free web site hit counter
Search CNIC
Kashiwazaki-Kariwa Earthquake
Japan's Nuclear Safety Shaken to the Roots

CNIC Report (23 July 2007)

At 10:13 am on July 16 a magnitude 6.8 earthquake struck just off the coast of Niigata Prefecture on the Japan Sea side of Honshu, Japan's largest island. As a result of the quake, four reactors (units 2, 3, 4 & 7) at Tokyo Electric Power Company's (TEPCO) Kashiwazaki-Kariwa nuclear power plant shut down automatically. At the time, unit 2 was being started up after a periodic inspection, while the other three units (1, 5 & 6) were still shut down for periodic inspection.

We express our condolences to the families and friends of those who lost their lives as a result of the earthquake. So far eleven people are reported to have died. We are also concerned for the well-being of those who survived, some of whom are members of CNIC. Their suffering is likely to continue for some time. Perhaps the only fortunate thing amidst this misery is that the nuclear reactors shut down. If they had not done so, there could have been a nuclear disaster in which a huge amount of radioactive material was released into the environment.

Fire in a transformer

Photos of burnt out transformer (top) and deformed road nearby (bottom) by Kazuyuki Takemoto (see Who's Who 111). Kazuyuki is a member of CNIC's board of directors. His car port was flattened by the earthquake.

As a result of the quake, a fire occurred in a transformer outside a building in unit 3. The transformer was part of the external power supply system. It appears that a short in an electric circuit gave rise to a spark, which set fire to insulating oil. The ground beneath the transformer was raised by 30 cm, while the ground next to the transformer, where the electric cables ran, subsided by 15-25 cm. TEPCO believes that the relative movement caused a short circuit and that this gave rise to the fire.

The fact that it took two hours to put out the fire exposed TEPCO's lack of preparedness. However, it turns out that TEPCO is not the only power company that is ill prepared to respond to such situations. Responses by power companies to a survey by the Mainichi Shimbun revealed that although power companies prepare for fires, no special consideration has been given to the particular problems of fires which occur as a result of earthquakes (Mainichi Shimbun, 20 July 2007). This is incredible, given that throughout Japanese history, whenever there was an earthquake, fires caused at least as much damage as the earthquakes themselves.

Despite the potential seriousness of this fire, TEPCO failed to announce whether there was a continual external supply of power and whether the emergency generator started up. This is very important information, because if the external power supply and diesel generators had failed, it would have been impossible to maintain a continual flow of coolant to the reactor. Even after automatic shutdown, the fuel in the reactor core is still extremely hot. Failure to maintain a continual supply of coolant could result in a core meltdown, leading to the release of highly radioactive material into the environment. There would then be a double disaster: a nuclear disaster on top of an earthquake disaster.

Radioactivity released to the environment
Water containing radioactive material was released from unit 6 into the sea. TEPCO originally said the quantity of radioactivity involved was 60,000 becquerels, but it later increased its estimate to 90,000 becquerels. One would not expect this amount to have any impact on the environment or on human health, but TEPCO took far too long to report the release. The following account is based on a report published on the asahi.com web site on July 21.

At around 12:50 pm, a pool of water was found inside the reactor building, but outside the controlled area. Within about an hour and a half, workers had analyzed samples and found that they contained radioactive material. However, because it was outside the controlled area they doubted their results. They collected new samples and repeated the analysis twice. It wasn't until about 20:30 pm that TEPCO finally reported the release to the authorities. It seems that TEPCO employees are still under the spell of the safety myth. They are still more inclined to believe the theory rather than the evidence before their eyes.

TEPCO also took a long time to work out how the leak occurred. Apparently the spent fuel pool overflowed and the liquid leaked out to the uncontrolled area through electric power cable tubing. Spent fuel pools frequently overflow during earthquakes, so the fact that radioactive material was released to the sea shows that TEPCO failed to adequately address this problem.

On July 17, TEPCO announced that iodine, chromium (Cr-51) and cobalt (Co-60) had been released to the atmosphere from the main exhaust stack of unit 7. At the time, TEPCO indicated that the release had stopped, but it subsequently corrected this, saying that the iodine (I-131, I-133) release continued until July 18. TEPCO said that due to a failure to manually shut down the turbine grand steam exhaust fan after the reactor shut down, iodine and particulates that had accumulated in the condenser were sucked out and released through the exhaust stack. That is plausible, given that in boiling water reactors (unlike pressurized water reactors) the same water that cools the reactor is converted into steam to drive the turbines and then condensed in the condenser. It is, therefore, to be expected that some radioactive material will find its way from the reactor to the condenser. Nevertheless, release of these radionuclides automatically raises suspicions of damaged fuel assemblies. However, TEPCO stated that, given the low level of iodine in reactor water samples and off-gas monitor readings, there is no indication that fuel assemblies were damaged. TEPCO estimated that the total amount of radioactivity released from the main exhaust stack of unit 7 was 4 x 108 becquerels (radiation dose of 2 x 10-7 milli-sieverts from iodine and 7 x 10-10 milli-sieverts from particulate matter).

The earthquake knocked over several hundred drum cans in the storage building for low-level solid waste. Of these, around 40 were found without their lids on. About 16 liters of liquid leaked, but TEPCO says that no radioactivity was found and none was released into the environment.

63 problems identified
As of July 21, TEPCO had identified a total of 63 problems. Fifteen of these were related to radioactivity. The spent fuel pools of all 7 units overflowed, although only unit 6 involved a release of radioactivity to the environment, and ducts in the exhaust stack of all reactors were displaced. There was also some radioactivity found in a massive pool of water from damaged pipes in the fire extinguishing system of unit 1 (40 cm deep (1,670 m3), fifth floor basement, auxiliary building).

Besides these, the following are just a few of the problems that did not involve radioactivity. The water level of the spent fuel pools of units 1, 2 and 3 dropped, then recovered mysteriously. Work platforms in the spent fuel pools of units 4 and 7 fell onto the spent fuel storage racks. There were oil leaks from several transformers. Some pumps are out of order. Blow-out panels in turbine and reactor buildings came off. Regular power supply to the administration building was lost and the emergency power supply had to be activated.

However, beyond these readily identifiable problems, the question of how much the buildings, pipes and equipment were weakened by the earthquake remains unanswered. Detailed checks are required, but since the impact of the earthquake would not have been uniform, it will not be possible to provide complete assurance that the plant is capable of withstanding the next earthquake.

Earthquake Resistance Design
Under Japan's old earthquake resistance guidelines, the design basis for nuclear power plants (NPP) assumed a "maximum design earthquake" (S1) and an "extreme design earthquake" (S2), where S2 was greater than S1 (see NIT 112, 103). The "extreme design earthquake" was thought to be impossible in reality, but it was taken into account just to be on the safe side. However, the earthquake that hit on July 16 exceeded the S2 design basis earthquake. Indeed, based on the information released by TEPCO, for unit 1 the peak ground acceleration at the plant was 2.5 times greater than assumed for the S2. The peak ground acceleration in the east-west direction was recorded at 680 gal, compared to the design basis of 273 gal (see tables). (Only peak ground acceleration data is available, because the earthquake exceeded the monitoring system's recording capacity.)

Peak Ground Acceleration 16 July 2007 (Gal)
Location
North-South
East-West
Vertical
Unit 1
311
680
408
Unit 2
304
606
282
Unit 3
308
384
311
Unit 4
310
492
337
Unit 5
277
442
205
Unit 6
271
322
488
Unit 7
267
356
355
Peak Ground Acceleration S2 (Gal)
Location
North-South
East-West
Vertical
(static design)
Unit 1
274
273
(235)
Unit 2
167
167
(235)
Unit 3
192
193
(235)
Unit 4
193
194
(235)
Unit 5
249
254
(235)
Unit 6
263
263
(235)
Unit 7
263
263
(235)

It is believed that this earthquake was caused by movement of an approximately 30 km long and 25 km deep fault. This fault was not taken into account during surveys carried out for the design of the Kashiwazaki-Kariwa nuclear power plant. Instead, the S2 design basis earthquake that was chosen was a magnitude 6.9 quake at an active fault 20 kilometers away. This was the fault which caused an earthquake in the same region in 2004. On that occasion, reactor number 5 shut down automatically, but peak ground acceleration did not exceed predictions.

From analysis of the distribution of the after shocks from the 16 July 2007 quake, it is now believed that there is an active fault extending directly under the Kashiwazaki-Kariwa NPP. Since Japan's earthquake resistance guidelines do not permit NPPs to be built directly above active faults, Kashiwazaki-Kariwa would not have been chosen to host a NPP had this been known at the time. However, it now appears that this should have been recognized. After the recent earthquake, Professors Takashi Nakata (Hiroshima Institute of Technology) and Yasuhiro Suzuki (Nagoya University) analyzed the data in TEPCO's license application and concluded that it indicated a fault five times longer than one identified by TEPCO (Asahi Shimbun, 20 July 2007). Between 1979 and 1985, using sonic testing, TEPCO found 4 small faults off the coast of Kashiwazaki-Kariwa, but it concluded that they were either not active or not important. However, Nakata and Suzuki said that the data indicated that three of these small faults were connected and were in fact one long 36-kilometer fault and that the fault was probably active.

Professor Nakata is the geomorphologist who last year proved that there is a previously unidentified active fault near the Shimane nuclear power plant (NIT 114). Earthquake studies for Japan's nuclear power plants have largely ignored the work of geomorphologists. They are taken more seriously now under new earthquake resistance guidelines approved in September 2006, but their role is still vague. It seems that the nuclear industry does not like them, because they are too good at finding active faults.

Clearly Japan's earthquake safety standards are inadequate. This fact can no longer be disputed. Given that the size and location of this earthquake was not predicted by TEPCO's survey, nor in the government's screening process, it is essential that an independent geological survey be carried out of the surrounding area, both on land and at sea. However, it would appear from the evidence that has already come to light that Kashiwasaki-Kariwa is not a safe site for a nuclear power plant and that the plant should be shut down permanently.

In just two years, three earthquakes (off the coast of Miyagi Prefecture on 16 August 2005, off the Noto Peninsula in Ishikawa Prefecture on 25 March 2007, and now this one) have exceeded the S2 design basis earthquake assumed at the time the plants were built. Since the earthquake resistance guidelines were revised, Japan's nuclear power companies have been carrying out geological surveys and safety checks on the basis of the new guidelines. However, the July 16 earthquake demonstrates the inadequacy of these measures. For example, detailed surveys are only required for a 5 km radius around the plant and power companies have until 2012 to complete safety checks. By rights, all nuclear power plants should be shut down until these surveys and safety checks have been completed. The Nuclear Industrial and Safety Agency (NISA) should review its policy of allowing reactors to continue to operate while these surveys and checks are carried out. However, at a press conference on July 18, NISA's spokesperson, Akira Fukushima, repeatedly dodged questions directed at this problem. As usual, the government's priority is to keep nuclear power plants operating, regardless of safety concerns.

Lessons to be learnt
This was not the biggest earthquake that could possibly hit Japan. This one was just a warning. There were enough failures this time to enable us to imagine what might happen if a bigger earthquake struck. In particular, we saw how the confusion caused by the earthquake led to errors and lapses of judgment. We saw equipment failures which in themselves might be manageable, but which, when compounded with the many other failures that earthquakes inevitably cause, could have been catastrophic.

Unfortunately, there is no indication that the government will prioritize safety over the narrow economic interests of the power companies. The following quote starkly illustrates the obstacles to change:

The official said that during the discussions on new standards in 2006, it was clear that above a level of about 6.7, "there would be a lot of backfitting required" to keep reactors operating. Were Japan in the aftermath of this week's quake to make a quake of 7.0 the design basis event, he said, for some plants the amount of upgrading needed could be economically prohibitive. (Nucleonics Week, Volume 48, Number 29, July 19, 2007)

Clearly it is important to change the priorities of the central government. Mr. Amari Minister of Economy, Trade and Industry ordered that the reactors remain shut down until safety has been confirmed. One wonders what criteria he might use to determine when they are safe again, but it is hard to imagine that his criteria would be based on good sense and sound science. Possibly the best hope lies at the local level. The mayor of Kashiwazaki City invoked his right under fire safety provisions to order that the reactors remain shut down. The issue will then come down to the mood of the local population. The earthquake shook the people of Kashiwazaki-Kariwa in more ways than one. What will it take to convince them that it is safe to operate the reactors again?

By Hideyuki Ban (CNIC Co-Director) and Philip White (NIT Editor)

Return to NIT 119 contents



CNIC

CNIC
Citizens' Nuclear Information Center
Akebonobashi Co-op 2F-B, 8-5 Sumiyoshi-cho,
Shinjuku-ku, Tokyo, 162-0065, Japan
TEL.03-3357-3800
FAX.03-3357-3801
Map
http://cnic.jp/english/
Email