Bishop Hill Chalk up another for low climate sensitivity
Jul 18, 2013 Climate: Sceptics Climate: sensitivity Nicola Scafetta has a new paper in Energy and Environment, which finds a figure for climate sensitivity of 1.35, some what lower than even the torrent of EBM papers over the last year or two.
Global surface temperature records (e.g. HadCRUT4) since 1850 are characterized by climatic oscillations synchronous with specific solar, planetary and lunar harmonics superimposed on a background warming modulation. The latter is related to a long millennial solar oscillation and to changes in the chemical composition of the atmosphere (e.g. aerosol and greenhouse gases). How- ever, current general circulation climate models, e.g. the CMIP5 GCMs, to be used in the AR5 IPCC Report in 2013, fail to reconstruct the observed climatic oscillations. As an alternate, an empirical model is proposed that uses: (1) a specific set of decadal, multidecadal, secular and millennial astronomic harmonics to simulate the observed climatic oscillations; (2) a 0.45 attenuation of the GCM ensemble mean simulations to model the anthropogenic and volcano forcing effects. The proposed empirical model outperforms the GCMs by better hind-casting the observed 1850-2012 climatic patterns. It is found that: (1) about 50-60% of the warming observed since 1850 and since 1970 was induced by natural oscillations likely resulting from harmonic astronomical forcings that are not yet included in the GCMs; (2) a 2000-2040 approximately steady projected temperature; (3) a 2000-2100 projected warming ranging between 0.3°C and 1.6°C , which is significantly lower than the IPCC GCM ensemble mean projected warming of 1.1°C to 4.1°C; (4) an equilibrium climate sensitivity to CO2 doubling centered in 1.35°C and varying between 0.9°C and 2.0°C .
There is a preprint here. No doubt there will be the usual barrage of comments arguing that it should be ignored because it's in Energy and Environment. Let us see what a critique based on logic turns up.
Reader Comments (53)
To Paul_K
1)
The model proposed in the paper, which is shown in Figure 16 does not make any quadratic secular trend assumption.
Figure 5 simply shows the temperature data detrended of the observed nonstationary accelerating trend and simply updates a figure shown in one of my papers in 2010. This is standard thecnique of analysis when non stationary data are studies in first approximation, which is the point there.
In fact, in cases of non stationarities, people detrend linear trends or in the case of an evident acceleration as in the studied case, people can take off a quadratic polynimial trend as explained in the paper. Figure 5 is therefore just a simple visual representation of the data after the upward trending is removed to highlight that the period 1880-1940 is modulated very similarly as the period 1940-2000.
I fear that you are interpreting figure 5 out of contest.
Moreover, your criticism "However, it is poor methodology, and just for the record, the quadratic is not “as orthogonal as possible”, although it may be so if one only considers the set of low-order polynomials." is baseless because you cannot accuse a methodology of being "poor" without making explicit the methodology that you consider "good" or "rich", so that a proper comparison could be made between the putative "poor" and the putative "rich" methodology. Tell us, what is your "rich" methodology?
The second order polynomial is quite orthogonal to a sequence that contains two oscillations, as observed in the data because a two-oscillation record would be collinear to a 4-order polynomial and a 3-order polynomial would still be sufficiently collinear (50-75%) to such record. The calculations can be done perfectly because on a 162 year long sequence a 2-order polinomial would be perfectly hortogonal to a sequence with two oscillations of 88-year period. Because the oscillation observed is about 60-year, that is fine enough.
A 2-order polynomial is sufficiently hortogonal to a sequence with two oscillations and would not alter them significantly while capturing the upward acceleration trend better than a linear polynomial. So, you cannot use polynomial larger than the second order for the purpose, and 2-order is better than 1-order. So, the best in this case is using the 2-order.
People could use a Fourier band pass filter but Fourier band pass filter also requires the data to be stationary to properly work and also contains some arbitrary aspect, in any case almost no significant difference would be observed.
Because you are not revealing your "rich" methodology in opposition to my putative "poor" methodology, I fear that you would have complained anything I would have done.
2)
The solar model using the tidal potential as a forcing assumes that the sun responds to the tidal potential by feeling mostly the envelop modulation which has a 61-year modulation peaking in the earlier 1940s and in the earlier 2000s: see figure 7B and compare with the temperature maxima in Figure 5B. And see also figure 13.
Scafetta N., 2012. Does the Sun work as a nuclear fusion amplifier of planetary tidal forcing? A proposal for a physical mechanism based on the mass-luminosity relation. Journal of Atmospheric and Solar-Terrestrial Physics 81-82, 27-40.
Infact the envelop correlates with available solar data for centuries aas demonstrated here:
Scafetta N., 2012. Multi-scale harmonic model for solar and climate cyclical variation throughout the Holocene based on Jupiter-Saturn tidal frequencies plus the 11-year solar dynamo cycle. Journal of Atmospheric and Solar-Terrestrial Physics 80, 296-311.
Your criticism is based on ignoring the sun or in assuming that the sun responds to the tidal potential in a perfect linear way, which is evidently not the case.
Also in music, your brain is able to feel the beat modulation of a sound, doesn't it?
According your argument it should not.
you write:
"If your postulate about a relationship is correct, you should be able to show that (a) the modulated amplitude of your speed of the sun series (after your scaling to temperature) is sufficient on its own to explain the mean variation of the filtered temperature series at 61-year periodicity and (b) the summation of your (scaled) speed of the sun input combined with your (scaled) high frequency cycle input yields series in both mean and variance which are compatible with these series in the instrumental dataset. "
that is mostly done in my 2010 paper:
Scafetta N., 2010. Empirical evidence for a celestial origin of the climate oscillations and its implications. Journal of Atmospheric and Solar-Terrestrial Physics 72, 951-970.
In the present paper note that the speed of the sun also peaks in the early 1940s and early 2000s, as the temperature peaks in Figure 5B. About the 20-year cycle look at Figure 8A where Fourier filters around 20 years are applied to the temperature and the speed of the sun and compared, and the coherence is evident.
So, I fear that you are still misunderstanding the paper also because you are interptreting the paper as if it has to respond to all possible questions one may have. The paper is not supposed to respond all questions, such as the exact physical mechanisms and all details one may think about.
The paper shows that the climate system is made of oscillations that can be recognized as having an astronomical origin, A model based on these oscillations performs better than the IPCC models in interpreting the climatic patterns, so, it produces 21-st century projections that must be considered more reliable by policy makers.
And invites people to further investigate the exact physical mechanisms and other details.
So Anthony Watts chose not to alert his readers to this because it reminded him of barycentric nuttiness.
What's next?
Should the editor of E&E sack himself and apologise to Mr Watts?
Or is the real story that Mr Svalgaard didn't clear it?
at this web-site:
http://people.duke.edu/~ns2002/#astronomical_model_1
you find the latest updated figures comparing both HadCRUT3 and HadCRUT4 global surface temperature records and the IPCC model projections. The figures also show a comparison with the model that I propose which is made of specific harmonics + a significantly reduced anthropogenic effect, which works much better than any IPCC GCMs in reconstructing past temperatures and projects a significantly lower 21st century warming.
All details are found in my publication
Scafetta N., 2013. Solar and planetary oscillation control on climate change: hind-cast, forecast and a comparison with the CMIP5 GCMs. Energy & Environment 24(3-4), 455–496.
http://people.duke.edu/~ns2002/pdf/Scafetta_EE_2013.pdf