Based on the EXAR study published in 2021, which examines the flood risk for the Aare River, ENSI requested nuclear installation operators to update their flood risk analyses by the end of 2022. As the EXAR study contained indications that morphological processes might play a more significant role than previously assumed, the operators also had to carry out analyses on bank erosion. This was particularly necessary because the EXAR study only conducted a general evaluation of possible morphological effects, with the evaluation being purely indicative. Detailed soil tests to determine the resistance to possible morphological effects and the preparation of safety verifications were not part of the EXAR baseline study, these had to be carried out by the operators.

All operators submitted their analyses on time. At the Paul Scherrer Institute (PSI), improvement measures based on the morphological effects were derived from the project. Furthermore, detailed measurements of the soil conditions and bank linings, along with a comparison of the resistance values with the modelled loads have shown that no safety-relevant buildings are endangered by erosion at the operational nuclear power plants (NPP) Beznau, Gösgen and Leibstadt, as well as the central interim storage facility ZZL and the decommissioned Mühleberg NPP.

The verifications of the individual nuclear installations and their assessments by ENSI are summarised for each installation below.

• Beznau nuclear power plant

  The EXAR study indicates a flow depth of about 0.4 m over the Beznau NPP site for the 10,000-year flood scenario. For a 1,000-year flood event, the Beznau NPP site remains dry in the area of the safety-relevant buildings.

  For the 10,000-year flood scenario, the turbine hall and the auxiliary buildings are flooded as per design. This event can however be controlled by the safety systems which have an original flood design height of 1.65 m. The original design height has been significantly increased in recent years as a result of various improvement measures.

  Where the bank linings are concerned, it should be noted that weaknesses were identified at individual points in the condition analyses. Beznau NPP assumed in its modelling that the corresponding points are erodible and has shown that no safety-relevant buildings are damaged by erosion. Nevertheless, ENSI requested that Beznau NPP check in detail what maintenance measures would be possible for the corresponding points of the bank lining.

  If the Aare water inlets are assumed to be clogged, which Beznau NPP has not ruled out, the steam generator feed is provided by safety systems that are independent of the Aare River water and the Aare cooling. Overall ENSI was satisfied that Beznau NPP meets the technical and radiological safety verification criteria for the 1,000 year and 10,000-year flood risk scenarios.

  In addition to updating the deterministic flood safety verification, Beznau NPP has updated its probabilistic flood analyses for power operation. In this context, cross-design scenarios were analysed with respect to water levels as well as morphological effects. For scenarios involving complete failure of all Beznau weir openings, Beznau NPP conservatively assumed that the ‘north and south diesel buildings’ would fail due to erosion. These buildings are buildings that were installed as part of the backfit project for provision of an autonomous emergency power supply. As the emergency systems remain available in all scenarios and the scenario in which all weir openings fail is very unlikely, flood events only have a small impact on the frequency of damage to the reactor core – according to the present analysis, they contribute less than 1 % to the total core damage frequency at Beznau NPP.

• Gösgen nuclear power plant

  For the Gösgen nuclear power plant, the EXAR study indicates a flow depth of around 0.65 m for the 10,000-year flood scenario, while it is no more than a few centimetres for the 1,000-year flood scenario. A major factor contributing to these water levels is that the pedestrian bridge, which crosses over the Aare in Däniken, to the north of Gösgen NPP, can become clogged with wood. The simulation performed by Gösgen NPP results in a somewhat greater flow depth of up to 0.85 m for the 10,000-year flood scenario and 0.45 m for the 1,000-year flood scenario due to the use of a more detailed model.

  Where bank erosion is concerned, it should be noted that Gösgen NPP has also analysed the scenarios from the EXAR study that go beyond the plant design and has concluded that no safety-relevant buildings could be damaged by erosion.

  Where control of flood scenarios is concerned, it has been demonstrated for shutdown path 1 (shutdown using safety systems cooled using the Aare) that it can be protected with newly installed dam beam systems up to a flow depth of about 1.5 m. However, as a blockage of the two Aare water reservoirs was not deterministically excluded, the safety verification is based on shutdown path 2 (shutdown using the groundwater-supplied emergency system). Thanks to the leak-tightness of the emergency building, no measures to counter water ingress are required even at significantly higher flow depths.

  ENSI was satisfied that Gösgen NPP meets the technical and radiological safety verification criteria for the 1,000 year and 10,000-year flood risk scenarios.

  In addition to updating the deterministic flood safety verifications, Gösgen NPP has also updated the probabilistic safety analyses (PSA) for power operation and as part of this has analysed the beyond design basis scenarios. In line with previous results from the flood PSA, floods contribute only slightly to core damage (according to the present analysis about 1.2 %). Overall, ENSI considers the update as generally plausible, nevertheless, it considers some of the assumptions made to be optimistic. With this in mind, ENSI has stipulated a few requirements. At the same time it should be noted that the plant improvements made by Gösgen NPP have not yet been incorporated into their PSA model. Overall, ENSI presumes that even after taking into account the need for improvement, floods only make a minor contribution to the core damage frequency.

• Mühleberg nuclear power plant

  In contrast to the other nuclear power plants, BKW, the company operating the Mühleberg nuclear power plant, had to update the analyses for the decommissioning phase 1 operating state. During this phase (now completed) the reactor was shut down, all fuel elements were in the spent fuel pool and the decay heat had to be removed using the systems provided for this.

  The safety verification documentation includes information on bank erosion, flooding of structures, systems and components, building resistance to floating and the deterministic and probabilistic safety verifications. BKW has been able to show that the impact on the plant remains so limited that the technical and radiological safety verification criteria for the 1,000-year and 10,000-year flood scenarios are satisfied at all times. Fuel damage due to a flood event can be excluded.

  The plant is now in decommissioning phase 2, meaning that the plant no longer contains any fuel.

• Leibstadt nuclear power plant

  In contrast to the other power plants, the EXAR study did not derive any new peak outlet flows or water levels because the Rhine River was not subject of the study.

  As the safety-relevant plant elements of the Leibstadt NPP are about 20 m above the normal water level of the Rhine and are therefore flood-proof, there was no need to determine any precise water level. Nevertheless, Leibstadt NPP had to demonstrate the stability of the slopes.

  Starting from conservative assumptions with respect to the water level and the soil characteristics – already known from the seismic safety verifications – the slope was demonstrated to be stable. Thus, the safety-relevant buildings of the Leibstadt NPP are neither endangered by the water level nor by morphological effects. The effects of the flood that are assumed for the safety verification are therefore limited to assumptions that the water reservoirs of the secondary cooling water system become clogged.

  In this case, the plant can, without any change from the existing safety verification, be transferred to the cold shut-down state and then be kept in this state using the existing safety systems. Consequently, the Leibstadt NPP complies with the technical and radiological criteria for the investigated flood scenarios of accident categories 2 and 3.

  Leibstadt NPP need not update its PSA within the scope of this order because there are no changes with respect to the hazard assumptions. Nevertheless, where flood risk is concerned, previous assessments suggest that it will only make a minor contribution to the core damage frequency.

• Paul Scherrer Institute

  The EXAR study for the Paul Scherrer Institute shows that the site will remain dry for both the 10,000-year and the 1,000-year flood event.

  No negative effects of the flood were identified for the 1,000-year scenario. For the 10,000-year scenario, ENSI identified a deficiency in the safety verification concerning the buoyancy safety of the buildings. In its statement, ENSI requested PSI to complement the safety verification in this respect.

  In addition, in the 10,000-year flood scenario, PSI has identified two buildings for which conservative modelling shows that they may be at risk due to bank erosion with an assumed outflow of approximately 4,399 m³/s and with assumed jamming of the PSI bridge with debris across its whole width. The buildings concerned are OPRA and OALA.

  As a consequence of the analyses, PSI has drawn up a procedure for the OPRA building according to which the radioactive material must be transferred to an unaffected building in the event of an imminent extreme flood. Based on an inspection, ENSI is satisfied with the feasibility of such a transfer should it become necessary.

  For the OALA building, PSI plans structural bank protection measures to upgrade flood resistance. The safety of this building against the design basis extreme flood scenario can be confirmed once these measures are implemented.

• Central Interim Storage Facility Zwilag

  For the Central Interim Storage Facility, the EXAR study shows that the site will remain dry for both the 10,000-year and the 1,000-year flood event. No negative effects of the flood were identified for the 1,000-year scenario. For the 10,000-year scenario, ENSI identified a deficiency in the safety verification concerning the buoyancy safety of the buildings. ENSI has requested ZZL to complement the safety verification in this respect.

  Where morphological effects are concerned, ZZL has shown that slope stability would not be at risk due to flood events and that no Central Interim Storage Facility buildings would be damaged by bank erosion.

More information

• New study on extreme flooding of the River Aare is also relevant for nuclear power plant sites » ENSI
• Das ENSI verlangt neue Sicherheitsnachweise zu Hochwasser » ENSI (German only)
• Hazard information for extreme flood events on the rivers Aare and Rhine (EXAR)