ML19220C716
| ML19220C716 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 04/23/1979 |
| From: | Bancroft M PUBLIC CITIZEN LITIGATION GROUP |
| To: | |
| Shared Package | |
| ML19220C706 | List: |
| References | |
| ACRS-SM-0087, ACRS-SM-87, NUDOCS 7905140072 | |
| Download: ML19220C716 (14) | |
Text
4 TESTIMONY OF MICHAEL BANCROFT PUBLIC CITIZEN LITIGATION GROUP BEFORE THE SENATE SUBCOMMITTEE ON NUCLEAR REGULATION HEARINGS ON THE THREE MILE ISLAND ACCIDENT APRIL 23, 1979 101 198 7 9 0514 o oq
~f
My nar.e is Michael Bancroft.
I have a Ph.D. in physics and have been a physics professor.
I am now a lawyer with the Public Citizen Litigation Group and specialize in the area of nuclear bealth and safety.
Mr. Chairman, I appreciate this opportunity to address the Subcommittee on Nuclear Regulation about the lessons of the frightening accident at Three Mile Island, Unit 2 (TMI-2).
I would like to discuss the warning signs available to the NRC before the accident, the protective actions taken and not taken after the accident, and where we go from here to enhance public safety and confidence.
Dr. Joseph Hendrie, Chairman of the Nuclear Regulatory Commission (NRC), has testified to this Subcommittee that there cannot be an acceptable nuclear power program in this country if there is an appre-ciable chance of Three Mile Island happening again.
Therefore, im-plementation of the recommendations I will present would not only enhance the public health and safety, but also benefit the nuclear industry.
My main recommendations are that there should be:
(1) greater NRC attention to problems encountered in starting up nuclear power plants and to generic sr.fety problems and a greater NRC willingness to assure that all required safety systems in operat-ing reactors are effective, (2)
NRC action to require workable evacuation plans and ade-quate routine and emergency offsite monitoring as a condition for allowing a nuclear power plant to operate, (3) a plan for dealing with the TMI-2 plant, followed by an NRC decision based on hearings and an environmental impact statement, all before workers are exposed to radiation in trial cleanup cperations, (4) a federal monitor in the control room of nuclear power plants at all times, having the authority to control operation of the plant, and (5) an amendment to the Price-Anderson Act to repeal the limita-tion on the liability of the nuclear industry to the victims of a nuclear accident for compensation of their injuries and losses.
\\o\\
\\99
. 1. Warn.ing Signs of the TMI-2 Accident Between March 28, 1978, when TMI-2 achieved initial criticality, and December 30, 1978, when it wa, declared by its owners to be in commercial service, the plant was plagued by numerous malfunctions and long periods of shutdown.
Altogether, TMI-2 was shut down for repairs for 195 of the 274 days in this startup period, or 71% of the time.
Several of the problems which have been cited as con-tributing to the accident on March 28, 1979 had occurred previously, sometimes repeatedly.
Among them are numerous losses of feedwater to the steam generators, including several from failure of the main feedwater pumps.
Two separate shutdowns were caused by pressure release valves which opened and failed to close.
Prior to February 1979, TMI-2 suffered 12 accidental reactor trips, including four in which the emergency core cooling system (ECCS) was activated.
This history is fully set out in Public Citizen's report " Death and Taxes: An Investigation of the Initial Operation of Three Mile Island No.
2."
(Attachment 1.)
The report documents that the owners of TMI-2 obtained approximately $40 million in tax benefits by declaring the plant in commercial operation before the end of 1978.
The plant was declared commercial at 11 pm on Saturday, December 30, 1978.
The final tests required by TMI-2's license conditions and more repairs were done in the previous two days.
This last minute rush suggests that the plant was put 'nto commercial service without getting to the root of the recurring problems.
The NRC, which is already too passive in monitoring the startup process and authorizing continuation to the next level of operation, dis-claims any interest in the tax considerations which affect the de-cision to put a plant into commercial service.
(See recommendation 1 on page 3 of Public Citizen's letter to President Carter accom-panying " Death and Taxes," Attachment 2.)
\\o\\
200
. Events at other nuclear power plants manufactured by Babcock and Wilcox (B&W) of similar design to TMI-2 showed that there are generic safety problems at B&W plants.
The NRC was alerted to these B&W safety problems by its inspection staff early this year, but no action was taken before the TMI-2 accident.
A drass rehearsal for TMI-2 was held at Toledo Edison's Davis Besse Unit 1 plant on September 24, 1977.
A loss of feedwater to the two steam generators led to a rapid increase of the primary coolant temperature and pressure.
The pressurizer relief valve opened to relieve the excess pressure.
Thereafter, it opened and closed repeatedly and finally stuck open.
Large amounts of steam poured through the open valve, decreasing the primary coolant pres-sure and activating the ECCS.
Unaware that the pressurizer relief valve was open and seeing the pressurizer level indication increas-ing, the operator shut off the ECCS six minutes after the start of the accident.
The pressure continued to drop until steam formed in the primary c alant system.
Based on a high pressurizer level reading. the operator stopped one primary coolant pump in each loop 9 mir.utes into the accident.
Fortunately, 21 minutes after the start, the open pressure relief valve was discovered and the line snd leak were blocked by another valve.
Within an hour of the onset of the trouble, flow was restored to the steam generators and the primary coolant temperature and pressurizer level were brought under control (Attachment 3.)
On January 8, 1979, an NRC Region III inspector, J.S.
- Creswell, wrote a memo describing problems at Davis-Besse 1 which he wanted brought to the attention of Atomic Safety and Licensing Boards weighing licensing of new B&W reactors.
Item 3 of the memo states:
Inspection and Enforcement Report 50-346/78-06 documented that pr :ssu i zer level had gone of fscale for approximately five minutes during the November 29, 1977 loss of offsite power event.
There are indications that other B&W plants nay have problems maintaining pressurizer level indications during transients.
In addition, under certain conditions such as loss of feedwater at 100% power with the reactor coolant pumps running the pressurizer may void completely.
A special analyris has been performed concerning this event.
This analysis is attached as Enclosure 1.
Because of pressurizer level maintenance problems the sizing of the GOi pressurizer may require further review.
)0tt u u-
. (Attachment 4.)
These items were forwarded to the licensing boards on February 28, 1979.
On March 28, 1979, Norman Moseley of the Office of Inspec-tion and Enforcement followed up with the NRC headquarters staff's evaluation of each item.
The problem with determining coolant level in the precsurizer is acknowledged, minimized, and put down for further study.
The loss of pressurizer water level indication could be con-sidered to deviate from GDC 13, because this level indica-tion provides the principal means of determining the primary coolant inventory.
However, provision of a level indication that would cover all anticipated occurrences may not be practical.
As discussed above, the loss of feedwater event can lead to a momentary condition wherein no meaningful level exists, because the entire primary system contains a steam-water mixture.
It should be noted that the introduction to Appendix A (last paragraph) recognizes that fulfillment of some of the criteria may not always be appropriate.
This introduction also states that departures from the Criteria must be iden-tified and justified.
The discussion of GDC 13 in the Davis-Besse FSAR lists the water level instrumentation, but does not mention the possibility of loss of water level indica-tion during transients.
This apparent ommission in the safety analysis will be subjected to further review.
(Attachment 5.)
There is no indication from these documents that any action was taken to resolve or ameliorate the problem then existing at operating B&W plants.
Babcock and Wilcox was aware of problems in controlling its reactors due to faulty readings of indicators in the control rocm.
On August 9, 1978, B&W informed Davis-Besse of trouble that occurred on March 20, 1978 at the Rancho Seco plant owned by the Sacramento Municpal Utility District.
This event involved over and under-pressure of the primary coolant due to erratic control of feed-
\\Q\\
. water to the steam generators.
B&W described the situation thusly:
The plant response was not the usual transient in that the ICS [ Integrated Control Systen] responded to the erroneous input signals rather than actual plant conditions, and resulted in a Reactor Protection System (RPS) trip on high pressure.
Subsequent to the Reactor Trip, the erroneous signals to the ICS contributed to the rapid cooldown of the RCS.
Plant operators had extreme difficulty in determining the true status of some of the plant parameters and in con-trolling the plant because of the erroneous indications in the control room.
(Attachment 6)
In the letter of advice to operators of B&W reactors, B&W emphasized the central role it assigns to the pressurizer level indicator as a key variable to monitor and control in an emergency situation.
Recognizing that no procedure can cover all possible combina-tions of NNI [Non-Nuclear Instrumentation] failures, the operator's respcnse should be keyed to certain variables.
If the operator realizes that he has an instrumentation problem (as opposed to a LOCA or steam line break, for example), he can limit the transient by controlling a few critical variables:
a.
Pressurizer level (via HPI or normal Makeup Pumps) b.
RCS pressure (via Pressurizer heaters, spray, E/F relief valves, etc.)
c.
Steam Generator level (via feed flow, feedwater valves, etc.)
d.
Steam Generator pressure (via turbine bypass system)
The pressurizer level and RCS pressure assure that the Reactor Coolant System is filled; the Steam Generator level and pres-sure assure adequate decay heat removal.
(A tcchment 6.)
The NRC had a copy of this letter and knew both the problems with pressurizer level indications and the key role B&W assigned to reliance on that indication do react to an emergency.
Yet the NRC chose to relegate this problem to the class of generic safety prob-lens that deserve attention in due course.
The NRC permits nuclear plants to operate in spite of generic
)Q\\
. safety problems on the basis that there is a safety margin from other systems-- the philosophy of " defense in depth."
However, in practice, the NRC uses the supposed saf3ty margin as a shell game to avoid the necessity to take decisive at tion, such as ordering a plant shutdown, when a vital safety system is not effective.
For instance, the NRC granted Metropolitan Edison (Met Ed) an exemption to its standards of performance of the ECCS of Three Mile Island Unit 1 on April 27, 1978, which was renewed on March 16, 1979 (44 Fed. Reg. 19079: March 30, 1979), allowing Met Ed to make the agreed-on modifications at its next extended outage, rather than requiring them immediately.
The NRC cannot ure the defense in depth philosophy both ways-- both as the basis of claims of a safety margin through redundancy to allow for multiple mishaps, and as the basis for diverting attention from safety defects in any particular system and allowing a waiver to operate in spite of those defects.
2.
Response to the Accident The TMI-2 accident has pointed up grave deficiencies in the NRC's knowledge and control of the early, most dangerous phase of the TMI-2 accident.
The NRC was dependent on Met Ed for at least the first two days for its information on the condition of the reactor and releases of radioactivity.
During this period, the utility deceived the public b;. making unfounded reassurances.
Lacer in my testimony, I will discuss the proposal to have a continuous federal presence in the control room of operating reactors with clear author-ity to take centrol of operations if necessary to protect the public.
a.
Radiation monitoring 101 204 Becauce of the inadequate system of routine, offsite monitoring of radioactive releases and the belated realization by public offi-cials of the seriousness of the accident, there was virtually no moni-toring of offsite radiation exposures until Saturday, and no systematic monitoring even now.
Before Saturday, in the time of the most intense radiation doses to the public, our knowledge is based on readings of Met Ed's offsite cumulative dosimeters, of which there are 17 or less, and occasional instantaneous rate measurements which were taken with mobile equipment.
This network is inadequate to determine the
_7_
doses received by the public around Three Mile Island during the most dangerous three days.
The NRC is attempting to compile and analyze all the radiation measurements to come up with the best estimates of total radiation doses received by the public.
I was told by Dr. Harold Peterson of the NRC's Radiological Standards Branch, who is *0rking on this ana-lysis, that they could not now exclude variations of a factor of three in public doses over the ones that were estimated as maximum from the monitors that were there in the first days.
This ignorance of public dose will make iu very difficult, if not impossible, to detect and determine the health effects of this e::posure in the years ahead and undermines bland assurances that have been given to the neighbors of Three Mile Island.
This is the latest instance in which scientists and the public have been deprived of the information needed to prove a suspected health danger.
To prevent reoccurence of this ignorance in future accidents, the NRC should require a more comprehensive -routine of fsite ra diation monitoring by its licensees, encourage the states and the EPA to do their own monitoring around nuclear facilities, and should order a licensee to put out a fine network of cumulative radiation detectors around the plant at the first sign of an imminent large release of radioactivity to the environment.
This system should be worked out immediately, for instance by Professor X.2. Morgan, and put into operation around TMI-2.
sO1
'05 b.
Evacuation planning and decision i
I The extent of the fuel damage, the spread of radiation within the plant, and the difficulty in cooling the reactor core, if these had been known to the NRC and state officials in the early stage of the accident, would have called for an evacuation of the vicinity of Three Sile Island.
On Friday, March 30 and Saturday, March 31, at least three senior NRC staff members, Harold Collins, Joseph Fouchard, and Roger Mattson urged an NRC recommendation to Governor Thornburgh to evacuate an area around TMI-2.
The NRC Commissioners displayed a collective indecision about evacuation, as well as an inability to get significant information on the available means to evacuate and hcw long different evacuations would take.
-S-Indecision and slow acquisition of knowledge by the NRC prevented a trial of the effectiveness of the Pennsylvania evacuaticn plans.
This may not happen in the event of another nuclear accident.
Preparation is needed to make evacuation understood and available in an emergency.
Assurance of the existence of workable evacuation plans is a basic prerequisite to protection of the public in the vicinity of a nuclear power plant.
State and local officials have the responsi-bility for ordering and carrying out any evacuation.
Nonetheless, it is the responsibility of the NRC to determine that such evacuation can be carried out smoothly and rapidly before allowing a nuclear plant to operate.
Under present proceduras, the NRC approves a utility's safety plan if it merely contains adequate provision for notifying state and local radiological emergency officials of the occurrence of an acci-dent.
Pre-announced tests of these communications networks have shown defects in even this first step in response to a nuclear accident.
The NRC has a program for assisting, evaluating, and approving state emergency response plans.
Even this approval is based on meeting a curtailed list of 70 " essential" elements, which mainly deal with the existence of organizations, authority, plans, and surveys.
The NRC has approved the emergency plans of only ten states, not including Pennsylvania or otner states with major concen-traticns of nuclear facilities, such as Illinois.
The General Accounting Office has recommended that as a precondi-tion to NRC licensing of a nuclear plant, the state it is in must have an NRC-approved emergency plan and the licensee have an agreement with state and local officials to conduct annual tests of the plans.
(" Areas Around Nuclear Facilities Should Be Better Prepared for Radiological Emergencies," EMD-78-110, March 30, 1979..)
In 1975, the Public Interest Research Group and 30 other citizen groups active in 20 states petitioned the NRC to require annual evacuation drills, including actual evacuation of a sector of the population, and distribution of evacuation information to the public within 40 miles of the plant.
I am providing a copy of the PIRG petitien for the record.
On the local level, many citizen groups peti'.ioned their public utility commissions for evacuation drills and distribution of evacuation information.
q
\\ o,\\
u
_9_
On July 7, 1977, the PIRG petition was denied by the NRC.
Informing the public was rejected because it would be " subject to misinterpretation and would be of little help" or "would be too complex for the public to understand."
Evacuation drills were re-jected on the strength of a calculation, based on the now discredited Rasmussen Report, WASH-1400, that "the mortality risk from the evacuations is about 200 times greater than the mortality risk frcm the potential reactor accident."
(42 Fed. Reg. 36327; July 14, 1977.)
In response to the GAO recommendation, the NRC rejected requiring an approved state evacuation plan on the basis that the NRC already assures that nuclear plants are adequately safe, so that emer~ ncy planning is a marginal addition.
(Attachment 8.)
Three Mile
-land shows that effective evacuation plans are not icing on the cake, but an essential precondition for the protection of the public.
I support the GAO proposals, including distribution of "information about the potential hazard, emergency actions planned, and what to do in the event of an accidental release" to the neighbors of nuclear facilities.
( Attachment '7. )
This distribution should be done periodically.
In addition to building these monitoring and offsite emergency response capabilities into the licensing process, the NRC should require them of operating plants within six months.
In the meantime, the NRC should undertake an analysis of operating plants for the like-lihood and consequences of large radioactive releases to the public.
The NRC already has a system of rating plants' safety to form the start of this process.
For high priority plants, the NRC should require sooner implementation, if not shutting down the plants until these protective measures are achieved.
3.
Dealing With the TMI-2 Plant
\\01 20"1 The crisis at Three Mile Island has had an unpleasant habit of lingering. There are recent reports of new radioactive releases, including releases of highly dangerous iodine.
NRC spokesmen have routinely coupled announcements of the amount of iodine present off-site with a statement that such amounts pose no threat to public health.
As far as we know, any amount of radiation exposure presents some health risk.
Such statements mislead the public and undercut the credibility of the NRC.
. Once the reactor is cooled down and the radiation completely contained, the problem remains of what to do with the TMI-2 mess.
I am concerned about further releases during the cleanup process and about what is done with the radioactive waste mater:al.
However, I am most concerned about the large radiation doses to workers poten-tially involved in dealing with TMI-2.
The worker exposure limits set by the NRC are unreasonably lax in light of the numerous recent scientific studies which show that the adverse health effects of low-level radiation may be more than ten times greater than was believed when the present standards were set.
The NRC has not yet acted on the petition of the Natural Resources Defense Council, filed in September 1975, requesting a ten-fold strengthening of the worker exposure standard.
The NRC has promised hearings on this subject later this year, but new standards n;ay not go into effect in time to help workers empicyed in the TMI-2 aftermath.
Current NRC regulations allow most workers (who do not have a history of occupa-tional exposures near the maximum of the nominal standard) to be s ub-jected to 3 rems per calendar quarter, or 12 rems per year.
What I fear is that Met Ed will undertake exploratory cleanup operations with no clear plan, evaluation of the options, or selection of an objective.
Instead of just sending in workers to see what can be done and proceeding by trial-and-error, the NRC should require Met Ed to devise and submit a plan of operation for NRC approval.
There is a good chance that the plant can never be made operable.
If so, it is in need of decommissioning and should be treated as such.
It would be callous and inhumane to allow ':et Ed to expose workers to large doses of radiation in restoration operations if some measure-ments and calculations would show that it is not practical to restore TMI-2.
This is not a hypothetical problem.
The Humboldt Bay nuclear plant in California has been shut down for almost three years to investigate whether it can withstand expected earthquakes.
There is a good chance it will not operate again.
Yet, in 1977, Pacific Gas and Electric exposed workers to large amounts of radiation doing maintenance on a possibly doomed plant.
2,0 b
. The NRC should take control of dealing with the aftermath of the TMI-2 accident.
It should authcrize Met Ed to take the measurer.1ents and samples needed to evaluate the situation inside the TMI-2 contain-ment.
The NRC should then hold hearings on ".et Ed's planc for TMI-2 as a license amendment and do an environmental impact statement on the consequences of the proposed action.
I am also concerned that Met Ed may employ one-time or migrant contract atomic laborers in dealing with the TMI-2 cleanup.
NRC regulations allow the practice of giving a worker his full quarterly dose of 3 rern in a sincle day or less.
Imposing this risk of radia-tion in return for a single day's pay, perhaps at a low unskilled labor rate, is inconsistent with the cost-benefit rationale for allowing so much greater er.posure of workers than of the public.
Finally, I am conceraed that there will be a proposal to bury some of the radioactive wastes on Three Mile Island.
The Governor of South Carolina has ruled that South Carolina will not accept any TMI wastes.
A bad precedent is developing of dealing with radioactive waste when no one wants it anywhere else, by burying it perforce where it is, in spite of the fact that the site was not chosen with that possi-bility in mind.
Nuclear wastes are with us.
We must dispose of existing waste in the best way possible and deal with the fact that no one wants nuclear garbage dumped near them in making the decision whether to create more.
4.
Continuous Federal Presence and Authority in Nuclecr Control Rooms 101 209 Candidate Jimmy Carter, in his first debate with Gerald Ford on September 24, 1976, statec:
A full-time federal employee, with full authority to shut down the plant in case of any operational abnormality, should always be present in control rooms.
There should be an NRC monitor in the control room at all times a nuc-lear plant is operating.
The NRC monitor should have the authority to order the plant to shut down and to take charge of operations in an emergency.
(See attachment 2, recommendation 3 on p.
3.)
The NRC monitor would serve three functions.
First, to prevent plant operation in violation of NRC regulations and get a sense, for enforcement purposes, of the capability and deficiencies of the plant operators.
Secondly, the monitor would communicate emergency develop-ments to the appropriate NRC authorities and experts via a " hot line."
Finally, the monitor would have the authority to order certain reactor operations to take place or not take place if he judges it necessary to protect the public health.
Exercise of this authority would be an extraordinary practice, subject to the monitor's judgement of his capa-bility as compared to the normal operator.
The monitor with this authority is needed to assure the public that it is ultimately pro-tected by comeone who holds its interest foremost, not having the company's extraneous interests in protecting its property, its image, and its purse from liability.
Continuous monitoring by federal inspectors is a commonplace in the meat slaughtering and processing industry, under the federal meat inspections laws, 21 U.S.C. 59 601 el seg.
If federal inspectors are required to be present at all times a slaughterhouse is operating, is it not prudent to have one present during operation of a more heavily federally regulated industry with vastly greater potential fcr public mayhem?
5.
Repeal of the Price-Anderson Liability Limitation There is no justification for this unique federal subsidy to the nuclear industry, which would result in victims of a nuclear disaster receiving, for their loss, only a few cents on the dollar.
Three Mile Island demonstrates afresh that the nuclear industry must back up its safety claims with its full assets rather than putting almost the entire risk of a nuclear catastrophe on the neighbors of the plant.
Repeal of the Price-Anderson liability limitation is the most signi-ficant change that is likely to come out of the Three Mile Island experience.
It is needed as a matter of elementary fairness to the public and to put the nuclear industry on an even footing with its competitors by requiring it to show that it believes in it own product.
,m
\\01 L.a
s
. Ralph Nade.- has recently called for eliminating the Price-Anderson liability limits and acdressed the unavailability of nuclear insurance to homeowners in letters to the federal government and the National Association of Insurance Commissioners. (Attachment 9.)
In upholding the constitutionality of the Price-Anderson Act last year, the Supreme Court left open the possibility that those with uncompensated property damage due to a nuclear accident and operation of the Price-Anderson Act could sue the federal government under the "taking" clause of the Fifth Amendment.
This means the federal sub-sidy to nuclear power could be bigger than Congress intended, but the benefit would only go to those with property damage, not to people for their personal injuries.
Look at what the people around Three Mile Island have already suffered.
The risk of future ill effec's from radiation exposure, the anxiety of the accident and the well-founded belief that they were not getting the whole story, and probable depression of the value of their property due to the accident.
Neighbors of a nuclear plant continue to bear the risk that a major accident will kill them, injure them and their offspring, or disrupt their lives, at the least.
On top of all this, it is not fair to deny them their legal remedy to sue for full compensation of their injuries, which is available to all other members of the public subject to an industrial accident.
Thank you.
I would be happy to answer any cuestions you may have, 10\\ o\\\\
'