I think that the creation of the website is an excellent initiative. Undoubtedly, that the study of the main causes of the three major accidents included in the website will help to reach the goal of reducing to the minimum possible the possibility of another major nuclear accident in any other country.

This is one step in the correct direction, but it is not enough. Other steps should be carried out with the aim of adding to the website a deep and objective evaluation of what happened in these three nuclear accidents and the main recommendations that should be implemented by governments, the nuclear industry, the nuclear regulators, nuclear research centers and the international and regional organizations in charge of the promotion of the use of nuclear energy for electricity generation and the application of approved nuclear safety measures.

My comment is: The Fukushima accident must be perceived in the perspective that it is the last one in the chain of already occurred severe nuclear accidents. For me, an amazing fact is, that nobody ever mention and in no available databases occurs THE FIRST severe accident with an extensive core melt, which happened in Slovakia in 1977. The first accident was in 1977 in A1 Bohunice NPP, Slovakia - one small unit (less than 300 MWe) with HWGCR reactor KS150 carbon dioxide cooled, heavy water moderated type with reported 25% core melt, where also some releases were approved. (see the publication: [BB] – KURUC, J., Matel, L.: Thirtieth Anniversary of Reactor Accident in A-1 Nuclear Power Plant Jaslovske Bohunice, Minulosť a súčasné trendy jadrovej chémie, 2007, Omega Info, ISBN 978 80 969290 9 2). This one should be included into the major nuclear accidents list too, once we want to be objective and to try to cover all the aspects related to root causes and consequences of the accidents.

Dear Jirina. Many thanks for the link. I knew a little bit about this incident, but I will read carefully the report mentioned in your comment.

Thank you for your prompt answer and I am very pleased that this kind of activity started. Unfortunately, from what I read at the site you inform about, are mostly summaries about "what happened", but there is no TECHNICAL ANALYTICAL views with pointing out the weaknesses discovered, and what does it mean really with respect to current safety and practices used in safety/risk evaluation of NPPs. My opinion is:

- wrong safety goals currently used for severe accidents - no risk limits are used, only frequencies (see one of my publications: J.VITÁZKOVÁ, E. CAZZOLI: Common Risk Target for severe accidents of nuclear power plants based on IAEA INES scale. Nucelar Engineering and Design, Vol.262, Pages 106-125; ISSN 0029-5493, September 2013)

- defence in depth concept should be substantially re-evaluated, because design of NPPs did not count on severe accidents in general, thus no additional improvements are sufficient to cover the loads coming from severe accidents, whatever initiators we consider (we can never cover ALL possible initiators and their combinations which potentially might occur) - see my contribution in book Risk Analysis IX, you can find it for instance on:

http://www.witpress.com/978-1-84564-792-6.html or http://www.amazon.com/Analysis-Transactions-Information-Communication-Technologies/dp/1845647920

or

http://www.amazon.com/Analysis-Transactions-Information-Communication-Technologies/dp/1845647920

or the full text on:

http://books.google.sk/books?id=omODAwAAQBAJ&pg=PA35&lpg=PA35&dq=The+principle+of+Defence-in-Depth+in+the+perspective+of+Probabilistic+Safety+Analyses+in+the+wake+of+Fukushima&source=bl&ots=jy2qWb8id_&sig=0ohSa_aj2A6S32BRZQGQcnPu_-U&hl=sk&sa=X&ei=USmHU9CEGKX20gWjwIHgCw&ved=0CCwQ6AEwAA#v=onepage&q=The%20principle%20of%20Defence-in-Depth%20in%20the%20perspective%20of%20Probabilistic%20Safety%20Analyses%20in%20the%20wake%20of%20Fukushima&f=false

And of course there are many problems related to safety/risk assessment related to harmonization of practices used etc.

Thank you, once more, for your effort and active attitude.

Best regards, Jirina VItazkova

Problem is that nuclear safety is not an ordinary business (not to say it is not even a business). And making great efforts on the big events such as accidents are probably good communication tools to show the public that big efforts are made to improve nuclear safety in order to make nuclear energy acceptable. Indeed it is good to always have a deep analysis of accidents and errors but I believe there are trends to forget the other main sources of danger and to spend most resources on these big accidents and not the day to day efforts to maintain nuclear safety. Even worse to 'liberalize' the nuclear safety activity as 'profit based' market making the regulatory bodies and/or technical advisors competing.

If Three Miles Island was a real sign that the safety analyses of nuclear power plants had real flaws and did not consider important sources of possible threats (small break loss of coolant accidents, indications of measurement devices, preparation of operators, ...), Tchernobyl and Fukushima stressed the divergence between what is recommended (and imposed) and what is really done either in a control room or during the design, construction and maintenance of the nuclear power plant.

The basic philosophy of nuclear safety was defined in 1947 by The Reactor Safeguard Committee of the Atomic Energy Commission in the USA for a reactor called SIR, when the Committee considered the first proposal for a contained reactor. At the same time, the role of geographic isolation in protecting the safety of the general public was emphasized.

In 1950 in WASH-3 the maximum accident for a reactor was defined as a break of cooling system of a reactor (LLOCA).

In 1957 the IAEA was established and 10 years later, in 1967, the document “General Design Criteria” was published, where the principles of defence in depth and deterministic approach to nuclear safety were defined. The US AEC developed and proposed Appendix A to 10 CFR (Code of Federal Regulations) Part 50 covering nuclear power plant design criteria.

The major weaknesses like containment fragilities and leak tightness, containment bypasses, interfacing systems LOCA and human interactions have been identified either by deterministic consideration or by the use of very crude PSA techniques early in the process of performing risk/safety analyses and despite this the basic designs were never modified either by fail-safe deterministic preventive or mitigative features such as very large containments with respect to possible accident loads, measures to eliminate the possibility of explosions from different sources or deliberate accidental airplane crash consequences. All design improvements regarding the issues listed above have been palliative, because they cannot completely eliminate the threats and loads.

WASH-740, "Theoretical Possibilities and Consequences of Major Accidents in Large Nuclear Power Plants" (also known as "The Brookhaven Report") estimated maximum possible damage from a meltdown at a “large” nuclear reactor (then “large” was a plant operating at 500 MWTh). The report was published by the U.S. Atomic Energy Commission (USAEC) in 1957. The conclusions of this study estimated the possible effects of a "maximum credible accident" for nuclear reactors then envisioned as being 3,400 deaths, 43,000 injuries and property damage of $7 billion. The only “possible” accidents that were contemplated were an ATWS and a LOCA. The estimate of probability was one in a hundred thousand to one in a billion per reactor-year.

When WASH-740 was revised in 1964-65 to account for the larger reactors then being designed (specifically, the Zion plant), the new figures indicated that there could be as many as 200,000 deaths, 400,000 injuries, and $17 billion in property damage, the consequences extending from Chicago into Ontario (Canada).

WASH-740 and its update were performed without the use of computers, PSA models and sophisticated theories, without 50 years of experiments, without previous evidence from actual severe accidents, with very crude assumptions, without 50 years of very expensive investments into the nuclear safety field, but the bottom line conclusions on potential weaknesses and challenges have not changed compared with contemporary analyses.

WASH-1400 known as “Reactor Safety Study” was a report issued in 1975 for US NRC. In the report the course of events that might arise during severe accident was analyzed (accident progression, Level 2) and also the potential radiological consequences of these events and their probabilities were discussed (Level 3). In the report the technique of Probabilistic Risk Assessment was developed using a fault tree/event tree approach. The report concluded that the risks to individuals posed by nuclear power stations were acceptably small compared with other tolerable risks. The report also concluded, that using the methods and resources and knowledge available at the time the probability of a complete core meltdown is about 1 in 20,000 per reactor per year.

The WASH-1400 PRA methodology was generally followed as part of the safety-assessment of all modern plants. Nevertheless, the study was later reviewed, the methodology used there was broadly endorsed as the best available, but it was cautioned that risk figures were subject to large uncertainty. Scientists later raised the issue that WASH-1400 estimates of fatalities due to severe accidents were based only on deaths during the first 24 hours after an accident, although radioactive Cesium released in an accident would remain active for decades, and could expose large populations to adverse effects. It was then fully accepted by the NRC that the uncertainties in WASH-1400's estimates of the probabilities of severe accidents were, in general, greatly understated and should be carefully considered in future PRAs/PSAs (PRA and PSA are interchangeable expressions for the same process of safety analyses)..

Later, the WASH-1400 report was replaced first by NUREG-1150 in 1991 following a period of intensive research and inspired in part by the Three Mile Island accident in 1979. Here severe accidents of five NPPs were analyzed including consequences/releases to the environment:

PWR Surry (TMI2), BWR-4 Peach Bottom, PWR Sequoyah, BWR-6 Grand Gulf and PWR Zion. Comparing these results with assessments of other plants based on measurements or PSAs, or accident progression, they correspond well. If they are compared to real longterm consequences, they correspond well to them too (see e.g. MINISTRY OF UKRAINE OF EMERGENCIES, ALL-UKRAINIAN SCIENTIFIC RESEARCH INSTITUTE FOR CIVIL DEFENCE OF POPULATION AND TERRITORIES FROM TECHNOGENIC AND NATURAL EMERGENCES: Twenty-five Years after Chornobyl Accident: Safety for the Future, National Report of Ukraine, Kyiv, 2011, UDK 621.311.25:621.039.586/(477.41-21), BBK 31.47 (4Ukr-4kij)-08 D22).

All the mentioned analyses identified already problems which may occur and also we know what to expect according to analyses and real accidents.

The MAJOR problem is, that the plants were NOT designed to withstand the conditions of severe accidents and NO partial improvements can improve this substantially. For instance currently popular venting systems which are installed to NPPs can maybe save containment but then they put up side down the philosophy of containment itself as the last physical barrier of defence-in-depth, since we release the radioactivity to the environment anyway, even though we call it controlled venting. The other problems are for instance single failures which are identified within PSAs violating major IAEA safety principles (see IAEA Safety Standards for protecting people and the environment, Fundamental Safety Principles, Safety Fundamentals No. SF-1, IAEA, 2006, ISBN 92-0-110706-4, ISSN 1020-525X, Vienna), and guaranteed failures occuring before core damage (i.e. not covered in PSAL1) - e.g during extreme seismic events, again violating basic IAEA safety principles....

Apparently, the originally considered probabilities of severe events are much higher in the reality. For me, the problems should be solved REALLY technical way, not ONLY political way or administrative way, as it is currently performed ....

Dear Jirina.!

In 80th years of the last century I took part in export of spent nuclear fuel on A1 Bohunice NPP to "Majak". In my opinion, failure 1977 has been provoked by nonprofessionalism engineering and technicians of the NPP. At failure local radiating pollution took place only.

Further, infringements of production schedules of storage of the spent fuel in pools of endurance has led to dissolution of Mg environments of nuclear fuel elements in water storehouses and polymerization of diphenil oxide in storehouses with the organic coolantr. Thus local pollution took place.

Failure on A1 Bohunice NPP hardly should be carried to « major nuclear accidents ».

Dear Sergey, I dare not to agree with you, as far as you opinion about  Bohunice A1 as a tpye of accident. TMI2 neither was connected with large consequences and reported is ONLY local contamination, but the MAJOR was because of the extent of the core damage (about 50% reported) and plant degradation. From this point of view the Bohunice A1 accident is comparable with TMI2 with the difference, that the core damage extent was reported about 25% - the plant degradation, and the power of the unit was smaller than TMI2. Until now the decommissioning has been performed, and for further decades not finished.