M. Jensen
Swedish Radiation Protection Institute
V
+46 87297239 F +46 8 7297162
E-mail: mikael.jensen@ssi.se
ABSTRACT
Spent nuclear fuel from nuclear power generation will remain radioactive for millennia even though 99% of the radioactivity will have decayed within the first millennium. Certain information about the waste may be of value to keep for long time periods for future generations which may - intentionally or inadvertently - come into contact with the radioactive waste.
The function of a nuclear waste archive must be related to its purpose, that is the safety issues of the repository. There are several ways to interpret the notion of safety, and the unusual long time periods involved, require a thorough analysis of what the archive should achieve.
We find that analogies exist, but they are not immediately relevant as input to a discussion of an analysis which must embrace both the archive safety and the timely retrieval and use of the information to achieve the ultimate goal - safety. Work to describe this whole chain has barely begun.
Present day waste management should identify records to be part of a future post-closure archive for radioactive waste repositories.
INTRODUCTION
Data stored for retrieval and use after a period longer than 100 years is clearly something unusual. Many archives, and libraries, contain material older than 100 years, but the reason for the material being in the archive is usually a general use, for researchers to benefit from studying the material etc.
For some records, there may have been an explicit purpose of later retrieval at the time the records entered the archive, but after several hundred years, the reason for the custodians duty often have changed, such as with ancient land ownership records.
In this perspective, archiving nuclear waste disposal records for the purpose of use throughout the lifetime of the waste, is a unique task. There are several reasons to reflect on very long term record storage. The issues involved are both mind-bending, philosophical and entertaining, especially when looked at from a distance. However, it is the belief of the author that at the time of closing a repository, these questions will boil down to controversial, legal and very practical problems. They deserve to be looked at seriously, and as early as possible.
SAFETY IN THE DISTANT FUTURE
The need for operational safety is an obvious concept for all nuclear installations. We also want repositories to be safe in the future, but there are some philosophical issues which may have to be addressed before we can tailor-make the need for the long term records needed for the repository safety. We need to examine the concept of safety itself. What does it mean for a repository to be safe?
This question is related to some important issues.
Safe or Kept Safe?
Do we assume that the repository perhaps would need to be improved at some time in the future? It is possible that the repository performs the way it was designed but that the standard for safety changes with time.
Do we assume that it could fail to provide continued safety and we would need to take remedial actions of some kind? If the performance of a car is monitored and repairs are carried out when needed, the owner may still claim to have a safe car.
Safeguards
Safeguards is used to prevent - geographically - the spread of fissile material. Do we also want to prevent governments which are distant, not only in space but in time, from using fissile material? If inaccessibility of plutonium is important, is this aspect of safety then more important than preventing humans from receiving small doses from an outflow from the repository?
Safety Done and Undone
Even if a moment is reached in a licensing process where a repository is considered safe or safe enough, a number of human actions, and in particular the human intrusion scenario, will always represent a possible loss of safety, following the principle that what humans can do, they can also undo.
It is obvious that the various nuances of the safety concept will correspond to somewhat different set of record collections, the priorities being different. To prevent intrusion, information about the repository's location is an important parameter. The alternative where safeguards is the priority could lead to a very different repository design in the first place. Perhaps a very deep borehole would be the most favorable option, and it could be argued that no information should be available at all.
Of all hypothetical future groups, concerned by the repository, perhaps the group of people involved in remedial action would need to know most. Both the location the content and detailed design would be of interest.
HOW FAR AWAY IS THE DISTANT FUTURE?
We have a different view of the time periods involved, long (100 years), very long (1 000 years), and the astronomically long time periods for the decay of the most long-lived nuclei in spent fuel.
We may feel that the time period 100 years can somehow be overviewed, in the sense that we feel that we have some experiences, that some relevant data are available for this time period, and that at least some transactions are taking place in society in the scale. Hong Kong was returned after 100 years and some long term bank transactions cover time periods at least in the same order of magnitude.
The astronomically long periods cannot be discussed with much realism. Still, it may be an alternative to treat all periods as the first 100 - 1 000 years. There are several reasons for this, and such reasons were given to justify taking a (relative) short term view, for a repository of spent fuel, in a study by experts from the Nordic Countries (1).
The Radionuclides' Decay
The radioactivity of spent fuel decreases sharply during the first 1000 years. The danger from the dominant nuclide, cesium-137, is obvious to potential intruders during the first part of this period. After 1000 years, the radioactivity of cesium-137 in the repository is 10 orders of magnitude lower than at the time of repository closure, and the total radioactivity in spent fuel is 2 orders of magnitude lower (taking Sweden as a reference regarding fuel composition and cooling off periods).
Philosophical Reasons
We are ignorant of both capabilities and needs of a society 1000 years (and more) into the future. For that reason alone, we are forced to design the information system to serve the interests of the societies in the nearer future.
Most predictive mechanisms decrease in accuracy with time. They are more accurate in their projections for a near future than for a more distant one. It is meaningless to speculate about people a million years from now, but it is good sense to design the archive system for the near future. Furthermore, some possible interactions or links can be described between near and distant future societies:
AN ARCHIVE SAFETY ASSESSMENT
An attempt has been made to identify important threats and possible countermeasures to a long term archive, within the Nordic Study (1), in the form of a two-day seminar in Oslo, Norway (2).
Some qualitative comments from the seminar are worth recalling:
EXISTING DATA
It is a natural process in the design of a nuclear waste archive concept, to first discuss the concept of safety, related to the purpose of the archive, and thereafter to proceed with investigation of the design of a successful strategy. The process itself is very much the same as for the classical performance assessment of a repository.
The next step is to collect data in a systematic way, to be used as building blocks in the assessment. Unfortunately, very little data exists about information survival. Some attempts were made in the Nordic Study in the form of a study of the Vatican Archive, and of the German Archives in the 20th Century (3, 4).
It is shown that some of the Vatican Archive's major losses were connected with transports to and from Rome. The papacy was transferred to Avignon in 1308. Many documents were left behind and some of them destroyed during their transfer to France. Five centuries later, 1810, Napoleon moved the archives, in 3239 chests, to Paris. One third of the archive was lost before they were brought back. One conclusion from the German Archive Study is that, in some cases, surprisingly small losses was suffered by the German Archives by the Allied's bombing. In both studies, it was found that some of the largest losses were suffered due to theft of paper as a valuable commodity. No definite conclusion are drawn by the Nordic Study regarding an archive safety strategy. The goal was primarily to open a field of research by offering stimulating information, rather than to reach definite conclusions.
WORK WITHIN IAEA
Within the International Atomic Energy Agency, 2 documents are now being drafted, both regarding maintenance of records for nuclear waste repositories. One document focus on near surface repositories, and the other on geological disposal sites ( 5, 6).
In addition to offering quality assurance guidelines for records management systems, both drafts point to the need to early identify the content of a post-closure records assembly.
In Ref. 5, the idea is promoted, that a national system of records are assembled and, in addition, a shorter version compiled for an international archive.
At the outset, the implementor, as well as the company operating the repository and the competent authorities, is in possession of a vast amount of information about the nuclear material and its history. This is referred to by IAEA, as well as by the Nordic study, as the primary information set.
Certain essential information could be extracted from this primary information in order to establish independent archives, called a second level information set. Such a process requires that relevant records for future use is identified, and that no later than at closure of the repository, such an archive is set up, to be guarded by the authorities. Perhaps the National Archives are the most obvious choices.
The most compact level of information would be well suited to be placed in an International Archive. Such highly condensed information is referred to as a third level information set.
AN IMMEDIATE TASK
What can and should be done now, is for all those bodies who in their records management systems have records, relevant to a post-closure archive, to identify those records.
For the implementor just to place information in an archive, is like placing a book in a library, with no special record of this entry and the right place for the book. A misplaced book in a library is a lost book. For these processes to be functional, there must exist national laws and international conventions which define and regulate the processes. Furthermore, the information placed in the archives and about those archives should be an integral part of the safety assessment.
Most of the post-closure records are parts of the operational records. A continued efficient management of current records is therefore a natural link to the construction of a well organized post-closure archive.
REFERENCES