ML12228A655

New York State Pre-Filed Revised Evidentiary Hearing Exhibit NYS000420, Pre-filed Rebuttal Testimony of Dr. Franois J. Lemay, Ph.D. in Support of Contention NYS-12C (Jun. 29, 2012)

ML12228A655
Person / Time
Site:	Indian Point
Issue date:	06/29/2012
From:	Lemay F International Safety Research, State of NY, Office of the Attorney General
To:	Atomic Safety and Licensing Board Panel
SECY RAS
References
RAS 23294, 50-247-LR, 50-286-LR, ASLBP 07-858-03-LR-BD01
Download: ML12228A655 (56)

Text

NYS000420 NYS 000241 Submitted: June 29, 2012 1 UNITED STATES 2 NUCLEAR REGULATORY COMMISSION 3 BEFORE THE ATOMIC SAFETY AND LICENSING BOARD 4 -----------------------------------x 5 In re: Docket Nos. 50-247-LR; 50-286-LR 6 License Renewal Application Submitted by ASLBP No. 07-858-03-LR-BD01 7 Entergy Nuclear Indian Point 2, LLC, DPR-26, DPR-64 8 Entergy Nuclear Indian Point 3, LLC, and 9 Entergy Nuclear Operations, Inc. June 29, 2012 10 -----------------------------------x 11 PRE-FILED WRITTEN REBUTTAL TESTIMONY OF 12 DR. FRANÇOIS J. LEMAY 13 REGARDING CONSOLIDATED CONTENTION 14 NYS-12C (NYS-12/12A/12B/12C) 15 On behalf of the State of New York (NYS or the State),

16 the Office of the Attorney General hereby submits the following 17 rebuttal testimony of François J. Lemay regarding Consolidated 18 Contention NYS-12C (NYS-12/12A/12B/12C).

19 I. INTRODUCTION 20 Q. Please state your full name.

21 A. François Jean Lemay.

22 Q. Dr. Lemay, could you briefly summarize the testimony 23 you provided on December 21, 2011?

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 1

1 A. International Safety Research (ISR) was retained by 2 NYS in connection with the States Consolidated Contention 12C.

3 In my December 21, 2011 testimony, I explained ISRs analysis 4 and conclusion that Entergy and Nuclear Regulatory Commission 5 (NRC) Staff underestimated the total economic cost of a severe 6 nuclear accident at Indian Point (IP). This underestimation is 7 primarily a result of their use of Sample Problem A input 8 values, which are not specific to the conditions at IP because 9 they include costs and times for decontamination that are 10 unrealistic given current known decontamination data and the 11 complexities of an urban to hyper-urban area such as that 12 surrounding IP.

13 Q. What is the purpose of this rebuttal testimony you are 14 now providing?

15 A. NYS has asked me to respond to Entergys March 30, 16 2012 and NRC Staffs March 30, 2012 testimony on Consolidated 17 Contention 12C.

18 Q. Have you read the testimony submitted by Entergy and 19 by NRC Staff on NYSs Consolidated Contention 12C?

20 A. Yes. I have reviewed Entergys Statement of Position 21 (ENT000449), Entergys Pre-Filed Written Testimony (ENT000450) 22 (Entergy Testimony), and the supporting exhibits filed by Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 2

1 Entergy on March 30, 2012. I have also reviewed NRC Staffs 2 Statement of Position (NRC000039), NRC Staffs Pre-Filed Written 3 Testimony (NRC000041) (NRC Staff Testimony), and the supporting 4 exhibits filed by NRC Staff on March 30, 2012.

5 Q. What are your overall conclusions regarding NRC Staff 6 and Entergys testimony?

7 A. Neither Entergy nor NRC Staff have provided a 8 documented basis for the cost of the decontamination value used 9 in their calculation of the economic costs associated with a 10 severe accident at IP. The decontamination time value selected 11 by Entergy represents a scenario that appears unreasonable and 12 unrealistic, based on past experiences and especially in light 13 of the currently unfolding experience at Fukushima.

14 Even after performing additional calculations to account 15 for some of the relevant points raised by Entergy and NRC Staff, 16 ISR concludes that the values used by Entergy lead to an 17 underestimation of the offsite economic cost risk (OECR) by a 18 factor between 3 and 7 for IP2. Furthermore, ISR concludes that 19 a similar underestimation factor range applies to IP3.

20 Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 3

1 II. PREPARATION FOR REBUTTAL TESTIMONY 2 Q. Have you reviewed any additional documents since your 3 December 21, 2011 testimony in preparation for this rebuttal 4 testimony?

5 A. Yes.

6 Q. I show you Exhibits NYS00422A through NYS000431. Do 7 you recognize these documents?

8 A. Yes. These are true and accurate copies of the 9 documents that were referred to, used and/or relied upon in 10 preparing this rebuttal testimony. They include documents 11 prepared by ISR in support of this testimony, documents prepared 12 by government agencies, peer reviewed published articles, recent 13 reports related to Fukushima, and documents prepared by Entergy, 14 Sandia National Laboratories, Pacific Northwest Laboratory, NRC 15 or the utility industry.

16 Q. How do these documents relate to the work that you do 17 as an expert in forming opinions such as those contained in this 18 testimony?

19 A. These documents represent the type of information that 20 persons within my field of expertise reasonably rely upon in 21 forming opinions of the type offered in this testimony.

22 Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 4

1 III. SCOPE OF CONSOLIDATED CONTENTION NYS-12C 2 Q. Dr. Lemay, Id like to start off by clarifying the 3 scope of work ISR completed for NYS. Did NYS ask you to perform 4 an independent Severe Accident Mitigation Alternatives (SAMA) 5 analysis for IP?

6 A. No, the ISR report (NYS000242) was not intended to be 7 an alternative SAMA analysis.

8 Q. What was the scope of your analysis in connection with 9 NYS 12C?

10 A. The purpose of our analysis, as stated on the first 11 page of the ISR report (NYS000242), was to assess Entergys use 12 of the [MELCOR Accident Consequence Code System 2] MACCS2 code 13 to estimate the economic costs. NYS asked ISR to review and 14 analyze Entergys use of the MACCS2 code to estimate the 15 economic costs associated with a severe accident at IP, and to 16 review NRC Staffs evaluation of Entergys SAMA analysis, which 17 relied upon those economic costs, as part of the Final 18 Supplemental Environmental Impact Statement (FSEIS). NYS asked 19 ISR to assess whether the MACCS2 input values related to 20 economic costs at IP were reasonable and, if not, to provide a 21 reasonable range of input values.

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 5

1 Q. Does the ISR report (NYS000242) offer a critique of 2 the MACCS2 code itself or suggest that the code should not have 3 been used?

4 A. No. ISR has not commented on the use of the MACCS2 5 code itself or any limitations of the MACCS2 code. Instead, 6 ISRs analysis focuses on the inputs to the MACCS2 code that are 7 used to calculate the costs associated with a severe accident.

8 Q. How did ISR conduct its analysis of the MACCS2 code 9 input values used for IP?

10 A. ISRs analysis is focused on the effect of the 11 critical input parameters on the total economic cost of a severe 12 nuclear accident. After performing a sensitivity analysis to 13 identify the input parameters that had the largest impact on the 14 OECR, ISR compared Entergys input values with values calculated 15 from data produced by other analysts in the field. In other 16 words, ISRs Report (NYS000242) sought to benchmark Entergys 17 input values against others values and to provide a range of 18 reasonable, site-specific, and appropriate input values. This 19 concept of benchmarking is not foreign to the IP SAMA Analysis 20 since it was used by Entergy to compare its SAMA candidates with 21 those for other operating plants that have submitted license Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 6

1 renewal applications as discussed in Appendix G of the FSEIS 2 (NYS00133I at G-1).

3 Q. What do you mean by benchmarking?

4 A. Benchmarking is often used to validate codes and 5 methodologies when an exact solution or experimental data is 6 available. It is also used when there is no unique solution to a 7 problem, or when considerable uncertainty is associated to a 8 result.

9 In that context, benchmarking consists of establishing 10 points of reference by comparing ones current practices with 11 what others in the field are doing. In the nuclear industry, 12 benchmarking is an essential exercise because it provides for an 13 important exchange of information amongst experts in the field, 14 leading to the use of the best data and methodologies.

15 Q. How did ISR compare Entergys values to literature-16 derived values?

17 A. We relied on our experience and also reviewed the 18 literature to identify and calculate a range of site-specific 19 input values that experts in the field of accident mitigation 20 would use to calculate the economic costs associated with an 21 accident at IP. For many sensitive input parameters, ISR found 22 that the input values used by Entergy and accepted by NRC Staff Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 7

1 were outside the range of reasonable values identified in the 2 literature. Because of this, Entergy and NRC Staff 3 underestimated the total OECR by a factor of at least 3 to as 4 much as 7.

5 Q. Did either Entergy or NRC Staff perform a benchmarking 6 exercise for Entergys inputs to the MACCS2 code?

7 A. No, neither Entergy nor NRC Staff have attempted to 8 benchmark (i.e., quantify the reasonableness) the MACCS2 input 9 parameters used to calculate the OECR. This is inconsistent 10 with Entergy and NRC Staffs approach to other parts of the SAMA 11 analysis.

12 Q. Please explain how Entergy's lack of benchmarking for 13 the MACCS2 input parameters is inconsistent with their work on 14 other parts of the SAMA analysis.

15 A. As NRC Staff explained, their SAMA analysis for Indian 16 Point uses probabilistic risk assessments (PRA).

17 The PRA for a commercial power reactor has traditionally 18 been divided into three levels: Level 1 is the evaluation of the 19 combinations of plant failures that can lead to core damage; 20 Level 2 is the evaluation of core damage progression and 21 possible containment failure resulting in an environmental 22 release for each core-damage sequence identified in Level 1; and Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 8

1 Level 3 is the evaluation of the consequences that would result 2 from the set of environmental releases identified in Level 2.

3 All three levels of the PRA are required to perform a SAMA 4 analysis, and the MACCS2 code is used to perform the consequence 5 analysis in the Level 3 portion of the PRA.

6 NRC and Entergy expended considerable effort to verify the 7 source term and core damage frequency used in accident 8 assessmentLevel 1 and Level 2. This is explained in Appendix G 9 of the FSEIS (NYS00133I at G-1 to G-10). Entergy conducted peer 10 reviews and benchmarking to verify the reasonableness and 11 robustness of the Level 1 and Level 2 PRA as described in 12 Attachment I to NL-08-028 (ENT000460). These initiatives 13 allowed Entergy to quantify the uncertainties on the key 14 parameters of the Level 1 and Level 2 PRA and to account for 15 these uncertainties with multipliers. Specifically, Entergy 16 applied a multiplier of 8 to account for internal and external 17 events (i.e. PRA Levels 1 and 2) and the corresponding 18 uncertainty (Entergy Testimony A62).

19 There is no evidence that Entergy's calculation of the cost 20 and duration of the decontamination for IP has been documented 21 and that Entergy performed the same peer review or benchmarking 22 for the Level 3 PRA, with the exception of the VALWNF value, Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 9

1 which was later revised by Entergy to address the loss of 2 tourism and business. Entergy has not explained why they 3 conducted the benchmarking exercise for the Level 1 and Level 2 4 PRA, but not the Level 3 PRA.

5 Q. What effect does the lack of Level 3 benchmarking have 6 on the SAMA analysis?

7 A. Without this benchmarking, Entergy and NRC have not 8 quantified the reasonableness and the uncertainty of the key 9 parameters they used to calculate the OECR. In short, Entergys 10 SAMA analysis is not defensible because the portion of the SAMA 11 analysis that relies upon the MACCS2 code lacks a documented 12 cost basis, a robust peer review, and a benchmarking exercise.

13 Q. But in A6b, NRC Staffs experts claim that the 14 existing margin in the SAMA analysis account for 15 uncertainties. Their experts then go on to conclude that this 16 margin accounts for uncertainties in all three levels of the 17 PRA. Do you agree with this assertion?

18 A. No. The multipliers for the Level 1 and 2 analyses do 19 not account for uncertainties in the Level 3 assessment, such as 20 those for the decontamination cost or time. As I just 21 explained, Entergy and NRC have not attempted to quantify the Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 10

1 uncertainty of the Level 3 PRA and therefore, cannot account for 2 the magnitude of the uncertainty.

3 IV. RESPONSE TO ENTERGYS AND NRC STAFFS CLAIM THAT FUKUSHIMA 4 REPRESENTS A WORST-CASE SCENARIO AND IS NOT RELEVANT 5 Q. Dr. Lemay, please explain how the MACCS2 code models 6 severe accidents and ultimately calculates OECR.

7 A. The MACCS2 code models eight different categories of 8 severe accidents, called release categories. As I explained in 9 my December 21, 2011 testimony and report, the OECR calculated 10 by the MACCS2 code is a frequency-averaged cost. The OECR is 11 obtained by adding the total offsite economic cost for each of 12 eight release categories after weighting them by their 13 respective frequencies.

14 Q. Are the eight release categories defined by the user?

15 A. Yes. Typically, the eight release categories are 16 selected by the user to represent a range of severe accidents, 17 from lower consequence/higher probability accidents to higher 18 consequence/lower probability accidents. In the case of the IP 19 SAMA analysis, the eight release categories were defined by 20 Entergy.

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 11

1 Q. Dr. Lemay, how do the eight release categories, as 2 modeled by Entergy compare to last years nuclear disaster at 3 Fukushima?

4 A. The bar chart below was created by ISR for this 5 testimony. It shows the activity of Cs-137 released during the 6 Fukushima accident compared to that of each release category 7 used by Entergy (ENT000464).

Activity of Cs-137 released 1.00E+17 Fukushima 1.00E+16 Activity released (Bq) 1.00E+15 1.00E+14 IP2 1.00E+13 IP3 1.00E+12 1.00E+11 NCF EARLY EARLY EARLY LATE LATE LATE LOW LATE HIGH MEDIUM LOW HIGH MEDIUM LOWLOW Release Category 8

9 The source of the Fukushima release used in the bar chart 10 are the reports by the Investigation Committee on the Accident 11 at Fukushima Nuclear Power Stations of Tokyo Electric Power 12 Company, Chapter 5, December 26, 2011, attached hereto as 13 Exhibit NYS00422A-NYS00422C; and Institut de Radioprotection et Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 12

1 de Sûreté Nucléaire, Note dinformation, 22 mars 2011, attached 2 hereto as Exhibit NYS000423.

3 As shown in the chart of the 8 release categories modeled 4 by Entergy, the activity of Cs-137 released at Fukushima is 5 actually six to ten times lower than the activity that Entergy 6 modeled for the simulated release category EARLY HIGH.

7 Q. Do you agree with Entergy and NRC Staff's criticism of 8 your work saying that Fukushima would be considered a worst-9 case scenario accident?

10 A. No. Fukushima is severe, but not worst-case. As I 11 previously explained and as seen on the bar chart, Fukushima 12 fits within the range of severe accidents defined by Entergy.

13 Q. Dr. Lemay, why did you use data from the Fukushima 14 accident to estimate consequences at IP?

15 A. The accident at Fukushima gives us the most recent 16 information available on the timeline and the magnitude of 17 decontamination efforts following a severe accident. It is a 18 real-world point of reference for assessing the cost of a severe 19 accident at IP.

20 Q. What impact do the eight release categories have on 21 OECR?

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 13

1 A. Some of the eight release categories have a relatively 2 small economic impact, while some have a relatively large 3 economic impact. Tables 5 and 6 from ENT000464, reproduced 4 below, show the contribution of each release category to the 5 total OECR.

6 7

8 9

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 14

Contribution of release modes to Contribution of release modes to the total OECR for IP2 the total OECR for IP2 Late medium Other Late medium Other Late high Late high Early medium Early high Early medium Early high 1

2 As can be seen from the numbers circled in red, release 3 category EARLY HIGH contributes over 60% to the total OECR for 4 IP2 and IP3. EARLY HIGH, EARLY MEDIUM, and LATE HIGH, 5 contribute over 90% to the total OECR.

6 On the other hand, release categories NCF, EARLY LOW, 7 LATE LOW, and LATE LOWLOW combined contribute less than 1%

8 to the total OECR for IP2 and IP3.

9 Q. Do you agree with NRC Staff, that input parameters 10 that represent an average of all the release categories are 11 appropriate to use, as they suggest in NRC Staff Testimony A6d?

12 A. No, particularly if the use of average parameters 13 gives unrealistic results for the release categories that 14 contribute the most to the OECR. In general, Entergy selected 15 input parameters that are appropriate for the release category 16 being modeled (release fractions, release duration); however, 17 for TIMDEC and CDNFRM, they used the same values derived from Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 15

1 Sample Problem A for every release category. The suggestion by 2 NRC Staff that it is acceptable to average input parameters over 3 all release categories is wrong. In fact, the relative 4 contribution of all release categories is determined when the 5 Offsite Economic Cost is multiplied by the frequency to obtain 6 the OECR.

7 If Entergy and NRC Staff insist on using a single value for 8 the input parameters related to the cost of decontamination, it 9 would be appropriate to use input parameters that more closely 10 align with the more severe end of the release spectrum because 11 the three most severe release categories make the largest 12 contribution to the total OECR.

13 V. RESPONSE TO ENTERGY AND NRCS TESTIMONY REGARDING THE 14 HISTORY AND PEDIGREE OF NUREG-1150 15 Q. In their testimony, both NRC Staff and Entergy argue 16 that the only appropriate values for decontamination time and 17 costs which should be used to analyze offsite economic costs in 18 the area around IP come from NUREG-1150 (NRC Staff Testimony at 19 A39, Entergy Testimony at A51). Are they referring to the 20 Sample Problem A values discussed in the ISR report 21 (NYS000242)?

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 16

1 A. Yes. The Sample Problem A values are the same as the 2 values for the Surry nuclear power plant in NUREG-1150.

3 Q. What is your understanding regarding NRC Staff and 4 Entergys rationale for only using values from Sample Problem A?

5 A. NRC Staff and Entergy focus on what Entergy calls the 6 pedigree of NUREG-1150 (Entergy Testimony at A26, A72, A76, 7 A95). They argue that the Sample Problem A values found in 8 NUREG-1150 are widely recognized and accepted in the nuclear 9 community. Both parties also justify the appropriateness of the 10 values found in Sample Problem A by stating that NUREG-1150 was 11 subject to extensive peer review and public comment (Entergy 12 Testimony at A26; NRC Staff Testimony at A39).

13 Q. Were the Sample Problem A values from NUREG-1150 14 developed specifically for IP?

15 A. No. As I explained in my previous testimony 16 (NYS000241) and the ISR report (NYS000242), these values were 17 developed for the Surry site surrounded by farmland in rural 18 Virginia.

19 Q. NRC Staff (A39) supports the use of values from Sample 20 Problem A by discussing the fact that NUREG-1150 was properly 21 vetted because it was subject to public comment. Have you Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 17

1 evaluated whether any public comments were received on the 2 economic cost portion of NUREG-1150?

3 A. Yes, the following comments were included in NUREG-4 1150d (NYS00252D at D-31 to D-32):

5 [PUBLIC] COMMENT: The models used in calculating 6 the cost of a severe accident lack many factors 7 that should be taken into account. Many of the 8 assumptions are questionable and unfounded. The 9 models have not been benchmarked. Some 10 interpretations and conclusions that were made in 11 draft NUREG-1150 are questionable. The cost 12 estimates need to be more thoroughly documented 13 to understand and evaluate the calculations.

14 [NRC] RESPONSE: The present version of NUREG-1150 15 provides a limited set of risk-reduction 16 calculations, principally related to the 17 potential benefits of accident management 18 strategies in reducing core damage frequency. It 19 does not assess the costs of these or other 20 improvements. Such analyses are more properly 21 considered in the context of specific regulatory 22 action.

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 18

1 * *

• 2 [PUBLIC] COMMENT: Decontamination costs used in 3 the calculations may be based on decontamination 4 of test sites in deserts instead of agricultural, 5 residential, and commercial property.

6 [NRC] RESPONSE: The draft NUREG-1150 cost/benefit 7 analyses reflected the conventional NRC methods 8 for assessing costs and benefits. Because 9 cost/benefit analyses are more properly 10 considered in the context of specific regulatory 11 activities, they are not provided in this version 12 of NUREG-1150.

13 Q. Do the public comments and NRC's responses to comments 14 on NUREG-1150 shed any light on how economic costs associated 15 with a severe accident should be calculated?

16 A. Yes, NRC's responses to the comments I just cited 17 suggest that the authors of NUREG-1150 expected NRC Staff to 18 require site-specific assessments of the costs of 19 decontamination in the context of specific regulatory 20 activities, such as those currently being conducted at IP. The 21 responses from the authors of NUREG-1150 do not justify the use Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 19

1 of the Sample Problem A inputs at IP or any other particular 2 reactor.

3 Q. What is NUREG-1150s justification for the Sample 4 Problem A decontamination times of 60 days and 120 days for 5 light and heavy decontamination, respectively?

6 A. NUREG-1150d, page D-30 (NYS00252D) states:

7 A reduction by a factor of three was assumed 8 to require 60 days of decontamination work; 9 a reduction by a factor of 15 was assumed to 10 require 120 days of decontamination work.

11 The decontamination efforts were assumed to 12 commence at the end of the 7-day emergency 13 phase.

14 NUREG-1150 has no further justification for the use of the 15 decontamination times in Sample Problem A.

16 Q. What other support does NUREG-1150 have for the Sample 17 Problem A cost and decontamination time values?

18 A. NUREG-1150 (NYS00252A) states on p. 2-20 that the 19 reader seeking extensive discussion of the methods used is 20 directed to [r]eference [NUREG/CR-4551 (NRC000057), a companion 21 study published in December 1990] and [r]eference [NUREG/CR-22 4691, the MACCS manual] which discusses the computer code used Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 20

1 to perform the offsite consequence analysis (i.e., the MELCOR 2 Accident Consequence Code System (MACCS), Version 1.5).

3 Q. Did you review NUREG/CR-4551, referenced by NRC Staff 4 (A39, A61, A81) and Entergy (A35, A36, A70, A72, A76, A77, A88, 5 A107), for documentation on how the Sample Problem A 6 decontamination times and costs were derived?

7 A. Yes, the document only contains the Sample Problem 8 input files for each of the five sites (including the Surry 9 Site). NUREG/CR-4551 discusses many of the MACCS2 input 10 parameters, but does not discuss how the decontamination time 11 and cost of decontamination were obtained.

12 Q. Do NRC Staff and Entergy cite any other documents to 13 support the pedigree of NUREG-1150 and the decontamination times 14 they used from Sample Problem A?

15 A. Yes. While NUREG-1150 and NUREG/CR-4551 provide no 16 detailed explanation for decontamination times, in A81 of NRC 17 Staffs testimony, Staff cites an additional document published 18 in April 1984 as support. This document is NUREG/CR-3673, 19 Economic Risks of Nuclear Power Reactor Accident (NRC000058).

20 NUREG/CR-3673 does describe a timeline for the duration of 21 decontamination, but its description in Figure 4.2 (NRC000058 at 22 4-5) is not entirely consistent with the description contained Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 21

1 in NUREG-1150. In NUREG/CR-3673, the first 30 days after the 2 start of the accident sequence are used for the collection of 3 dose-rate information for decision making. Thus, 4 decontamination starts at 30 days and ends at 120 days. In 5 NUREG-1150, decontamination starts at 7 days. This discrepancy 6 illustrates that there is no single documented basis for the 7 cost and time of decontamination.

8 Q. Did you make any observations regarding the timeline 9 for the duration of decontamination which was discussed in 10 NUREG/CR-3673?

11 A. Yes, I have concluded that the assumptions made in 12 NUREG/CR-3673 are unreasonable for a severe accident at IP.

13 To conclude that average clean-up at Surry would take 90 14 days with approximately 46,000 workers, NUREG/CR-3673 (NRC00058 15 at 4-20) assumed labor comprises half the cost of 16 decontamination and that a worker would cost $30,000 per year.

17 Applying the same methodology as NUREG/CR-3673 to the cost 18 of decontamination calculated by Entergy for the EARLY HIGH 19 release category at IP2, a total of 1.5 million workers (363,000 20 worker-years) would be required to decontaminate the affected 21 area in 90 days. It is not reasonable to conclude that 1.5 22 million workers would be available and therefore, used to Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 22

1 decontaminate an affected area after a severe accident.

2 Therefore, the assumptions regarding the timeline in NUREG/CR-3 3673 are invalid.

4 Even if we assumed that decontamination will take a full 5 year, 363,000 workers would still be required. This too is 6 unlikely and leads me to conclude that the average time for 7 decontamination would be more than a year.

8 ISR created an exhibit containing the details of the 9 analysis I just explained, which is attached hereto as Exhibit 10 NYS000431.

11 Q. Does NUREG/CR-3673 provide any support for the Sample 12 Problem A decontamination cost?

13 A. On page 4-15, NUREG/CR-3673 (NRC00058) gives 14 approximate costs of decontamination that, once adjusted for 15 CPI, match the values used in Sample Problem A. It further 16 states:

17 The cost estimates used in this study for 18 various levels of decontamination effort in 19 an area are taken from a detailed review of 20 decontamination effectiveness and costs 21 performed at Sandia National Laboratories 22 (SNL) [Os84].

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 23

1 There is no further discussion of the cost estimates in the 2 NUREG/CR-3673 document. It thus appears that the costs 3 contained in Sample Problem A are documented in the document 4 referred to as [Os84].

5 The references section of NUREG/CR-3673 lists [Os84] as 6 Ostmeyer, R.M., and G.E. Runkle, An Assessment of 7 Decontamination Costs and Effectiveness for Accident 8 Radiological Releases. Albuquerque, N.M.: Sandia National 9 Laboratories, to be published. NRC00058 at 8-8.

10 Q. Does Os84 shed light on what data the Sample Problem A 11 values were based upon?

12 A. The document [Os84] upon which the Sample Problem A 13 cost estimates are based, as stated in NUREG/CR-3673, does not 14 appear to exist in a published form and therefore was not likely 15 to have been subject to peer review or public comment.

16 Therefore, it is not a reliable source upon which experts in 17 this field would base any findings.

18 We conducted a search and could not locate [Os84]. We 19 asked NYS for a copy, but NYS was not able to locate the 20 reference.

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 24

1 Q. Has Entergy or NRC Staff offered any document that 2 shows the calculations and/or data that support the use of the 3 values found in Sample Problem A?

4 A. No.

5 VI. EFFECT OF NRCS PRIOR SITE-SPECIFIC ANALYSIS OF 6 DECONTAMINATION FOR INDIAN POINT 7 Q. Other than the ISR report (NYS000242), are you aware 8 of any site-specific analysis of the economic costs associated 9 with a severe accident at IP?

10 A. I am not aware of any site-specific analyses performed 11 by Entergy.

12 However, while researching the pedigree of NUREG-1150 we 13 learned that in the 1980s NRC Staff contracted with Battelle 14 Pacific Northwest Laboratory (PNL) to conduct a case study of 15 the economic costs associated with severe accidents at IP. This 16 site-specific case study is described in Chapter 5 of NUREG/CR-17 5148 Property-Related Costs of Decontamination, dated February 18 1990 and attached hereto as Exhibit NYS00424A-NYS00424BB.

19 Q. How did you locate NUREG/CR-5148?

20 A. When attempting to respond to NRC's testimony 21 regarding the appropriateness of Sample Problem A and its roots Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 25

1 in the pedigree of NUREG-1150, ISR researched the origins of 2 the economic cost parameter input values reported in NUREG-1150.

3 NUREG-1150 contained a citation to a document which 4 appeared to form the basis of the Sample Problem A values 5 entitled: NUREG/CR-3413 Off-Site Consequences of Radiological 6 Accidents: Methods, Costs and Schedules for Decontamination, 7 which was completed in November 1984, published in August 1985, 8 and is attached hereto as Exhibit NYS00425A-NYS00425G.

9 NUREG/CR-3413 explains the results of a PNL study commissioned 10 by NRC following the accident at Three-Mile Island. It further 11 describes a database and computer program called DECON that was 12 developed to conduct a decontamination analysis of a large, 13 radiologically contaminated area. DECON was designed to be used 14 with CRAC2, a predecessor to the MACCS2 code. The methodology 15 used by DECON appears to be very similar to that of CONDO 16 (NYS000250).

17 ISR was interested in learning more about the database and 18 DECON program, but could not find any use or mention of the 19 DECON code past the mid-1980s. Through internet searching, ISR 20 located an email address for one of the NUREG/CR-3413 authors, 21 J.J. Tawil. On May 2, 2012, ISR sent Dr. Tawil an email 22 inquiring about the history of the DECON program and what became Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 26

1 of it. Our email exchange with Dr. Tawil is attached hereto as 2 Exhibit NYS000426.

3 Dr. Tawil wrote:

4 I think the primary difficulty was that my 5 last project for the NRC was to characterize 6 the off-site consequences of reactor 7 accidents . . . for three reactor sites, one 8 of which was Indian Point . . . . I think 9 the NRC was a little shocked at the 10 magnitude of the off-site consequences of an 11 SST-5 at Indian Point and decided not to 12 publish the report.

13 After receiving this email from Dr. Tawil, we shared it 14 with NYS. Its my understanding that the NYS librarian was able 15 to locate NUREG/CR-5148 based on the information in Dr. Tawils 16 email and a reference to a to-be-published 1990 document in 17 Sandia Site Restoration.

18 Q. How do the decontamination values in the NUREG/CR-5148 19 IP case study compare to ISRs analysis of decontamination costs 20 using CONDO?

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 27

1 A. The methodology employed by PNL using the DECON code 2 provides further support for using data to develop site-specific 3 MACCS2 input values.

4 It is not possible to directly compare the decontamination 5 costs contained in NUREG/CR-5148 with those calculated by 6 Entergy because they use different habitability criteria, 7 different decontamination factors (DFs) and different source 8 terms.

9 However, the detailed calculations performed by Dr. Tawil 10 (NYS00424A-NYS0424BB) show that in urban areas, the cost of 11 decontamination of building contents can exceed the cost of 12 decontamination of land and structures (NYS000424G, Figure 4.3, 13 at 4.26 - 4.28). He also finds that in urban areas, 14 decontamination of building contents is labor intensive and 15 labor costs are a large fraction of the cost of decontamination 16 (NYS000424B - NYS000424E, Section 2.4, at 2.8 - 2.71).

17 Q. Has either Entergy or NRC Staff explained why they did 18 not rely on the database and DECON program developed by PNL or 19 NUREG/CR-5148 IP case study to develop site-specific inputs for 20 the IP SAMA analysis?

21 A. No.

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 28

1 Q. Given the history and lack of documented support for 2 the Sample Problem A values, is it reasonable for Entergy and 3 NRC Staff to rely on Sample Problem A and ignore the data and 4 literature cited in the ISR report and the NUREG/CR-5148 IP 5 site-specific case study commissioned by NRC?

6 A. No. At a minimum, Entergy and NRC Staff should have 7 benchmarked the Sample Problem A values from NUREG-1150, to 8 verify their robustness when applied to determine the OECR at 9 IP. Such a benchmarking exercise cannot replace a detailed 10 calculation of the cost or time of decontamination, but it can 11 give confidence that the values are reasonable.

12 ISR conducted such a benchmarking exercise. However, 13 Entergy and NRC Staff have dismissed the wealth of other 14 relevant literature and data upon which ISR relied:

15 Experience at Chernobyl (NYS000249, NYS000250, 16 NYS000251, NYS000263) 17 Experience at Fukushima (NYS000264, NYS000265, 18 NYS000266, NYS000267, NYS000268, NYS000269) 19 Data from CONDO Report (NYS000250) 20 Data from RISO (NYS000251, NYS000253) 21 Data from Luna Report (NYS000255) 22 Data from Site Restoration Report (NYS000249)

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 29

1 Data from Reichmuth Report (NYS000256) 2 Q. Is Entergys statement in A72 that [a]lthough the 3 basis for the NUREG-4551 cost estimate for farm decontamination 4 is not fully explained in that document, we know of no more 5 appropriate decontamination data that is readily available to 6 licensees to use in a SAMA analysis reasonable?

7 A. No. As shown in the ISR report, many other data 8 sources exist that Entergy could have relied upon for 9 benchmarking purposes or to generate site-specific input values.

10 VII. RESPONSE TO ENTERGY AND NRC STAFF TESTIMONY REGARDING ISRS 11 MACCS2 INPUT CALCULATIONS 12 Q. Are there any topics discussed in NRC Staff or 13 Entergys testimony that ISR does not contest?

14 A. Yes, in their testimony, NRC Staff and Entergy discuss 15 some points that are not part of Contention 12C or that have no 16 impact on the conclusions of reached by ISR. In reaching its 17 conclusions, ISR did not rely on challenges to the following 18 points:

19 MACCS2 code (Entergy Testimony at A26, A37 to A52; NRC 20 Staff Testimony at A60): Contention 12C does not 21 contest the use of MACCS2 for SAMA analysis. ISR 22 agrees that MACCS2 input parameters represent suitable Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 30

1 averages for the cost and decontamination time in a 2 grid element.

3 Dose reduction factor (Entergy Testimony at A87; NRC 4 Staff Testimony at A52): and deposition velocity (NRC 5 Staff Testimony at A41 to A43). ISRs calculations 6 used the same dose reduction factors (3 and 15) as 7 Entergy uses. ISR did not modify or contest the use 8 of the deposition velocity chosen by Entergy for its 9 calculations. In fact, ISR agrees that MACCS2 uses 10 dose reduction factors that are related, but not 11 identical to DFs.

12 Cesium (Entergy Testimony at A126; NRC Staff Testimony 13 at A20 and A21): ISR recognized and agreed that Cesium 14 is the most important radionuclide of consideration 15 for decontamination following a severe accident at a 16 nuclear reactor.

17 Q. What is your response to Entergys argument (ENT000449 18 at 27-28) that the SAMA analysis estimates average consequence 19 results for the entire 50-mile radius region around the IP site 20 (an area of approximately 7,854 square miles), not just the 21 comparatively small region of New York City, which comprises 22 approximately 2% and that the State attempt[s] to scale up Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 31

1 certain cost estimates related to the New York City portion 2 without including commensurate scaling down of estimates for the 3 98% of the SAMA analysis region that is outside of New York 4 City?

5 A. We disagree with Entergy. Using the data from Site 6 Restoration as a starting point, there is no need to scale down 7 the cost of decontamination in the region outside of New York 8 City because the population density (790 pers per km2) is 9 comparable to the density around Albuquerque (700 pers per km2).

10 In our original calculation, the NYC area accounted for 2%

11 of the 50 mile area around IP. After removing the surface area 12 which could be accounted for by water and farmland, our revised 13 calculations show that the area with more than 10,000 persons 14 per km2 (i.e. what we classified as the NYC metro area) is 4% of 15 the total non-farmland area.

16 ISR is well aware that the SAMA analysis estimates average 17 consequences for the entire 50-mile radius region and we have 18 averaged the costs on the basis of land use area. To be clear, 19 ISR did not apply the same cost to the 4% non-farmland area that 20 represents NYC as to the remaining 96%. We only assigned the 21 high cost per km2 for decontaminating NYC to a very small area 22 (4%). We assigned the lower cost of decontaminating the Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 32

1 remaining urban or semi-urban area to the rest of the 50 mile 2 area around IP that is not farmland or water, an area that is 3 76% of the total area. The other 20% accounting for water and 4 farmland is not included in our calculation.

5 A. ISRs Response to Testimony Regarding the Calculation 6 Of Population Dose Risk (PDR) 7 Q. In A82, Entergy claims that ISR should have considered 8 not only OECR, but also the changes in Population Dose Risk 9 (PDR). What is your response?

10 A. ISR provided all of the MACCS2 output files, which 11 also contain the population dose results for the calculations of 12 PDR.

13 ISR agrees that changes in some of the input parameters may 14 affect PDR, and the costs associated with PDR, in a different 15 manner than how they affect OECR. In response to Entergy and 16 NRC Staffs arguments, ISR checked the impact of the input 17 parameter changes discussed in the ISR report on the dollar 18 value equivalent of PDR.

19 Q. What is the effect of the changes to input parameters 20 discussed in the ISR report on PDR?

21 A. For all of the MACCS2 runs that we used to justify our 22 conclusions, we have now calculated the present dollar value Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 33

1 equivalent of the total off-site population dose from the PDR 2 and the total OECR using the equations provided in the Entergy 3 report: Re-Analysis of IP2 and IP3 SAMAs (ENT000459). The total 4 present dollar value equivalent for ISR reports maximum input 5 values (NYS000242, Table 13), decreases from $17.8M to $17.1M if 6 the cost associated with the decrease in PDR is included.

7 In summary, the change in PDR as calculated in our MACCS2 8 runs has no impact on our conclusions. It changes the total cost 9 by less than 4%.

10 Q. Do ISRs conclusions apply only to IP2?

11 A. No. In the ISR report (NYS000242), ISR ran the MACCS2 12 code for IP2 to illustrate the changes in the calculated OECR 13 when the sensitive inputs are changed.

14 In A27 and A28, NRC Staff challenged this by saying that 15 IP2 has a higher source term than IP3. As I explained above, the 16 source term is the activity released during the accident and in 17 MACCS2, it is specified as a release fraction of the core 18 inventory in units called Becquerel. For the SAMA analysis, the 19 activity of Cesium-137 released is the parameter of interest.

20 Its true that IP2 has a higher source term than IP3, but 21 its also true that IP3 has higher release mode (or release 22 category) frequencies. The release mode frequency is a measure Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 34

1 of the likelihood of this particular release mode (or release 2 category) and it multiplies the off-site economic cost to obtain 3 the OECR. The effect on OECR by the higher source term for IP2 4 is offset by the effect of the higher release mode frequencies 5 for IP3 and, therefore, the resulting OECR values for IP2 and 6 IP3 are comparable. They are $2.12E+05/yr for IP2 and 7 $2.61E+05/yr for IP3.

8 In response to Entergy's and NRC staff's testimony, ISR re-9 ran the MACCS2 simulations using the ATMOS input file for IP3 10 and ISRs proposed input values (i.e. in the CHRONC input file) 11 to verify that there is no material difference between IP2 and 12 IP3. The results for IP3 were comparable to IP2. The 13 underestimation of the OECR calculated by Entergy resulted in a 14 factor between 3.2 and 7.6 for IP3, compared to 3.0 and 6.9 for 15 IP2. These are the revised factors after ISRs calculations were 16 adjusted to account for the comments made by Entergy and NRC 17 Staff.

18 B. ISRs Response to Testimony Regarding Nonfarm 19 Decontamination Cost (CDNFRM) 20 Q. Did ISR evaluate whether assumptions regarding CDNFRM 21 raised by Entergy and NRC Staff would affect the ultimate 22 calculation of OECR?

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 35

1 A. ISR determined that although much of the testimony 2 presented by Entergy and ISR had no bearing on our analysis, 3 some of the testimony required further analysis in order to 4 determine whether the ultimate calculation of the OECR would 5 change based on the assumptions Entergy and NRC Staff provided 6 in their testimony. We performed further analysis of several 7 parameters (e.g., the cost of decontamination per unit area) 8 used to calculate CDNFRM.

9 Q. What is your response to NRC Staffs argument in NRC 10 Staff A64 and A65 that the area ISR used to calculate CDNFRM 11 improperly includes surface water and farmland?

12 A. In response to Entergy and NRC Staffs point that 13 surface water and farmland should not be included in the area 14 used to calculate CDNFRM, ISR recalculated the values for 15 CDNFRM, subtracting surface water and farmland from the area.

16 As shown in Table 13 of the revised tables from the ISR report 17 that ISR created for this testimony, attached hereto as 18 NYS000430, the impact of this change is a decrease of 13% at the 19 low end of the range for the OECR. There is no change at the 20 high end of the range. Thus, even when surface water and 21 farmland are removed from the area, Entergy and NRC Staff have 22 still underestimated OECR by, at minimum, a factor of 3.

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 36

1 Q. In NRC Staff A64 and A65, Staff claims that ISR 2 incorrectly accounted for parkland in the calculation of CDNFRM.

3 Is this true?

4 A. No, ISRs approach to parkland is correct. Based on 5 the Entergy MACCS2 site input file, parkland area was not 6 accounted for as farmland and therefore must be included in the 7 only other category, as nonfarm land area.

8 In our calculations, we implicitly categorize parkland as 9 the area outside the NYC metro area, which is either semi-urban 10 or urban. This is justifiable because in order to retain the 11 value of the parks, the cost of decontamination (principally 12 tree felling and removal) and replacement of trees, soil and 13 grass is likely comparable to a semi-urban or urban area on a 14 per square kilometer basis.

15 This approximation is valid since park land accounts for 16 about 10% of the urban or semi-urban area outside NYC metro area 17 and is therefore, not a large component of the cost.

18 (1) ISRs Response to Testimony Regarding the Use of 19 Site Restoration and the Use of the Reichmith Study 20 Q. What is your response to Entergys Testimony at A121 21 that ISR should not have included the cost of compensation in Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 37

1 calculating CDNFRM in Approach A (Site Restoration/Luna) and 2 Approach B (Reichmuth)?

3 A. We removed the compensation costs and re-calculated 4 CDNFRM for Approaches A and B, as shown in the tables attached 5 hereto as NYS000430. The resultant effect on the OECR is a 6 decrease of the upper bound by 5%.

7 Q. What is your response to Entergys assertion in A121 8 that ISR ignores differences between New York and Albuquerque 9 and, thus, improperly extrapolates data from Site Restoration?

10 A. In A121, Entergy asserts that [d]econtamination costs 11 would be expected to be generally proportional to [interior]

12 square footage for a residence. Entergy claims that the 13 interior square footage for housing is greater in Albuquerque 14 compared to New York City (NYC), and that there should be a 15 reduction in the decontamination costs using Site 16 Restoration/Luna for NYC. Entergy further states the same would 17 be expected for commercial property.

18 Reference ENT000469 shows that average homes in the five 19 boroughs of NYC are smaller than in Albuquerque. There is no 20 evidence, however, that the multi-story residential and 21 commercial buildings that are common in the NYC area would have Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 38

1 less interior square footage than in Albuquerque, which has 2 fewer multi-story buildings.

3 Furthermore, Entergy (A121, A129) claims that the housing 4 density in the area outside the NYC metro region is about 50%

5 that of the City of Albuquerque, so there should be a reduction 6 in the decontamination costs for this region.

7 The authors of the Site Restoration Report state that they 8 visited several residential, commercial, and industrial sites 9 in and near Albuquerque, NM. (NYS000249) Therefore, if 10 housing density is used to assess the cost, the whole area 11 around Albuquerque City must be considered. From the New 12 Mexico: 2000 Population and Housing Unit Counts (Census Bureau),

13 attached hereto as NYS000427, the housing density of the 14 Albuquerque census county division is approximately 700 15 units/square mile, which is comparable to the 790 units/square 16 mile reported by Entergy for the area outside the NYC Metro 17 region.

18 Q. What is your response to Entergys testimony in A129 19 that the data from Site Restoration is inappropriate for the 20 calculation of CDNFRM since it is based on a plutonium 21 dispersion scenario, for which decontamination is more expensive 22 than for a nuclear reactor accident?

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 39

1 A. Our use of Site Restoration was intended to be a point 2 of reference in a benchmarking exercise that illustrates various 3 approaches to determining the cost of decontamination ($/km2) and 4 the resulting CDNFRM ($/person). For estimation purposes, the 5 two parameters extracted from Site Restoration: the 6 decontamination cost per area and the corresponding DF, are 7 irrespective of the geographic extent and severity of the 8 accident.

9 The data from Site Restoration is one way to calculate 10 CDNFRM, and the ISR report explains how the Site Restoration 11 costs can be adjusted to be more applicable to a nuclear 12 accident (NYS000242, Approach A at 16-18). For example, the ISR 13 report assumed that the costs of a nuclear reactor accident 14 would be one to two times higher than the costs of a plutonium 15 dispersion scenario. This is because Cs-137, as well as the 16 remaining gamut of radionuclides released (including Pu-239) 17 from the reactor, must be considered when decontaminated, in 18 addition to Pu-239.

19 Q. What is your response to Entergys testimony in A122 20 and A134 that the data from Site Restoration focuses on a small 21 geographical area, and did not account for economies of scale 22 that could be realized for larger areas?

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 40

1 A. It is reasonable that economies of scale are possible 2 during the decontamination of large areas, such as farmland.

3 However, NUREG/CR-5148 (NYS00424A-NYS0424BB) shows that a large 4 part of the cost of decontamination in urban areas is related to 5 the labor intensive tasks of cleaning the buildings contents.

6 There appears to be no practical way to scale these tasks 7 economically.

8 In addition, the logistics of planning a large scale 9 decontamination program, including budget approval, approval for 10 waste repositories, and set up times would be long, contentious 11 and difficult. For comparison, it has taken one year to 12 finalize plans and budgets for preliminary remediation efforts 13 in the Fukushima Prefecture as shown in the Road to Recovery, 14 attached hereto as NYS000428.

15 Q. What is your response to Entergys testimony in A135 16 that consideration of on-site waste disposal and waste volume 17 reduction, as discussed in Site Restoration, would lead to 18 significantly lower estimates of decontamination costs?

19 A. There are significant issues concerning on-site waste 20 disposal for a severe nuclear accident. The approval and 21 planning for on-site waste repositories is a long and 22 contentious process (NYS000264). In a hyper-urban region such as Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 41

1 NYC, on-site disposal is likely impossible due to the lack of 2 suitable space. For the rest of the 50-mile SAMA region, there 3 would be long drawn out negotiations to identify suitable 4 locations for on-site disposal. Wherever such repositories are 5 located, they would result in a decline in property values, 6 which is currently not considered in the MACCS2 calculation.

7 Therefore, we believe it may be more cost-effective to employ 8 off-site waste disposal.

9 The waste volume reduction of 50-60% considered in Site 10 Restoration is for farmland or rangeland decontamination. Our 11 use of Site Restoration data is solely for the determination of 12 CDNFRM, which is only for nonfarm land. Therefore, this waste 13 volume reduction does not apply to our calculations.

14 (2) ISRs Response to Testimony on the Use of CONDO 15 Q. What is your response to Entergys testimony in A143 16 that when using the CONDO approach, the decontamination 17 techniques that ISR chose are not consistent with the DFs for 18 light and heavy decontamination?

19 A. We disagree. ISR selected decontamination techniques 20 from the lists contained in CONDO (NYS000250 at Table A7 of 21 Appendix A) that most closely represented light and heavy 22 decontamination. Some techniques, such as felling and removal Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 42

1 (DF=10) of trees are essentially the only option for both levels 2 of decontamination. Vacuuming, cleaning and washing (DF=5) for 3 building interiors are also the only option for both levels of 4 decontamination. After removing these techniques and other 5 outliers such as removal (DF=1.4) for plants and shrubs and 6 removal and replacement (DF=50) for paved areas, light 7 decontamination techniques have a DF of 3 to 5, and heavy 8 decontamination techniques have a DF of 10.

9 Q. What is your response to Entergys testimony in A145 10 that ISR's approach to adding a hyper-urban designation for 11 the NYC metro region and classifying the remaining area as urban 12 is unjustifiable when using the CONDO methodology?

13 A. The CONDO methodology classifies environments 14 according to their population density: rural (<25 per km2); semi-15 urban (25 to 1000 per km2); and urban (>1000 per km2) (NYS000250 16 at 54). With the revisions made to the nonfarm land area I 17 discussed on page 36 of this testimony, there are two population 18 grid elements (designated as the NYC Metro area) that have a 19 population density that exceeds 10,000 per km2, based on the 20 Entergy MACCS2 site input file. The remaining grid elements 21 have a population density between 25 and 10,000 per km2.

22 Therefore, ISR determined that it was necessary to add a hyper-Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 43

1 urban designation for the NYC Metro area, given its 2 extraordinarily high population density.

3 To define a range, we calculated the minimum value based on 4 the assumption that the NYC Metro region is urban, and the 5 remaining area is semi-urban. This is supported by the 6 aforementioned classifications provided in CONDO.

7 Q. What is your response to NRC Staffs testimony in A69 8 that MACCS2 assumes that contamination is spread over a flat 9 plane and that it implicitly addresses 3-D structures through 10 the use of per capita costs like CDNFRM and, thus, ISR should 11 not have included multipliers in the CONDO approach to address 12 3-D aspects of decontamination?

13 A. First, it appears that NRC Staff is describing 14 contamination by gravitational settling. Gravitational settling 15 only applies to very large particles (>100 m), and is only 16 important close to the release point.

17 In the MACCS2 model, airborne contamination is transferred 18 to a surface through a transfer coefficient called the dry 19 deposition velocity. This transfer coefficient is a function of 20 the particle size, the chemical form of the particles, the 21 chemical affinity of the surfaces and the roughness of the Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 44

1 topography. This transfer coefficient applies to vertical and 2 horizontal surfaces, and even to leaves in the trees.

3 Second, the cost of urban decontamination ($/km2) is 4 proportional to building density and internal space density. All 5 four ISR approaches (A to D) began with a calculation of the 6 decontamination cost per unit area of nonfarm land ($/km2) with 7 consideration of 3-D buildings and structures. In order to 8 account for floor and building density, multipliers must be 9 used. The use of multipliers is consistent with CONDO and other 10 costing tools such as DECON, which is presented in NUREG/CR-3413 11 and which I discussed on pages 26-28 of this testimony.

12 With the appropriate multipliers, the cost per unit area of 13 nonfarm land ($/km2) is multiplied by the area of nonfarm land 14 and then divided by the population to obtain a value for CDNFRM.

15 In the absence of a documented methodology for the costs used by 16 Entergy to determine CDNFRM as discussed in Section V of this 17 testimony, this is the reasonable and most appropriate method to 18 calculate CDNFRM.

19 Q. What is your response to NRC Staffs testimony in A74 20 that ISRs assumption of uniformity for the exteriors and 21 interiors of buildings renders ISRs cost estimates unrealistic 22 and unreasonable?

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 45

1 A. NRC Staff is correct that contamination would not be 2 uniform on building roofs, exterior walls, and interior space; 3 but it is not cost effective to differentiate between the 4 portions of surfaces which would require decontamination and 5 those that would not in any given building. It is more likely 6 that bulk decontamination, such as hosing or vacuuming, would 7 take place first followed by verification for hot spots. Our 8 assumption is reasonable given the likely operational efforts 9 and strategies which would be employed in the event of a severe 10 accident.

11 Q. In A140, Entergy suggests that your use of 12 spreadsheets based on the CONDO database, instead of the code, 13 may have errors, inaccuracies or biases. Is this plausible?

14 A. We have included these spreadsheets along with any 15 assumptions in Annex C of our report (NYS000242 at 46-56). As 16 such, they are available for review and criticism. Any other 17 claims regarding our calculations using the CONDO data were 18 addressed in preceding responses of this testimony.

19 (3) ISRs Response to Testimony on the Use of RISO 20 Q. What is your response to Entergys critique of the 21 RISO approach in A149 and A150?

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 46

1 A. For the RISO approach, Entergy claims in A149 and A150 2 that two decontamination techniques/surface pairs were excluded 3 as compared to the CONDO approach, and of the decontamination 4 techniques considered, some are compatible with heavy 5 decontamination instead of light decontamination. Furthermore, 6 Entergy claims in A150 that the cost data is not evident from 7 the RISO reference.

8 We excluded plants and shrubs, and treesboth of which were 9 not available in RISO and did not significantly contribute to 10 the total cost in the CONDO approach because they were less than 11 1% of the total cost. The only techniques with a heavy-12 decontamination-like DF were vacuuming, cleaning, and washing, 13 which are the only RISO techniques for interior decontamination.

14 All other RISO techniques referred to in our report have a DF 15 less than 3, which is representative of light decontamination.

16 The cost data was based on the RISO document but is explicitly 17 described in the RODOS Report: Estimation of the unit costs of 18 decontamination techniques, attached hereto as Exhibit 19 NYS000429.

20 Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 47

1 C. ISRs Response to Testimony Regarding Decontamination 2 Time (TIMDEC) 3 Q. Has ISR reviewed its estimates for TIMDEC in light of 4 testimony provided by Entergy (A102 and 103) and NRC Staff 5 (A57)?

6 A. Yes, ISR has considered this testimony by NRC Staff 7 and Entergy. As I discussed earlier in this testimony, the only 8 real support NRC Staff or Entergys testimony provides for the 9 Sample Problem A TIMDEC values is NUREG-1150 and NUREG/CR-3673 10 (NRC000058). As I also discussed earlier in this testimony on 11 pages 22-23 of this testimony, the Sample Problem A TIMDEC 12 values are not reasonable, and NUREG-1150 describes an 13 idealistic decontamination scenario based on these unrealistic 14 values, without justifying their use or showing how they were 15 calculated. Additionally, I explained how the discussion of 16 decontamination time in NUREG/CR-3673 is based on unreasonable 17 assumptions.

18 In preparing this rebuttal testimony, ISR also reviewed 19 reports regarding decontamination following Fukushima that have 20 been published since December 2011. In light of these recent 21 reports on decontamination efforts in the Fukushima Prefecture 22 (NYS000428), there is support for changing the lower bound of Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 48

1 TIMDEC to 1 and 2 years for light and heavy decontamination, 2 respectively.

3 According to these reports, full-scale decontamination for 4 the outer edges of the plume deposition began one year after the 5 accident. Since it is on the outer edges of the plume 6 deposition, it is representative of light decontamination.

7 Amongst other things, the delays have been caused by the lengthy 8 time it has taken to develop a decontamination plan, which is 9 dependent on detailed radiation surveys and procurement of 10 suitable and efficient decontamination equipment and materials, 11 and by the time it has taken to gain approval of the 12 supplementary budget by the government. Another reason for the 13 delay is the need to secure approval from the local communities 14 for storage sites for decontamination waste. (NYS000265).

15 These delays would be expected following a severe accident at IP 16 and have nothing to do with the direct damage caused by the 17 Tsunami.

18 It is therefore possible that residents from some areas 19 will be allowed back to their home after a delay of more than a 20 year. Therefore, at the low end, a minimum TIMDEC of 1 year is 21 justifiable by the recent reports.

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 49

1 If light and heavy decontamination is conducted in series 2 (i.e., light decontamination first and then heavy) as is the 3 case for the Fukushima Prefecture, a minimum TIMDEC of 2 years 4 for heavy decontamination is also reasonable. In MACCS2, the 5 cost associated with TIMDEC is solely the decline in property 6 value during inhabitation. The impact of the change in TIMDEC on 7 the OECR is a 30% decrease at the low end of the cost.

8 Q. Both the NRC (A57) and Entergy (A102, A160) have 9 reiterated that the intention of the MACCS2 decontamination 10 model is to restore property to habitability based on the 11 defined dose criterion, not to entirely decontaminate the area.

12 Based on this, they infer that the ISR estimates of TIMDEC are 13 too large. Do you agree?

14 A. TIMDEC represents the average time from evacuation of 15 a population to return to their original home in a given grid 16 element. It is possible that within that grid element, some 17 people will return to their property more quickly than TIMDEC, 18 and that decontamination efforts may continue long after TIMDEC.

19 Given the large uncertainty in determining this time, we defined 20 a range of values based on real-world experience with actual 21 severe accidentsFukushima and Chernobyl.

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 50

1 As discussed in the previous answer, we now consider that 2 the minimum time between evacuation and return home for areas 3 requiring light decontamination is 1 year and 2 years for heavy 4 decontamination. This appears reasonable, from the current 5 experience in the Fukushima Prefecture. Maximum times of 15 and 6 30 years respectively represent upper bounds based again on the 7 decontamination plans in the Fukushima Prefecture (NYS000269).

8 An appropriate average value would be somewhere between these 9 minimum and maximum values.

10 Q. What is ISRs response to Entergys claim in A103 that 11 the costs associated with large TIMDEC values are due to per 12 diem expenses for relocated individuals and, thus, ISRs 13 analysis results in an intentional defeating of [the] MACCS2 14 decontamination optimization model?

15 A. Entergy misunderstands how the MACCS2 inputs of TIMDEC 16 and POPCST work. Interdiction costs are the sum of relocation 17 costs, such as per diem expenses for relocated individuals, and 18 the decrease in property value during decontamination. TIMDEC 19 is only a factor in the decrease in property value during 20 decontamination. Relocation costs are solely a function of the 21 POPCST value and the population. As discussed in the ISR report Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 51

1 (NYS000242 at 28-29) for the POPCST parameter, relocation costs 2 are assumed for a duration between 20 and 93 weeks.

3 Q. Entergy and NRC Staff make several arguments about 4 various decontamination techniques, including leaving 5 contamination in place and new methods. (Entergy Testimony at 6 A91; NRC Staff Testimony at A38) What is your response to this 7 testimony?

8 A. In A91, Entergy claims that leaving contamination in 9 place, but burying it may be a financially attractive 10 alternative that reduces the dose. This issue has no bearing, 11 positive or negative, on our calculations and final conclusions.

12 We agree that this technique could work well in farmland where 13 deep-plowing can bury the contamination, but it is less 14 applicable to the urban areas of NYC where concrete and paved 15 surfaces are predominant.

16 In addition, no matter what the decontamination technique, 17 it is difficult to achieve a high dose reduction factor (DRF=15) 18 where cesium is present, even if some surfaces can be 19 decontaminated perfectly (DF>100). As an example, it may be 20 easy to decontaminate the glass surfaces with a DF>100. However 21 it may not be possible to decontaminate the brick to a DF>5. The 22 actual DRF achieved near the building will be much less than Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 52

1 DF>100 and closer to the DF=5. The lowest DF determines the 2 achievable DRF, so even though the glass is thoroughly 3 decontaminated, the dose reduction factor will not be as high 4 due to the remaining cesium contamination in the brick.

5 D. ISRs Response to Testimony Regarding Other Entergy 6 and NRC Staff Comments About MACCS2 Parameters 7 Q. What is your response to Entergys testimony in A153 8 regarding the VALWNF parameter, specifically that the increased 9 factor associated with lost tourism and business bounds the 10 increased factor associated with scaling up the 1997 SECPOP2000 11 values to 2004 values?

12 A. The values taken from SECPOP2000 represented the 1997 13 value of possessions such as houses, automobiles, etc. Entergy 14 used the 2004 Gross County Product (GCP) to additionally account 15 for lost tourism and business in the calculation of VALWNF. In 16 general, the GCP is the total value of goods and services 17 produced. NYS00270A-NYS00270B.

18 From the above, SECPOP2000 values and GCP values account 19 for two separate things and they cannot bound one another. Both 20 values are needed in the calculation of VALWNF, and both should 21 be in 2004 dollars.

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 53

1 Q. Regarding the POPCST parameter, Entergy suggests in 2 A157 that the long-term relocation cost should be based on 3 historical unemployment durations instead of the current term 4 for unemployment benefits. What is your response?

5 A. The unemployment as a result of a severe nuclear 6 accident, potentially affecting up to 20 million people, would 7 not be comparable to traditional, historical unemployment.

8 Given that the unemployment in 2005 had an average duration of 9 18 weeks (Entergy Testimony at A157), a range of 20 to 93 weeks 10 would be reasonable for unemployment triggered by a severe 11 nuclear accident.

12 Q. What is your response to Entergys testimony in A99-13 A101 regarding the reasonableness of the calculations performed 14 with the MACCS2 code as modified by ISR?

15 A. Entergy incorrectly testifies that the modified MACCS2 16 code did not run properly. The original set of runs performed 17 with the MACCS2 code modified by ISR failed to run to completion 18 because a single carriage return character was missing at the 19 end of the input files. When that carriage return character was 20 added, the input files ran to completion without error and gave 21 the results presented in the ISR Report.

Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 54

1 Furthermore, ISR is well aware of the verification 2 requirements when performing nuclear safety calculations. In 3 accordance with ISR's rigorous internal QA Standards, we ran all 4 the sample problems provided by Sandia and all input files 5 prepared by Entergy using the MACCS2 executable provided by 6 Sandia and the version modified by ISR in order to ensure that 7 the model was running properly. Our results were identical to 8 those published by Sandia and by Entergy. The modified code 9 provides correct results.

10 IX. CONCLUSION 11 Q. Does the testimony from NRC Staff and Entergy affect 12 ISRs overall conclusions about the economic cost estimates used 13 in the SAMA analysis for IP?

14 A. No. The analysis conducted by ISR shows that 15 Entergys input parameters to the MACCS2 code significantly 16 underestimated the total economic cost of a severe nuclear 17 accident at IP.

18 I have reviewed all the exhibits referenced herein. True 19 and accurate copies are attached.

20 Pre-filed Written Rebuttal Testimony of François J. Lemay Consolidated Contention NYS-12C 55