Westonpa has mentioned Murray Salby before in the climate change debate. My attention was drawn to him by Luboš Motl's Reference Frame Blog.
Murray Salby is theorising that CO2 concentration in the atmosphere related to the integral of a temperature anomaly, multiplied by a fudge factor.
"Presentation Prof. Murray Salby in Hamburg on 18 April 2013"
CO2 concentration is definitely an integral effect and relates to the underlying phenomena somehow (including human emissions), but does what he is saying lead us to progress in thinking more productively about exactly how this is occuring?
What he seems to be saying that when the average earth temperature (with soil moisture being of secondary influence) is above a certain reference temperature, net CO2 is emitted from the earths surface, and when it is less, net CO2 is absorbed. He doesn't say how he chooses his reference temperature from which to measure the temperature anomaly. Which basically means (as he admits) there are considerable intrinsic uncertainties to his analysis.
He uses the air temperture measuements as a proxy for ground surface temperature measurements. Ground surface temperatures vary a great deal more between night and day than air temperature. I suspect the size of these variations between night and day (the very high frequency element) has an influence to play on net CO2 absorption or emission as well.
He says this occurs on all timescales and analyses the ice core data using frequency domain mathematical analysis to show this.
His main conclusion in this regard seems to be that CO2 concentrations are systematically underestimated in a way that can be mathematically related to ice core depth, at both the high frequency end (due to diffusion) and at the low frequency end (due to non-conservative effects he doesn't explain).
According to his analysis there is a sort of band pass filter effect going on that changes with ice core depth, which leads to a systematic underestimation of our reconstructions of the historic atmospheric CO2 concentrations. It is very hard to discount this possibility without understanding the physics and chemistry of the situation very well and showing the effects he refers to (non-conservative effects and diffusion effects) are small.
However does the intrayear variations, rising and falling as they do throughout the year in cyclic form, follow this law as well? He leaves analysis of this out.
Is there anyway of testing his ideas on short timescales? Any thoughts?