Global Warmed

Despite a song that claimed it was a fact that there were 9 million bicycles in Beijing, there really aren’t. They’ve moved to cars and are probably already regretting it.


Greetings from Ho Chi Minh City, which had a different name the last time I visited. It’s been a while. They don’t have 9 million bicycles here, either. But it seems as though they have at least that many motorcycles. Moving seemingly at random. To cross the street just look down at your feet and trust their keen eyes and reflexes. It works just fine–most of the time.

It’s warm outside as I write this in an air conditioned room in a holdover French hotel. At 10:00 this morning it was 93 degrees Fahrenheit, or 34C, if you prefer. That’s 4C lower than the average temperature for April, so I suppose I should feel relieved.

There are 6.65 million people living here. As temperatures have warmed over the past 30 years, how has it affected the Vietnamese here in the city formerly known as Saigon?

Their life expectancy has increased from 42 to 79 years of age. Average income has climbed from $100 per person per year to $1,130 between 1986 and 2010. Poverty has decreased from 58% in 1993 to 29% in 2002.

Aside from the statistics, what anyone can see in a taxi ride from the airport to the hotel is people living a very normal, if very different life. Whatever problems they face, and they face a myriad, climate and climate change is not very high on the list. They look animated, healthy, vibrant and completely engaged in life.

Perhaps if temperatures once again start climbing after a 16-year break, our great-grandchildren might end up living like the Vietnamese. At the risk of once again uttering heretical statements, on a Saturday afternoon it looks to this observer as if it might end up an improvement.

10 responses to “Global Warmed

  1. OFF TOPIC: Occassionally on this blog some one refers to a “left” that I don’t even recognize. Here are 2 left sites that have articles on CAGW.
    The first carries a wider range of viewpoints on AGW than any other site I frequent. The second is very anti CAGW. I am not saying that I agree with these sites. I am saying that they are the real thing. This is what the real left is saying on CAGW.

  2. Tom,

    glad you’re keeping warm in Ho-Chi-Min city.

    Just wondering, tho, when did the “16-year” hiatus actually start? It seems like we’ve been calling i the “16-year” hiatus for an awfully long time. When do we go to 17 years? 🙂

    • Hiya Jimmy–you’ll have to go after the chart-obsessives on that one. My personal belief is that you want 30 years at a minimum to call a trend (well, to be honest it would be 38), but hey, what do I know? It’s just a living…

      • Aye, I sometimes wonder if climate science predictions are like stock market predictions: whatever it’s doing, it’s safe to say it’ll keep doing that, at least until it changes.

        But I really want to know the exact date and time that the yearly global average temperature starts and ends. It’s no fun cheering if you don’t know when the game ends.

      • Tom,
        According to the stats experts at Lucia’s, it is really about ~13 years to statistically falsify the climate trend.

      • I actually disagree with the experts, then. I got taught differently, I guess. I would not take less than 38 for a long term trend. But, hey–that’s just me. Well, me and the people who trained me.

      • Nullius in Verba

        The time required to recognise a trend is something that ought to be calculated from the characteristics of the signal and noise.

        Trend calculation assumes the data has the form of a linear trend plus random noise with a certain power spectrum. A linear trend add a component that looks like a 1/f^2 graph that spikes upwards at f = 0, and the noise looks like a wiggly line that dies away to zero as you get closer to f = 0. Taking a finite segment of data ‘blurs’ the spectrum, blurring detail shorter than one over the length of the time series. So by estimating a trend, what you’re doing is looking for the bit close to zero in the spectrum where the noise is much smaller than the 1/f^2 spike belonging to the trend. And ‘enough data’ means little enough blurring that you can clearly see this region.

        If told what the noise function is, and in particular how rapidly it approaches zero as frequency gets smaller, a statistician can work out how much data you need. But if you don’t know what the noise function is, and have to estimate it from the data itself, then it’s hard to tell whether lumps in it are part of the signal or part of the noise. Make one assumption (like AR(1) noise that approaches zero fairly fast) and you’ll get one answer. Make another assumption (like ARIMA(3,1,0) that doesn’t approach zero at all, but is really only an approximation to something that approaches it quite slowly) and you get a much bigger answer. With one assumption you can pick up trends in a decade or two, with the other assumption it might take centuries.

        That was what the famous “unit root” question was about. The usual statistical tests say the noise approaches zero so slowly that we don’t have enough data yet, even after 150 years, to detect a trend. But those tests can’t distinguish signal from noise (there being no objective difference between them – ‘signal’ is just those components you are interested in and ‘noise’ is those components you aren’t interested in).

        This means that the tests merely confirm whatever statistical assumptions you start with. If you assume there’s a trend, you’ll find one. If you don’t, you won’t. The data fits, either way. The truth is that the data cannot tell us. You need some other external source of information (like the physics) to tell you what noise model to use, and then the evidence for your conclusion is mostly about that external evidence you used, not the data/trend.

      • Aye, I’m with NiV, hard to know what amount of time is necessary to call a “trend” in the statistical sense. My interest lies in a simpler problem: when does the period start and end?

        I’m not sure the length of such a trend can – or needs to be – specified. Here at the 16-yr mark, even the most faithful climacolytes are wiggling a little in their chairs. If 16 yrs makes them wiggle a little, then 17 years will make them wiggle more; at 20 years we’ll have to start providing sedatives. There’s a scientific battle to be fought for sure, but the political battle doesn’t depend directly on the science. It depends on how people percieve the science and the scientists. The longer the flatline goes on, whatever its significance, the less pressure there will be for foolish policies. One year makes a difference.

  3. The WSJ likes your revenue neutral carbon tax. I guess that make you a conservative officially.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s