Share This!
Text SizeAAA Share Email

UNION OF CONCERNED SCIENTISTS UPDATE ON JAPAN'S NUCLEAR POWER CRISIS TELEPRESS CONFERENCE
MARCH 25, 2011 11:00 A.M. P R O C E E D I N G S

OPERATOR:  Ladies and gentlemen, please stand by, your Japan nuclear reactor update will begin shortly.  Once again, please stand by while all participants are connected. Good day, ladies and gentlemen, and welcome to the Japan Nuclear Reactor Update.  At this time, all participants are in a listen-only mode, but later today we will conduct a question and answer session, and instructions will follow at that time. If you should require assistance during today's conference, you may press star, then zero on your touchtone telephone to speak with an operator.  Also as a reminder, this conference call is being recorded. Now I would like to introduce your host for today's conference, Elliott Negin.

MR. NEGIN:  Thank you. Good morning everyone, this is Elliott Negin, I'm the media director here at the Union of Concerned Scientists.  Thanks for joining our call this morning. Just to remind you again, the Union of Concerned Scientists is an independent science-based advocacy group that has been a nuclear industry watchdog for 40 years.  We are not for, or against, nuclear power.  Our goal has always been to ensure that the industry operates its reactors as safely as possible. If we do not get to your question during this morning's briefing, please email us at media@ucsusa.org, and we will get back to you as soon as we can.  If you have any trouble getting everything down that you need from today's briefing, there will be a transcript and an audio file on our website later today. We may already have answers to some of your questions posted on our Frequently Asked Questions feature, and answers to the more technical questions are on our allthingsnuclear.org blog.  Late yesterday, we posted more information on the blog about the risks of storing spent fuel in pools at reactor sites. Now, we will not be holding our daily 11:00 a.m. press briefing this weekend, but we will start up again on Monday.  At this point, we plan to hold these briefings through next week, but we will be sure to let you know if there are any changes in our schedule.

Now, after our speakers are done this morning, and we open up the phone to your questions, please ask only one question, and if necessary, one follow-up.  And please, mute your phone after you ask your question so the sound of your typing doesn't disturb everyone else. Now, this morning, our speakers are David Lochbaum and Edwin Lyman, who will update us on the latest developments in Japan. David is the director of UCS's Nuclear Safety Project.  He is a nuclear engineer and he worked at U.S. nuclear plants for 17 years.  He also has worked as a safety trainer for the Nuclear Regulatory Commission. Dr. Edwin Lyman is a senior scientist in the UCS Global Security Program.  Ed has a doctorate in physics and is an expert on nuclear plant design and the environmental and health effects of radiation. I will now turn the phone over to David Lochbaum.

MR. LOCHBAUM:  Thank you, Elliott, and good morning. Yesterday, there were reports that three workers received radiation exposures from water that they were walking through in order to repower some equipment in the Unit 3 turbine building at Fukushima.  It's not clear what the source of this radioactivity might have been. The turbine building is normally isolated from both the Unit 3 reactor core and the Unit 3 spent fuel pool. There were reports of damage to fuel in the Unit 3 reactor core, and also to damage to fuel in the Unit 3 spent fuel pool.  But it's not clear at the moment from the data we've seen how the water in the basement of the Unit 3 turbine building became contaminated. We haven't seen any clear signs whether it came from the reactor core or from the damaged fuel in the spent fuel pool.  They were adding water from above and from fire trucks to the spent fuel pool in Unit 3 and it's possible that water drained from that location down to the turbine building. There are also pipes that connect the reactor vessel in the primary containment to the turbine building that were normally isolated in situations like this, but might have been a pathway for the radiation to get from the containment to the turbine building.  We're just not sure from the data so far. There were also reports yesterday that it's possible that the reactor vessel for Unit 3 may have been breached.  There's also inconsistent signs, there's some data put out by the Japanese Nuclear and Industrial Safety Agency that show pressures inside the Unit 3 reactor vessel, the Unit 3 containment building, and those pressures would seem to suggest that the reactor vessel is still intact, because the pressures are different.  If there had been a substantial breach, those two pressures would have equalized over time.  So, it's contradictory data. I think that's reflective of the situation over there.  There's quite a bit of damage in quite a few areas.  There's not a lot of instrumentation available.  There's not a lot of access available for workers to go through the facility and more accurately assess conditions in lieu of the instrumentation that's spotty at best.  So, I think it's reflective of it's going to take a while to fully identify what's happened there, what happened and why I think remains to be seen. That's pretty much what I know from yesterday.  Thank you.

MR. NEGIN:  Thank you, David. Now we have Dr. Edwin Lyman.

DR. LYMAN:  Thank you, Elliott, good morning.

I would just like to make one observation, and that is the amount of iodine 131 that has been estimated by the Austrian Meteorological Agency as about 20 percent of the release that occurred at Chernobyl after the first four days of the accident, it's my estimate that that is consistent with the release of iodine from the gap that is between the fuel pellets and the cladding.  So, that is roughly about five percent of the total iodine inventory that I would estimate was at the Reactors 1, 2 and 3 at the time they were shut down.

So, that number is actually consistent with limited damage to the uranium fuel itself. That estimate stopped after the first four days of the accident, so we don't know beyond that, but I would like to point out that the longer the Japanese authorities are able to postpone any significant breach to the reactor vessel or the containment, the longer there is for the iodine 131 to continue to decay, and even if there's a significant breach that's just emerging at this point, it's already been almost two weeks, or actually it has been two weeks, almost, about 14 days since the accident occurred, and that would be almost two half lives of iodine 131, which would mean the inventory is down by about four from where it was.

So, even though the total amount of iodine 131 at the time of the initial accident was enormous, many, many times the amount that had been released at Chernobyl, at this point, even if all the iodine in all three reactors were released, I believe it would only be a few times what was released at Chernobyl. So, I don't know if that's good news, but certainly the delay of the ability of the authorities to manage this crisis may have really helped in ultimate radiological consequences.  Thank you.

MR. NEGIN:  Thank you, Ed. We will now open up the phone to questions.

OPERATOR:  Ladies and gentlemen, at this time, if you wish to ask a question, please press the star, and then the number 1 key on your touchtone telephone.  If your question is later answered, or you wish to remove yourself from the queue for any reason, you may press the pound key to do so. Once again, to ask a question, please press star, then 1. Our first question.

REPORTER:  Good morning, gentlemen, thanks again, as always. On the point that you just brought up, Dr. Lyman, about the iodine, how did that apply to the cesium, though?  I realize, of course, that the iodine -- the longer the delay, the less impact the iodine release could have if there are, indeed, breaches in the reactors, but could you please comment on the potential for further cesium releases?

DR. LYMAN:  Yes, the picture isn't as good for cesium 137, which, as we discussed before, has such a long half life that there's been no real change in the inventory over the last two weeks.  Cesium is not quite as volatile as iodine, and although in a core melt, the likelihood that most of the cesium in the fuel would be released to the reactor vessel, it's likely that both iodine and cesium would be released almost entirely into the reactor vessel.  However, cesium would condense faster on cooler parts of the plant, so there may be more played out cesium, especially if other parts of the plant are relatively low temperature.

So, the actual release of cesium to the environment, I think, is highly unclear, though you also do buy some time by delaying release, the worst releases are always if there's a containment failure within a few hours of the start of the accident.  So, it's very, very hard to predict, but there are other mechanisms that could release the cesium even if the cores completely melt.

REPORTER:  And just a quick follow-up on the point of cesium.  At this point, knowing what you know, and knowing what you don't know, do you think that this accident ultimately will or will not result in some sort of permanent quarantine for the immediate vicinity of the plant?

MR. LOCHBAUM:  You know, there's no such thing as a permanent quarantine a priori, because it's still an economic decision.  You know, no matter how contaminated the site is, if you spend enough money, you can probably decontaminate to the extent it can be habitable again.  So, it's really an economic decision. The Soviets tried it, it wasn't feasible at Chernobyl, and created essentially a permanent exclusion zone.  It's likely that there would be some region where contamination would be so extensive that thorough clean-up would be prohibitively expensive.  Again, that will be an economic decision once the full extent of the contamination is resolved.  But as the dose rates look now -- another factor is it's not clear the reported dose rates, what isotopes are being attributed to them.  We need more information about how they decay over time and then the distributions to know how much of that is long-lived contamination.  But as of, you know, now, or two days ago, there's certainly a plume mapped out by DOE that would certainly require an extensive remediation, and, you know, again, whether that would lead to permanent condemnation would be an economic decision.

REPORTER:  Thank you very much.

OPERATOR:  Our next question.

REPORTER:  Hi, thanks for doing this.

I'm seeing these sort of ominous reports from the Japanese Prime Minister, you know, warning of a breach in the reactor vessel of Unit 3, can you give us some kind of context to understand why, you know, how risky that is and just give me some sense of, you know, we've been hearing about this for weeks now and I actually, I mean, just give me a sense of what you guys know about this breach and what it could mean.

MR. LOCHBAUM:  This is Dave Lochbaum, I'll start and then Ed can supplement. First of all, we're hearing reports of a reactor vessel breach on Unit 3, but the data is inconclusive as to whether that's actually what's going on.  There could be other factors that could cause the reactor vessel integrity to be lost, for example, one of the relief valves could have stuck open. If it is, indeed, a reactor vessel breach, the consequences, or the most likely reason for that would be that the core has been damaged to the point of melting.  Some of the molten mass fell down to the bottom of the reactor vessel and caused a hole, burned a hole or created a hole in the bottom of the reactor vessel, which became the breach. If that's the case, it's bad, because first of all, it indicates that the level of fuel damage went beyond blistering and cracking of the fuel rods, to the point of melting of fuel, which is a more severe form of fuel damage.  And it was followed by a loss of integrity of the reactor vessel. If the reactor vessel remains intact, you always have the option or the ability to put water back into it to cool the fuel, even if it's damaged.  If the reactor vessel has been breached, you face challenges in putting enough water in it and keeping the water in it rather than having it just drain back out through a hole in the bottom.

In the emergency procedures world, if you do lose the integrity of the reactor vessel, the option is to fill up the entire containment above the point where the breach has occurred so that you try to still cover up the fuel, even if the fuel has been relocated, part of the fuel may have relocated, and the reactor vessel integrity has been lost. So, you still have methods to deal with that, if that were to occur, but they're obviously much more severe than if you keep the reactor vessel intact and allow the fuel to remain where it was placed originally. Ed, is there anything to add to that?

DR. LYMAN:  Just that, you know, until the core starts to degrade, as I mentioned at the beginning, a relatively small amount of radioactivity is actually released to the atmosphere, or to the reactor vessel and the coolant, and as the core starts to degrade, much higher quantities can be released. So, it's already been reported that the core has been exposed up to halfway for many days, I think that the expectation is that at least the part that was above the water line experienced extensive damage, and so the reactor vessel itself now probably might have up to, I don't know, 30 or 40 percent of some of the more volatile radionuclides like iodine and cesium in the vessel, then if the core melts through the bottom of the reactor vessel, and falls on the floor of the containment, it then can react to the concrete basement and that can generate additional gasses which will help sweep some of that material up and potentially out through a breach.

So, when all's said and done, something like 70 or 80 percent of the iodine, cesium and the fuel could actually be released to the containment atmosphere, and if the containment is breached, that's available for release.  So, as bad as the releases have been so far, they could increase by several fold, ultimately, if this proceeds any further.

OPERATOR:  Our next question.           

REPORTER:  Hi, folks, can you hear me? David, I see you're on a panel at the Senate Energy Committee on Tuesday with the NRC and NEI, and I wanted to get your thoughts on what you might urge the committee to do, if anything, in evaluating and acting on the Japanese situation, and also your initial take on the NRC's plan to do these 90-day and six-month reviews.  Thanks.

MR. LOCHBAUM:  Sure.  Well, I'll take the second part of that first.  I think the NRC's 90-day and its longer term reviews are the right way to handle this.  There's still a lot of information to figure out as to what happened and why, so that fits more into the longer term review and lessons learned application, but there's also some pretty short-term issues that came up that don't need to wait for the longer term review to do that. So, I think the NRC's two-phased approach is the right way for the agency to deal with lessons learned from Japan and what it means here in the U.S.

Which leads to the first part of the question, I think next week we're going to try to propose some of the things that should be considered by the NRC as they incorporate lessons learned from Japan to make the U.S. reactors less vulnerable to that type of situation.  Some of the things we're going to recommend are in the area of station blackout, when reactors lose their normal power and the back-up power and are faced with coping with just battery power, in Japan, that duration was eight hours, and the event lasted longer than eight hours, putting them in distress.

Rather than say that four hours or eight hours is too long or too short, we're going to suggest that we need to look at what if whatever the duration is, the event lasts longer than that, how long will it take for additional assets to be brought to the site.  In some locations, it might be relatively quickly, in some of our plants there's a military base in close proximity, with a fairly high reliance that some assets could be moved to the reactor site more quickly.  Many of our reactors are remotely located, where it might take longer to get those temporary generators and additional batteries to the site.  I think it would be prudent to review that and see what those timelines are.

If there's a high assurance that additional temporary generators and batteries could be arrived on-site within the current battery capacity time, then there's less likelihood that the operators would be faced with no power or no options.  If, on the other hand, it's going to take longer to get those assets to the site, it might be a good idea to extend the life of the on-site batteries, providing greater likelihood that the operators aren't left in the dark.  So, that would be one of our recommendations.

The other one, and the more easier to fix issue is dealing with spent fuel issues and spent fuel hazards.  In U.S. reactors, many of our spent fuel pools are only filled with spent fuel, making the risk or the hazard as high as we could possibly make it.  Everybody who has looked at the issue, the National Academy of Sciences, Alvarez and colleagues, including Ed Lyman, UCS and others, have repeatedly and consistently said, we need to reduce the inventory of fuel stored in the spent fuel pools by accelerating the transfer to dry cask storage.

We're going to reiterate that request or that recommendation, try to get the spent fuel pools thinned out so that the inventories aren't as large, which in turn significantly reduces the risk if cooling is lost or water inventory is lost from the pools.

So, hopefully, instead of keeping repeating the recommendations, we'll actually see some actions taken this time.

REPORTER:  Following up, will you make you any recommendations about Congressional action, especially on nuclear loan guarantees or anything like that?

MR. LOCHBAUM:  Not at the hearing next Tuesday, we are not likely to touch upon that issue, unless we're asked about it.  That's something that's a longer term issue, we don't see that on the near term.

REPORTER:  Thanks.

OPERATOR:  Our next question.

REPORTER:  Thank you very much.

You mentioned the discrepancy in the readings that you're seeing in trying to determine whether or not there's been a breach, so what indications or things will you be looking for to help determine whether or not there's been a breach of the vessel?  It seems sort of odd to me that we now think the Japanese might be overstating the case, but what will you be looking for to help determine that?

MR. LOCHBAUM:  A couple of things, if you're adding water in, and the water is -- and the relief valves -- excuse me, let me go back.

The bottom of the reactor vessels are at a much lower elevation than the relief valves are, so if you're adding water in and water is draining out as quickly as you're adding it, but the level in the reactor vessel is not up to where the relief valves are, then it eliminates the relief valves being open as a potential pathway for the water to escape the vessel.

That doesn't necessarily mean there's a breach in the bottom of the reactor, but you try to go through a number of things that could explain the indications you're seeing, to eliminate some of the pathways and include some pathways and soon you're left with only the one pathway that fits the data you have.

Part of the problem we're having is it's not getting all the data, the water levels, the pressures, the flow rates of water going in and so on, to be able to do the analysis and either eliminate pathways or include pathways. The data we're getting is very sketchy and it's been impossible thus far for us to do that analysis and come up with what seems to be the most likely scenario to explain the data.

We're still looking for that data. Once we have it, if we're able to do the analysis, we will put forward what we think our scenario is, but at the moment, it's very difficult to do.  It's hard to connect the dots when you have such few dots.

DR. LYMAN:  This is Ed Lyman, one other thing, if there were a large breach in the vessel and the core actually melted through and fell into the containment floor, there would probably be a rapid rise in containment pressure for, you know, at least 30 minutes to an hour or so.  So, I don't think they've detected that yet.  I guess if the containment already were breached, it may not be as rapid, but there would be some indications of a serious change, a serious and rapid change in these parameters.

REPORTER:  Thank you.  Are you hopeful that in 90 days that we will really know enough?  I mean, these things just tend to stretch out and out.  Are you going to be able to stick to 90 days?

MR. LOCHBAUM:  Well, I think in 90 days we will be able to make some short-term things, in terms of the lessons learned and station blackout power availability and extensive management.  Longer term, there's some questions about the hydrogen collected in the reactor buildings and caused the detonations.  I think those will not likely be resolved in 90 days but should be able to be captured in the longer term phase of what the NRC is undertaking.

REPORTER:  Thank you very much for all those explanations, it's very helpful.

MR. LOCHBAUM:  Thank you.

OPERATOR:  Our next question.

REPORTER:  Thank you, Ed, and David for this, and I apologize because I came in later, so this may have been covered already, but we've seen smoke and steam coming out of some of the reactor buildings, and Ed, you mentioned that if the core has melted and the fuel has reacted with the concrete floor that there would be gas released.  So, can any dots be connected between the smoke and the reactor containment breach, can we draw any analysis there?

DR. LYMAN:  Yes, I'm afraid I couldn't really venture a guess.  I guess there's still a lot of different causes or sources for that steam and smoke, so that's really hard to make any kind of estimate.  Sorry.

REPORTER:  Just another question, I sense that during the last few days that the U.S. nuclear industry has been talking a lot about the additional safety measures they have taken since Three Mile Island and since the September 11th attacks, they didn't say directly, but my sense is they're trying to say because of these additional measures, we're better equipped than Japan to cope with some of these emergency situations.  What do you guys think of that?

MR. LOCHBAUM:  Well, I think with the situation they have, they have to grasp at straws to try to explain why what happened there can't happen here.  They're basically similar reactor designs operating under similar regulations, so if our reactors were faced with a similar challenge, the outcome would be similar.

You know, there are certain things that were done since 9/11 that help, but I don't think that's a panacea that makes any of our reactors invulnerable to the type of problem that Japan faced.

DR. LYMAN:  This is Ed.

I would just like to comment on the post-9/11 measures, and as we've discussed in these calls previously, first of all, the plans are secret.  So, no member of the public actually knows what those plans involve. Second, what we do know is that because those were addressing what NRC considers to be a beyond design basis event, mainly an aircraft attack on a currently operating plant, we do know that the equipment, any equipment that would be staged or designated for that post-accident management would not meet the highest nuclear safety standards that are required for protecting its design basis events.

So, for instance, Nuclear Energy Institute already admitted that that equipment is not safety-related and would not be seismically qualified.  So, if you have an event other than an aircraft crash, let's say a large earthquake, it's not clear that those plans and that equipment would actually survive to be available, and putting U.S. reactors in the same spot that they were in in Japan.

REPORTER:  One last question, when you guys say you want to recommend about the battery life and we need to re-evaluate that, when I was talking with NRC, they say each station, they have the battery life that lasts about four to 12 hours and that each station has to then decide whether or not within that time frame they can recover either off-site or emergency power supply.  Are you guys recommending anything beyond what they have been doing?

MR. LOCHBAUM:  Yeah, I think what we're recommending is that the U.S. reactors have done the same thing the Japanese reactors did.  They base their battery duration on probabilities that they would either fix or repair a diesel generator, or re-establish connection with the electrical grid within that time frame.

The incident in Japan showed what happens when the accident doesn't follow that script, and lasts longer than the math or the calculations that were done on the front end.  I think what we're going to recommend, what we will recommend is that to better protect plants from situations like happened in Japan, they ought to look at what happens if, whatever that duration is, whether it's four hours or eight hours or whatever, what happens if they last much longer than that.

There should be some assurance that you can get other temporary generators to the site, or additional batteries to the site, before that time ends, again, whether it's eight hours or four hours.  If you can't get those additional assets to the site in that time, it would be prudent to extend the battery life so that the operators aren't left with no power to operate the pumps that they're given.  Japan showed what happens when you have plenty of pumps but no power.

It would be foolish for us to stick to this notion that an accident is only going to last four hours or an accident is only going to last eight hours, and leave millions of people in harm's way if the accident lasts longer than that. So, the U.S. Government may stick to that notion, we would hope they would be smarter than that, but we'll see.

REPORTER:  Thank you.       

OPERATOR:  Our next question.

REPORTER:  Thank you very much for taking my question.  I have another question because my first one was already answered, what would be the worst case scenario if we have a reactor breach, and, you know, all this plutonium should just go to the sea of the ocean because it affects the sea life and the atmosphere.  So, what can we impose -- what can we see as the worst?

DR. LYMAN:  This is Ed Lyman.

I mean, worst case meaning that there's a breach of the reactor vessel, the core falls into the containment, it spreads out across the floor, and this would require the containment floor to be completely dry, which I'm not sure that would be the case, but if it were completely dry, it would spread out to the corners of the containment, or it could actually contact the containment liner and melt through the liner, and then you have a pathway directly from the core material to the environment.

So, then it depends on essentially how much of the radioactive isotopes that were contained in the core enter the atmosphere of the containment and then how much leak out from the containment.

There are numerous modeling and simulations over the years show that a high fraction of the isotopes like cesium and iodine would be released from the core material in this situation, and enter the atmosphere in the containment.  There are a range of other isotopes, radioactive barium, tellurium, and strontium, all that have varying properties, and would be released to varying extents less than 100 percent.  It could be on the order of five, ten, 20 percent, it depends.

Then there are the lowest volatile elements that include plutonium and certain lanthanides, and certain other actinides, like americium and uranium.  Uranium actually under certain conditions could be released on the order of one to 10 percent, that was demonstrated in experiments over the last ten years, plutonium and the lowest volatile isotopes would be less than one percent, probably.

The ultimate consequences could exceed those at Chernobyl, because of the total inventory of radioactive material in the three reactors and potentially three spent fuel pools is many times what was in the core at Chernobyl. But the key is how much, what are the released fractions, and that's still highly uncertain.

But in this case, which is essentially a late containment failure, very late, weeks after the reactors originally scrammed, analyses typically show that there would be some -- well, first of all there's radioactive decay, like I said at the beginning, iodine, some other short list isotopes, significantly reduced, and to the extent that other parts of the reactor cooler, you might have played out, but it really has to do with when the timing of the containment failure in relation to the vessel breach.

So, if the vessel breaches and the containment failure is still delayed significantly, then you have more played out and less environmental release.

REPORTER:  Thank you.

OPERATOR:  Our next question.

REPORTER:  Thank you for taking my call, and I would like to ask you about the food safety and how long will the radioactive materials affect on the Japanese produce and is it okay as long as, you know, we wash that vegetable, we can eat it, is it ripe, or should we essentially destroy those vegetables?

DR. LYMAN:  This is Ed Lyman.

I think we don't have the expertise really to second guess those kinds of determinations.  Certainly with regard to iodine, as I said before, that is a limited problem, and will be cleared in a few months. To the extent there's contamination with cesium, and other longer lived isotopes, frankly, I don't think there are any good answers about how affected produce can be decontaminated.  There are both national limits in Japan, international limits for destruction of contaminated produce, and I think those will have to be followed.

The problem is going to be when the contamination is within limits that the authorities say is safe.  I think people need to understand more.  I think I've said before, understand the potential doses associated with those levels and make their own decisions whether they should consume the produce or not.

But I would think whether or not it can be washed off, because I've been watching NHK and seeing what people are saying, I would be very wary of any unverified information about how to make produce safe at this point.

REPORTER:  And I have one follow-up, and what do you think about the FDA's ban on Japanese produces and do you think it is above appropriate measures and how long do you think it will last?

DR. LYMAN:  I'm sorry, which measures are you referring to?

REPORTER:  The FDA's ban on Japanese imports.

DR. LYMAN:  Oh, FDA.  I haven't looked, obviously it's not just the real risk, but it's also public perception and it's the same, you know, whether it's Mad Cow Disease or melamine or lead, and unfortunately, one of the economic outcomes of this type of event which let's say those people who sit around and model how bad these things can get don't really consider is the perception of contamination which can be as damaging as the actual, or more damaging than the actual health risk, and that will have an economic impact whether or not there's actually a significant health risk.

So, unfortunately, it's going to take a long time for Japan to restore confidence in the safety of its exports.

REPORTER:  Thank you so much.

OPERATOR:  Our next question.

REPORTER:  I'm curious about what these particular sequence of events caused and sort of the failure of imagination dimension about all these things.  We don't expect tsunamis, of course, in the middle of the country or anything where we have nuclear reactors, and so we don't think this kind of thing can happen there, but are there other things that we haven't anticipated, whether it's cyber attacks or some other cataclysm that give pause to the idea that we've designed this reactor to certain events and therefore it should be fine?

MR. LOCHBAUM:  This is Dave Lochbaum, I'll take a shot at answering that.

I'm an engineer, by both education and experience, and I like engineers, engineers are good people, and at the risk of stereotyping engineers, we're not really good at what-if questions, except in looking backwards.  What do we need to do to make sure this doesn't happen again.  We're not quite as imaginative looking forward as to what new wrinkle might happen that we need to guard against.

So, with that in mind, this event, as past disasters have done, we have studied in quite depth to figure out what systems worked and what systems didn't work, what procedures worked, so on.  So as to make the repairs or the upgrades to lesson the likelihood that this happens again.

But the real question you asked was how do you anticipate the next disaster and take steps to prevent it.  And that's a challenge. It's easy to do, the difficult part, if it's done, and the harder challenge is then to get somebody to pay for the fixes for a problem that hasn't occurred.

It would be relatively easy now to get owners to pay for fixes for batteries and spent fuel maintenance and so on, that were known in advance, but nobody wanted to pay for them. This country is very good at closing the barn door, once it's opened, but not until then.

So, it's very difficult when you're dealing with low probability, high consequence events to pay for things except after the fact. I think that's true not only for the nuclear industry, but the airlines and space travel and so on.  It's just a byproduct of high technology.

DR. LYMAN:  This is Ed, if I could just chime in for a second.  Originally, you know, reactors were designed on a design basis.  There were a set of rules that reactor designs were intended to show they could resist.  Then that was re-evaluated after TMI and the NRC realized that that are certain vulnerabilities to the so-called severe accidents that the reactors were not designed to protect against.

For instance, the containment building was not designed to protect against certain events that new studies showed could potentially occur in a severe accident.  For instance, a hydrogen explosion, a large steam explosion, a rupture of the reactor vessel at high pressure, none of those things were considered.

So, in some cases, it's a retrofit.  In some cases, they did analyses to convince themselves that it didn't matter, even though they hadn't designed against it, the containments were safe anyway.  And it's been that approach.

As new information comes up, the challenges to the original notion of how safe the design basis is, there's always been tension about how much more the NRC should require.  And the ultimate result is that according to the industry's own estimates, there's still, you know, roughly let's say about a 0.1 percent chance per year that another Three Mile Island core damage will happen in the United States, just based on internal events alone.  That is if a pipe breaks in the plant, not caused by an earthquake or something like that.  Most utilities have not even calculated the potential addition to that risk from external events like earthquakes.  So, they don't even have an assessment of the additional damage.  But few have, and find that they're roughly comparable to the impact of internal events.

Then there are the risks during shutdown.  It turns out that when a reactor is shut down, the risk of a severe accident is even higher than it is when it's operating.  For years, the industry resisted actually trying to assess those risks, and as a result, very few of them have actually performed what's called a shutdown or a low power probability risk assessment.  But again, some have, and they find that, again, it would double the risk.

So, the short answer is, even according to the industry's numbers, the risk is about one percent per year that a TMI will happen in the United States, and the question is, is that acceptable or not.

OPERATOR:  Our next question.

REPORTER:  Yes, hi, I wanted to ask about possible recommendations you might make on the BWR MARK I, and in one of the answers you gave you said maybe this indicates that they need even more back-up than other plants.  I wonder if you would make that recommendation and would you suggest a pause on the uprates or the prioritization for these plants in terms of movement to dry cask?  That's my question.  And I apologize, I don't know how to mute my phone.

MR. LOCHBAUM:  This is David Lochbaum. In the Senate briefing next Tuesday, we're going to recommend that steps be taken to better manage spent fuel risk at all plants, not just on water reactors, and not just MARK Is.  Those recommendations would have the greatest value at BWR MARK I containments and MARK II containments, but there's value to be gained in all U.S. reactors.

Similarly, we're going to make some recommendations to better protect against events where loss of power leads to fuel damage, both in the reactor cores and the spent fuel pools. Those lessons are equally applied to both boiling water reactors and pressurized water reactors, so we're going to make the recommendation and hope they lead to improved conditions at both types of reactors and not try to prioritize one versus the other.

REPORTER:  So, do you have any BWR-specific recommendations that you'll make?

MR. LOCHBAUM:  Not at the moment.

REPORTER:  Okay, thanks.

OPERATOR:  Our next question.

REPORTER:  Hi.  I think Ed Lyman said yesterday that he was maybe going to do some of his own modeling based on dose rates to figure out whether -- how much cesium 137 had been emitted from the plants.  I was wondering if you had anything more to say about your own modeling and your own sense of the cesium 137 release from the plant.

DR. LYMAN:  Yes.  No, I haven't made any progress on that work over what I said yesterday, which was the dose rates mapped out by DOE on the 22nd seem to be consistent, very roughly, we're talking about within a factor of five or even ten with the estimates of the Austrians Meteorological Agency.  So, it's not out of bounds, and I will, if I have additional insights, I'll report them.

REPORTER:  Thank you.          

OPERATOR:  Our next question.

REPORTER:  Hi, guys, thanks for holding these briefings.

Two questions, sort of technical in nature.  There's been some reports that a build-up of salt might complicate where they've used the sea water for emergency cooling might complicate long-term cooling at the units, I was wondering if you guys could offer any thoughts or background on that.

And I've heard you all say that the potential MOX in one of the reactors could be a great concern, whereas some of the industry folks have been saying less so.  Could you expand a little bit about why MOX fuel is a big concern at this reactor as well?  Thanks.

MR. LOCHBAUM:  This is Dave Lochbaum, I'll take the first part of that question and defer the MOX question to Ed.

Salt, as the sea water evaporates, the amount of salt concentrates in the reactor vessel, or in the spent fuel pool.  Left unattended, that salt could cause some problems, it plates out on the fuel, it could blanket the fuel and slow down the heat transfer from the fuel pellets through the metal cladding to the water on the outside, and that could cause fuel to get hotter than it might otherwise get.

The other potential might be for salt to -- if it does plate out, and salt is soluble, so generally speaking, it's in a dissolved form, but it does get oversaturated and start plating out, it could start impeding the flow of water through the fuel assemblies themselves.  You would have to -- quite a bit would have to evaporate to get to that point, but that is something that could occur.

Our latent problem is salt water is corrosive and it could start damaging some of the equipment, the pumps and controls that are being used to add water to the reactor core in the spent fuel pool, and impede the ability to continue adding that water, circulating that water.  So, that's a related challenge with using sea water.

DR. LYMAN:  Yes, with regard to the question of MOX, we've done quite a lot of analysis of MOX fuel over the years, and in France, about 30 percent of the core of over 20 of their light water reactors has MOX fuel in it.  That's a really significant amount of fuel. And, so, when you're talking about core ratings that large, then the properties of MOX that are different from those of uranium fuel become more apparent.

For one thing, MOX starts out with more plutonium in it, and it's a different structure than uranium fuel because you're mixing together two different atoms that have different sizes. So, you end up with what's called a heterogenous microstructure, and you end up with clumps of plutonium that heat up more than the uranium part of the fuel, and that can cause problems in certain accidents because you have hotspots.

Another problem with plutonium is it affects the ability to control the plant, it makes control rods less effective, it makes boron injection less effective, if you want to try to shut down a reaction.

There are also studies that show that because MOX fuel has this heterogenous microstructure, it actually releases more fission products at a lower temperature than uranium does, so the fuel is kind of leakier than uranium.  And overall, because you're starting out with perhaps a percent plutonium in that fuel, the inventory plutonium in the core is greater than it would be if it's just uranium, and also the inventory of other alpha emitters, which have similar radiotoxicity to plutonium, are also significantly increased, sometimes by a factor of 20 or more, and so that in the event of an accident, where there's actually release of fuel particles, then that could have an impact.

However, in this case, there's only about five percent, only about five percent of the core is MOX, and so a lot of these effects would not be significant or apparent or have an impact on the evolution of this accident.

So, I -- even though, in general, I'm very concerned about these MOX at larger core fractions, in this case, largely because of the efforts of members of the public to force the Japanese to slow down their MOX program, because of serious safety issues like these, that there's a limited only small amount of MOX in that core and that will hopefully limit any additional risk associated with it.

REPORTER:  Thank you very much.

OPERATOR:  Our next question.

REPORTER:  Hello, thank you very much for taking my call.

I have several questions with regard to spent fuel handling.  What is your recommendation on the long-term solution to spent fuel storage now that Yucca Mountain has been taken off the table, because of seismic anti-radiological risks, should we be urgently seeking another location?  And could you comment on should we be reprocessing our own spent fuel into MOX, as I believe is being planned for South Carolina?

DR. LYMAN:  This is Ed Lyman, I can start.

We still believe that, in fact, the most appropriate and the only way, only option that we have for the long-term management of the spent fuel that we've already generated in this country is through geologic repository somewhere.  We have never had a position on Yucca Mountain, because we don't have geological expertise, but clearly there were some technical problems and a lot of political problems with the way that site was chosen.

So, we do support a new process that will, to the extent possible, be more open and inclusive in selecting a new repository site, but the alternative of leaving spent fuel essentially above ground in reactor sites forever is probably not sustainable.  And at this point, it's just choosing the least bad option for a serious problem.

That said, we do think that with appropriate safety and security, that dry cask storage for an interim term can be managed at reactor sites, and so we do have some time to be able to get a repository selection process on track.

With regard to reprocessing, we are completely opposed to reprocessing, not only would it not help with waste management at all, because it not only results in high level waste that will also have to go into repository, but you also generate large quantities of low-level waste and other types of radioactive waste for which the U.S. doesn't have a current policy on how to dispose of.

So, it would actually make the waste problem worse.  And reprocessing produces separated plutonium, which is a proliferation threat, and so you would be spending enormous amounts of money to separate large stockpiles of plutonium, which pose a proliferation of nuclear terrorism threat and then you would have to spend even more money turning that back into MOX fuel and using reactors which has additional risks that I was just talking about.  So, you would be spending an awful lot of money with no benefit and a lot of extra risk, that doesn't make sense to us.

REPORTER:  Thank you.

OPERATOR:  Our next question.

REPORTER:  Hi there, thanks so much for all your information.

You've addressed the issue of the worst case scenario of a breach of a reactor vessel, from looking at pictures at Unit 3, it seems like a considerable amount of debris from the explosion has sunk down into the rest of the building, and what would be the impact of this debris falling into the spent fuel pool?  Do we have any knowledge of what the spent fuel pool water levels are?  And can you talk a little bit about the hazard associated with meltdown in the spent fuel pool area?

MR. LOCHBAUM:  This is Dave Lochbaum.

One of the concerns about debris falling into the spent fuel pool would be initially whether just the weight of that debris, damaged fuel that's stored in the racks below, the spent fuel pool is analyzed for a bundle being dropped and hitting the racks and the fuel that are in them.  The fuel pool is not really analyzed for the roof or other parts of the building falling down into the pool.  So, because of the heavy weight, it could cause more damage just from the impact.

The second consequence or concern from debris would be if a large piece were to fall across the top of the fuel racks, the fuel in those racks is cooled by water that flows upward, vertically, through the racks, and gets mixed to remove the heat.  Anything that plugs the top, stops the flow of water upward, and impedes the cooling water effect flowing by the heat fuels in the racks.  So, that would be a secondary effect.

There was some signs earlier in the week, or last week, the water level in the Unit 3 spent fuel pool had dropped below the top of the irradiated fuel stored in those racks.  It appears that since then with all the addition, the work that was done spraying water onto the refueling floor, dropping water from helicopters above, that the water levels have been recovered to at least the top of the irradiated fuel in the racks.  We're not certain it's been completely refilled, but it looks like the fuel is no longer uncovered, perhaps.

That's also somewhat being confirmed by the reduction in radiation levels that are reported at the site, that's confirmatory evidence that the water level has been recovered somewhat in the spent fuel pool, but other than that, we don't have any definitive answers.

In terms of the last part of the question, the consequences, since you can see the building, the walls are gone, if there is damage to the fuel, the spent fuel pool, there's not any barriers left for preventing that radioactivity from reaching the environment. So, there's a lot of fuel in the spent fuel pool, there's not many barriers between it and the environment, if that fuel is already damaged or becomes damaged in the future.

REPORTER:  As a follow-up, I notice pictures of fire trucks extremely close to the walls of that building, is there a hazard to those workers if the damage that you're describing may have occurred?

MR. LOCHBAUM:  There is some damage, but the walls in the floor of the spent fuel pool are at least four to five feet thick of concrete.  Concrete is a pretty good shield against radiation from the spent fuel inside the pools, and because the open, the top of the pool is not covered, particularly now that the building is gone, if the water level drops, the radiation tends to be aimed upwards, shining upwards.  There may be some reflection bounced back from clouds and other material in the atmosphere, to pose an indirect challenge to the workers, but they're being monitored, and periodically over the past week, workers have been evacuated, due to high radiation levels. So, I suspect if those monitoring indicated that the workers on the fire trucks were exposed to high radiation, they would be pulled back from those locations.  There were reports last week of remote fire truck operations, once the hose is aimed and the water is turned on, there's no real need for the workers to stand there and watch it.

OPERATOR:  Our next question.

REPORTER:  Just following up a little bit on the batteries, were those regulations strengthened at all after 9/11 and do you all happen to know if the cost of such batteries are hugely expensive or, you know, if the industry should pick them up or how dramatic those costs might be?

MR. LOCHBAUM:  This is Dave Lochbaum.

The baseline rules that require the batteries to be put in in the first place, and determined how long they had to be, were not changed by 9/11.  There were some 9/11 upgrades that led to reports that some owners have temporary generators to replace or supplement the batteries, but the actual durations, whether it's four hours or eight hours, that wasn't revisited after 9/11.

As far as the cost, if you've ever seen those battery banks, they look like car batteries, just a bunch of car batteries hooked up together to provide that power.  It's not an astronomical cost, and I think anybody looking at the events in Japan would say that the cost of not doing that is probably much higher.

REPORTER:  Great, thank you.

OPERATOR:  Our next question.

REPORTER:  Hi, guys.  I'm back.

This may be an ignorant question, but something that's just been puzzling me for some time, they have been pumping sea water into the reactor cores for 12, 13 days now, and yet the level never seems to go up.  The reports are consistently that at least half of the fuel is exposed in the reactor core.  Help me understand why, after all this pumping, we actually haven't covered the fuel.

MR. LOCHBAUM:  That's not a question that I haven't been asking myself.  The flow rate yesterday was about 50 gallons per minute that they were adding either in the pumps they restored to the reactor feedwater line.  That's a relatively modest rate.  I don't understand fully why, as you pointed out, given how long they've been pumping water, why aren't the levels changing.  If nothing else, if it's by passing the core by some means and going into the containment, you would think you would start seeing containment pressure go up and the temperature go down, and it's not seeing those kind of results.

So, the data that's coming out is not giving a clear picture as to what's going on there.  I don't know if the data is bad, and just hasn't been updated for a while, or if the data is real and something that I can't explain is going on.  I wish I could help, I just haven't got that insight yet myself.

REPORTER:  I think for the first several days, I think I was under the assumption that they actually didn't have any idea, and that some of those numbers were assumptions, but they put out new numbers every single day, and they do change, which indicates that they are at least attempting to measure them, but it doesn't make any sense, given the thousands and thousands of tons of water that they've poured in that they haven't apparently made any progress in covering the fuel.

MR. LOCHBAUM:  Well, one of the challenges that was identified some 20 years ago, by a colleague of mine, Paul Blanch in Connecticut, was that the water level instrumentation for boiling water reactors is tenuous at best.  It's not that you have, like, a level gauge that tells you that you've got X number of inches, or millimeters of level in the reactor vessel.

In a boiling water reactor, when the plant is operating, the water level is boiling, that's where it came from, from its name.  So, the way they infer what water level is, is they compare the density or the weight of the water drawn from a sample of the reactor to a reference leg that's not boiling, and when the boiled water boils or the water level drops in the reactor vessel, there's a relative change in that weight versus the weight at the constant level reference leg.

It works fine when the plant is operating normally, but in accident conditions, the heat inside the dry well will boil off the water in the reference leg, which could make your indicated water level 16 feet different from what it really is.

So, attempts have been made to make the water level instrumentation more reliable, we may learn from this that those attempts fixed yesterday's problem but not the one we're facing today.  So, that could explain why even though they're getting water level instrumentation, it has no reference to reality.

REPORTER:  Would that be your best guess is that these levels actually are fairly meaningless?

MR. LOCHBAUM:  Well, they're reporting two different levels, an A and a B for most of the reactors on Unit 3, they're saying the A and the C.  There are three different water level instruments, that's the A, B and the C channels, and typically if you see a problem like I just described you would see one of them go off first, because they don't all react that same way because they're located physically in different places.

So far, they've been hanging in pretty close.  One might say 1740 millimeters and the other is 1750.  If you started seeing those kind of casually, you might start seeing one go either up or down and the other one hanging there like it was yesterday.  So, we're not seeing those aberrations.  So, it's not much more than a guess, based on the data, because the instrumentation may be suspect, so the values we're getting may or may not reflect reality.  So, I think they're faced with that problem and we are as well.

REPORTER:  Okay, thanks.

OPERATOR:  Our next question.

REPORTER:  Good morning, guys.

Ed, a couple of minutes ago you were talking about the design basis on reactors, and the risk assessment that was used, and that when the reactor was shut down, the risk would actually double.  I'm curious, though, as to what you and Dave think about the validity of the risk assessments that the industry has been using.  It seems to me they are always coming up with odds that are so high, X, Y, Z situation can never happen.  Do you think that their basic assumptions need to be changed?

DR. LYMAN:  This is Ed, let me start.

First of all, I may have misspoke.  The actual risk during a shutdown is maybe ten times higher, but since you're only shut down for less than a month per year, on average, on an annual basis, it's like twice as high.

With regard to the accuracy of the probabilistic risk assessment, the NRC has not, until recently, required that they meet very high quality assurance standards.  The way PRA is actually used in regulations is on a pretty informal basis, and the NRC didn't see the need to require additional quality for those calculations unless they were being used in a kind of very serious fashion.  So that they have a kind of graded approach where the more important the obligation, the more information they want to know about the PRA.

But it's only been in the last few years where there's been effort to try to come up with quality control standards for how you do these studies, and for the uncertainty estimates, which are very important, and those haven't even been fully implemented by the industry yet.  There are professional standard setting bodies, they have been arguing how to set a standard for PRA for years, and the recommendations that are coming out of that are that it doesn't make sense just to look at internal accidents, but you also need to look at things like seismic risk and shutdown events, because that's actually those risks can be separable to the internal events.  But industry has been resistant to expanding these PRA, or expanding the scope of the PRA, while they're still using it in certain applications.

So, we've been saying for several years now, it doesn't make sense to base any decisions on PRA results if you're not looking at all the potential accidents that could happen, including seismic, and that's fallen on deaf ears at the commission.

So, the uncertainty that's associated with any of those numbers have not even been quantified.

MR. LOCHBAUM:  This is Dave Lochbaum, I would agree with everything Ed said and add to it by saying a few years ago I presented a paper to the Commission on the risk studies that are done in the industry, if you add up every minute that every power reactor operates in the United States, you would get somewhere around 3600 reactor years today.  Over that time, we've melted down Farley Unit 1 and Three Mile Island Unit 2.  Given that track record, you come up with something like two meltdowns every 1800 reactor years.  But if you look at the industry's numbers, they're ten or 100 times less likely than past history has been.

So, what's wrong with the math?  Why doesn't it seem to reflect our actual history? I think it speaks to some of the issues that Ed's talked about, in that there are some, not fanciful, but optimistic characterization of certain events, throwing out bad data in the past when it's necessary, that's led to numbers that the math may be right, but the reality doesn't seem to be matching.

REPORTER:  External events to the calculation will help some, but if their basic assumptions and formulas or algorithms aren't correct or based on the assumption that it can't happen, will that change anything?

DR. LYMAN:  This is Ed.

I mean, there are a lot of different sources of uncertainty, when they do these calculations.  One is basic uncertainty that the data that you put into it when you're modeling, you know, for instance how frequently will a particular component break.  There are other sources of uncertainty.  You have to take into account operator actions.  They put in guesses as to what's the likelihood that someone won't successfully do something or have to do.

Then there are the uncertainties about what if an event occurs that you didn't think of.  That's not as complete as uncertainty.  In fact, it's pretty consistent that IN the last several years 30 percent of the significant precursors that occurred at U.S. plants, that is events that could potentially lead to a serious outcome, 30 percent of those were not actually modeled in that plant's probabilistic risk assessment.  So, they're missing, you know, at least 30 percent of the things that could happen that could be serious.

So, I mean, it's probable if you account for all these uncertainties properly or at least give an estimate, and that's what they have not yet done.

MR. LOCHBAUM:  An example of that would be Davis-Besse and the damage that their reactor vessel had.  That plant's PRA had like a zero percent chance that the reactor vessel would fail, and yet they came within two to 11 months of the reactor vessel failing, because it was considered not complacent enough to cause the leakage over six years and cause the head to degrade like that.  So that the math, again, showed us that there was no chance that the reactor vessel would fail, yet they very nearly did.  So, it looks good on the math.

I've told the Commission that those PRAs should be put in the science fiction portion of their library.

REPORTER:  Is that the same as the PRA for Indian Point that once said it would rule out deliberate crashes of aircraft because that's just too unlikely to occur?

MR. LOCHBAUM:  Yeah, it's so unlikely as to not even be factored in.

REPORTER:  Thank you.

OPERATOR:  At this time, I see no further questions in the queue.

MR. NEGIN:  This is Elliott Negin again, I want to thank you all for participating in our phone call this morning.  I just want to remind you, we will not be hosting a phone briefing tomorrow morning, Saturday, or Sunday morning.  We will be back at 11:00 a.m. on Monday morning, Eastern Daylight Time, to discuss what happened over the weekend in Japan, and also the implications for the U.S. nuclear industry.  Thank you very much for participating.

If you have any questions today, following this briefing, please email us at media@ucsusa.org, and we will do our best to get back to you as soon as we can.  We will not be fielding requests over the weekend.  We will resume answering your requests on Monday morning.  Thanks again.

OPERATOR:  Ladies and gentlemen, thank you for your attendance during today's conference.  This does conclude today's program, and you may now disconnect.

(Whereupon, at 12:22 p.m., the telepress was concluded.)

Powered by Convio
nonprofit software