Why don't people realize that long term data is often predictable when the short term is not?

Question

Discussions around here often focus on this year or that. Or claims are made that, since we can't always predict the weather, we can't predict climate change.

But, in science, it is frequently true that short term data is not predictable, but long term is.

A classic example is radioactivity. Take a piece of lead-210. Look at individual atoms with an electron microscope. It is utterly impossible to predict whether or not any atom will decay in the next hour (although many of them will).

And nothing in science is more certain than that, in 22.3 years, half of the atoms will have decayed.

Global warming data is very much like that. Individual years are not predictable, but the long term average is. This graph is educational:

http://data.giss.nasa.gov/gistemp/graphs/Fig.A2.lrg.gif

MT ZION - Excellent thinking, and quite right, except for your guess as to what the data shows. The averaged data shows exactly the steadiness you're looking for, followed by a sharp increase. The right hand edge of this graph is basically a vertical line up about 0.2 degrees past the 2004 point. In the past 10,000 years, the average varied slowly by a few tenths of a degree. In the last 30 we've pushed it up about half a degree, which is way past unusual.

Oops, forgot the graph.

http://www.globalwarmingart.com/wiki/Image:Holocene_Temperature_Variations_Rev_png

The black line is the average, of course.

Ben O - It's not surprising that global warming is proceeding faster than this very conservative body predicted.

"It's a consensus document and it is conservative particularly from the perspective that it cuts off the science in early 2005 and a lot has happened since then. Of course, being a consensus document, a lot of the material that I think is reasonably well-supported also gets weeded out through that process. If the IPCC says it you better believe it and then leave room to think it is actually a lot worse than they have said." Tim Flannery

Answer 1

It comes from a lack of knowledge of statistcs.

Just as an unbiased coin flip is 50/50. Some people would assume that 10 heads in a row would means heads are more likely. Some would say that now a tails is due. Neither is correct. The next flip is still 50/50.

Long term the coin will flip 50/50. Short term variances prove nothing.

We as human want instant evidence. Scientific research depends on the statistical significance of the data. The more varied and erratic the data, the more important the long term data becomes. Weather/climate is a perfect example (much better that my coin flip thing.)

Answer 2

Because it doesn't support their opinion. It's as simple as that. Most people come to this site with their own values and beliefs. The problem with values and beliefs is that they aren't objective. So, showing someone that already has strong beliefs data that doesn't support those beliefs is a non-starter. It's almost impossible to change their core values and beliefs, no matter how valid the data.

Answer 3

That sounds like an apology on behalf of climate scientists for being wrong about 2007.

Apology accepted.

By the way, that link is an excellent graph for supporting AGW. It shows 2005 being hotter than 1998 and shows the current rate of warming to be much higher than the IPCC could.

Bob,
It's also not surprising that you personally are stating that global warming is occuring faster than the IPCC says it is.

Answer 4

A temperature graph renders sort of a fractal image. No matter how close or far you stand, the graph looks just as 'noisy'. If you look at the graph you referenced the temps move up and down but seem to hug a trend line. Were this graph possible to reconstruct backwards through the centuries, and even millenia, with the same methodology used to construct the last hundred years, a hundred year period broken into years would look just as noisy as a hundred thousand year period broken into thousand year increments.

Answer 5

It does get frustrating when people keep assuming that short-term weather and long-term climate are the same.

You could make any number of analogies. If you gamble in a casino there's no way of knowing if you'll come out ahead or behind on any given bet or any given day, but if you keep betting, in the long run you will almost surely lose money.

dogs' coin flipping exmple is good. If you flip a coin, there's no way of knowing if it will come out heads or tails. If you flip it 100 times, you'll come pretty close to 50 heads and 50 tails.

With the stock market it could fluctuate wildly on a day-to-day basis and be completely unpredictable, but in the long run it will go up.

The same is true of weather and climate. Any given day or month or year is virtually impossible to predict in terms of temperatures or weather, but in the long run we know that human forcings will dominate and the temperature will steadily increase, as it has for the past 30 years.

Answer 6

Bob is correct as science is all about averages and not one off individual events.

Answer 7

Because people believe what they want to believe, regardless of the evidence or facts.

the prevalence of theists would be a good example here.

Answer 8

Bob - That's a great way of saying "Don't Hold Us to Facts, Just Trust Us."