To trend or not to trend
David Whitehouse through the Global Warming Policy Foundation has recently released a report stating that “It is incontrovertible that the global annual average temperature of the past decade, and in some datasets the past 15 years, has not increased”. In case it is unclear, both the author and institute are considered sceptics of man-made climate change.
The report focuses on arguing the observation that if you look at only the past decade then there is no statistically significant change in global average annual temperature. Understanding what this does, or doesn’t, mean requires considering two related statistical concepts; 1) significant versus non-significant effects and 2) sample size and power.
Detecting a change is not the same as detecting no change. Statistical tests, indeed most of science, generally operates around Karl Popper’s falsification. Null hypotheses are set-up, generally a statement or test of the form ‘there is no effect’ and the alternative hypothesis is set-up in the contrary ‘there is an effect’. We then set-out to test these competing hypotheses. What is important to realise, however, is that technically one can never prove the null hypothesis, only gather evidence against it. In contrast however one can prove the alternative hypothesis. Scientists generally word their results VERY precisely. As a common example, imagine we want to show there are no sharks in a bay (our null hypothesis). We do some surveys, and eventually one finds a shark. Clearly our null hypothesis has been falsified, as finding a shark proves that there are sharks in the bay. However, let’s say we do a number of surveys, say 10, and find no sharks. We don’t have any evidence against our null hypothesis (i.e. we haven’t found any sharks..yet), but we haven’t ‘proven’ there are no sharks, only that we looked and didn’t find any. What if we increased it to say 100 surveys? That might be more convincing, but once again we can never prove there are no sharks, only demonstrate that after a large number of surveys (even 1,000, or 1,000,000) its highly unlikely there are any. In other words, as we increase our sample size, we have more ‘power’ (a statistical term) to be confident that they represent the underlying truth.
And so in the case of David Whitehouse’s claim we see similar elements. Just because an analysis of the last decade of global temperatures does not find a statistically significant trend, does not prove there is none. It may mean there has been no change, but it might also mean that the dataset is not large enough to detect it (i.e. there is not enough power). Furthermore, by reducing your dataset (i.e. only looking at the last 10 years rather than 30) you are reducing your sample size, meaning you are MORE likely NOT to detect an effect. A cunning statistical sleight of hand to make evidence of a trend disappear.
I lecture these basic statistical concepts to my undergraduate class and demonstrate it graphically. If you put a line over any ten years of data, it probably could be flat, only once you accumulate enough data, say thirty years, does the extent of the trend become clear.
Demonstrates the difficulty in detecting long-term trends with noisy data
This point is actually noted by the report (e.g. see Fig. 16).
Essentially, the only point that the report makes is that if you look at a small part of the dataset (less than a few decades), you can’t make a statistically robust conclusion, since you will be within a low power margin of error. Most importantly, we must be able to detect trends early even when the power to detect them may be low. And as I have stated in earlier posts, changes in variability are as important a metric as changes in the average, and the former, which is predicted from climate change, will make detecting the latter, which is also predicted, even more difficult.
James Russell is a quantitative ecologist jointly appointed in the School of Biological Sciences and the Department of Statistics. He was the 2012 Prime Ministers Emerging Scientist prize recipient. See all posts by James Russell »
But you forget to mention that co2 levels have risen significantly in the ten years.If the chicken little brigade were to be right ten years ago we would now be having 10 million climate refugees instead of zero or snow would now be a rare and exciting event. Instead we argue if the last ten years hiatus is significant or not. The hypothesis of runaway global warming is surely on shaky ground. How many more years of co2 increases and flat temperatures will we need to lay the green madness to rest and go back to cheap electricity and research funds spent on things like clean water and the eradication of malaria.
11 years ago
Your statement about snow shows a lack of understanding about global climate change. A rise of 2C in global average temperature won’t eliminate snow; in fact, in some locations we will see a tremendous increase in snow (see: New England, eastern Canada, etc.). More precipitation in the still-frozen winters will lead to more snow, not less. You should look up some of the recent reports put out by Hansen (NASA): summed up, the reports say global climate change will primarily contribute more *extremes*, in both hot and cold, than some naive “everything is warmer” behaviour.
And as far as global temperatures go, a large forcing effect on the temperatures is the sun (see “Dependence of global temperatures on atmospheric $CO sub 2$ and solar irradiance”, Thomson, 1997, PNAS), and we’ve been having a strangely prolonged minimum for the last decade. Coincidentally, this ties directly into the period which you are claiming shows that global climate change is not happening. When the sun hits a true maximum again, all that extra CO2 we’ve pumped into the atmosphere is going to accelerate the temperature climb. The forecasts call for a completely ice-free Arctic in the next couple of years, something we have never seen. The glaciers are melting at an unprecedented rate. At some point, denying the obvious isn’t possible anymore, even if you want to be a skeptic.
Finally, “chicken little”? Really? When 98% (see survey by PNAS: http://www.pnas.org/content/107/27/12107.full) of climate scientists agree that global climate change is the correct scientific conclusion, it’s not the 98% who need to defend their position, it’s the remaining 2% of denialists. Most of the big names in climate change denial aren’t even scientists anyway, which is somehow indicative.
11 years ago
It wasnt my quote about snow it was from Dr David Viner a senior research scientist at the CRU in 2000. It shows how badly wrong this field gets it. Willing to put out scare stories but didnt have the skill to predict the last ten years hiatus in the face of rapidly rising co2. Did you find anything to cut and paste about the 10 million climate refugees that we supposed to show up by 2010 but didnt make it? The only thing thay went up was property values in Tulavu. The notorious pnas survey. What a sad joke. I would love statschat to run the slide rule over that “survey”. And as for that crackpot Hasen equating coal trains with jewish death trains he has lost all perspective. Vested interests and noble cause corruption.
11 years ago
http://thebentangle.files.wordpress.com/2011/11/skepticsvrealistsv3.gif
I’m sceptical of your stats reasoning in this post though. Finding no sharks does support the null hypothesis. It just doesn’t make it certain.
11 years ago
“The report focuses on arguing the observation that if you look at only the past decade then there is no statistically significant change in global average annual temperature.”
No it doesn’t.. from start, middle and end it hammers the point that there have been 15-16 years of no significant warming. In particular it notes that the start of the current 30 odd years of warming consist of a warming period from about 1980 to 1997 followed by a “plateau” 1997-2012where no statistically significant warming occurred.
He further shows that there have been many historic periods where warming stalled for about a decade but none of the climate models can account for a fifteen year stall, ie, this is important in the history of climate modelling and observations since about 1850.
His use of a decadel range is really in response to the Met Office (and others) now using decades to show warming since 1980, but he then shows you get a different picture if you break up temps into 5 year groups that show warming is not related to human defined decades but is instead a more chaotic system in action.. he is arguing *against* defining warming, stasis or cooling by decade and that climate warming hysteria is based, not on a 30 year period, but just 17 years.. a near similar period to the warming plateau.
On a somewhat rueful note I observe the many, many statements of “Worst drought in 70 years!”, “Biggest cyclone since 1880!” or “highest temperature since 1879!” as proof that its all happened before.. and *before* CO2 was considered a problem.
JC
11 years ago
The figures of “worst since some time in the 19th century” are usually qualified by “when records began”; so they hardly constitute agregate proof of “it happened before”. (Sure /some/ of the records go back further, but not most of them.)
Arguing that the temperature rise from 1980 to 1997 is not significant because it’s only 17 years is missing exactly the point James Russell is trying to make: it’s easy to prove H1 and hard or impossible to prove H0.
Unless I’m reading things wrongly, pretty much /any/ interval that includes 1980 to 1997 shows a statistically significant temperature rise, especially the 30-year benchmark.
(It seems to me that the current “temperature stability” is an illusion: there has been a continuing build up of thermal energy, it’s just been going elsewhere: melting ice, which I point out is a finite and dwindling resource.)
11 years ago