Thursday, October 29, 2009

The Future of New Brunswick Power

Of these two sets of comments by Silver and Powers, I'm tempted to agree with Silver more.

Not because he has worked as a consultant for utilities but because his points seem more valid and his chief complaint seems extremely well considered as well. Powers complaint concerning 'sovereignty' may be valid, and his use of historical experience is useful in understanding his point of view, but is not particularly relevant to New Brunswick necessarily.

One item that I would like to note is that Hydro Quebec is going to assume control of New Brunswick's Point Lepreau after refurbishment is complete. This, combined with Quebec's refurbishment plans for Gentilly-2 will give Hydro Quebec a foot in the door of the nuclear industry rather than simply a toe. With plans for AECL to be sold this is a significant factor to be considered.

Quebec gets almost all its electricity from hydroelectric projects so at first glance the need for nuclear power would be minimal, but if Quebec has plans to expand its market to the USA then it will need more electricity than it already has and 100% of that has to be carbon-free to be politically palatable. Since it is selling electricity for profit, expensive solar panels and subsidized wind power are not options since those would minimize or eliminate profits rather than increase them or would lead to an increase in Quebec's hydro rates.

No, the only option is for Quebec to acquire more nuclear power. But building in-province would be politically dangerous. Charest has seen the backlash in Saskatchewan and Alberta from some, so pushing for more nuclear power is risky in your own province. So why not acquire and expand the nuclear sites in New Brunswick.

The 'expand' part would be tricky. To do so, Hydro Quebec would have to purchase the CANDU 6 rights from AECL at the very least. But considering they just put down a $10 billion offer for New Brunswick Power, might we see them make a bid for AECL to get the rights to build more nuclear reactors of their own?

Time will tell.

Wednesday, October 28, 2009

The Boy Who Cried Flu

I think Margaret Wente nails this one right on the head.

The media-overexposure of H1N1 is dangerous in its own way as people are now realizing. Part of the problem is that people in the media and government seem to have forgotten a fundamental human nature, the nature to be suspicious.

If a vendor is trying desperately to sell you something that you are hesitant to buy, are you more or less likely to dig in your heels as they push you harder to buy it? For me, I'd be less likely to buy it because I'd suspect that there is a reason they are so desperate to sell me it. Poor quality, price is too high, ancient curse, I'm not sure, but if they're so desperate to get rid of it there must be a reason and its probably not to my benefit.

People are rational beings, not irrational crazies as some would like to believe. We were told months ago that the H1N1 was a 'pandemic' and were warned that 'millions could die'. Media kept a running tally of the sick and dying. Internet sites plotted the cases on fancy maps, showing the ever expanding spread of the virus. Health groups were feverishly putting out new information every day. People watched with fear and trepidation as H1N1 reached Canada and... nothing really happened.

Some people got sick. Some people died. As Wente points out, somewhere between 700 and 1400 people die from the flu every day around the world. In comparison, only 5000 people have died from H1N1 worldwide.

People aren't idiots, they see that their friends and family get sick with H1N1 and recover. Just like they do every year when they catch the flu. The few cases that do die are few and far between. So more and more they come to believe that the government and media are exaggerating the danger and believe less and less what they say about it.

Pushing harder only causes these people to dig in their heels and take positions that typically would be untenable.

The real tragedy of all this is when a real serious pandemic strikes, people may have become so cynical about the government and media that they won't believe them at all.

UPDATE: Just noticed Thomas Walkom's article saying something similar.

Tuesday, October 27, 2009

H1N1: Australia Edition

Somewhere in a news report I was reading recently, the fact that Australia had reported an increase in the confirmed flu cases over 8 times the normal seasonal flu was highlighted as a reason for the near-hysterical response by some governments and people.

That was an impressive number I thought, and quite convincing when held in isolation, but then I remembered that the Australian flu season is over now, and they did not have the benefit of a vaccine either. So it got me to thinking, that Australia is a similar nation to Canada in terms of infrastructure (health care, etc..) and population (large nation with areas of dense population density and areas of sparse population density). It might serve as a useful analogy to see how dangerous H1N1 really is.

Figure 1: Confirmed Flu Cases in Australia by Week [1]

As shown in Figure 1, there really has been an increase in the number of confirmed flu cases in Australia, 8.4 times the 5 year average in fact according to the Australian government. But these are only 'confirmed' cases this year, and in years past there's no guarantee that people didn't experience flu like symptoms only to decide not to report them or confirm them with a laboratory. In effect, because the testing rates this year are not similar to testing rates in the past its a useless measure.

A more useful measure perhaps is how often general practitioners have patients who are experiencing 'influenza like illnesses' (ILI). Australian government records helpfully record that the rate of observed ILI cases in 2009 is actually lower than in 2007 and similar in magnitude (although not in time) to 2008.

Figure 2: Rate of ILI in Australia by General Practitioners [1]

Alright, one might argue, but what if people are so sick they're not going to their family doctor and are instead heading straight to the emergency room, this wouldn't be reflected in the rate in Figure 2. This is true, however, records show that the number of presentations at emergency rooms in Australia with ILI is similar to 2007 but significantly higher than 2008. The 2007 rate was attributed to what the report says is a public response to a few cases of child deaths associated with influenza. Interestingly enough, there might be something to this, the number of hospitalizations in Australia from influenza is typically around 2000/year, whereas this year it is around 4000.

What can we take away from this? Well, the rate of confirmed influenza has increased by 8 times, but the number of hospitalizations has only doubled. So we can either confirm the hypothesis that they are indeed testing more often or confirm that influenza is less dangerous percentage wise because the rate of ILI hospitalizations over the number of lab confirmed influenza has actually decreased. Moreover, its also possible that with all the talk about H1N1 going around, that doctors are taking fewer chances and are hospitalizing patients more often than normal. After all, their fundamental rule is to do no harm.

Figure 3: Rate of ILI Presentations at Emergency Rooms in Australia [1]

Finally, just in case you weren't convinced by all that, it would be expected that if influenza was as widespread and as dangerous as some are making it out to be, that there would be a significant increase in the absenteeism reported. As seen in Figure 4, this is absolutely not the case.

Figure 4: Absenteeism reported in Australia [1]

The message I took away from all this information was simple. Get your flu shot if you want, but lets not panic, and lets not get carried away worrying about H1N1.

[1] Australia Influenza Surveillance Summary Report, Australian Government [link]

Friday, October 23, 2009

On The Future of AECL

Ironically, the one major issue that seems to be dominating the Canadian technology/political news is the one issue that I am loathe to touch with a ten foot pole.

The future of AECL has been in doubt for several months as the federal government made it clear that they intended to sell AECL to private companies. At first brush, this is not necessarily the 'death knell' that many people make it out to be.

AECL is having a terrible time marketing itself to the world for a number of reasons.

1) CANDU reactors are more expensive than other kinds of reactors.

2) The benefit that using natural uranium has been steadily decreasing over the years as the cost of enriching uranium decreased and AECL has been unable to capitalize on new fuels (like Thorium) the way it would have liked.

3) Frequent conflicts with the CNSC as changing safety regulations place them on uneven ground trying to keep up with the safety requirements placed on them. This means they have not been able to market a single standard reactor.

4) Management fiascos like the one at Point Lepreau in New Brunswick are the norm rather than the exception. (see MAPLE for another example) Doesn't matter who is to blame for the delays or what the reason is, the fact that they cannot deliver on a product in the time they say the will makes these events fiascos.

The CANDU reactor is more expensive than other reactors, but that alone shouldn't be enough to undo AECL. Neither should the changing realities about UO2. Even the conflicts with the regulator nor management fiascos on their own shouldn't bring them down. But taken together it is a potentially lethal mix.

With or without privatization, the CANDU line of reactors is nearly dead. Of the scores of new reactors being constructed or planned not one is a CANDU. Even Ontario is having trouble swallowing the cost of a new CANDU.

But one thing is for sure, the research side of AECL is in deeper trouble than the commercial side. Without any plans for new research reactors what will the research half of AECL be able to do? Spinning off the commercial CANDU end of AECL might preserve that half (or might not) but without a research reactor, the research end of AECL is doomed.

What is needed politically is clear action. The time for striking committees and getting advice should be over and the time for political willpower to push one direction or another is upon us. The longer the delay in clarity, the less likely anyone is to even buy AECL and AECL will be left to die a slow, agonizing death.

Tuesday, October 13, 2009

About Those Polls...

Alright, so Raphael and I have restarted our poll tracker. So I'll discuss the boring technicalities of why I compile the polls the way I do. Some of what I do may seem simplistic, but I have put some thought in behind why I am doing it that way.

The Politics of Binning

The first thing that I do is 'bin' the polls based on their polling end date. Binning the polls is a way of accumulating more data to obtain a smaller numerical error but a larger time 'error'. During the previous election, because of the large number of polls being released virtually everyday, I had the luxury of playing with the methodology a bit. Because of the relative scarcity of polls now, I am binning them into weeks.

Generally polling companies rarely release polls that have been conducted over the weekend, although this is not a hard and fast rule, the one polling company that is continuing to release polls week after week (Ekos) is always polling during the weekdays and not on the weekends. So the bin boundary was set to be Friday evening.

A Weighty Issue

That means that every Friday I accumulate all the new polls together from the previous week and perform a straight weighted average. Now, you might be thinking I've taken the lazy man's way out and am ignoring the intricacies of polls. Decided versus Leaners Included. Reliability of polling companies. And so on and so forth.

The truth is I could create a complicated and massive system designed to account for all these issues. I could use the previous two election results to measure the difference between using 'leaners included' and 'decided only' methods. I could estimate the reliability of polling companies (as has been done elsewhere). But fundamentally, such methods are ignoring a few key details.

First of all, they are ignoring the dreaded 'margin of error'. Statistically speaking, you can't get your results more accurate than the margin of error, which says that 95% of the time the value will be within the value plus or minus the margin of error. So a poll that pegs the Tories at 36% with a margin of error of 3% the day before the 2008 election wouldn't have been 'wrong' or a 'bad poll'. It is still within the statistical margin of error.

Secondly, they assume that polling companies are static and that they don't change their polling or weighting methods. If a polling company knows that their polling is systematically wrong one way or another then they will change their polling weighting methods to try to obtain a more accurate one. A biased polling company doesn't get any work.

Finally, I invoke the power of Occam's Razor (otherwise known as the KISS - Keep It Simple Stupid! principle). To prove that a more complicated method of analysis is needed, the standard is placed higher than simply "because we can". Other groups have tried to find systematic biases and errors in polling results only to find their final results off when push comes to shove. The reason, I believe, is because they are using a complicated method, where a simple method would derive similar results in terms of accuracy.

I'll discuss this further later.

Saturday, October 3, 2009

The Future of Warfare

The lights go out.

That'll be the first thing that you notice. You'll look outside and will probably notice that the lights at your neighbours and the streetlamps are out too. Odd though, there isn't a thunderstorm, must have been some grid problem again, you'll think. You're used to power outages though, you've remembered to keep a flashlight handy.

But why won't your cell phone turn back on? That's bizarre. You'll also find that your car probably won't work either. While you don't know it now, its because the on-board computer systems are fried.

Now you may start to panic. Maybe it will cross your mind that you've just experienced an EMP attack, but maybe not. As night falls, and the lights have not yet come back on you'll be wondering how long before the electricity comes back. The answer may surprise you.

Months to years.

In 1999, during a tense moment in Russian-US discussions over NATO actions in the former Yugoslavia, a Russian Duma member declared that "if Russia wanted to harm the US without incurring nuclear retaliation, it would simply launch a single nuclear-tipped missile from a submarine and detonate it high in the atmosphere over the American continent... pulses of electromagnetic energy ... would wipe out the electricity generating and transmission system across a huge portion of the continent." [1]

The damage could take years to repair.

The Americans are well aware of the potential repercussions of such a strike. In 1962, a nuclear test code-named Starfish Prime detonated a 1.4 megaton nuclear warhead 400 km above the Johnston Atoll, causing electrical disruptions 1400km away in Hawaii. [2] To put it in perspective though, North Korea's latest nuclear test was only between 10 and 20 kilotons.

An EMP strike is a very real possibility and one that is being considered by many nations in the world, both how to commit an attack, and how to defend against one. Its hard to imagine a world without electricity for months but its been the fodder for science fiction works for years. While such an attack would produce no immediate casualties, its hard to imagine that it would not produce a nuclear retaliation, if that were still possible. MAD by another name I suppose.

[1] Kramer, D. "US electricity grid still vulnerable to electromagnetic pulses" Physics Today, September 2009.

[2] Day, C. "Very low-frequency radio waves drain Earth's inner radiation belt of satellite-killing electrons" Physics Today, August 2008

Thursday, October 1, 2009

Blowing in the Wind

I say this often, but I can't say it often enough. Wind power is not an alternative energy source. It can not nor will it ever be able, on its own, to replace coal, nuclear, oil or any other consistent source of electricity.

I hear talk however about 'combining' wind with electricity generation from hydroelectric projects. Which again, is just foolishness. The problem stems from the fact that when the wind stops blowing something needs to replace all of the electricity generated by wind not just part of it.

Some environmentalists use the following logic: (the numbers are just for fun - a nuclear station produces far more than this in a year)

Hydro produces 100 kWh in a year
Nuclear produces 50 kWh in a year
Wind produces 50 kWh in a year
Demand is 150 kWh in a year

Wind + Hydro = Demand

Voila! Wind can replace nuclear power, they will say. The problem comes when you look on a day-to-day or even minute-to-minute basis. At one particular instant this may be the case: (again, numbers are ridiculously low and just for fun)

Hydro can produce up to 100 kW
Wind produces 10 kW
Demand is 150 kW

Wind + Hydro < Demand

What happens when electricity supply is less than demand? Well, you stop reading this blog for one thing, because your electricity goes out.

And yet, I still have to deal with environmentalists who insist that wind power can replace nuclear power. For good measure I'll quote an Obama scientist working under Steven Chu, Steven Koonin. In an interview with Physics Today he said: "Wind is now 2% of electricity generated in the US... it will probably get to 20%, but then you start getting into issues of intermittency and transmission... beyond that, I think there are two material options: nuclear fission power and carbon capture and storage." [1]

[1] "Physicist Steven Koonin takes on a new role as DOE's 'technical conscience'" Physics Today, Sep 2009