by Judith Curry
So, exactly how long has it been warming?
The IPCC AR5 made a very strong statement regarding the attribution of recent climate change:
It is extremely likely that more than half of the observed increase in global average surface temperature from 1951 to 2010 was caused by the anthropogenic increase in greenhouse gas concentrations and other anthropogenic forcings together. The best estimate of the human induced contribution to warming is similar to the observed warming over this period.
I have been arguing that the IPCC’s attribution arguments are unconvincing unless they can also explain the early 20th century warming, and the longer period of overall warming prior to the 20th century.
Early 20th century warming
Consider Figure FAQ 10.1 from the AR5:
It is seen from the figure with both natural and human forcing that climate models simulations agree with observations very well during the period 1970-2000. There is significant disagreement during the following periods:
- 2000-present (the so-called hiatus)
- 1940-1970 (the so-called grand hiatus)
- 1910-1940
The IPCC AR5 doesn’t have much to say about the early 20th century warming or the grand hiatus (Section 10.7.1.1):
The AR4 concluded that ‘A substantial fraction of the reconstructed Northern Hemisphere inter-decadal temperature variability of the seven centuries prior to 1950 is very likely attributable to natural external forcing’. The literature since the AR4, and the availability of more simulations of the last millennium with more complete forcing, including solar, volcanic and greenhouse gas influences, and generally also land use change and orbital forcing) and more sophisticated models, to a much larger extent coupled climate or coupled earth system models, some of them with interactive carbon cycle, strengthens these conclusions.
Well the devil is in the details – further, if we are to be convinced by the AR5 attribution of what is essentially a strong warming period of 30 years, then unexplained periods temperature variability of 30 years are significant.
The AR4 Chapter 9 provides some illumination:
Modelling studies are also in moderately good agreement with observations during the first half of the 20th century when both anthropogenic and natural forcings are considered, although assessments of which forcings are important differ, with some studies finding that solar forcing is more important while other studies find that volcanic forcing or internal variability could be more important.
It is ‘reassuring’ that all the AR4 models do a reasonable job of reproducing these features, when they are allowed to select their external forcings to . . . reproduce these features. In the AR5, the selection of external forcing data was more objective, which accounts for slightly poorer agreement between the CMIP5 simulations and the observations.
Lets take a look at the early warming period, from 1910-1940. This period accounts for ~40% of the total observed warming since 1900. How much of that warming can be explained by CO2?
From NASA GISS, the CO2 concentration in 1910 is 300 ppm, and in 1940 is 311 ppm. So of the 100 ppm increase in CO2 since 1910, only 11% of this increase occurred during the period 1910-1940 which comprises 40% of the surface temperature increase since 1900. (aside: 25% of the anthropogenic CO2 has been emitted since 1998, which is a period of warming hiatus).
And while you are considering the above figure, check out the ‘hiatus’ in CO2 increase that occurred 1940-1960; its difficult to ignore that this could be driven by the grand hiatus in warming. (Note: Ralph Keeling was very interested in this feature).
Clearly something else other than CO2 has been the predominant cause of the warming 1910-1940, and climate models do not include this effect since they don’t reproduce the magnitude of the warming.
In terms of explaining this period of warming, the stadium wave argues: 1910-1940 (warming), 1940-1975 (cooling), 1975-2001 (warming), 2002- present (cooling) – against a background secular warming trend.
Earlier historical records
Insight into the length of the secular warming trend is provided by historical temperature records prior to the 20th century, I highlight two data sets here.
The first is the Central England Temperature record (Wikipedia) going back to 1659:
And Tony Brown’s CET reconstruction going back to 1538.
Berkeley Earth has attempted a historical reconstruction of global land temperatures back to 1760 (which is dominated by observations in Europe and eastern North America prior to 1900).
Substantial variability is seen on timescales of 30 years. As for a secular warming trend, the Berkeley Earth analysis shows a warming trend back to 1800, with considerable variability in the late 18th century. The CET analysis shows considerable variability particularly in the 17th century
Some of the variability can be attributed to large volcanoes that occurred around 1800 (AR5 fig 18.12)
This was also a period of the grand solar minima: 1645 – 1711 (Maunder) 1791-1825 (Dalton).
400(?) hundred years of warming
Figure 5.7 from the AR5 compiles a range of paleoclimate reconstructions for the past two millennia:
The borehole reconstruction (bold red) shows a striking secular increase since 1500. The other reconstructions show substantial variability, but all show an increase since 1700 and most of the NH reconstructions show an increase since 1600.
AR5 Section 10.7.1 provides some convoluted reasoning for arguing that all these variations are pretty well understood, and that internal variability and solar variability have small influences.
Consider AR5 figure 10.19, where the orange curves represent the climate model simulations and the other curves represent ‘selected’ reconstructions.
Now compare figure 10.19 with figure 5.9 above. What to conclude from the differences? Well the take away message seems that given the large range of paleoclimate reconstructions, you can cherry pick them to agree ok with your model simulations.
Implications for attribution arguments
So, what does all this mean for IPCC’s ‘extremely likely’ attribution statement for the warming since 1950?
- There have been large magnitude variations in global/hemispheric climate on timescales of 30 years, which is the same duration as the late 20th century warming. The IPCC does not have convincing explanations for previous 30 year periods in the 20th century, notably the warming 1910-1940 and the grand hiatus 1940-1975.
- There is a secular warming trend at least since 1800 (and possibly as long as 400 years), that cannot be explained by CO2, and is only partly explained by volcanic eruptions.
The combination of these two points substantially reduces the confidence that we should place in attribution statements of warming since 1950. Getting the ‘right’ answer (i.e. explaining the warming from 1970-2000) for the wrong reason (i.e. CO2) simply has not been eliminated as a serious consideration, and hence an ‘extremely likely’ conclusion is highly inappropriate.
So, what could be the cause of a 200 – 400 year period of secular warming? The obvious places to look are to the sun and the ocean. Sean Lovejoy dismisses the idea of a ‘grand oscillation’ on these time scales. Sun-climate connections are receiving renewed interest, as evidenced by the NAS Workshop Report The Effects of Solar Variability on Earth’s Climate. During my visit to Oxford last summer, I met with oceanographer David Marshall, who reminded me that the influence of the oceans on climate starts to get interesting at timescales around 1000 years.
In terms of trying to unravel forced from unforced, and natural versus human caused variability, it seems that sorting out what is going on during the grand hiatus (1940-1975) is particularly important, with a major clue provided by the CO2 hiatus (1940-1960).
We need to keep working the historical data record to build credible observational data sets back further in time. And we need more and better paleoclimate proxies, which means we need more robust (calibrated proxies) and more scientists in the field actually collecting samples.
The politically driven push to manufacture a premature consensus on human caused climate change and create an argument based on bootstrapped plausibility has misdirected climate science for the past two decades. The hockey stick attempted to wipeout secular variations prior to the 20th century, but even Mike’s Nature trick spliced the early 20th century warming as an integral part of the blade. At most, only a small fraction of the early 20th century warming was caused by CO2 (this issue was recently addressed in a post by Vaughan Pratt).
And finally, someone in the IPCC needs to read more than one chapter; there are a lot of inconsistencies in the different chapters.
Filed under: Attribution, Data and observations