ML20211P804

Causes & Consequences of Chernobyl Nuclear Accident & Implications for Us Regulation of Nuclear Power Plants

ML20211P804
Person / Time
Issue date:	07/09/1986
From:	Harold Denton Office of Nuclear Reactor Regulation
To:
Shared Package
ML20211P741	List: ML20211P737 ML20211P804
References
NUDOCS 8703020436
Download: ML20211P804 (29)

Text

~

i i

a THE CAUSES AND CONSEQUENCES OF THE CHERNOBYL NUCLEAR ACCIDENT AND IMPLICATIONS FOR U.S. REGULATION OF NUCLEAR ~ POWER PLANTS

• Harold R. Denton Di rector, Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Comission.

Washington, D.C. 20555 -

l 1

• An invited paper as a follow-up. to the International Energy Seminar on, "Chernobyl: The Implications for Nuclear Power and the International Nuclear Regime," sponsored by the International Energy Program of the Johns Hopkins Foreign Policy Institute, Washington, D.C., July 9, 1987.

8703020436 070224 PDR REVGP NRCCHER PDR

! . .__ _ . . . _ , - .- , .- . - _ ~ , _ . .-.

w THE CAUSES AND CONSEQUENCES OF THE CHERNOBYL NUCLEAR ACCIDENT AND IMPLICATIONS FOR THE U.S. REGULATION OF NUCLEAR POWER PLANTS Harold R. Denton I. CAUSES OF THE CHERNOBYL NUCLEAR ACCIDENT Until the tragic nuclear accident of the Unit 4 reactor at Chernobyl in the Soviet Ukraine beginning on April 26, 1986, there were well over three thousand years of commercial nu-clear power plant operation without significant loss of life to the public due to a nuclear accident.1 Even after the Chernobyl accident (with estimates of thousands of delayed cancer fatalities over the next 50 years throughout the USSR and Western Europe), views were expressed as to the irrelevancy of the Soviet accident to the outlook for a similar worst case accident happening to reactors operating in the Western World. The design of the RBMK graphite-moderated, pressure tube reactor units at Chernobyl is simply too different from the more prevalent type of reactor designs in the West to draw meaningful inferences from this accident regarding their safety along with the presumption that safety regulation in the USSR might not be so stringent as in other nations.

While not discounting the possible correctness of these views, responsible government offi-cials in the United States and other affected countries felt it prudent to withhold offi-cial judgments on what safety lessons, if any, are to be learned from the Chernobyl acci-dent until adequate factual information and analyses could be generated. As in the case of the TMI accident, public concern around the world over nuclear safety had been aroused by the Chernobyl tragedy. Such anxieties are not readily allayed even with the availabil-ity of accident information at the level it exists today, let alone the sketchy facts that emerged in the first few months following the accident. Moreover, operator errors played a significant role in both the TMI and Chernobyl accidents--causal factors that are not wholly related to reactor designs or their differences. And, of course, since this was the first nuclear accident with large off-site releases of radionuclides, there is an apparent potential for gaining valuable insights from the successes (or inefficiencies) of Soviet actions to terminate the accident and mitigate its consequences. The latter includes emer-gency preparedness and evacuation measures, medical treatment, protection against contami-nated food and water supplies, and a variety of post-accident recovery measures.

In addition to providing an overview of the causes of the Chernobyl accident and actions taken to mitigate its consequences, the scope of this paper will also include methodologi-cal issues in assessing the health consequences of the accident and provide a preliminary view of some implications of the Chernobyl accident for the regulation of U.S. nuclear power plants.

1The Report of the President's Commission on the Accident at Three Mile Island estimates that less than one public cancer fatality will likely result from the low radiation doses received by some 2 million persons living within 50 miles of the TMI nuclear plant.

1

Chronolony of Events in the Accident Scenario NRC's Severe Accident Policy Statement defines a " severe nuclear accident" as an accident

, in which substantial damage is done to the reactor core whether or not there are serious offsite consequences.2 It is almost impossible to have such an accident unless a number of equipment or human. failures occur in sequence or concurrently. Each unique combination of failures that theoretically could yield a severe accident is called a scenario. Compu-ter models and quantitative risk assessment methods are used in the United States and several other countries to compute the on-site and off-site consequences of each scenario and the probability of its occurrence, albeit with a broad range of uncertainty. If the severe accident scenario goes beyond large-scale core melt and breach of containment, then it also must include weather and radionuclide dispersion and deposition modelling as well as assumptions regarding time of day, momentary population distribution, emergency shelter-ing or evacuation measures, medical treatment and other factors to complete the basis for computing consequences. Accordingly, there are literally hundreds of theoretical severe accident scenarios that are in the realm of possibility, although some are of such very low probability as to provide a negligible contribution to the total (or overall) risk of the fuller spectrum of accident scenarios.

I Positive and timely operator actions can sometimes terminate a sequence of accident precur-sor events, thus preventing a severe nuclear accident. In most cases, passive and auto-mated engineered safety features are relied upon to assist in the defense-in-depth capabil-ity of accident prevention or consequence mitigation, given that one or more failures have already occurred. Redundancy and diversity of engineered safety features are employed to

reduce the probability that a chain of failures will produce a severe nuclear accident.

Among such functional safety features included in the Light Water Reactors (LWRs) prevalent in the United States and many other countries are:

, (1) Reactor trip or scram system that rapidly inserts reactor control rods to stop the fission process and terminate core power generation.

(2) Emergency core cooling system that cools the core, thereby keeping the release of radioactivity from the fuel into the containment building at low levels.

(3) Post-accident radioactivity removal systems to remove radioactivity from the contain-ment atmosphere.

(4) Post-accident heat removal systems to remove decay heat from within the containment building, thereby preventing overpressurization of the containment building.

(5) Containment integrity systems to prevent radioactivity within the containment build-ing from escaping into the environment.8 The systems reliability of the above ESFs, except possibly containment integrity, is i greatly enhanced by the diversity and redundancy of various components and sub-systems.

In addition to containment, defense in depth is also achieved by maintaining the integrity f

e 8NRC Policy on Future Reactor Desians: Decisions on Severe Accident Issues in Nuclear  !

Power Plant Regulation, NUREG-1070, U.S. Nuclear Regulatory Commission, Washington, D.C.,

July 1985.

~

8 David Okrent, Nuclear Reactor Safety: On the History of the Regulatory Process (Madison,

Wisconsin

The University of Wisconsin Press, 1981).

J 2

i i

---

=---w---- _ _ _ _ _ _ - , , , _ _ , , _ , . _ _ . - _ _ _ _ _ , _ _ _ _ _ _ , _ _ _ , , _ _ _ _ _ _ - _ - _ , , , , - _ _ , , _ . , , _ . . , _ _ . _ _ _ _ , _

of several other barriers between the radioactive materials in the LWR reactor core and the human environment: the metal cladding of the reactor fuel rods; the water in the reactor coolant system; and a high-strength reactor vessel that provides the reactor coolant pressure boundary. The RBMK design of the Chernobyl-4 reactor unit contains some, but scarcely all, of the above LWR design features providing defense-in-depth capability to limit the risk of certain severe accident scenarios. Notably absent in the RBMK design is a reactor vessel and a containment with adequate strength for overpressure at the top of the reactor for the type of accident that occurred at Chernobyl.

The following is a brief chronology of the more salient events in the unique accident sce-nario that took place during the ten-day period between April 26 and May 6, 1986.4 Of necessity, some of the more technical and complex aspects of the scenario are omitted.

The accident took place during a safety-related test being carried out on a turbogenerator at the time of a scheduled reactor shutdown. This was meant to test the ability of a turbogenerator, during station blackout, to supply electrical energy for a short period until the standby diesel generators could supply emergency power to avert a severe accident.

e From 01:00 to 13:05 on April 25, reactor power of Unit 4 was reduced from 3200 MWt to 1600 MWt in accordance with the start of the test procedure; turbine generator No. 7 was then removed from service as planned.

e At 14:00, the emergency core cooling system (ECCS) was disconnected to prevent inadver-tent actuation during the test (the ECCS is designed to ensure the core remains cooled during postulated loss-of-coolant accidents). At this time, the test was delayed at the request of the load dispatcher, and the reactor plant was left in service for an addi-tional nine hours.

e After the load demand was lifted at 23:10, power reduction was resumed in preparation for the test (test specifications required the experiment to be performed at a reactor power level between 700 and 1000 MWt). However, despite operator efforts, reactor power subse-quently dropped below 30 MWt, a level far too low for stable reactor operation prescribed l

in the test procedures, e The operators were able to stabilize reactor power at 200 MWt by 01:00 on April 26. How-ever, as a result of xenon buildup in the fuel, which is a natural occurrence that intro-duces large amounts of negative reactivity during prolonged low power operation after high power operation, the operators had to manually withdraw the control rods beyond safe l

operating limits to increase power.

I l

4Information for the chronology and the following section on root causes of the accident l was primarily obtained from three sources:

USSR State Committee on tihe Utilization of Atomic Energy, "The Accident at the Chernobyl Nuclear Power Plant and Its Consequences," Information compiled for the IAEA Experts' -

Meeting, August 25-29, 1986, Vienna, 1986.

International Nuclear Safety Advisory Group (INSAG), " Summary Report on the Post-Accident Review Meeting on the Chernobyl Accident," August 30-September 5, 1986, IAEA, Vienna, September 24, 1986.

USNRC, " Report on the Accident at the Chernobyl Nuclear Power Station," NUREG-1250, l

Report of an Interagency Task Force, U.S. Nuclear Regulatory Commission, January 1987.

l l 3 l

l

i

-d e Despite the inability to meet established test power conditions, the decision was made to proceed with the test. At 01:03 and 01:07, two standby main circulating pumps were started and joined with six pumps already running; the water flow in the reactor core was then excessive, violating maximum flow limits. The increased water flow through the core caused a reduction in steam generation and a drop in steam pressure and liquid level in the steam drum separators, e At 01:19, to prevent an automatic shutdown of the reactor under these varying level and pressure conditions, the operators blocked the reactor protection (scram system) signals related to the pressure and liquid level in the steam drum separators. The feedwater flow to the steam drum separators was then increased (but excessively) to restore the depressed water level, thus lowering the core inlet temperature and further reducing steam production. Particularly at low power levels, a reduction in core voids (reduced steam generation) produces a negative reactivity insertion in plants of RBMK-1000 design causing a reduction in reactivity. Within 30 seconds, the automatic control rods had fully withdrawn from the core to compensate for the reduction in reactivity. The opera-tors then assisted the automatic control rod system by withdrawing the manual control rods but overcompensated for the reactivity reduction causing the automatic rods to move back into the core.

e At 01:22, the decision was made to proceed with the test. The reactor protection signals associated with the turbine stop valves on both turbines were blocked (in violation of l test procedures) to prevent the automatic shutdown of the reactor when these valves were I closed. A computer printout showed that the available excess reactivity margin had dropped to a level requiring immediate shutdown of the reactor, a requirement that was '

ignored to permit test completion. By this time the majority of the control rods were probably rendered ineffective for power control.

I e At 01:23, the stop valves of turbine generator No. 8 were closed to begin the test and 4

the four main circulating pumps on the generator busbar began to coast down. At this point an uncontrolled power excursion began. Shortly after the start of this power increase, the unit shift mana@r gave the order to scram the reactor using the highest level of emergency shutdown fLiction available. This would allow insertion of all con-trol and shutdown rods into th6 core. After the scram was initiated a number of severe l shocks were reportedly felt in;;the control room and the operator observed that the con-I trol rods had failed to fully insert. The control rod drives were then de-energized in j the hope that the rods would fall into the core under their own weight, which they might not have done.

l I

e In only 4 seconds after the uncontrolled power; excursion began, the power level of the reactor reached 100 times (according to Soviet estimates) the 3200 MWt design capacity i

of the reactor. According to the INSAG analysis (Footnote 4), with such an extreme surge of energy addition to the fuel (thus overheating it), the hot fuel and other fragments would have interacted with the surrounding water. The subsequent steam produc-tion resulted in a runaway pressure increase. The overpressure and the heat production ruptured a number of kuel channels. During thia first explosion, fragmented material was ejected and the roof fif the reactor hall was blown open. The reactor space, which is designed for the assteed rupture of only one fuel channel, overpressurized and the upper plate with a weight of about 1000 tonnes was lifted abruptly off the top of the reactor.

At this moment, all fuel channels were ruptured, the control rods lifted, and the hori-zontal pipes sheared off. The second explosion happened about two to three seconds after j the first explosion. It is not yet clear whether the hydrogen produced, reacting with air, was the cause, or whether it was the result of a second power excursion. About 25%

l 4

I

e of the graphite blocks and material from the fuel channels were ejected. The inventory of the system was blown into the reactor core, the reactor hall and the space below the core. The water-containing shielding tanks broke.

e All of the key events and activities following this moment (01:24) are described in Part II below that deals with accident management and other measures to limit or reduce the consequences of the accident.

Root Causes of the Accident: Plant Design, Safety Objectives and Procedures, and Operator Actions and Attitudes The RBMK-1000 reactor design utilized in each of the four units at Chernobyl produces 3200 megawatts of thermal power (MWt), yielding 1000 megawatts of electrical power (MWe).

The RBMK design uses graphite as a neutron moderator and light water as the coolant. Pres-sure tubes, contained in vertical channels in the graphite, either contain low-enriched uranium oxide fuel or are used as locations for control rods and instrumentation. It is also distinguished by extensive use of computerized control systems, but a slow scram sys-tem. Its reactor safety systems provide for emergency core cooling, main coolant loop overpressurization protection, reactor space overpressure protection, mitigation of radio-active releases, steam pressure suppression, and hydrogen gas removal (NUREG-1250).

The Soviets regard the RBMK as their " national" design and the following advantages were influential in its design:

e Engineers already had extensive experience with graphite-moderated, boiling water-cooled reactors.

e Major components could be fabricated at existing manuf'acturing plants.

e The modular design meant that reactor size was not limited by considerations related to fabrication, transportation, or installation of components.

I e A serious loss-of-coolant accident larger than that considered as design basis was thought to be virtually impossible because of the use of numerous pressure tubes rather than a single pressure vessel.

e Fuel use was very efficient.

e A very high plant capacity factor could be achieved with the use of online refueling.

In view of the many safety features in the RBMK design, it is not surprising that the

! Soviet's report at the Post-Accident Review Meeting in Vienna on August 25-29 emphasized operator errors during the test as the primary cause of the Chernobyl accident (Foot-note 4). The Soviets felt the operators lacked a proper vigilance toward safety, in part because the previous excellence of performance by the operators of Unit 4 created an atti-tude whereby plant personnel felt that close adherence to procedures was unnecessary. _How-ever, our own experience and analysis of the human factors where errors could contribute to the risk of severe at.cidents reveals the importance of operator training, particulary for those accident scenarios exceeding the spectrum of design basis accidents. Especially valuable is the use of simulators to test operator response to the preliminary stages of various realistic severe accident scenarios as well as making appropriate accident manage-ment decisions following a loss-of-coolant accident or LOCA, a technique which it is be-lieved the Soviets might not have sufficiently emphasized. Moreover, appropriate attention to man-machine interfaces in safety analysis can lead to designs (including control room 5

l

displays) that make certain human errors less likely or less prejudicial to the performance of safety functions, or provide greater margins of time to recover from operator error.

In the case of Chernobyl-Unit 4, the plant design placed a heavy dependence on adherence to administrational controls and procedures for safe operation. The following major opera-tional events or errors and administrative or management control breakdowns contributed importantly to the accident (NUREG-1250):

(1) Overall management control of the test and its integration with plant operations were not clearly established. The test was directed by an engineer with expertise in the turbine generator / electrical area and not in nuclear safety.

(2) The test procedure did not receive an adequate safety review. It is reported that there was a possibility that the accident might have been less severe if the ECCS had not been blocked. Also, necessary safety precautions and instructions in the proce-dure were evidently not adequate.

(3) The operators may have felt an undue sense of urgency to complete the test since the test would have been delayed for a year had it not been performed.

(4) The power reduction for the test was interrupted for nine hours at the load dis- -

patcher's request. This delay changed the initial core conditions from that contem-  !

plated in the test procedures and contributed to an unstable safety regime.

(5) The operators did not follow the test procedure.

e The test was started at a low power level that violated both the test procedure and station operating instructions. At 6% power, instead of 22 to 31% specified in the test procedure, little steam was being generated, so the eight circulation pumps produced a flow rate above allowable limits. With the high flow rate and low power level, the water inlet temperature to the core was very close to saturation. Under these conditions, an increase in power caused a much greater increase in steam voids and reactivity than normal.

e The reactor scram signal for the trip of the second turbine generator was blocked, which violated safety procedures and was not called for by the test procedure.

e The operators failed to maintain in place the minimum " equivalent" of 30 control rods required as excess reactivity margin.

(6) Other safety systems were also defeated.

e Blocking the steam separator pressure and water level scrams allowed reactor opera-tion despite unstable conditions.

e Control rods were withdrawn well beyond safety distance limits specified by plant procedures. This error rendered the emergency protective (scram) system ineffective.

(7) The plant operators and station management did not demonstrate an adequate under-standing of the safety implications of their actions. Their willingness to conduct the test at a very low power level, with abnormal and unautharized control rod con-figuration and core conditions, and with safety features bypassed indicates an in-sufficient understanding of the reactor and its behavior under these conditions.

6

. I There are a number of design characteristics of RBMK-type reactors that may have contrib-uted to the Unit 4 accident or exacerbated its severity (NUREG-1250):

(1) The RBMK-type reactors have a limited capability to accommodate the structural (stress and strain) effects of thermal transients. '

(2) The R8MK-type reactors have slow acting control rod systems. It takes about 20 seconds for a control rod to become fully effective (versus the 4 seconds of the l

rapid power surge in the accident).

(3)

The RBMK-type reactors operate close to a condition that is beyond departure from nucleate boiling, and at which there is a limited margin until the onset of a heat transfer crisis, fuel center line melting and fuel failure.

(4)

RBMK-type reactors use local area compartmentalization to accommodate the effects of breaks of the primary coolant pressure boundary. This concept relies on the pressure containing capability of each room, cell, or vault to contain material re-leased in the event of an accident. The " compartments" are not designed to accommo-date more than one primary coolant pressure boundary breach based on failure of the largest pipe in the compartment.

! (5)

! There is no design capability (e.g., provisions to preclude " pipe whip") to provide protection failures. It against an initial failure of a pipe resulting in subsequent additional is, therefore, plausible to postulate that " pipe whip" of an initial piping failure could, because of the lack of piping restraints, cause additional failures that would exceed the design capabilities of the " compartments."

The combination of limited capability to accomnodate thermal transients, a slow acting con-trol rodpeaking power system, on the the limited fuel thermal margin provided to the fuel, the effect of flux and the lack of adequate containment, the capability for common cause failures in excess o,f the design basis, and the lack of design characteristics to l accommodate beyond-design-basis events are RBMK-type reactor safety characteristics that would not be acceptable in the United States.

II.

SOVIET ACTIONS TO TERMINATE THE ACCIDENT AND MITIGATE ITS CONSEQ Terminating the Accident and Containing the Release of Radionuclides to the Off-Site Environment l

At 01:24 on April 26 when the steam " explosion"5 irrevocably destroyed the reactor and its operability, the accident.about 30 fires broke out on the surrounding structures from the hot ejecta of The most worrisome of these was on the roof of the turbine building which was commonly shared between Units 3 and 4.

At 02:54 (90 minutes after the explosion),

firefighting units arrived on the scene Chernobyl (11 miles distant). The heroic efforts from the towns of Pripyat (3 miles distant) and of the firefighters, who received burns and dangerous levels of radiation (some of which proved fatal) brought the fires in the turbine building under control by 03:34. All fires, other than those in the reactor core, SThis explosion is not to be confused with the type and force levels of ato:nic bomb ex-plosions. An atomic bomb type of explosion is simply not possible in a severe accident with a commercial nuclear power plant. This also applies to the type of fuel and fissioning process of the Chernobyl plant design.

7

- 1

'l l

were extinguished by 05:00 and Unit 3 was shut down. However, it was not until the morning of April 27 before Units 1 and 2 (physically separated form the other units) were shut down.

According to the INSAG analysis (sup_ra), the accident management actions taken at Chernobyl were generally quite successful. _The first attempt to supply water to the core from emergency auxiliary feed pumps to quench the core debris was apparently unsuccessful and quickly abandoned. The subsequent staps, namely, dumping materials into the reactor well (boron, carbide, dolomite, sand and clay, and lead), supplying cold nitrogen to bring down the temperature in the core space and to reduce the oxygen concentration, and construction of a flat heat exchanger beneath the foundations of the reactor building, stabilized the situation at an early stage (about nine days after the initiating power surge) of the accident.

The uniqueness of the worst case type of nuclear accident that took place at Chernobyl meant that practically all measures to achieve a short term stabilization of the accident and longer term post-accident recovery measures, including reactor entombment, are of a pioneering, experimental character. Particularly imaginative and highly successful was the prompt commandeering of military helicopters to dump (beginning April 28) a variety of materials, totalling 5000 tonnes, through the open roof of the Unit 4 reactor building in order to terminate the atmospheric release of radionuclides being distributed over Western USSR and many other European nations due to the changing pattern of winds. It was necessary to fly the helicopters at sufficiently low levels that the pilots were ex-posed to undesirably high levels of radiation. The amounts of materials dropped on the reactor core in this manner and their accident recovery functions are as follows:

(1) Forty tonnes of boron carbide to ensure against recriticality of the remaining reactor fuel.

(2) About 800 tonnes of dolomite to generate carbon dioxide that could provide " gas blanketing" and could also contribute to dissipating the internal heat within the core space by absorbing the energy of the burning graphite.

(3) Clay and sand (1800 tonnes) to introduce an immediate filtration for radionuclides being released and to quench the fire.

(4) About 2400 tonnes of lead to absorb heat by melting and to provide a liquid layer that would in time solidify to seal and shield the top of the core vault (INSAG, supra).

Tt.e course in Figure 1.of The the radionuclide releases and the success of measures to halt them is shown release of radionuclides from Chernobyl did not occur as a single acute event. Rather, only about 25% of the release took place during the first day of the acci-dent; the rest of the release occurred as a protracted process over the following nine days ending on May 6. Throughout this period, and particularly on the first day, the release was accompanied by large amounts of energy which elevated the radionuclide plume to great heights. The height of the plume on the first day following the explosion reached about 1200 meters (3900 ft.) and a height of 200 to 400 meters during the following nine days.

Following the initial burst of release energy in the first few hours after the accident, the release rates sharply declined and continued a more gradual decline of the cooling period through May 1. This cooling was abetted by the aerial drop of the materials noted above which continued in heavy force during the period April 28 through May 2 and at re-duced levels for some days thereafter. A heat-up period continued from May 2 through May 5.

8

15 i

14

• Soviet estimates of daily radioactive releases normalized to residual radioactivities as of 13 - May 6 to allow for radioactive decay of nuclides API I having rapid decay rates. Uncertainty bounds-

+ 12 -

m are estimated at .150%. (Excludes noble gases) c) 11 -

m ED 10 -

- Cold Nitrogen Introduced

$c) ^0 9- To Reactor Vault h 8- -Helicopter Deposits of 7, Covering Materials Begin u U15 e

$ 6-0_, .2 5 -

o cc " 4-gx " '

- May 6 g 3-

= 2- "

o 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Days After Start of Accide11t l Figure 1.--Daily releases of radioactivity from the Chernobyl Nuclear Plant i

t following the accident in the early morning of April 26, 1986.

This was principally due to the heat arising from residual decay of radionuclides in the l reactor core debris. The Soviet experts estimated this raised the fuel temperature to above 2000'C. Also, it is hypothesized that carbidization of uranium dioxide made it easier for volitalized fission products to escape.

Starting on May 6, a sudden decrease in the release rate to about 1% of the initial rate occurred and it continued to decrease thereafter. Soviet experts attribute this to special measures taken, especially the introduction of cold nitrogen on May 4 and 5 into the reac-tor vault, preventing further oxidation reaction of the fractured graphite, and the forma-tion of more refractory compounds of fission products as a result of their interaction with the material deposited. It is believed that the effectiveness of the cold nitrogen injec-tions was increased due to the liquefaction of the core debris resulting from the excessive temperatures of fission product decay heat that relocated the molten mass into the lower pipe runs where it solidified.

Although radioactive releases to the off-site environment fell to ten thousand curies per day by the fourteenth day and later to only two curies per day, it was deemed necessary to entomb the reactor to prevent significant health effects from cumulative small doses over an extended period of several hundreds of years. This required a major technical innovi-tion. Design goals were established to achieve a radiation level not exceeding 5 millirems per hour at the roof and 1 millirem per hour at the walls of the sarcophagus as well as protection against natural events such as tornadoes, earthquakes, and extreme temperatures at an annual probability of once in ten thousand years or greater (INSAG, supra). Criteria l

related to construction techniques included: minimization of construction time; minimiza-tion of radiation doses to construction workers; and the use of the simplest, most reliable, j and well established means of construction.

9 l

l The massive entombment structure was completed by November 15 and the construction was car-ried out without any human presence on the plant site.s The work was done by robotics with I television cameras mounted on the construction cranes so that operators could direct their movements by remote control. The heavy walls of the structure utilized 410,000 cubic meters of concrete and 7000 tonnes of reinforcement bars. They were constructed in 12-meter steps to an overall height of 60 meters (200 ft.). Sensors have been placed next to the damaged reactor as well as on the roof to monitor the condition of the radioactive material and the escape of radioactivity. The entombed plant is ventilated by the injection of air with the outlet shaft provided with filters to maintain low radioactivity releases commensurate with those of a normally operating nuclear plant.

Emergency Preparedness and Evacuation Measures According to the INSAG report (supra), an overall conclusion to be drawn from the emergency response is that although it had to be initiated at the local level, the actual management of the total emergency situation and response required a rapid acceleration of resource commitment. Owing to the scale of the accident, such resources, and the authority for their commitment, could not be expected to exist at the local level. It must be recognized that for any accident of this severity, irrespective of the location or country of occur-rence, there would need to be a major commitment of manpower and equipment resources in order to regain control of the situation and to reduce the consequences for the population -

and environment.

Because of the unusual energy of the initial explosion and the additional heat of the many fires, the radionuclides released from the accident were thrust to a higher level than would be expected from most of the theoretical U.S. severe accident scenarios. A result is that the radionuclides were dispersed to more distant locations with consequent reduced early depositions within ten miles of the plant than might have been expected. However, local plant officials initially underestimated (or underreported) the severity of the immediate threat of the accident. According to Ramberg, Notwithstanding the silence of Soviet officials at the time, policymakers in Kiev and Moscow were informed within hours. Major General Gennadi Berdov, the Ukrainian deputy minister of internal affairs, initially took command of efforts to control the situation 90 minutes into the accident, sealing off the immediate area around the complex with local militia. A

' team of doctors and technicians from Moscow reached the Chernobyl environs at 8:00 a.m. on April 26. But it was not until a senior delegation of government and party members arrived at the site that evening that policy-makers at the highest level understood the seriousness of the disaster.7 Thereislimitedinformationonthespecificnatureofadvanceradiologicalemergencyplan-ning for the Chernobyl power station, although a Soviet paper presented at an IAEA workshop in 1980 describes a general planning framework that includes a plant location strategy, accident classifications, public safety measures, and an accident management organization structure for nuclear power plants.s At the top of the management organization is a "coor-dination center" involving both government authorities and plant personnel, divided into five sections and attending to one of the following problem areas:

sNuclear Engineering International, Vol. 32, No. 390, January 1987, p. 2.

78cnnet Ramberg, " Learning from Chernobyl," Foreign Affairs, Winter 1986/87, pp. 304-328.

s8ekrestnov, N. V. , and V. Ph. Kozlov, " Aspects of Accident Management at Nuclear Power Plants in the USSR," presented at International Workshop on Planning for Rare Events:

Nuclear Accident Preparedness and Management, January 1980.

10

e Constant surveillance of the operating conditions of the power plant e Radiation control e Dosimetric inspection of the territory around the plant and the environmental protection zone e Protection of the population and provisional evacuation, if necessary e Medical aid for the population and plant personnel, including iodine prophylaxis However, the Soviet delegation at the IAEA meeting in Vienna in August conceded that exist-ing emergency response plans had only limited value to the response teams arriving in Chernobyl and substantial ad hoc planning of evacuation and other emergency actions was required. One of the initial decisions was that a precautionary evacuation of the town of Pripyat (2-4 miles from the plant) should be carried out as soon as possible. On the morning of 26 April, people were instructed to remain indoors with windows and :.oors shut.

Schools and kindergartens were closed. The collection of meteorological and radiological monitoring data was organized and aerial radiological monitoring began. The military helicopters used were equipped with air samplers and radiation detection instruments and the crews were provided with personal dosimeters and respiratory protection.

As described in the INSAG report (ibid.), late in the night of 26 April, radiation levels in Pripyat started rising, reaching a value of the order of 10 mSv/h on 27 April.8 It soon became apparent that the lower intervention level for evacuation (250 mSv whole body

! dose) could be exceeded and eventually even the upper intervention level (750 mSv whole body dose) if the population remained in their homes and no other countermeasures were taken. Ad hoc evacuation plans, taking into account the actual situation, had to be de-vised. The evacuation of Pripyat (45,000 persons) started on the morning of 27 April after safe evacuation routes had been established for the 1100 buses on the basis of the first results of radiological monitoring, and all the necessary transportation means, equipment and escorting personnel were gathered and relocation centers had been defined, manned and equipped.

Of the on-site personnel, about 300 had to be hospitalized for radiation injuries and burns. Specialized treatment and care were given during the first few weeks. No member of the public from the off-site area had to be hospitalized for radiation injuries, although many attended hospitals for other reasons. Necessary provisions had to be made for the decontamination of people's skin and for the exchanging of clothing in some cases.

In the following days, the same protective actions had to be gradually applied to the other population (90,000) in an area of radius 30 km (or 18 mi.) around the plant. In addition, some 19,000 cattle had to be evacuated from the area in hundreds of trucks since many farmers refused to leave unless such provisions were made. However, the total number of livestock evacuated eventually grew to 50,000 from the Ukraine and 36,000 from Byelorussia.10

'One milliselvert (mSv) is equal to 100 millirems which is roughly equivalent to the natural background radiation received annually from cosmic and terrestrial sources in the United States and most other places around the world.

20 Theodore Shabad, " Geographical Aspects of the Chernobyl Nuclear Accident," Soviet Geography, No. 7, 1986, pp. 504-526.

11 l

l l

ibid.), when the Pripyat residents According got the order totoevacuate, Soviet news accounts a delegation went toassembled City Ha by Shabad u (ITto inquire abo t th e need for the emergency measure and was persuaded by General Berdov, the Ukrainian police official, that it was necessary. The buses, forming a caravan 10 km long, moved to the entrances of apartment buildings according to a prearranged plan. Residents thought the move was for only a short time, and traveled light as they climbed into the buses. The police searched all apartments to make sure everyone was gone. Of the total Pripyat population, about 10,000 children, including 7,000 school children and 3,000 children of preschool age were apparently taken directly to summer camps in the Ukraine and in other Soviet republics, while 25,000 adults were evacuated in about 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> to nearby rural areas in Kiev Oblast, mainly to Polesskoye Rayon, to the west. When the Pripyat evacuees arrived during the night of April 27-28, the population of Polesskoye Rayon, which was 35,000, virtually doubled. In the confusion many families became separated, and it took weeks to sort mat-ters out. When it was learned that the evacuation of Pripyat would be for a long period, residents were permitted to return for brief, supervised visits to pick up essential house-hold items, family documents and other needed articles that had been left behind in the rush. The evacuation of the remaining 90,000 from the 30-km zone took several days and was somewhat more difficult since it involved 71 villages and large numbers of livestock.

Soviet officials who supervised the evacuation reported that generally the behavior of the evacuees was exemplary and without panic (NUREG-1250).

Protecting Against Contaminated Food and Water Supplies Other emergency measures included the short and long term protection of the population against radiologically contaminated food and water supplies. Consumption of milk contain-ing more than one ten-millionth curie per liter of the radioactive isotope iodine-131 was banned. Also, standards for permissible levels of contamination in other food products were issued in early May. Special measures were taken to prevent or minimize radiation exposure via the ingestion pathway. Decontamination activities were extensive in the 30-km zone and other measures have extended beyond that zone, particularly to the city of Kiev (60 miles to the south). Another account, given at the Vienna meeting, indicates that the primary contamination of evacuated dairy cows was surface contamination and most animals were washed down (NUREG-1250). Those animals which had not been washed down or were in-jured during evacuation were slaughtered. In addition, intervention levels for I-131 in milk, 75% of which is exported from the area, as well as leafy vegetables, were established with the object of limiting the dose to a child's thyroid to 30 rem per year. Other, un-identified standards were selected for I-131 in meat, poultry, eggs, and berries. Later, still more foods were included and an overall dose limit of 5 rem for an individual in the first year was established (INSAG, of. c_i_t,.).

According to Shabad (Jo . cit.), programs to rehabilitate contaminated land for agricul-tural purposes should difTirentiate between areas close to the nuclear power station, where radioactivity is higher, and more distant land. The more highly contaminated area is unlikely to yield usable crops in the foreseeable future, and should simply be treated with calcium compounds that would fix them in the soil and prevent migration and dissemi-nation. Other areas might be sown with crops such as lupines that absorb radionuclides.

These crops would then be harvested and buried. Some of the less valuable land might be left idle altogether to revert to natural vegetation. In general, it would be advisable to modify the type of farming in any area exposed to radiation hazard with a view to planting crops that are more immune to radionuclides. For example, rye tends to absorb strontium to a lesser extent than wheat. Industrial crops are particularly suitable in areas with radiation pollution since they are subjected to processing after harvesting.

Potatoes, instead of being consumed as food, might be used to make starch since radionu-clides tend to become eliminated in the starch-making process. Another industrial crop 12

= _. -

4 Ie that might be cultivated in the contaminated zone is flax. As it happens, rye, potatoes and flax are the customary crops in the sandy soils of the Chernobyl area. Much experi-l mentation will doubtlessly be needed to demonstrate what measures will prove to be the most economical and safety-effective in restoring these agricultural lands'to productive use over subsequent decades.

There is also concern over contamination of the groundwater and surface water in the area i

i of the Chernobyl power station which also serve the Kiev region. These water supplies are i being monitored and remedial work has been done. In the early stages of the accident, well i

water rather than surface water was used by the residents of Kiev as a preventive measure.

' The public water supply has been used subsequently, but with constant sampling to assure that any contamination falls within protective action guidelines. A unique aspect of the

! Soviet emergency response was the seeding of clouds with silver iodide by aircraft to break them up and prevent rainfall in the region for a number of weeks after the accident--an l action that reduced deposition of airborne radionuclides in the area.

j The Soviets reported that, in June, construction of a series of hydraulic engineering j structures was initiated in order to protect the groundwater and surface water in the i

Chernobyl area from contamination (USSR, Footnote 4). Among such measures was the (now completed) strengthening of the dam for the cooling water channel at the Chernobyl site and the building of a holdup reservoir for rainwater along the whole periphery of the 2-km channel.11 Another protective measure was described by Yevgeny Velikhov, Vice President 1

of the Soviet Academy of Sciences, at a hearing on January 20, 1987 of the U.S. Senate Labor and Human Resources Committee.12 This involved the excavation and construction of a 50-meter deep concrete wall into the ground at the Chernobyl-4 unit in order to prevent

! contaminated groundwater and rainwater from leaking into the Pripyat River that empties j into the Kiev Reservoir. Also, a massive concrete slab has been constructed beneath the j Chernobyl-4 unit to protect against contaminated groundwater. Miners were brought from various regions of the USSR to excavate a tunnel 136 meters long in order that this con-crete barrier could be constructed at the base of the reactor.13

Other Post-Accident Recovery Measures 8ecause of expected power shortages in the Soviet Union in the winter and to reduce health i

effacts of the accident, site decontamination was expedited. The site, the turbine build-ing roof and the sides of nearby roads were sprayed with rapid polymerizing solutions to congeal upper layers of the soil to prevent resuspension of radioactive dust into the i

atmosphere. According to Velikhov (supra), about 20 centimeters (8 inches) of topsoil was removed from thousands of sqtare meters of surrounding land and placed in newly constructed i metal structures within the Chernobyl site. Other measures included removal of debris and contaminated equipment from the site and laying, where necessary, concrete slabs or filling j in with clean soil over hot spots of radioactivity.

! These decontamination efforts permitted the restart of the Chernobyl-1 unit on October 1 l and Unit 2 on November 6. Unit 3 is expected to restart by June 1987 ano the start of j construction of Units 5 and 6 will then begin. According to the INSAG report (ibid.),_the I

, 21" Stronger Supervision After Chernobyl," Nuclear News, Vol. 30, No. 1, January 1987,

, pp. 63-65.

I 180ennis Wansted, " Fifty Meters Deep: The Great Concrete Wall of Chernobyl," The Energy Daily,' January 21, 1987, p. 4.

j 18 Nuclear Engineering International, Vol. 31, No. 385, August 1986, p. 4.

13

, _. ~ _ _.__ _ _ _ _ _ _ _ _ _ _ _ _ _ , , _ _ _ , - _ -

___s

contaminated area within a 30-km radius was divided into three zones: a special zone (some 4-5 km around the plant), where no re-entry of the general population is foreseeable in the near future and where no activity besides that required at the installation will be permitted; a 5-10 km zone, where partial re-entry and special activities may be allowed after some time; and a 10-30 km zone, where the population may eventually be allowed in grama:1ricultural and cf radiological activities may be resumed, but which will be subject to a strict pro-surveillance. Access and egress controls for personnel and vehi-cles have been established at the zone boundaries to reduce the spread of contamination.

In early July, seven villages in Bragin Rayon within the 30-km zone had been decontaminated and could be resettled. Bragin, a town of over 6,000 population just outside the 30-km zone had over 1,000 homes and other structures washed down to remove radioactive traces. Some of the mossy older structures and fences that could not be easily decontaminated were dis-mantled.

The rehabilitation t(fort also involved the decontamination of thousands of trees, the uprooting of shrubs, and removal of a 5-10 cm layer of topsoil. The Byelorussian environmental protection chief, Viktor Kozlov, said peat soils retained more radioactivity and were to be decontaminated by being plowed under, as well as the use of irrigation and liming (Shabsd, op. cit.). To prepare for the runoff of snowmelt water in the spring of 1987, steps were take Tto bury radioactive garbage and to remove cow manure from fields since this ortjanic material tends to retain radioactivity. In the Ukraine and the Gomel Oblast of Byelorussia emergency housing programs got under way with plans calling for the construction of entire new settlements, accommodating at least 50 households each. A total of 4,000 r:ew homes were targeted to be ready in October, in time for the winter.

III. ASSESSING THE SHORT AND LONG TERM HEALTH CONSEQUENCES OF THE ACCIDENT Radiation Sickness and prompt Fatalities: The Role of Medical Treatment A review of the health and environmental consequences of the Chernobyl accident is pre-sented in Chapter 8 of NUREG-1250 (op. cit.). At high doses of ionizing radiation, many cells will be killed or functionally coiiip7omised resulting in possible severe damage to the individual. The effects of such damage appear rapidly and, at very high doses, may include death. After total body exposures of different magnitudes, the following effects are possible or expected: at greater than 50 rads (or rems), radiation sickness such as nausea, vomiting, and weakness may occur with 100% incidence expected at about 200 rads; at greater than 150 rads, in addition to radiation sickness, start of hematopoietic syn-drome with blood and immune system problems and some deaths within 60 days; at 300 to 500 rads, death in 50% of those exposed within 60 days; and at over 700 rads, nearly 100%

!. mortality is expected.

Some organs are at particular risk. For example, a 15-rad dose to the testes can tempo-rarily reduce fertility; at higher exposures, the severity and duration of reduced fer-tility is increased until at 300 to 700 rads, permanent sterility may result. In the ovary, a dose of 200 to 450 rads may cause sterility. A 200-rad dose to the lens of the eye may cause cataracts; for the skin, a dose of 250 rads may cause erythema, 700 rads -

loss of hair, more than 2000 rads - severe dermatitis or radiation " burns." The thyroid presents a special case because it concentrates radiciodines which have been inhaled or ingested. The dose from these radionuclides in the thyroid can greatly augment the thy-roid dose received from external irradiation and from other internal emitters. Total doses of 200 rads may cause impaired function, but loss of function is more likely for l- doses greater than 3000 rads. Complete destruction of the thyroid requires doses of 100,000 rads or more.

14 l

I

Energy deposited in a cell by ionizing radiation may not immediately affect vital cell functions but may damage the cell's genetic material, leading to an adverse effect ex-pressed at some later time. These effects are often referred to as statistical or sto-chastic effects. A stochastic effect is one for which the probability of occurrence in a person is proportional to the radiation dose received, but the severity is not. Induc-tion of cancer and genetic effects are the two types of stochastic effects associated with radiation exposure causing the greatest societal concern. Cataracts, operable thy-roid (non-cancerous) nodules, temporary or permanent sterility, and certain damages to the fetuses (teratogenetic effects) during the early months of pregnancy are also sto-chastic effects of significant concern. Estimating stochastic effects, particularly at low doses of radiation, is scientifically controversial since direct observations of these effects at doses under 50 rads or rems cannot be made. Estimates of these kinds frca the Chernobyl accident and the related issues will be discussed in the following section.

After the Chernobyl accident, acute radiation effects were diagnosed in 203 individuals, all of whom were either working at the reactor or were brought in to deal with the emer-gency. Twenty-nine persons were reported to have shown some acute effects within the first 30 to 40 minutes after the accident, and within the first 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br />, acute radiation sickness was diagnosed in these 203 individuals. Estimates of radiation doses received were based on clinical criteria, not on dosimetric data. Of 22 persons estimated to have received more than 600 rads, 21 died in 4 to 50 days; one died later. Of 23 estimated to have received 400 to 600 rads, 7 have died. All 53 persons estimated to have received 200 to 400 rads and all 105 estimated to have received 80 to 200 rads have survived.

At this time, 29 persons have died as a result of radiation exposures; in addition, one person died of severe burns and another was killed when part of the reactor building col-lapsed. Assuming that the Soviet dose estimates are reasonably accurate, the above data suggest a median lethal dose above 400 rads. -

The Soviet written report and presentations in Vienna indicated extensive and relatively prompt medical treatment measures shortly after the Chernobyl accident began (USSR, Foot-note 4). The medical and health section serving the plant was informed of the accident at about 02:00 on April 26. These medical personnel assisted the first 29 victims within the first 30 to 40 minutes, sending them immediately to the hospital. As an indicator of the speed and extent of the emergency medical response, the Soviets reported that by 06:00 on April 26, 108 people had been hospitalized and an additional 24 were admitted during the day. After initial diagnosis in local or regional hospitals, 129 patients suf-fering acute radiation sickness were sent to a specialized hospital in Moscow and 72

, patients were sent to clinical institutes in Kiev. Teams of specialists arrived within 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />, consisting of health physicists, radiology therapists, laboratory assistants, and hematologists. Within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, these specialists examined 350 persons and performed about 1000 blood analyses.

Ultimately, 203 persons were treated for acute radiation. syndrome resulting from gamma-and beta-ray exposure; none of these victims were members of the general public. Beta-ray exposure resulted in severe skin burns in 48 victims. Dr. Robert Gale, a U.S. physician, assisted Soviet doctors in giving bone marrow transplants to 13 of those who received sub-stantial radiation exposures, but 12 of these did not survive (INSAG, _op. cit.). Bone marrow transplantation has only been moderately effective according to Dr. Xngelina Gus'kova, one of the Soviet representatives at the Vienna meeting, and would not be l expected to play a significant role in any future major accident. Many of the deaths were hastened by burns resulting from beta exposure. Of the 203 hospitalized victims, 35 ex-ceeded doses of 400 rems (up to a maximum of 1200 to 1600 rems). Altogether, 1240 doc-tors, 920 nurses, and several thousand supporting assistants were mobilized to provide 15 l . _ _ . -

medical care services to these victims and others among the 135,000 evacuees. Each evac-uee was medically examined and blood tests were carried out. Long-term programs are being established for biological monitoring of exposed workers and the general population of evacuees.

Uncertainties in Assessing Delayed Cancer Fatalities In view of the many sources of uncertainty in preparing estimates of delayed cancer fatal-ities from the radioactive releases from the Chernobyl accident, it is scarcely surprising to find a wide range of estimates of these long-term fatalities. For whatever reasons, the Soviets chose, by their own admission, to provide non-conservative estimates of these fatalities (within their own territories) at the August meeting in Vienna (USSR, g. cit.).

It is generally agreed by the experts that the greater the radiation dose received by a5 individual, the greater the probability of a fatal cancer being induced during his or her remaining life span.

All of the 203 individuals who developed acute radiation sickness were estimated to have received in excess of 100 rems of radiation dose, and 29 of these have since died. Accord-ingly, each of the 174 survivors of this group have incurred a demonstrably higher risk of cancer fatality. Since cancer is primarily a disease of the elderly, there ma be a lapse of20to50yearsbeforetheincidenceofcancerfromtheChernobylradiation.{4 Less demon-strable is what increase in cancer fatality rates might be experienced by the 45,000 resi-dents of Pripyat and the 90,000 other evacuees from the 30-km zone surrounding the Cher-nobyl plant.

The Soviet report (g. cit.) states that doses for the vast majority of the population did not exceed 25 rems, altEigh some people in the most contaminated areas may have received 30 to 40 rems. However, the Soviet report also indicates that some inhabitants who re-mained in the area for 7 days or more before being evacuated would have received at least 60 to 80 rems from external radiation alone. Moreover, the averace dose for inhabitants of the zone 3 to 7 km around the plant is estimated to be 54 rems. That no acute radiation sickness symptoms were observed despite doses in excess of 50 rems might be explained by the fact that doses were protracted over a period of days. On the other hand, the tabu-lated values are once referred to as " maximum estimates," and this may help to explain some apparent inconsistencies - i.e., 30 to 40 rems may be intended as a more realistic es-timate of the maximum dose received off site. The report indicates that the dose estimates for the evacuees are preliminary and that more accurate estimates will be forthcoming.

The maximum collective external dose delivered to the 135,000 evacuees by direct radiation from the effluent cloud and by radionuclides deposited on the ground was estimated by the Soviets to be 1.6 million person-rems, or an average of about 12 rems per person. On the basis of that collective dose estimate and assuming a cancer fatality rate of about 2 per 10,000 rees, one would calculate about 320 excess fatal cancers in the. population attribut-able to the external radiation exposure. It would, however, be very difficult to detect this excess since, according to the Soviet report, in the course of normal events about 12% of the evacuees (about 16,000) would normally die of cancers from other causes. In addition to these fatalities from external radiation would be those resulting from inhala-tion or ingestion of radionuclides.

"A notable exception is leukemia, a relative rare form of cancer, which may emerge within a period of 2 to 12 years according to the experience data of the A-bomb survivors at Nagasaki and Hiroshima.

16

. i i ,

Although inhabitants of the 30-km zone were given potassium iodide (KI) to minimize uptake of radioiodines, the population received appreciable doses to the thyroid through inhala-tion and ingestion of the radioiodines I-131 and I-132. The uptake of I-131 in the 30-km zone and elsewhere along the path of the radioactive cloud was monitored extensively.

Doses to the 100,000 children who were monitored are of special concern, primarily because i

their smaller thyroid glands receive higher doses for a given intake and because of their high consumption of milk, a food in which I-131 fallout is likely to become concentrated.

It was estimated that most doses to the thyroid from inhaled or ingested radiciodine were less than 30 rems, although a few children may have received doses as high as 250 rems.

i Children receiving more than 30 rems to the thyroid were put under continuing medical ob-3

' servation, but the risk of hypothyroidism appears to be negligible below about a 1000-rad thyroid dose from I-131 (NUREG-1250, _op. cit.). If the average dose to the gland were about 30 rems, the collective thyroid dose would be about 4 million person-rees. If one adopts this collective dose and the thyroid risk coefficients presented in recent NRC re-ports,15 then about 100 excess thyroid cancers are estimated for this population, ten of them likely to be fatal.

~

The Soviets also estimated the collective radiation impact on the European part of the USSR which received the bulk of the radiation dosage. About 75 million persons exposed to varying amounts of radiation are found in four of the Soviets: The Ukrainian SSR, the Byelorussian SSR, the Moldavian SSR, and the Russian SFSR. Estimates of the external dose were provided in the Soviet report for various regions.of the European part of the Soviet Union, both urban and rural dwellers. Doses tended to be higher for people in rural areas because they spend more time outdoors and eat more locally produced food. The

! individual doses for the year 1986 ranged from 3 mrems to about I rem. The collective dose to all 75 million people over the next 50 years was estimated to be 29 million person-rems, about 30% of which would be received in 1986.

! The Soviets also estimated exposure through the food pathway. For food produced in 1986, there will be contamination from a variety of radionuclides, especially I-131, Ru-106, Ce-144, Cs-137, and Cs-134. The highest doses from Cs-134 and Cs-137 were believed to be l in the Ukrainian and Byelorussian regions of Poles'ye where an estimated 100,000 curies

of Cs-137 was deposited. A critical consideration is that the soil characteristics are j such that radiation uptake by plants is expected to be 10 or 100 times more than what it would be in other soil types. The collective dose delivered through the food pathway to

the population of the Poles'ye region for a period of 70 years after the accident was es-l timated to be 210 million person-rems. Exper'ts at the IAEA meeting in Vienna questioned i this figure because previous experience in estimating collective doses from release of l cesium to the atmosphere (e.g., from nuclear weapons tests) suggests that the dose via l food consumption is roughly equal to that from external exposure. Furthermore, prelimi-i nary whole-body scanning measurements suggest that cesium transfer through the food chain j

may be only about 10% of what was predicted for the region. According to the U.S. attendees of the meeting, there was general agreement among both Soviet and Western experts that the estimated collective dose given in the Soviet report was probably too high, perhaps by a factor of ten (NUREG-1250, op. cit.). The estimated effect of the Chernobyl accident on the exposed population of 75 miTTTon is quite substantial. Even if the Soviet report over-estimates the dose via the food pathway by an order of magnitude, one estimates a total collective dose of about 50 million person-rems. Assuming a cancer fatality rate of 2 per 10,000 rems, then about 10,000 fatal cancers (plus a comparable number of nonfatal cancers) would be projected over the next 70 years.

l 15E. Gilbert, " Late Somatic Effects," in Health Effects Model for Nuclear Power Plant Acci-dent Consequence Analysis, NUREG/CR-4214, U.S. Nuclear Regulatory Commission, 1985.

H. Maxon et al., " Thyroid Effects," in NUREG/CR-4214, ibid.

17 1

Beginning with the initial detection of radiation from the Chernobyl release at the Fors-mark nuclear power station in Sweden on April 28, radioactivity was monitored in air, on the ground, and in food throughout Europe. The quality and completeness of these data vary greatly by country. A summary of this data is presented in the June 12 report of the World Health Organization.18 In addition, more than ten European nations have pro-duced reports on the national effects of the Chernobyl accident, the most detailed of which are by Italy, Finland, and Sweden. The assessment of short- and long-term collec-tive external doses requires a population-weighted sum of doses from ground-deposited radionuclides over geographical areas that is sufficiently fine grained to reflect the significant variations in depositions resulting from wind and rainfall patterns. Rain can cause large increases in deposition versus dry areas only a short distance away. The effect on the food pathway due to deposition of cesium-137 with a 30 year half-life decay rate is especially important to the assessment of long term ingestion doses from agricul-tural products. In the short term, deposition of iodine-131 with an eight-day half-life is important to computations of ingestion doses.

Better data of the above kinds than presently exists as well as refinements in computer modeling are needed to improve the quality of health effects estimates from the Chernobyl radionuclide releases. As a tentative approximation, the average individual in Europe (outside the Soviet Union and those countries receiving very little of the fallout, namely, Spain, Portugal, Ireland, England, Denmark and most of France) is estimated to receive a 60-mrem dose from the accident, this dose being spread over a period of years (NUREG-1250,

g. cit.). For comparison, the average individual will receive about 100 mrem each year from Sackground radiation. Using this estimated average dose and a total population of about 350 million people in that part of Europe being considered, a collective dose of 20 million person-rems is calculated. Based again on a cancer fatality rate of two deaths per 10,000 rems, about 4000 excess cancer deaths outside the Soviet Union may be calculated to result from the accident. These deaths would be completely masked by the 70 million or so cancer deaths from other causes predicted in the population over the next 70 years.

Because of the very small doses of radionuclides reaching the United States (primarily I-131 in milk), only about two excess thyroid cancers are projected having a low probabil-ity of fatality (NUREG-1250, g. c_it.).

In the first month or so following the Chernobyl accident, data on radionuclide releases and deposition patterns was so sparse and poorly organized that few experts would hazard a guess as to the likely health effects of the accident. Thomas Maugh, a science writer for the Los Angeles Times, reported in less than a month after the accident the estimates of several scientists suggesting that as man as 5,000 to 40,000 cancer deaths could ulti-mately result from the accident.17 The highest of these was by John Gofman, professor emeritus of medical physics at the University of California, Berkeley. He estimated about 23,000 cancer deaths in the USSR and 16,500 deaths in Scandanavia and the rest of Europe.

The conservative estimate of 5,128 excess cancer deaths in the USSR and Europe were pre-pared by physicists Frank von Hippel of Princeton University and Thomas Cochran of the Natural Resources Defense Council, who cautioned that their estimates could be low by a

( factor of ten.

[

As reported by Maugh, the gaps between the two sets of fatality estimates result from marked differences in methodology. Von Hippel's and Cochran's estimates were based on computer simulations of radiation releases conducted at the Lawrence Liverrrare National 16 World Health Organization, Regional Office for Europe, Summary Report of Working Group on Assessment of Radiation Dose Commitment in Europe Due to the Chernobyl Accident, Bilthoven, June 25-27, 1986, Copenhagen, September 8, 1986.

17 Thomas H. Maugh II, "Chernobyl Death Toll May Run Into Thousands," Los Angeles Times, May 22, 1986.

18 l

l

Laboratory and the use of the same fatality rate of 2 excess deaths per 10,000 person-rems as used above in the Soviet estimates as derived from the BEIR III report of the U.S.

National Academy of Sciences. s By contrast, Gofman's approach was based on measured radiation levels in selected cities most affected by the Chernobyl accident, but also by his own estimates of dose-response that are about 17 times higher than the fatality rates per unit dose of the BEIR III report.

A more recent and detailed analysis and fatality estimates prepared by von Hippel and Cochran are found in the August / September 1986 issue of the Bulletin of the Atomic Sci-ences.ts It is noteworthy that this carefully reasoned report did not provide any bottom-line number for cancer fatalities. Rather, the following conclusion was reached for the expected fatalities for the exposed population of Europe, including European USSR:

[I]t is the addition of such small extra risks over many millions of individ-uals that results in our estimate of thousands to tens of thousands of extra cancer deaths...As a result of all the compounding uncertainties in the fac-tors involved, our estimates of the long-term health consequences of the Cher-nobyl accident are uncertain even as to order of magnitude.28 One of the lowest estimates for cancer fatalities expected to result from the Chernobyl accident resulted from an analysis of the European Economic Commission that put the number at 10M excess cancer deaths over the next 70 years, compared with a projected 60 million cancer deaths from other causes over the same period.20 In a Wintar 1987 publication by the National Academy of Sciences, National Academy of Engi-neering, and the Institute of Medicine, forum articles by Richard Wilson, Robert Gale,2 and Frank von Hippel continued the dialogue on the cancer fatality effects from Chernobyl.

Von ibid.)

werldHipfel havenoted been that "somewhat calculations he and different" as Cochran a result ofmade in data Soviet their supplied August article at the(Vienna meetir9 of the IAEA and further stressed that:

large uncertainties will remain, however, if only because of the continuing M certainty in cancer dose-effect coefficients. Our knowledge of these coeffi-Cents, especially the iodine 131 thyroid dose-effect coefficients, could be considerably improved by a proper epidemiological follow-up of those groups I

near Chernobyl who received the highest exposures.

Professor Richard Wilson, Professor of Physics at Harvard University, also commented on the uncertainty surrounding cancer dose-effect coefficients:

(T]he individual doses and the dose rates resulting from the Chernobyl plant emmisions are so small that the cancer risks are uncertain. Most estimates of risk use a linear dose-response relationship of one cancer (fatality] per 5,000

' to 10,000 person-rems, but animal data suggest that the linear relationship l overstates the risk when doses are given at a low rate...We can characterize the l

effects more accurately by comparing them with the health risks from burning

! fossil fuel, using similar calculations and assuming a similar dose-response i

I tsNational Academy of Sciences, The Effects on Populations of Exposure to Low Levels of Ionizing Radiation: 1980 (BEIR III), National Academy Press, Washington, D.C., 1980.

18 Frank von Hippel and Thomas B. Cochran, " Estimating Long-Term Health Effects," Bulletin of the Atomic Scientists, Vol. 43, No. 1, August / September 1986, pp. 18-25.

20 Nuclear Engineering International, Vol. 32, No. 390, January 1987, p. 3.

19

relationship at low doses although the biological endpoints are different. Such calculations suggest that the health effects in Europe of toxic pollutants caused by Soviet fuel burning each year are as bad as the total health effects caused by the isolated accident at Chernobyl.

In his forum article, Dr. Robert Gale presents the following calculations of excess cancers and cancer deaths using Soviet projections of radiation doses and then provides a commen-tary regarding their likely overestimation:21 For the 135,000 people in the evacuation zone, 1,000 excess cancers and 500 ex-cess cancer deaths; for the other 75 million people in the European Soviet Union, 5,000 to 50,000 excess cancers and 2,500 to 25,000 excess cancer deaths; for the other 5 billion people worldwide, 2,500 to 100,000 excess cancers, and 1,250 to 50,000 excess cancer dead.s. The estimated worldwide total would be 8,000 to 150,000 excess cancers and 4,000 to 75,000 excess cancer deaths. These ranges are approximate.

These data are controversial. In my opinion the number of excess cancers is likely to be overestimated by this approach for the reasons I described. Even the lower estimates may be too high and the actual number of cases may be con-siderably lower, perhaps by a factor of 10. These excess cancers would repre-sent a less than one percent increase from normal levels in the European Soviet population. Nevertheless, each person with an unnecessary cancer is important to us and to the Soviets.

In view of the likelihood that a substantial fraction of the projected cancer fatali-ties from the Chernobyl accident relates to individuals who will have received rela-tively minor incremental doses of radiation above normal background levels, it was sur-prising that none of the cited references gave explicit attention to the de minimis nature of these risks for many exposed individuals. The use of a de minimis standard of individual risk for regulatory decisions is drawing increasing attention from na-tional and international scientific bodies such as the (U.S.) National Council on Radiological Protection and Measurements, the (U.K.) National Radiological Protection Board, and the International Commission on Radiological Protection, as well as a number of U.S. Federal agencies and the U.S. Courts.22 Among the suggested applications of de minimis risk concepts are: the setting of regulatory priorities; as a " floor" for as .

low as reasonably achievable (AI ARA) considerations; as a cut-off level for collective 21" Forum on Chernobyl," Issues in Science and Technolony, Vol. 3, No. 2, Winter 1987, pp. 6-8.

22For a review of developments in applying de minimis risk concepts see:

Miller B. Spangler, "A Summary Perspective on NRC's Implicit and Explicit Use of De Minimis Risk Concepts in Regulating for Radiological Protection in the Nuclear Fuel Cycle," De Minimis Risk, Chris Whipple, Ed. (New York: Plenum Publishing Corporation, 1987).

Miller B. Spangler, "The Need for De Minimis Risk Standards in Regulatory Decision Mak-ing: An Individual or a Societal Risk Concept?" Environmental Health Risks: Assess-ment and Management, R. Stephen McColl, Ed. (Waterloo, Ontario, Canada: University of Waterloo Press, 1987).

20

.I dose assessments; for setting outer bounds of geographical zones; as a floor for defini-tion of low-level waste; as a presumption of triviality in legal proceedings; to foster administrative and regulatory efficiency; and to provide perspective for public understand-ing, including policy judgments.23 Although not explicitly mentioning de minimis risk concepts, several of the aforementioned references did provide commentaries of the small nature of the estimated increments to in-dividual cancer risk resulting from the Chernobyl accident, some fraction of which might justifiablyberegardedasdeminimisrisk. For example, von Hippel and Cochran noted that The associated individual risks are very small for all but those populations within a few tens of kilometers of the accident. However, these small risks, added up over hundreds of millions of individuals, could total thousands to tens of thousands of cancer deaths and generally less fatal thyroid tumors."18 A second explanation of this general kind was provided by J. Puskin of the U.S. Environ-mental Protection Agency, who was the author of the " Health and Environmental Consequences" chapter of NUREG-1250 (op. cjit.):

[T]he risk to an average individual in the population owing to the accident is relatively small. Assuming that the original Soviet estimate of dose received through the contaminated food pathway is high by a factor of 10, the estimated average individual dose from external and internal pathways would be about 0.67 rad, roughly equivalent to the dose received from background radiation over a period of 7 years. Based again on a risk factor of 2x10 */ rad, this dose would give an estimated lifetime risk of 0.013%, which is only about 0.1% of the stated Soviet baseline risk of fatal cancer (12-13%).

A third example is the Forum article by Richard Wilson:

Only for the group exposed to the highest amounts of radiation, the 24,000 people living between 3 and 15 kilometers from Chernobyl (excluding those living in Pripyat), were the exposures at a level--45 rems average--at which human data show adverse health effects. These people will have about a 3 percent increase in cancer incidence. This is likely to be compensated for by the increased health care that they will receive. For the 2 million l people living in Pyelorussia (downwind from Chernobyl), the Soviet estimate for increased lifetime dose is 0.7 rem. This is considerably less than the difference in the lifetime external dose a person receives on moving from New York to Denver. It is also less than the difference in the dose a person receives from inhaled radon if he or she moves from an average New England house to an average Pennsylvania house. Since few people, if any,

worry about these differences in natural background radiation, it would be

! inconsistent for the 2 million people in Byelorussia to worry about their exposure from Chernobyl and even more inconsistent for less-exposed West Europeans to worry.2 I

tsJoyce Davis, The De Minimis Regulatory Cut-Off concepts, Testimony before the Advisory Committee on Reactor Safeguards, U.S. Nuclear Regulatory Commission (February 9,1984),

23 pp.

21

IV. A PRELIMINARY VIEW OF SOME IMPLICATIONS OF THE CHERNOBYL ACCIDENT FOR THE REGULATION OF U.S. NUCLEAR POWER PLANTS Implications for Graphite-Moderated Reactors A few of the differences between the design features and safety concepts of the RBMK graphite-moderated, pressure tube reactor at Chernobyl and those of the Light Water Reactor (LWR) prevalent in the United States have been discussed in Part I. The Fort St. Vrain High Temperature Gas Cooled Reactor (HTGR) in Weld County, Colorado and the Department of Energy's (DOE) N-Reactor at the Hanford Reservation in Washington State are the only oper-ating power reactors moderated by graphite in the United States. As the N-Reactor is not licensed by the NRC and is under the authority of 00E, the Chernobyl implications for it are being assessed separately by that agency and others. The HTGR-type reactor has been under development in the United States and West Germany since the late 1950's. The current HTGR development efforts in the United States are concentrating on the Modular HTGR (MHTGR) concept which uses available HTGR technology in combination with inherent and passive safety features. The MHTGR concept is being proposed by DOE in conformance with the Commission's recently published "Statu4nt of Policy for the Regulation of Advanced Nuclear Power Plants" (51 FR 24643]. Thus, assessment of the Chernobyl implications and candidate issues has value both to Fort St. Vrain and the MHTGR.

Among the candidate issues examined by the U.S. Nuclear Regulatory Commission are those in- i volving operations, design, containment, emergency planning and severe accident phenomena.24 The only commonality of the 330-MWe Fort St. Vrain reactor and the MHTGR with the Chernobyl design is the use of a graphite moderator and gravity-driven control rods. Fort St. Vrain uses a helium coolant which is pressurized to 700 pounds per square inch (4.8 megapascals) and which flows downward through half-inch diameter holes in a fully ceramic (graphite) core. The reactor core and the entire primary coolant system, including steam generators and helium circulators, are enclosed in a prestressed concrete reactor vessel (PCRV) which, through use of inner and outer penetration seals and in conjunction with a filtered and vented confinement building, satisfies the NRC's General Design Criteria for reactor containment. The MHTGR will have a steel pressure vessel rather than a PCVR.

In the issue areas of operations, design, containment, emergency planning, and severe acci-dent phenomena, the assessments performed by the NRC found that the implications have gen-erated no new licensing concerns for HTGRs and that the overall conclusions and the spe-cific area conclusions are the same as for LWRs (see below),as Indeed, following the THI-2 accident, emergency planning needs for Fort St. Vrain were reviewed and it was concluded that a 5-mile radius for an emergency planning zone (EPZ) would be sufficient in comparison with the 10-mile radius required for LWRs. A study of the potentials for a Chernobyl-type fire and explosions derived from hydrogen and carbon monoxide " water gas" in Fort St. Vrain was initiated immediately with the news of the Chernobyl event. While the study is not yet complete, the findings are expected to conclude that the use of a helium coolant, the over-all negative reactivity coefficient, completely diverse alternate shutdown and cooling sys-tems and the protection offered by the PCRV against reactor fires, internal postulated ex-plosions and fission product release to the environs remove Fort St. Vrain from any vulner-ability to characteristics of the Chernobyl design.

24USNRC, " Assessment of the Implications of the Accident at Chernobyl for Nuclear Safety Regulation in the United States," NUREG-1251, Oraft for Comment, U.S. Nuclear Regulatory

l. Commission, February 1987.

25 I

Unless otherwise noted, the analyses and conclusions presented in the remainder of this part have been summarized from NUREG-1251 (ibid.).

22

k Administrative Controls Over the Safe Operation of Nuclear Power Plants: Dealing With Human Error and Accident Manaaement AdministrativecontrolsoverplantoperationsincludeNRCrulesandregulaf. tons, facility license conditions and Technical Specifications (TS) and plant procedures. The overall administrative control framework requires that safety-related activities at nuclear power plants be conducted in accordance with approved written procedures. These activities in-clude, for example, operations, tests, inspections, calibrations, maintenance, experiments, modifications, safety review and approval functions, and audits. The safety design basis of the plant is based upon assumed initial conditions for transients and emergencies.

These initial conditions (e.g., temperatures, pressures, control rod positions, and equip-ment availability) establish a " safe operating envelope "~ Effective administrative con-trols are necessary to ensure that reactor operations are conducted within this safe oper-ating envelope. Clearly, for administrative controls to be effective, they must be tech-nically accurate and complete, they must be understood by those responsible for implement-ing specific procedures, and management must ensure that they are enforced. A key finding from the Chernobyl accident is that such administrative controls in place at Chernobyl were

not effective in maintaining conditions within the safe operating envelope established by the RBMK designers. ,

In NUREG-1251, the administrative controls over plant operations in the United States were reviewed to determine if adequate controls are in place to maintain plant conditions within the safe operating envelope. This review included an assessment of procedural adequacy and compliance, approval of tests, bypassing of safety systems, availability of engineered safety features, operator attitudes toward safety, management systems, and accident manage-ment. The staff confirmed that some ongoing activities with a nexus to the Chernobyl acci-dent should continue. In addition, a few new issues mquiring staff attention were identified.

Emergency Operating Procedures (EOPs) are intended to ensure safe shutdown and to mitigate the effects of accidents and transients. Facility E0Ps are designed for coping with acci-dents and transients which initiate from within the safe sperating envelope. The ability

, of operators to successfully implement E0Ps depends upon plant safety parameters initially being within the safe operating envelope. As a result of the TMI accident, NRC required that new symptom-based E0Ps be developed. These new procedures have not been implemented by all facilities, and NRC audits have identified deficiencies in implementation at sev-

! eral facilities. Thus, significant effort by licensees is needed to complete implementa-

, tion of new E0Ps.

. Operator training needs to stress fundamentals of reactor safety, how the plant should

! function, and the underlying danger if plant conditions move outside the safe operating envelope. With adequate training and knowledge, personnel would be less likely to suc-cumb to pressures to speed up, take shortcuts, or defeat. safety functions knowing the

possible consequences. Operating experience and the Chernobyl event indicate more train-

! ing is needed in the areas of maintenance of safety parameters and plant conditions within the safe operating envelope, E0Ps and accident management. -

The Chernobyl accident has emphasized the need for contingency planning (assuming core damage has occurred) to ensure that appropriate controls, training, and planning have pre-j pared the plant staff to manage plant assessment activities, response actions, and emer-gency actions. Significant effort to prepare for events involving degraded core cooling and to upgrade emergency planning has been accomplished. However, more work needs to be

, done in training and procedure development for coping with severe core damage and for ef-

! fective management of containment.

23

- - , - - - - , - , , - - . - - , _ - , - , , , - - - - , , - - - - - - - - - - - - - - - - , ---.----r------ --- - - - - , . , -

- ,-,---..,-n.- - - - . -

Management attention and diligence is required to ensure that plant operations, testing and maintenance are conducted within the safe operating envelope. Management must focus on assuring that all of the administrative control systems are effective and enforced.

Audits, internal inspections and review of operating data and events must be performed to obtain feedback on the quality of safety activities. Control over reviews of changes, tests, and procedures must be conducted by qualified and knowledgeable individuals. Ex-perience has shown that the quality of some of these reviews has not been consistent and, in some instances, design changes have been made and testing conducted that place the plant outside the safe operating envelope. Industry action to improve the review process required by NRC has been undertaken; however, more needs to be done to focus the responsi-bility for safety. All plant personnel have a safety responsibility, but this responsi-bility is coupled with other functions. The staff believes that the benefits of a high level on-site nuclear safety manager, with no other responsibilities or duties, should be examined.

It should be stressed that the significance of lessons to be learned from the Chenobyl nuclear accident would have been very much greater, if it were not for the lessons we (and many other nations using nuclear plants similar in design to ours) have learned resulting from the nuclear accident at Three Mile Island and subsequent insights and learning expe-rience from the evaluation of plant operating experience data and research programs. Insti-tutional arrangements with foreign nations have magnified the shared benefits of these kinds of learning (NUREG-1070, op. c3.), i Desian Implications for Accident Prevention and Containment Performance The nuclear design of U.S. reactors, notably the absence of positive void coefficients and control rods that are fast acting and offer substantial shutdown margins, provides assur-ance against a Chernobyl-type superprompt critical power excursion. Nevertheless, in NUREG-1251 the possible need for confirmatory reviews of the acceptability of risks from other low probability reactivity-event sequences was assessed, including accident scenarios that could occur at low power and shutdown conditions that may not be bounded by analyses for full power. Safety assessment also was made of multiple-unit site implications includ-ing consideration of effects of shared shutdown-related systems and of radioactive release effects on operator safety at the other units. Moreover, consideration of fires focused on the adequacy of protection provisions for fires with radiation present.

Although the design of U.S. nuclear reactors provides assurance against a Chernobyl-type superprompt critical power excursion, it was concluded that the judgments which determined the identification of possible accident sequences analyzed in safety analysis reports and underlying design approvals should be reviewed for reconfirmation of their validity with the more sophisticated analytical tools now available. This could be accomplished either by the NRC or by licensees with NRC groundrules and review.

One of the unique aspects of the Chernobyl accident was that it occurred at a relatively low power (<7%). This has been a cause of some concern since low power operation is gen-erally considered to be a safer condition than high or full power operation. The aspect of low power operation considered in the NRC reyiew is whether the design basis events are presently being evaluatsd at their most limiting power level or whether more attention should be given to these events at low power. It was concluded that a systematic evalua-tion of accident initiators at low power should be carried out, especially in view of the fact that existing Probabilistic Risk Assessments have paid very little attention to low power conditions or testing in evaluating severe accident risk.

24

The radioactive gas and smoke release at Chernobyl, Unit 4 spread to the other three oper-

ating units at the site. The airborne radioactivity was transported to the other units by i a common ventilation system as well as general atmospheric dispersion paths. This raises the basic question of how accidents at one unit of a multiple unit site affect the remain-l ing units, and further questions of how these effects may be compounded when structures, systems and components are shared between units. The NRC review concluded that the Cher-3 nobyl experience should be taken into account in the current assessment of the adequacy

of protection of control rooms of other nuclear power units on a multi-unit site in the l event of an accident at one of the units. This should be done on the basis of recently i developed descriptions of radioactive release source terms.2s New plants should not share j systems forming part of the shutdown capability.

Firefighting capability, involving about 30 fires due to the Chernobyl accident, was em-l phasized by the Soviets among the lessons to be learned from their accident. The Soviets stressed the need for special,equipmert to lift firefighting equipment to roofs. Also of I concern was the need for protective clothing for fi"t fighters when fighting fires in radio-active environmeats. However, the NRC review of our own regulations, equipment, and pro-cedures to combat fires revealed that actions already taken to upgrade fire protection plus those in progress are judged adequate, except possibly provisions for firefighting with '

radiation present. It was reremmended that current provisions be reviewed for this partic-ular safety issue. "

. The Chernobyl accident brought new attention to containments and performance of contain-

! ments under severe accident condition's. Such challenges include phenomena such as in-j creased pressures from an uncontrolled hydrogen combustion or release of large quantities

of non-condensible gases from core-concrete interactions. Research programs and regula-tory initiatives to address the issue of containment performance during severe accidents I have been under way for some time at the NRC.27 The Severe Accident Policy Statement l issued in July 1985, stated that the Commission will strike a balance between accident j prevention and consequence mitigation encompassing actions that improve understanding of

{ containment building failure characteristics and design features or emergency actions that

decrease the likelihood of containment. building failures (NUREG-1070, Footnote 1).

l The Policy went on to state that although not specifically designed to accommodate all of the hostile environments resulting from the complete spectrum of severe accidents, con-

'. tainments can retain a large fraction of the radiological inventory from a portion of the l spectrum. For example, large, dry containments may be sufficiently capable of mitigating j 2sSee, for example:

l USNRC, " Reassessment of The Technical Bases for Estimating Source Terms," NUREG-0956,

Draft Report for Comment, U.S. Nuclear Regulatory Commission, July 1985.

USNRC, " Nuclear Power Plant Risks and Regulatory Applications," NUREG-1150, Draft Report j for Comment, U.S. Nuclear Regulatory Commission, February 1987.

27See, for example: -

W. T. Pratt and R. A. Bari, " Containment Reapon'se During Degraded Core Accidents Ini-

tiated by Transients and Small Break LOCA in'the Zion / Indian Point Reactor Plants,"

NUREG/CR-2228, U.S. Nuclear Regulatory Commission, July 1981.

USNRC " Nuclear Power Plant Severe Accident Research Plan," NUREG-0900, U.S. Nuclear

• Regulatory Commission, January 1983.

USNRC, " Estimates of Early Containment Loads From Core Melt Accidents," NUREG-1079,

! U.S. Nuclear Regulatory Commission, December 1985.

25 I

e the consequences of a wide spectrum of core-melt accidents; hence, further requirements m y be unnecessary or, at most, upgrading current requirements to gain limited improvements of their existing capability may be necessary. The Commission expects that these matters will continue to be subjects for study (e.g., in the NRC research program and in further plant-specific studies such as the Zion and Indian Point probabilistic risk assessments).

Integrated systems analysis will be used to explore whether other containment types exhibit a functional containment capability equivalent to that of large, dry containments. Al-though containment strength is an important feature to be considered in such an analysis, credits should also be given to the inherent energy and radionuclide absorption capabilities of the various designs as well as other design features that limit or control combustible gases.

Further research and plant-specific studies should improve our understanding of the con-tainment loading and failure characteristics for the various classes of facilities. The analyses should be as realistic as possible and should include, where appropriate, dynamic and static icadings from combustion of hydrogen and other combustibles, static pressure and temperature loadings from steam and non-condensibles, basemat penetration by core-melt materials, and effects of aerosols on engineered safety features. A clarification of con-tainment performance expectations will be made including a decision on whether to establish new performance criteria for containment systems and, if so, what these should be. Accord-ingly, based on the existing research and ongoing containment evaluation programs, new pro-grams or initiatives are not needed as a result of the Chernobyl accident that failed a containment of drastically different design.

Implications for Emergency Planning and Improved Understanding of Severe Accident Phenomena There are three problems in drawing a nexus between the Soviet response to the Chernobyl accident and emergency planning implications for U.S. plants:

(1) After the TMI accident, large resources were committed to improved emergency planning and response capabilities at each U.S. nuclear plant whereas at the time of the Cher-nobyl accident the Soviets apparently had very minimal plans to cope with the emergency problems.

(2) The specifics of the Chernobyl release are unique to the RBMK design whereas the amounts of radioactive material released from U.S. plants would, for most accident sequences, be considerably less. U.S. plants have substantial containments and severe accidents would generally progress more slowly than at Chernobyl with much longer warning times before any containment breach for all but a few low probability sce-narios involving large-scale core melt.

27500, for example:

W. T. Pratt and R. A. Bari, " Containment Response During Degraded Core Accidents Initiated by Transients and Small Break LOCA in the Zion / Indian Point Reactor Plants," NUREG/CR-2228, U.S. Nuclear Regulatory Commission, July 1981. -

USNRC, " Nuclear Power Plant Severe Accident Research Plan," NUREG-0900, U.S. Nuclear Regulatory Commission, January 1983.

J. Mishima et al., " Effectiveness of Engineered Safety Feature Systems in Retaining Fis-sion Products," NUREG/CR-3787, U.S. Nuclear Regulatory Commission, USNRC, " Estimates of Early Containment Loads from Core Melt Accidents," NUREG-1079, U.S. Nuclear Regulatory Commission, December 1985.

26 a

(3) The Soviets had to assemble 4,000 busses and trucks for the Chernobyl evacuation, whereas in the United States most of the public has access to private automobile transportation to be supplemented for emergency evacuation by bus and other trans-portation that is preplanned.

Regarding cost-effective measures for off-site decontamination, long-term relocation, and

- restoration of agricultural and other resource use, it would appear that there is much to be learned of value from the pioneering Soviet experience in these matters--both their successes and less effective innovations.

i j One issue being reviewed by the NRC is the adequacy of the size of the emergency planning zones (EPZs) around U.S. commercial nuclear power plants. The Soviets evacuated 135,000 persons from a zone of 30 kilometers (18 mi.) radius of the Chernobyl plant. The 45,000 residents of Pripyat (3 mi.) were initially sheltered as a protective measure and then evacuated the following day due to increased radiation doses. Logistics and contamination control problems along evacuation routes also influenced the timing of evacuation. Inges-tion pathway protective measures were taken in the U.S.S.R. both within the 30-km zone and well beyond. Similar measures were also taken in several Soviet Bloc countries, in Scandi-

navia and Western Europe.

Regarding EPZ size, a major NRC research effort began about 1981 to obtain a better under-l standing of fission product transport and release under severe accident conditions. (See Footnotes 26 and 27). The latest of these studies (NUREG-1150, o,p. cit.) will be used to identify, analyze and put into perspective the actions taken during We~ Chernobyl accident; i.e., how the Chernobyl source term and emergency actions relate to U.S. source term and emergency actions. NUREG-1150 results will have a substantial impact on decisions concern-ing the adequacy of U.S. EPZ size among other things. A preliminary conclusion is that the Chernobyl accident and the Soviet response does not reveal any apparent deficiency regarding U.S. plans and preparedness including the 10-mile plun.e exposure pathway EPZ size and the 50-mile ingestion exposure pathway EPZ size. These zones provide an adequate basis to plan

and carry out the full range of protective actions for the populations within these zones, as well as beyond them, if the need should arise. Notwithstanding this, the basis for the

! emergency planning zones needs to be reexamined in the light of new information gleaned from j source term research (NUREGs-0956 and 1150, oE. c_it.). _

Emergency response planning also involves medical treatment issues such as the adequacy of (1) the Federal policy on potassium iodide (KI) prophylaxis, and (2) medical services in reasonable proximity to U.S. nuclear power plants. The Soviets credited the use of KI by the Pripyat population with the permissible iodine content (less than 30 rad) found in 97%

of the 206 evacuees tested at one relocation center. They said there were no serious ad-verse reactions from the use of KI. The federal policy statement acknowledges the effec-tiveness of KI in certain circumstances; however, it concludes that the preponderance of information indicates that a nationwide requirement for the predistribution or stockpiling for use by the general public would not be cost-beneficial. Further, the staff believes that the present arrangements and future plans for medical services around U.S. commercial nuclear power plants are adequate.

~

The NRC also reviewed the ingestion pathway issues from the Chernobyl accident involving the adequacy of (1) U.S. standards for the ingestion of radioactive materials through food and water (and the mechanisms for adapting those standards to changing conditions) and (2) the U.S. plans and preparedness for taking measures to protect the public from the ingestion of hazardous levels of radioactive materials in food and water. The Food and Drug Administration has published action levels (47 FR 47073) to provide state and local 27 l

1

. \

l agencies recommendations for taking protective action in the event that radiation causes contamination of human food and animal feeds. These can be used to determine whether levels of radiation encountered in food after a radiological incident warrant protective action and to suggest appropriate action that may be taken if action is warranted. The states and local governments have primary responsibility for taking protective actions to protect the public from the ingestion of contaminated foodstuffs. The Protective Action Guides (PAGs) for foodstuffs and animal feed would apply during and after an accident, al-though they may be modified by the federal government or by state and local governments.

The federal mechanism for providing recommendations to state and local governments is the Federal Emergency Response Plan.28 The adequacy of the federal guidance cannot be tested against the Chernobyl accident because the specifics of the Chernobyl release are unique to the RBMK design. However, the adequacy of the federal guidance will be reviewed through evaluations provided by U.S. source term research (supra). To date it appears that the existing federal guidance will provide adequate protection for members of the general pub-lic from contaminated food.

Decontamination techniques employed by the Soviets, including those for personnel, appear to be similar to those used by the United States in support of the nuclear weapons testing program, the TMI-2 accident, and interdiction related to chemical spills. Desert areas and coral atolls have been decontaminated, but there is little U.S. experience in the large-scale decontamination of forests and orchards or croplands with the purpose of restoring viability and productivity to the land as now being pioneered by the Soviets. However, the effectiveness of Soviet decontamination and relocation efforts will also have to be examined as the data become available. The U.S. capabilities will have to be examined within parameters provided by U.S. source term research. Past and current research results indicate that at least temporary relocation may be necessary for populations at distances that will vary with the source terms, weather states, and other conditions that are unique to each of a large spectrum of severe accident scenarios.

Improved understanding of severe accident phenomena, especially radionuclide dispersion modelling will result from ongoing data monitoring and analysis in the USSR and elsewhere in Europe. Research on other aspects of severe accident phenomena, computer modelling of this phenomena and further advances in Probabilistic Risk Assessment will also contribute to a reduction in uncertainty that will serve to improve the quality of emergency response and p'reparedness planning for severe nuclear accident scenarios, especially those that are the dominant contributors to the overall level of risk.

2sFederal Emergency Management Agency, " National Radiological Emergency Preparedness /

Response Plan for Commercial Nuclear Power Plant Accidents (Master Plan)," published in 45 FR 84910, December 23, 1980.

USNRC/ FEMA, " Criteria for Preparation and Evaluation of Radiological Emergency Response Plans and Preparedness in Support of Nuclear Power Plants," NUREG-0654, U.S. Nuclear Regulatory Commission, November 1980.

28 l

l

- . , ._ _ . .