How Many Licks Does It Take to Get to the Center of the Climate Tootsie Roll Pop?
In this case the tootsie roll pop represents the determination of the human fingerprint of global warming and the licks are the number of years it takes to establish that warming trend. A new paper by B.D. Santer, et al. (2011) has shown that lower atmosphere temperature records must be at least 17 years long in order to discriminate between the weather “noise” and the climate “signal” (human-caused changes). The paper abstract may be viewed here.
The paper was motivated by recent Congressional testimony provided by Professor William Happer. Professor Happer claimed that current climate models are incapable of producing 10-year periods with little or no warming. This claim is demonstrably incorrect.
A special thank you to Dr. Santer for sending me a plain-english summary of the paper which I am using to write this blog post.
Some key points:
Many independent observations convincingly indicate that the Earth’s surface and lower atmosphere are warming.
Oceans are gaining heat, ice is rapidly melting, and the atmosphere is warming. (A wealth of details and illustrations regarding this warming may be viewed on my Web page titled: Modern Day Climate Change.) The Santer et al. (2011) paper analyzes satellite measurements of the temperature of the lower troposphere – the region of the atmosphere from the surface to roughly 5 miles above the surface. (See Skeptical Science for more information about how these satellites function.) Satellite measurements of atmospheric temperature are completely independent of surface thermometer measurements so there is no urban heat island effect nor micro-climate influence (A/C vents, buildings, roads, airports, etc.) because the satellite is in outer space. Consistent with the surface warming inferred from the thermometer record, the satellite data also indicate robust global-scale warming of the lower troposphere. This lower tropospheric warming is roughly 0.9 degrees Fahrenheit (0.5 degrees Celsius) over the full period of satellite temperature records (the last 32 years).
Intentional or accidental cherry-picking of data is confusing the public and policy-makers.
Imagine a cherry tree with 96 red cherries and 4 blue cherries. If somebody asked me or another scientist what type of tree this is I would state: “It is very likely to be a red cherry tree and I am investigating why there are a few blue cherries that do not appear to fit in.” A cherry-picker will just show you the blue cherries to convince you otherwise. (A few examples are illustrated in my post Chopping Down the Cherry Tree.)
The Santer et al.(2011) paper demonstrates that choosing a period less than 17 years to determine climate trend is cherry-picking. This is because short, decade-long temperature records are strongly influenced by the large “noise” of year-to-year climate variability. This “noise” is associated with natural phenomena, like El Ninos and La Ninas.
By looking at longer, multi-decade temperature records, the influence of year-to-year climate noise is reduced, and it is easier to identify a slowly-emerging “signal” caused by human-caused changes in atmospheric levels of greenhouse gases.
Recently, some scientists and politicians have argued that the relatively small surface and tropospheric warming we have observed since 1998 is at odds with our understanding of how the climate system should be responding to the increases in greenhouse gases arising from fossil fuel burning. There have also been highly-publicized claims that computer models of the climate system, when run with human-caused changes in greenhouse gases, cannot produce 10-year periods with little or no warming. This paper shows that such claims are incorrect. Climate models can simulate 10-year periods with minimal warming, even when they are run with historical increases in greenhouse gases.
Dr. Santer offers this excellent analogy:
We would not make inferences about decade-to-decade changes in the Dow Jones index after analyzing records of the minute-by-minute changes in the Dow’s activity over one or two individual days. The “noise” of day-trading tells us nothing about the causes of longer-term changes in the Dow.
The skeptics’ favorite satellite record (UAH) also shows significant warming of the lower atmosphere over its 32 year record
Santer et al. (2011) show that the UAH dataset now has a signal-to-noise ratio of nearly four. In other words, the global-scale warming of lower tropospheric temperature over the 32-year period from 1979 to 2010 is now nearly four times larger than computer model estimates of internal climate “noise” on the 32-year timescale. This means that purely natural changes in climate are highly unlikely to explain the overall warming trend in the UAH data.
Santer et al. (2001) also investigated whether computer model estimates of internal “noise” are systematically smaller than in observations, particularly on decadal timescales. If they were systematically smaller, climate “noise” would be a more plausible explanation for the warming trend we have observed in tropospheric temperature data.
They found no evidence of such a bias. On average, the 22 computer models that they examined actually overestimated the size of observed temperature “noise”. This suggests that the S/N ratios they estimated are conservative and that the true signal-to-noise ratio is probably even larger than four.
Bottom line: If somebody chooses a time period of less than 17 years to try to undermine the evidence that the world is warming you need to understand that one of two things are true:
1) They do not understand the difference between weather and climate
2) They are trying to fool you
Either way, whatever they say after that can be discounted and you can cross them off your list of experts. (If it is #1 please educate them!)