Annex E
A SUMMARY OF KEY POINTS TO EMERGE FROM A
REVIEW OF THE POSIVA SAFETY CASE ON BEHALF OF THE FINNISH RADIATION
AND NUCLEAR SAFETY AUTHORITY (STUK).
The most recent disposal safety case published
by Posiva—the Finnish disposal agency is:
"Expected Evolution of a Spent Nuclear Fuel
Repository at Olkiluoto" Posiva Oy—December 2006 (Revised
October 2007)
In April 2008, a review of this was produced
for "STUK", the Finnish regulatory agency:
"Review of Posiva 2006-05: Expected Evolution
of a Spent Nuclear Fuel Repository at Olkiloto" Michael Apted
et al (April 2008)
Key points from this review include:
RELEVANCE OF
DATA TO
SAFETY—NOT
CLEAR
safety importance of processes and data needs
to be set out (p1) and the definitions of the safety functions
of the different parts is vague (p7). The report does not set
out clearly which outcomes would lead to unacceptable safety hazards.
(p8)
"analyses of the safety importance for
many of the evolutionary processes and associated data are absent"
(p9)
"it remains unclear whether Posiva really
understands and can prioritise the safety-importance implications
of acknowledged uncertainties in the normal evolution processes."
(p5) (see also p10)
POSIVA HAVE
NOT DEMONSTRATED
"SAFE DISPOSAL"
although the Posiva proposal is expected to
lead to safe disposal—"this remains to be demonstrated
by Posiva" (p10)
"A main concern with the report is that
it represents a mostly qualitative analysis" (p9)
"there does not seem to be coordinated
efforts to explore if there are as-yet unrecognised processes"
(p5)
LOW LEVEL
OF CONFIDENCE
IN FLOW
DATA
"in the case of flow and salinity modelling,
the uncertainties are so substantial that the level of confidence
in the results seems low" (p5)
(NB—given the role of flow in the carriage
of radionuclides back to the surface this lack of confidence seems
particularly significant)
BASIS OF
SOME DATA
INADEQUATE
some of the data used is out of date (p2) and
some underlying reports have not been published (p7)—in
addition the justification of some conclusions is not cited "so
the conclusions are no more than working hypotheses"
(p7)
ASSIGNING VALUES
STUK referred to the "seeming arbitrariness"
(p8) of assigning values to missing information
(NB—this should be compared to the Nirex
(Nov '05) Viability report—in which parameters were chosen
in order that the EA target would be met.)
URANIUM
SKB have reported on the extreme safety importance
of assigning a low value to the rate that Uranium Dioxide dissolves—however
Posiva have only considered the chemistry of this reaction in
a qualitative manner (p2) (see also p9)
(NB. Page 48 of the Posiva Environmental
Impact Assessment—for the Expansion of the Repository for
Spent Nuclear Fuel 2008—states that uranium dissolves very
slowly in water in the conditions that would be expected. This
should be compared to the variability of measured Uranium concentrations
of 100 million)
BUFFER EROSION
the extreme safety implications of buffer erosion
(during the glacial phase) should not be understated (p2) furthermore
Posiva do not consider buffer erosion at all in the geochemical
part of the report (p9)
EXCAVATION DAMAGE
ZONE (EDZ)
"the issue of the EDZ (excavated
damage zone) appears to be underplayed" (p5)
(NB—the EDZ issue was particularly significant
in the 1990s Inquiry)
ALKALINE WATER
"The issue of hyperalkaline waters and
their geochemical interaction with bentonite and rock remains
open" (p6)
(NB—this is a particularly important issue
for the UK—as the Government plan that HLW should be "co-disposed"
with ILW. As ILW disposal is planned to be largely cement based
& cement is expected to give rise to a significantly alkaline
environment—then it can be seen the implications for HLW
disposal must be considered. The significance of this issue is
also outlined in the Oct '09 EU JRC report.)
Further points to be considered:
— system developments over time that do
not meet "normal evolution" (p1) (see also p10);
— design and emplacement malfunctions (p1);
— full implications of changes in the "temperature/water
system/mechanical/chemical" conditions (p1)—this area
of work should be "substantially improved" (p7);
— most recent information on the glacial
phase (which is presently inadequately considered) (p2) In fact
Posiva do not consider the evolution of subsequent glacial reports
anywhere in the report (p9);
— Posiva need to consider the issue of
natural resource exploration and exploitation. (p9);
— in numerous places Posiva refer to the
fact that decisions on materials and design parameters are not
yet fixed (p8); and
— possible interactions between bentonite
and iron or copper are not addressed" (p6).
SCIENTIFIC METHOD
NB—in analysing the Posiva safety case—STUK's
conclusions do not follow "scientific method".
Thus, STUK conclude:
"By evaluating the safety consequences
of such 'unexpected' conditions, some of these may be recognized
as inconsequential and others may require further study in order
to confidently establish that the overall safety requirements
are met" (p3)
It is well recognized by the Environment Agency
(in the UK) that "further study" into areas of uncertainty
may not result in the "confident establishment" that
a safety case may be met.
In fact it may result in the opposite conclusion—ie
that a safety case may not be met.
| 
