of the EPZ pursuant to the SPMC. To some extent, the SPMC implicitly acknowledges that differences exist because it contains ETEs (at IP 2.5, Attachment 4, p.20) which do not come from Volume 6 or the IDYNEV runs that produced the Volume 6 ETEs.
0@ Q.5. Does the SPMC contain an explanation of where those ETEs in IP 2.5 come from or how they were generated?
l
A.5. No. That is the problem. The SPMC's ETEs are provided without any explanation of how they were derived. The SPMC contains no separate volume akin to Volume 6, nor does it contain a supplement to Volume 6 explaining the assumptions and methodologies used in calculating its ETEs.
Q.6. Isn't it pretty clear that the SPMC's ETEs were developed by KLD using the IDYNEV model and the same link-node network described in Volume 6?
A.6. Yes, we have learned this through discovery but there is no way of knowing this from examining the SPMC. And that knowledge still doesn't explain how KLD came up with these ETEs, which are not consistent with those described in Volume 6. KLD has apparently used some new assumptions, inputs, and data to calculate the SPMC's ETEs. In no single place are all these new assumptions, inputs, and data presented or justified. Some of these assumptions can be found in the Applicants' ETE testimony from the NHRERP hearings; others are found in the NHRERP transcript (in the cross examination of Applicants' ETE panel).
But there is no single document which presents and justifies these assumptions.
Q.7. So long as the SPMC's protective action decision-makers have an accurate set of ETEs, why does the SPMC need to contain, to " publish" if you will, the assumptions and analyses which produced the ETEs?
I l
, t A.7. There is one legal reason and at least three practical ones. The legal reason is that NUREG-0654, Appendix 4, calls for a written study, a " report," describing how ETEs were calculated and presenting those ETEs in a particular way.
Publishing only the end result, the ETEs, does not, I believe, satisfy NUREG-0654, Appendix 4. The three practical reasons for having a published ETE study underscore the need for the legal requirement. The first practical reasons for having an ETE report, a written report, is that, as with any piece of scientific analysis, it permits others to examine the study to ascertain how reliable it is. Since ETEs are an important tool for PAR decisionmakers, and lives may depend on their accuracy, it is important that the analysis which produced them be readily available for critical analysis and review. We have been hampered in our analysis of the SPMC's ETEs for this very reason. The second practical reason for having a written ETE analysis is that it provides PAR decisionmakers with useful I knowledge about the ETEs so that they can apply them or adjust them with appropriate judgment and sensivity to the conditions at hand in a real emergency. For example, if the PAR decisionmakers for NHY's Offsite Response Organization ever find themselves considering PARS at a time grini to the staffing of the ACPs and TCPs, they may wish to know what the assumptions were in the ETE study about staffing times. A published report, either contained in the Plan or accompanying
1 it as an appendix, will allow decisionmakers ready access to this information. The third practical reason is to permit update work to be done in a prompt and certain fashion -- and without necessarily hiring the initial ETE consultant. As conditions change, e.g., population growth occurs or road improvements or changes are made, those who are responsible for ensuring that the ETEs are updated to realisticly reflect current conditions will, if they have an ETE study to review, h
have a way of knowing how much things have changed in assessing whether to hire a consultant to do some update work. And if they do decide to do an ETE update, having a thorough study explaining the existing ETEs will permit a new contractor (as well as the old one) to do the update work without re-doing the entire effort from scratch.
Q.8. Doesn't an ETE study for the SPMC exist, in effect, on computer tape in the form of the KLD IDYNEV runs which generated the SPMC's ETEs?
A.8. No. There are a whole series of assumptions, and numerous sets of data used to produce model inputs, which would not be reflected on the tapes. For example, while the Board has indicated that 31,000 vehicles would be an appropriate number of vehicles to input into the IDYNEV model in the " beach areas" for a day with " reasonably expectable peak occupancy,"
PID, 9.122, and an IDYNEV run which generates new Scenario 1 ETEs would indicate whqte the modler distributed the 31,000
vehicles along the beaches,1' that tape would not indicate why the 31,000 vehicles were distributed in this fashion. Some analysis must occur before the 31,000 vehicles can be divided among the numerous beach area origin centroids. Only a published ETE study, accompanying those runs, can provide this information. This is apparent from examining Volume 6 itself.
1 Most of the analyses it contains are not described on the 1
IDYNEV runs themselves; only the results of those analyses are
\ l shown. Volume 6 is a necessary document which, along with the IDYNEV runs, explains the derivation of the NHRERP's ETEs.
1 0.9. Well, then, how about Volume 6 in combination with ;
the IDYNEV runs for the SPMC's ETEs . . . taken together don't they constitute an ETE study for the SPMC7 l A.9. No. As I noted before, the analyses set forth in Volume 6 make a number of assumptions about conditions in the EPZ which either will not or may not apply for an evacuation conducted in the Massachusetts portion of the EPZ pursuant to the SPMC. This is not to say that Volume 6 has an analyses bearing on the SPMC's ETEs; it just cannot be relied upon as the CQmplete set of analyses supporting the SPMC's ETEs. What is needed is a supplement to Volume 6, or perhaps an amended / updated Volume 6, which describes the pertinent analyses underlying the SPMC's ETEs. Until this supplement is 1/ Runs reflected in IP 2.5 in fact use only 20 000 vehicles in the beach areas.
1 publish'ed, there is no ETE study for the SPMC's ETEs that satisfies NUREG-0654 or any of the practical reasons for having a published ETE study. Furthermore, NUREG-0654, Appendix 4, states that ETEs "should be updated as local conditions change l l
(e.g., change in type or effectiveness of public notification l system)." NUREG-0654, App. 4, p. 4-1. This type of change (sirens) and others have occurred from Volume 6 to the SPMC, i
and they necessitate that some new analyses be undertaken to justify the assumptions and inputs used in the IDYNEV runs that produce the SPMC's ETEs. l Q.10. At a couple of points you have made reference to a number of assumptions made in the Volume 6 ETE study about l conditions which will prevail in an evacuation of the Seabrook EPZ but which will not or may not apply for an evacuation l
conducted in the Massachusetts portion of the EPZ pursuant to the SPMC. What are these assumptions?
A.10. I will list eight of the most obvious and important ones. First, there is the size of the beach population on peak days. Volume 6 used a figure of 25,470 (see Volume 6 at 2-11, E-5). For modeling purposes, these vehicles were then allocated to the various beach area " origin centroids" shown on Figure 1-3 of Volume 6. (See also the Traffic Demand table for Scenario 1 and 2 in Volume 6, pp. M-1 through M-5.) The Licensing Board has now ruled that this number was too small and should have been 31,000 for a day with " reasonably
e expectable peak occupancy" and 35,000 for a day with the
" expectable maximum peak occupancy." PID, H9.122. These differences in the beach population will affect the ETE analysis significantly.
Second, the Applicants have now indicated that the plan will be revised such that Traffic Guides will not activate Access Control Points ("ACPs") for approximately two hours after the Order to Evacuate ("OTE"). Applicants' Rebuttal Testimony No. 9 at 33. Applicants are now also assuming that the Traffic Guides will not be able to man Traffic Control
, Points ("TCPs") immediately in fast-breaking accident scenarios; instead, it will take up to 3.75 hours8.680556e-4 days <br />0.0208 hours <br />1.240079e-4 weeks <br />2.85375e-5 months <br /> after an SAE l is declared to staff the ACPs/TCPs. Id. at 26. These assumptions both differ from the assumption made in the Volume 6 analysis that all traffic control measures are in effect throughout the evacuation. Volume 6, 10-70. These differences will affect the ETEs.
Third, the Volume 6 analyses assume that TCPs/ACPs will be I'
staffed by state / local police. The SPMC cntains two " modes" however, and in " Mode 2" the staffing will be done using civilians recruited by NHY who are not professional traffic handlers. In this mode, the ETEs are very likely to be longer than they would be in Mode 1, because non-professionals just can not be counted on to move congested flow traffic as efficiently as experienced police officers can. Some addition
factor'needs to be applied to the SPMC's Mode 2 ETEs to account for this.
Fourth, the configuration of the TCP/ACP diagrams in the SPMC differs in some significant respects from that found in Volume 6, and these changes will affect the ETEs for some ERPAs.
Fifth, I understand that the siren system was assumed to be a fixed-pole system in Volume 6 whereas the system being relied upon by the SPMC is a mobile siren system that needs to be driven into place at the Alert, then erected, and then sounded. The potential delays in notifying the Massachusetts beach population in a fast-breaking accident scenario (like the one used for the ETE planning basis in Volume 6) have not been analyzed and factored into an ETE analysis for Massachusetts, nor has there been an analysis of the potential impact on ETEs of hearing differing EBS messages directed separately at Massachusetts and New Hampshire beach-goers.
Sixth, the Volume 6 ETE study assumed that 25 percent of the population within the EPZ but outside the Region ordered to evacuate will voluntarily evacuate. Volume 6 at 10-3. The Applicants abandoned this assumption during the NHRERP l
hearings, however, and adopted the view that voluntary evacuation would be 25-50% in those portions of the EPZ not ordered to evacuate but within a distance from Seabrook equal to or less than other portions which had been ordeted to evacuate. Mass AG testimony on this topic was that the
volunt'ary evacuation in Massachusetts would be in the range of 56%. See Attachment 5, p. 10, to the Testimony of Zeigler et al., fol. Tr. 7849.
Seventh, the Volume 6 ETE study assumed that all the crowded beach areas, including those in Massachusetts, would be closed using a " beach closing" announcement at the Alert level I
and that the day-trippers would begin their trips from the l beach areas at that time. Een Vol'ume 6 at 10-13. The SPMC, however, does not even permit consideration of a beach closing in Massachusetts until the SAE level. SpMC, S3.6.1(E), p.
3.6-10. While the ETE analysis in Volume 6 assumes as its planning basis that there is a simultaneous declaration of an Alert and an SAE, and a simultaneous " beach closing" announcement in New Hampshire and Massachusetts, it appears that for many other accident scenarios the New Hampshire beach-goers may have their beaches closed first and may move over to the Massachusetts beaches after the Alert is declared but before the SAE is declared. No analysis of this on the Massachusetts ETEs has been attempted. Given the different emergency classification levels which trigger consideration of beach closings in Massachusetts and New Hampshire, it appears that the " planning basis" used in the Volume 6 ETE study, which assumes that the beaches in New Hampshire and Massachusetts are ,
1 always closed simultaneously for all summer scenarios, is not likely to be the typical situation encountered. A more I
l i
appropriate planning basis should be utilized which, while retaining the Volume 6 assumption that the accident is a rapidly escalating one, assumes that there is at least some difference in time between the beach closings in New Hampshire and Massachusetts.
Eighth, the size of the permanent resident population has grown since Volume 6 was published in August 1986, and it should be reflected in the SpMC's ETEs. It appears that the Ip l 2.5 ETEs were calculated using the three-year old data. I Q.ll. What is it, then, Dr. Adler, that you are I l
recommending be done with respect to an ETE study for i
A.ll. I am recommending, and it appears that Appendix 4 of NUREG-0654 requires this, that an appropriate ETE study be dann for an evacuation from the Massachusetts portion of the EpZ conducted pursuant to the SpMC. That study should assess the impact on ETEs of the conditions that will likely prevail in Massachusetts under both Mode 1 and Mode 2 conditions. A more appropriate ETE planning basis should be used in conducting the analysis. Upon completion, the analyses and assumptions that constitute this study need to be published as a " report," as required by NUREG-0654, and either incorporated into or appended to the SpMC. That written report should then be made available to all interested parties for their review.
)
.. i Q.12. Have you reviewed the Evacuation Time Estimates
-("ETEs"). included in the SPMC7 A.12. Yes, I have. They are listed in IP-2.5, Attachment 4.
Q.13. Are the ETEs shown in IP-2.5 Attachment 4'the same as those described in NHRERP Volume 67 A.13. No, they are not. The ETEs shown in IP-2.5 Attachment 4 are only for Region 8 A as defined in Vol. 6 and for a new region (Region 13) not defined in Vol. 6.1 In addition, the ETE's for Region 8 are different from those shown in Vol. 6.
Q.14. Why were the.ca Regions used for the Massachusetts ETEs?
A.14. We tssume they wished to isolate the Massachusetts ETEs from chose for the New Hampshire routes.
Q.15. Have you evaluated the IP-2.5 ETEs with respect to Contention JI-27 A.15. Yes, I have.
Q.16. Would you first summarize that Contention?
A.16. JI-2 states that the IP-2.5 Attachment 4 ETEs are unrealistic and in particular that realistic ETEs would be longer.
2/ Region 8 is designated the Inner -quta Peqion and contains ERPA A (Hampton Falls, Seabrook, and Hainptoi. Beach) and ERPA B (Amesbury and Salisbury). Egg Volume 6, Table 10-3, p. 10-5.
3/ Region 13 is depicted on p. J-188 of Appendin J of the j SPMC. It contains ERPAs A through E, i.e., all NH towns within {
5 miles and all Mass. towns.
1
Q.i7. Do you believe that those ETEs are unrealistic and that realistic ETEs would be longer? l A.17. Yes, I do.
Q.18. Would you explain the basis for this opinion?
A.18. There are several bases for this opinion, as listed 4
in Bases A through I of Contention JI-2. Basis A refers to inefficiencies in traffic control personnel. The IDYNEV runs used to prepare the IP-2.5 Attachment 4 ETEs assume implicitly that the civilian traffic guides staffing SPMC traffic control posts ("TCPS") in Massachusetts are as efficient as New Hampshire State Police would be in directing traffic through those TCPs. Prudent planners would not make this assumption.
~
The tasks that civilian guides will be called upon to perform are ones for which they have had only short training and no real experience. (My testimony on the SPMC's training for traffic guides will be filed next week.) In addition, at key TCPs, the guides will undoubtedly be asked by evacuees for specific routing instructions and/or for information about the emergency. It is unlikely that they will be able to dispatch such evacuees as rapidly as would a State Police Officer.
Furthermore, guides must, at several of these key TCPs, act as a traffic signal would, periodically stopping traffic flow from one direction in order to allow flows from another directiori.
The IDYNEV model runs assume that these guides alternate tiow directions on a 75 second cycle and that only 2.5 seconds are i
' lost each time the flow direction is changed. Nowhere, however, are traffic guides instructed that they should maintain a 75 second cycle nor aru they provided stopwatches to monitor flows. For all these reasons, I must conclude that these traffic guides will not be able to move the traffic in Massachusetts as efficiently as state / local police woul7 Q.19, Would you describe your evaluation of Contention JI-2, Basis B7 A.19. Basis B states that the SPMC's ETEs are based on incorrect assumptions concerning the number of cars that will flow across Massachusetts roads. I evaluated this Basis together with the related Basis E, which states that the number of vehicles evacuating from and through Massachusetts is underestimated. There are five issues that are raised by these bases: 1) that Massachusetts beach-area transient populations are underestimated, 2) that EPZ-resident returning commuters and shoppers are not accounted for in the model, 3) that flows of New Hampshire beach transients through Massachusetts are not represented in the model. runs, 4) that through traffic levels on I-95 and I-495 as represented in the model do not correspond to procedures regarding establishment of Access Control Posts
("ACPs"), and 5) that the Massachusetts permanent resident population is underestimated for 1989.
Q.20. Would you explain why you believe Massachusetts beach-area transient populations are underestimated in the SPMC's ETEs?
l
)
l
A.2b. The ETEs shown in IP-2.5 Attachment'4 reflect beach population estimates as derived from 1987 Avis-Airmap counts showing 29,293 total vehicles. This total is approximately.
2,000 less than the 31,000 vehicles specified in the Board's PID. Also, the SPMC's ETEs are based on an incorrect distribution of the vehicles counted in the beach areas, i.e.,
the distribution shown on the Avis photos. The Avis Airmap distribution is, I believe, seriously flawed in that it underrepresents the vehicle population at Salisbury Beach.
Vehicles actually observed (7211) at that area of Salisbury in the July 1987 Befort flights conducted for the Massachusetts Attorney General total over 1,000 more than the number projected in that area from the Avis Airmap counts (6119). The Avis Airmap projections imply that only 21% of the total beach population is in Salisbury whereas the actual Befort vehicle counts show that 30% are in Salisbury. By comparison, the NHRERP Volume 6 estimates place Salisbury's beach parking capacity at 32% of the EPZ's beach total. See Volume 6 at J E-5. KLD's parked vehicle counts based on 1985 overflights, as (
l reported in Volume 6, id., similarly show 32% of the total i
i counted beach vehicles to be located along Salisbury l 1
Beach.A# Thus, the projections for Salisbury Beach from the Avis A/ By comparison, for Seabrook Beach, the Avis Airmap counts show 10% of the EPZ beach population while the Befort I
1 overflights show 9%, and the Vol. 6 estimates are 10%.
l
i Airmap counts are anomalous and likely understate actual Salisbury Beach populations as a proportion of the total beach population. Since the EPZ total beach vehicle population is also underestimated in the data used to prepare the IP-2.5 Attachment 4 ETEs, the resulting times are unrealistic estimates, particularly for the Massachusetts portion of the EPZ.
Q.21. Turning to the second issue you mentioned about the
" number of cars" that will flow across Massachusetts roads, have you previously prepared for the Board in connection with a
this proceeding a discussion of the effects of returning EPZ-resident commuters and shoppers, and is that discussion equally applicable to the ETEs presented in IP-2.5 Attachment 47 A.21. Yes. My affidavit dated March 27, 1989 does address this issue in some detail and it is applicable to the IP-2.5 Attachment 4 ETEs.
Q.22. As to the third issue you mentioned, would you i
explain what flows of New Hampshire beach transients through Massachusetts are not represented in the model runs used to prepare IP-2.5 Attachment 4 ETEs?
A.22. At TCP A-SE-06 in Seabrook, the intersection of Rt.
1A and Rt. 286, (at NHRERP, Vol. 6, I-13), the guide is instructed to do the following, " Facilitate traffic movement from southbound Route 1A onto westbound Route 286. If westbound traffic on Route 286 becomes congested, then send l _ _ _ _ _ _ __
q l- 1 L '
l l
'traffid' southbound on Route 1A. When' Route 286 traffic moves away from the intersection, revert to guiding traffic onto Route 286. Thus, the preferential movement'is onto Route 286, -
however, keep' traffic moving out of Seabrook Beach even if 3
J l ~ Rout'e 286:is congested." l However, despite this explicit' instruction and despite the fact that the IDYNEV model shows' Route 286 becoming congested within minutes after a New Hampshire beach closing message, and staying congested for hours, the IDYNEV model used to prepare I the IP-2.5 Attachment 4 ETEs routes none of the Seabrook traffic south on Rt. IA into Salisbury. Were it to do so, as l
it should, the Massachusetts ETEs would be increased.
I Q.23. Turning to the fourth issue about the " number of cars" that will flow through Massachusetts, wotid you. explain your earlier statement that through traffic' levels on I-94 and I-495 as represented in the model do not correspond to procedures'regarding establishment of ACPs?
A.23. Yes. The IDYNEV runs used to prepare the IP-2.5 Attachment 4 ETEs include 3,000 vehicles assumed to be on I-95 at the start of the evacuation. According to Applicants' Direct Testimony No. 7, dated September 12, 1987, fol. Tr.
5622, p. 72:
I This estimate of 3000 through vechiles represents those who entered the EPZ prior to the implementation of access control and have not as yet completed their travel by
)
the time the Order to Evacuate (OTE) is given. At the OTE, the number of through l
1 I
l
_ - _ _ _ _ _ _ - - _ - _ i
~~
vehicles could be substantially.less than
, the number which occupy the network at the time the access controls are applied. This reduction in vehicles reflects the fact that many of these vehicles will have exited the EPZ between the time the access control was applied and the time the OTE is given.
The testimony later indicates that 4,600 vehicles are shown on the interstate system in films taken on July 4 and 5, 1986.
Thus, use of the 3,000 through vehicles as assumed in these runs is based on the explicit assumption that acess control is in place before the_OTE. In fact, in the 1988 FEMA Evaluated Exercise, the key interstate ACPs in Massachusetts were not staffed until over an hour after the SAE was declared.
Further, we now learn through the Applicants' Rebuttal Testimony No. 9 dated February 29, 1989 at p. 33 that, " Traffic guides will be instructed to activate the ACPs appromately two hours after the OTE, providina that traffic volume'has declined to the extent that the activity of screening motorists will not result in the formation of long queues of inbound vehicles."
If this is indeed the policy for staffing ACPs, the assumption I of only 3,000 through vehicles is unrealistic. Actual through-traffic counts at the Hampton toll booth and on I-95 at the New Hampshire / Massachusetts state line indicate that directional volumes frequently exceed 6,000 vehicles per hour on Fridays, Saturdays, and Sundays in July. If ACPs were not established until two hours after the OTE, some number of these '
vehicles would continue to flow through the EPZ along I-95 and I i
i
- c.
\. =
L I-495.To test the effects of-this additional flow,.we performed Scenario 1 runs'with IDYNEV which added along I-95 a.
flow of 4,000 vehicles per hour-in'each direction (northbound and southbound) for one hour and forty minutes after the beach closing (1:15 after.the OTE). Under these conditions, traffic l ..
flow onto I-95 at key on-ramps was impeded, and the ETEs were l
extended by slightly over one hour. Thus, this through traffic caused by failure to promptly staff ACps at or before the OTE, as originally assumed in Volume 6, could have a significant effect on ETEs and cannot simply be ignored as in the Ip-2.5 Attachement 4 ETEs.
Q.24. As to the fifth issue, how will 1989 resident population figures affect the ETEs?
A.24. As I indicated, it appears that the Volume 6 1986 estimates of the permanent resident population were used to calculate the Ip 2.5 ETE. Dr. Albert.Luloff has filed testimony concurrently with this testimony in which he states that the permanent resident population figures should be used in any properly done ETE study for the SPMC.
Q.25. If the factors that you described above were accurately accounted for, would the resulting ETEs be realistic?
A.25. No, I do not believe that they would be. The ETEs as projected by the Applicants, without these factors accounted for, already approach and exceed ten hours for some scenarios.
Se.e NRC Staff Ex. No. 1, fol. Tr. 6742. The additional factors i
describsd above will add at least one hour to the ETEs, meaning that a summer evacuation could extend from two o' clock in the afternoon until the early morning hours of the next day. Over a period of time that protracted, lack of food, lack of sleep and restrictions on other bodily functions will undoubtedly ,
l affect driver performance and behavior in ways that simply can not be extrapolated from other observed conditions. See my corrected testiony fol. Tr. 7181, 76-77.
Q.26. Have you examined JI-2, Basis I?
A.26. Yes, I have. Basis I asserts first that the SpMC contains no ETEs for the transit dependent population and those in special facilities. This is true. There are no such ETEs in the SPMC at all. This, in an of itself, violates NUREG-0654, Appendix 4, in my opinion. As I read Appendix 4, there are separate requirements for calculating the transit dependent ETEs and the special facility ETEs. As to the "non-car-owning population dependent upon public transport" (Appendix 4 at 4-9), i.e., the transit dependent population, Appendix 4 requires an " estimate of the time required to evacuate that seament" of the population. Id. (emphasis supplied). As to the special facilities, however, ETE
"[e]stimates" (plural) "shall be made," and "[e]ach special f acility shall be t.reated on an indiyldual bani:. " Id at 4 9 4-10. Thus, while a single ETE needs to be calculated for the transit dependent population (for each Scenario and Region),
0 the ETE'for each of the special f acilities needs to be included I in the ETE study. None of this is contained in the SPMC, however.
Q.27. Are these ETES for the transit, dependent populations and special facilities in Massachusetts contained in Volume 67 I A.27. Volume 6 contains no individualized ETEs for the special facilities in either Massachusetts or New Hampshire.
So Volume 6 does not resolve this NUREG-0654 deficiency. But Volume 6 does contain, in Section 11, some calculations which purport to cupport the transit dependent population ETEs contained in Tables10-10a through 10-10d off Volume 6, pp.
10-71 through 10-74.
Q.28. Are those calculations and transit dependent ETEs contained in Volume 6 of the New Hainpshire plan adequate for
.SPMC purposes?
A.28. No, there is no assurance that they are. First, I would ask that you re-read my testimony which I filed in the NHRERP hearings (fol. Tr. 7181) where I set forth (at 61-68) my.
analysis why those transit dependent ETE calculations contained in Volume 6 are not reasonably reliable in the context of the NHRERP. Most of that analysis applies here, too. In the i context of the SPMC, however, those transit-dependent ETE !
calculations in Volume 6 are even less reliable. There are a l number of reasons for this. First, the bus transportation plan 1 ;
L analyzed in Section 11 of the NHRERP was one which assumed that l
l l
l
n
, buses would first report to staging areas. See Volume 6 at 11-19, 11-20.- The SPMC uses a different transportation plan for buses: route' guides are'to be dispatched to travel to each of the bus yards, where the bus driver' instructions are to be given and the dosimetry is to be handed out. The buses will !
. then be driven in convoy fashion back into the EPZ, led by the route guides, and they will not stop at the staging area.
SPMC, IP 2.5. This process is very likely to take longer than the bus mobilization scheme analyzed in Volume 6, but neither the SPMC nor Volume 6 attempts to analyze how long the SPMC's bus mobilization process will take.
Q.29. Why is the SPMC's bus mobilization process likely to take longer then the process analyzed.in Volume 6.
A.29. In Volume 6 it is assuraed that buses should be able to start arriving in the local communities within 1-1/2 hours after the Order to Evacuate and that within one additional-hour "a sufficient number of buses necessary to evacuate the schools and the transit-dependent will be on-hand and will have started evacuation activities." Volume 6 at 11-20. But if we assume ,
the same ETE planning basis used in the Volume 6, where the SAE occurs only about 25 minutes prior to the Order to Evecuate, then the 1-1/2 to 2-1/2 hour arrival assumption appears to be quite improbable. This is because the route guides are themselves are not notified to report to duty un+ il the SAE level is reached. SPMC, Figure 2.1-1. They are then to report
_J. _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . _ _ _ _ _ _
1 first t6 the staging area for briefing, assignment, and equipment issuance. SPMC, IP 2.10, p. 25, Attachment 3. They then must drive to their assined bus yards, meet up with the buses, and then ride on the buses back into the EPZ, and then drive to their particular assignments. Given the extremely distant location of many of the bus yards, Egg Testimony of Maureen Mangan and John Poalillo on Behalf of the Attorney General James M. Shannon, Regarding the Actual Availability of the SPMC's Manned Vehicles and Drivers, filed February 21, 1989, these tng trips (to the bus yards and back) plus the mobilization time for the route guides to report to and be briefed at the staging area will very likely take considerably longer than 1-1/2 1/2 hours for many of the buses. So a
" sufficient number of buses necessary to evacuate the schools and the transit-dependent" will not likely be available within 2-1/2 hours of the Order to Evacuate.
Q.30. Are there other reasons why the Volume 6 transit-dependent ETE calculations cannot be relied upon in the context of the SPMC?
A.30. Yes, many of the Massachusetts bus routes analyzed in Volume 6 have been modified, and there is obviously no assessment in Volume 6 of the time it will take the buses to circulate on these new routes.
Q.31. What is it that needs to be done to produce a reliable set of ETEs for the transit-dependent segment of the population in the context of the SPMC.
A.31.' An entirely new analysis needs to be conducted using ;
l appropriate SpMC-specific assumptions, and that analysis needs i to be documented in a written report, or as a portion of the NUREG-0654 ETE report. Then that report needs to be f I
disseminated for review and critique by all interested parties. !
Q.32. Isn't it nevertheless reasonable to assume that the transit-dependent ETEs under the SpMC will always be shorter than the general population ETEs?
A.32. No. Until a proper analysis of the Massachusetts transit-dependent ETEs is done, and until a proper set of general population ETEs has been calculated, it would be imprudent to make this assumption. Perhaps it is reasonable for the summer scenarios, for which the ETEs are very long, to make this assumption. But for the off-season scenarios, such as Scenario 8 (off-season, mid-week, evening, good weather),
the transit-dependent ETEs could well be longer.
Q.33. Have you reviewed Contention JI-3. dealing with implementation of a beach porelation monitoring system?
I A.33. Yes, I have.
Q.34. Why is such a system necessary?
A.34. The size of the beach population at the start of the evacuation has a direct and significant effect on ETEs E but that population rises and falls dramatically from day to 5./ The impact of beach vehicles counts on ETEs is felt not just for the beach area ETEs but for the 5-mile and 10-mile ETEs shown on the Ip 2.5 ETEs as well. The impact is systemic, not localized. j
E day and'even over the course of a single day. According to estimates accepted in the Board's PID, the beach vehicle count can exceed.30,000 on summer weekends, or approximately one-third of the total evacuation traffic. The IP-2.5 Attachment 4 ETEs are specified for ten discrete scenarios, but those scenarios evaluate only three beach transient population !
l sizes: 100% of peak, 75% of peak and none. There are two !
related questions which must be answered in order to accurately ,
determine ETEs for a particular evacuation: 1) At the time of ,
an evacuation, how many beach area vehicles are, in fact, present? and 2) What is the ETE for that number of beach area vehicles?
Q.35. Would you describe how a system to answer those ;
l questions might work?
A.35. There is a full range of possible designs for the system. Here, I will simply explain the basic structure of the system and describe some illustrative examples of how the system could be implemented. There are two basic components on which the structure is based. First, there are several "real-time indicator variables" that can be used to determine the beach vehicle population at a particular point in time.
One obvious indicator variable is the net vehicle accumulations 1 l
as measured by traffic counters arranged in a cordon on the '
l roads feeding the beach area. Inductance loop counters (with l
i associated modems for transferring data to a central I
l l
l l - - - - - - - - - - _ _
, .\
t microcomputer)'could be located in a cordon on:
- 1) Route 1A in Salisbury, MA;
- 2) Route 286 in Seabrook, NH; i
- 3) ' Route 51 in Hampton, NH;
- 4) Route 1A in Hampton, NH.
j The system could be programmed to record hourly traffic volumes ,
f I in both directions at each counting station. The hourly data I l
could then be transferred by phone to the central computer at pre-determined intervals or on demand. Transfer of the data
, and programming of the counters could be achieved from the central computer by phone using standard telemetry software.
The bi-directional traffic data could then be summed to .
I determine net vehicle influx for use in calculating the beach population. The cost of equipment for this system would be approximately $20,000. The accuracy of this calculation would depend both on the placement of the counters and on the length of time between calibrations of the system.
Short of implementing this full system, there are numerous other indicator variables which could be statistically correlated with beach vehicle populations and which then could be used to estimate the actual beach vehicle population at the time of an evacuation. For example, the traffic flow on one of the major beach feeder roads (i.e., Rt. 51), the number of tolls collected at the Hampton toll booth, or the number of vehicles parked in the state park lots (Hampton and Salisbury) are each likely to be good indicators of the size of the beach population at a prticular point in time. More generally, it
would be"possible to conduct beach vehicle counts on several days using simple 35mm overflight photography. The effects of up to seven variables such as weather, time of day, day of week, month, holiday period, and water temperature could be statistically correlated with the actual measured counts on eight overflights.N' A smaller number of overflights could be performed if the number of variable were reduced. From this exercise, a simple calculation template could be constructed that took as inputs the weather, time of day, and-other variables whose effects had been measured. This template could then be used to estimate beach vehicle populations at the time of an evacuation, i
The second component of this structure is one which '
translates beach vehicle estimates into ETEs. The ideal way to make this translation would be to have IDYNEV installed so that a new run using actual beach vehicle estimates could be completed at the time of the emergency. Alternatively,
" sensitivity tests" of the type described in NHRERp Volume 6 (p. 10-16) could be performed with a range of beach vehicle populations from 0 to 35,000 in increments of 5,000 to 10,000 vehicles (i.e., 3 to 6 runs), and a table could be constructed of ETEs from those runs, with a simple interpolation methodology described for calculations involving beach 1/ The overflights conducted by consultants for the l Massachusetts Attorney General cost approximately $2,000 per '
flight including air time, photography, and photo-interpretation (vehicle counts).
L- - - . - - . - - - - - - - _ _ . - - - _ _ _ _ _ _ _ _ _ _ _ . - - - - - - - - - - - - - - ----- -------___
-l 6?
populations.between the levels for which'IDYNEV runs were completed.
Together.these two system components, the first which estimates beach vehicle populations.and the second which
' estimates ETEs based'on those populations, would provide a substantially more realistic basis for calcu'lating likely ETEs at the time of a summer-season emergency. <
Q.36. Does this conclude your testimony?
A.36. Yes, it does.
1 i
i r
l i
C
! L '.R ? D ,
' ; Nr0 '
UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION 89 APR 5 P 13
0 F F '..
06Lht i g i. ,,Li
- u. . -
)
In the Matter of ) Docket Nos. 50-443-OL
) 50-444-OL PUBLIC SERVICE COMPANY ) (Off-Site EP)
OF NEW HAMPSHIRE, _E _T _A _L . ) '
)
(Seabrook Station, Units 1 and 2) ) April 3, 1989 i
)
CERTIFICATE OF SERVICE I, Allan R. Fierce, hereby certify that on April 3, 1989, I made service of the within TESTIMONY OF DR. ALBERT E. LULOFF ON BEHALF OF ATTORNEY GENERAL JAMES M. SHANNON REGARDING JI-2 AND JI-21 (PERMANENT RESIDENT POPULATION) and TESTIMONY OF DR. THOMAS J. ADLER ON BEHALF OF JAMES M. SHANNON, ATTORNEY GENERAL POR THE COMMONWEALTH OF MASSACHUSETTS, CONCERNING CONTENTION JI-1, JI-2 (ETEs) AND JI-3 (DATA COLLECTION), by Federal Express as indicated by (*) and on April 4, 1989 by first class mail to the remaining members of the list.
- Ivan W. Smith, Chairman *Kenneth A. McCollom Atomic Safety & Licensing Board 1107 W. Knapp St.
U.S. Nuclear Regalatory Stillwater, OK 74075 Commission l East West Towers Building 4350 East West Highway Bethesda, MD 20814
- *Dr. Richard F. Cole
- Docketing and Service Atomic Safety & Licensing Board U.S. Nuclear Regulatory U.S. Nuclear Regulatory Commission Commission East West Towers Building Washington, DC 20555 4350 East West Highway Bethesda, MD 20814 .
l l
l
h)
- M e<v:n
?- j
- Robert.R'.. Pierce, Esq. .
.
- Thomas G.LDignan', Jr.,~Esq.
Atomic SafetyL&' Licensing? Board.
i Katherine'Selleck, Esq. 1 U.S.. Nuclear. Regulatory Commission- Ropes & Gray? i East West Towers Building One International Place i 4350 East West Highway Boston, MA 02110
.Bethesda, MD 20814 H.' Joseph Flynn,:Esq. .
=*Sherwin E. Turk, Esq.
Assistant General Counsel U.S. Nuclear Regulatory Office of General' Counsel- Commission j Federal;Emer'gencyLManagement Office-of the General Counsel' Agency 15th Floor 500 C Street,JS.W. . 11555 Rockville Pike Washington,.DC 20472 Rockville, MD 20852 Atomic Safety & L'icensing Robert A. Backus, Esq.
Appeal Board Backus, Meyer & Solomon U.S. Nuclear Regulatory 116 Lowell Street Commission P.O. Box 516 Washington', DC 20555 Manchester, NH 03106 Atomic Safety & Licensing Board Jane Doughty U.S. Nuclear Regulatory Commission Seacoast' Anti-Pollution League Washington, DC 20555 5 Market Street Portsmouth, NH 03801 Charles P. Graham, Esq. Barbara St. Andre, Esq.
Murphy & Graham Kopelman & Paige, P.C.
33 Low Street 77 Franklin Street Newburyport, MA 01950 Boston, MA 02110 Judith H. Mizner, Esq. R. Scott Hill-Whilton, Esq.
79 State Street Lagoulis, Hill-Whilton ,
2nd Floor & Rotondi Newburyport, MA 01950 79 State Street Newburyport, MA 01950
~
i Dianne Curran, Esq. Ashod N. Amirian, Esq.
Harmon, Curran, & Towsley 145 South Main Street Suite 430 P.O. Box 38 2001 S Street, N.W. Bradford, MA 01835 l Washington, DC 20008 Senator Gordon J. Humphrey Senator Gordon J. Humphrey U.S. Senate One Eagle Square, Suite 507 Washington, DC 20510 Concord, NH 03301 (Attn: Tom Burack) (Attn: Herb Boynton) 1 o________ __
D:f u-2
' George Dana Bisbee, Esq. Phillip Ahrens, Esq.
Assistant Attorney General- Assistant Attorney General Office of the Attorney General Department of the Attorney 25' Capitol Str<eet General Concord, NH 03301 Augusta, ME 04333 Sandra Gavutis, Chairperson Calvin A. Canney Board of Selectmen City Manager RFD 1, Box'1154 City Hall Rte. 107 126 Daniel Street Kensington, NH ~ 03827 Portsmouth, NH 03801 Gary W. Holmes, Esq.. Richard A. Hampe, Esq.
Holmes & Ellis Hampe & McNicholas 47 Winnacunnet Road 35 Pleasant Street Hampton, NH 03842 Concord, NH 03301 Robert Carrigg, Chairman J.P. Nadeau Board of Selectmen Selectmen's Office-Town Office 10 Central Road Atlantic Avenue. Rye, NH 03870 North Hampton, NH 03862 William S' Lord
. James H. Carpenter, Alternate Board of Selectmen Technical Member Town Hall -' Friend Street Atomic Safety & Licensing Amesbury, MA 01913 Board Panel U.S. Nuclear Regulatory Commission Washington, DC 20555 JAMES M. SHANNON ATTORNEY GENERAL COMMONWEALTH OF. MASSACHUSETTS Allan R. Pierce' fl. L Assistant Attorney General Nuclear Safety Unit Department of the Attorney General One Ashburton Place Boston, MA 02108-1698 (617) 727-2200 .
I DATED: April 3, 1989 l
1 I
_ _ _ _ _ _ - _ _ - - _ _ _ _ _ - _ _ _ _ - _ - - - _ _