UNTESTED CHINESE NUCLEAR TECHNOLOGY COULD IMPERIL THE CITY OF 20 MILLION.
A debate has started about the wisdom of building two large Chinese-supplied nuclear reactors in Karachi. The fundamental concern is that the nearly 20 million people living in Karachi—about one out of every 10 Pakistanis—could be at risk from these reactors.
The two reactors, worth $4.8 billion apiece, are to be supplied on a turnkey basis by the Chinese National Nuclear Corp. A soft Chinese loan of $6.5 billion apparently proved irresistible to the cash-strapped Pakistani government. This brand of reactor, known as the ACP-1000, has not yet been built or tested anywhere. The Pakistan Atomic Energy Commission (PAEC), which will operate these reactors, insists the reactors will be safe.
But unlike Pakistan’s officialdom, which is determined to rapidly expand nuclear power generation, some Chinese industry insiders are fearful of the nuclear rush in their own country. With billions of dollars at stake, they suspect that money, schedules, and outsourcing to unqualified subcontractors may become more important than nuclear safety. The former vice-president of the China National Nuclear Corp. recently stated that, “Our state leaders have put a high priority on [nuclear safety] but companies executing projects do not seem to have the same level of understanding.” It is one of these Chinese companies that has designed and will build the Karachi reactors.
Nuclear supporters in Pakistan point to the International Atomic Energy Agency (IAEA) and highlight that it visits nuclear power plants and makes sure they are safe. But the fact is that after the Fukushima accident, while addressing his board of governors, and the world, on March 21, 2011, the director-general of the IAEA stated categorically that, “We are not a nuclear safety watchdog … responsibility for nuclear safety lies with our member states.”
Risky Business
Advocates for the Karachi reactors claim that the so-called K-2 and K-3 plants (each with a generation capacity of 1,100 megawatts) are based on a Chinese adaptation of a long-established reactor type but with added safety features. Although the reactor is still in the process of design, they expect it to work safely and well.
But will this first of a kind reactor actually behave as it should? According to IAEA sources, Pakistan has not requested a safety review of the ACP-1000 design even though it is committed to buying two reactors of this type. This is odd. Sensible people would not even buy a used car without driving it to see if everything works, and no airline would consider buying a new jetliner without extensive flight-testing. Nuclear reactors have systems far more intricate than those inside the most complex passenger aircraft. This is for good reason; the consequences of a reactor failure could be immeasurably worse than an airplane crash.
Unlike Pakistan, China exercises caution in nuclear matters. It has asked for assistance from the IAEA’s Design and Safety Assessment Review Service with the ACP-1000. It plans to build two such reactors in China, and subsequently hopes to get permits for exporting similar units to Europe and North America. However, the IAEA review will not be a detailed, independent, international technical assessment of this reactor’s safety. According to the IAEA, the Generic Reactor Safety Review requested by China and scheduled to begin this May, “is focused on checking the status of the documentation (completeness and comprehensiveness)” compared to IAEA-recommended standards. The IAEA is explicit that this review is not “intended … to constitute any kind of design certification.”
As an importer of nuclear technology, China is a more discerning buyer than Pakistan. Today, a debate rages in China over the purchase of new, so-far-untested American nuclear reactors. A former vice-president of the China National Nuclear Corp. has gone on record saying China should not buy the new untested American reactors because, “This is very advanced technology, but it has not been commercialized in a nuclear reactor anywhere, so it needs to be proven over time.”
On the other hand, China National Nuclear Corp. seems to have no qualms about selling its untested reactors to Pakistan. Indeed, K-2 and K-3 construction was initially supposed to be coincident with the construction of two prototypes of the ACP-1000 reactor design in China. But nuclear industry sources have suggested that China may now abandon these prototype reactors in favor of a different design. The final decision is yet to be taken. Should this happen, the Chinese reactors in Karachi may well be the only ones of their kind built anywhere.
Instrumentation and control systems are another potential worry. For the prototype reactors planned for China, these systems will be supplied by the French nuclear company AREVA and the German industrial giant Siemens. A contract was signed in December 2013, but these European companies may not be allowed to sell the same systems for the Karachi reactors. If the Chinese nuclear industry has to produce such systems, they will be the first of a kind.
A troubling precedent suggests the need for caution.
In the 1990s, China designed and built a prototype nuclear reactor at Qinshan. An accident in 1998 due to a design flaw shut it down for a year. Initially the Chinese nuclear designers and operators could not understand or fix the problem, and had to contract a U.S. company for the repair work. One part of the reactor had to be redesigned. After this China did not build any more reactors of the Qinshan design for itself, but happily sold this type of reactor to Pakistan; four such reactors will eventually operate at Chashma. Pakistan was lucky that the reactor accident happened before the first Chashma plant went online. There was enough time for it to be redesigned to avoid the problem that led to the breakdown at Qinshan. But will Pakistan’s luck always hold?
Site and Sound
Reactor supporters are not worried about an earthquake or tsunami risk to the reactor site. According to PAEC officials, the K-2 and K-3 site has been carefully studied and the IAEA is claimed as having approved the site. This is not as reassuring as it sounds.
At PAEC’s request, the IAEA did indeed review, in 1998, a seismic study carried out at the Karachi Nuclear Power Plant (KANUPP), adjacent to the site of K-2 and K-3. But IAEA sources tell Newsweekthat the IAEA has not specifically carried out an assessment for the new Karachi site.
From the documents made available for this piece, the PAEC seems to assume that the largest possible earthquake that might happen off the Makran Coast, shaking Karachi and unleashing a tsunami, would be an 8.3-magnitude event. This was the size of the 1945 Makran earthquake.
But new scientific research, published in 2013 by a team from Britain’s National Oceanography Centre and Canada’s Pacific Geoscience Centre, finds that the largest earthquake in the Makran area may, in fact, be a lot bigger than what the PAEC has assumed. The lead author of the study concludes: “Past assumptions may have significantly underestimated the earthquake and tsunami hazard in this region.” The new research suggests the Makran area is capable of producing earthquakes as large as 9.2-magnitude ones. An earthquake of this size would be significantly larger than the 9-magnitude earthquake that hit Fukushima in March 2011. It would release over 20 times the energy of an 8.3-magnitude earthquake, assumed by the PAEC in its studies of the earthquake risk at the Karachi reactor site.
An earlier study, in 2007, of the earthquake risk to Karachi by a group of researchers from the United States and Pakistan looked at historical earthquakes over a period of more than 1,000 years and found the risk to be poorly understood. The study concluded that “Considering the number of known active [earthquake] faults that menace Karachi from almost every direction … it seems possible if not probable that hazard is higher than that assigned by recent national and global hazard maps.”
Even in the 21st century, earthquakes remain notoriously unpredictable. Just a month before the Fukushima disaster, Japan’s nuclear regulatory authority approved a request to run the nuclear plant there for another 10 years. This request had been supported by studies on the risk to the site from earthquakes and tsunamis. Everything looked okay. But, of course, it wasn’t.
Fukushima was hit by a once-a-millennium event the nuclear industry’s earthquake experts did not anticipate. The experts were clearly wrong about how big the biggest earthquake could be, where it could be, and the tsunami’s strength. This was despite the fact that Japanese earthquake and tsunami researchers form a large and well-funded scientific community with the world’s most advanced instruments and computer models, and have many hundreds of years of carefully collected earthquake and tsunami records to build on.
On the other hand, Pakistan has limited capability to monitor earthquake activity. In 2012, the National Seismic Monitoring & Tsunami Early Warning Center in Karachi complained that, “nearly half of the 62 seismometers working in the country are not transmitting real-time data to the national seismic-activity monitoring network.” They had been disconnected for lack of money to pay the monthly connectivity cost of between Rs. 2,000 and Rs. 5,000 each.
Remember Chernobyl?
The vulnerability of nuclear reactors to lax attitudes about safety and a lack of experience by operators became catastrophically evident in the 1986 Chernobyl disaster. An IAEA investigation and report determined that “a poor level of safety culture” and the “lack of feedback of operating experience and the inadequacy of communication between designers, engineers, manufacturers, constructors, operators and regulators … were critical factors in the events leading up to the Chernobyl accident.”
How vulnerable is Pakistan to the kind of human and institutional factors that led to the Chernobyl accident? Pakistanis are often willing to accept high levels of risk and place low priority to safety. The approach to problems in the general public is often unscientific; many are satisfied to place their faith in God as protector. Would these broadly shared social attitudes be reflected in how nuclear power plant operators handle problems?
A further concern is how well the Pakistani operators will understand and manage the new Chinese reactors. The operators, who will be from the PAEC, are likely to have less-than-complete knowledge of the imported nuclear plants because there was no real local input into the design or manufacture of key components and software.
Along with natural disasters and operator error, in Pakistan there is the need to worry about the risk of deliberate sabotage or terrorist attacks on nuclear reactors. Although PAEC officials dismiss the possibility—and one hopes they are right—the problem needs to be taken seriously.
Well-organized and well-armed religious terrorists, often with insider help, have successfully attacked even tightly guarded military institutions. The list includes places that would expect to be attacked in wartime and so should have been heavily defended—including the Pakistan Army’s General Headquarters, the Navy’s Mehran base, and the Air Force’s Kamra base. If security forces cannot protect their own bases, it is hard to see how they could successfully defend a nuclear power plant.
