Home > Deconstructing Watts, holocene, paleoclimate > A response to Middleton: Ice Cores vs. Plant Stomata

A response to Middleton: Ice Cores vs. Plant Stomata

2011 January 11

David Middleton posted a pretty interesting piece about high frequency CO2 changes and differences in Holocene mid-latitude CO2 as derived from leaf stomata and that found in ice cores. It seems it was latter picked up by WUWT. Learned lots, but the conclusion of the piece is over-reaching. I had a short chat with Middleton about it. It was clear I would have to be more precise to make my point.

Here is the chart of David Middleton’s that is in question. It seeks to compare recent (1000+ year) temperature history (as defined by Moberg 2005) and CO2 levels as defined by Kouwenberg. Note that there is no error envelope around either series – but the envelope around both is pretty huge. But we will proceed without it.

Moberg and Kouwenberg

Instead of eyeballing the alignment, I have taken his chart and slipped the CO2 series back to line up the dashed line at the end of the LIA circa 1610 CO2 with the corresponding dashed line of temperature at the beginning of the LIA about 1120 temperature. Now we can make some rate calculations for CO2 response to temperatures.

Moberg and Kouwenberg shifted

For pixel interpolation, I found the cells to be about:
28 wide
36 tall

The shift (or lag) displayed is approximately 240 years.
Values rounded to two significant digits.

For comparisons, I divided the series into 4 sections: A, B, C, D.
Section C was defined by Middleton.

Section A

End = 940
Start = 770
Years = 170

End = -0.31
Start = -0.31
Temp = 0

End = 254
Start = 307
CO2 = -53

In section A, for NO change in Temp, we have a change of -53 ppm CO2

Section B

End = 1120
Start = 940
Years = 180

End = -0.02
Start = -0.31
Temp = +.29

End = 318
Start = 254
CO2 = +64

In section B, for a +0.29 in Temp, we have a change of +64 ppm CO2
64/.29 = 220 ppm CO2 for every 1C increase

Section C

End = 1610
Start = 1120
Years = 490

End = -0.72
Start = -0.31
Temp = -0.41

End = 281
Start = 318
CO2 = -37

In section C, for a -0.41 in Temp, we have a change of -37 ppm CO2
-37/-0.41 = 90 ppm CO2 for every 1C increase

Section D

End = 1745
Start = 1610
Years = 135

End = -0.61
Start = -0.76
Temp = +0.15

End = 357
Start = 281
CO2 = +76

In section D, for a +0.15 in Temp, we have a change of +76 ppm CO2
+76/+0.15 = 510 ppm CO2 for every 1C increase

So to recap, CO2 response to 1C change for each section is:
Section A: (no response or slightly negative)
Section B: 220 ppm
Section C: 90 ppm
Section D: 510 ppm

If sections A, B, and C offer a clue as to CO2 response to temperature in the preindustrial world, then the natural CO2 accumaltion due to warming from 1610-1745 should be in the range of 0ppm to 33ppm (220ppm/C * 0.15C). The actual increase in section D is 76ppm which suggests that more than half the CO2 increase by ~1980 (1745 + 235) is from sources other than “a response to rising temperatures 240 years earlier.”

If a CO2 response to temperatures from 240 years earlier is the primary component of the rise in current CO2, we should expect a drop in CO2 accumalation beginning around 2040 corresponding to the drop in temperature just after 1800.

  1. David Middleton
    2011 January 11 at 4:54 am

    Interesting analysis.

    The problem in trying to perform a quantitative analysis is that there are only two major warming/CO2 pulses with which to work. More stomata chronologies are needed; as are multi-proxy Northern Hemisphere temperature reconstructions that go back more than 2,000 yrs.

    There’s no way to know how variable this relationship is. I made a similar chart for the early Holocene using Wagner et al., 1999 stomata and Alley’s GISP2 Central Greenland reconstruction. The lag time appeared to be less than 200 yrs. But, Alley is a local reconstruction; not hemisheric like Moberg.

  2. 2011 January 11 at 6:19 am

    Yeah, I think you have more work to do before you can claim that you can account for all the modern rise of CO2 levels through warming since 1750 – because that is a quantitative claim.

    There’s no way to know how variable this relationship is,

    Sure there is. Maybe not right now, but you should be able to get your hands around the problem. With better models of Holocene temperatures and more complete data about Holocene CO2 levels, you should be able to improve your analysis – maybe even to the point where is will support your claim.

    Another approach is to figure out where the Holocene CO2 is being stored and release with this 240 year cycle. Oceans? Biomass? If you can build a decent model, you can constrain some of the possibilities and fill in some of the gaps.

  3. David Middleton
    2011 January 11 at 12:06 pm

    My bet would be on the oceans.

    I think the answer would obvious if I could plot the Carbonate Compensation Depth against atmospheric CO2 over the last 10,000 years. I just don’t have the data to do that.

    My blog posts are all “works in progress.” Kind of like a cyber-scratch pad. It’s always good to have input from different perspectives. Thanks for taking the time and effort.

  4. 2011 January 13 at 3:02 pm


    No need to know the CC Depth, as that is not important for atmospheric CO2 levels. Only the temperature of the upper ocean mixed layer is important. With Henry’s Law, an increase/decrease of 1°C increases/decreases the pCO2 of the upper 100-200 meter of the oceans with 16 ppmv. Thus a global ocean surface temperature change of 1°C over the MWP-LIA-CWP cycle would give a maximum CO2 change of 16 ppmv in the atmosphere, as that fully compensates the change in the upper oceans. In fact less, as the other side: vegetation will give an opposite change in CO2 sequestering. The ice cores (high resolution Law Dome: less than 40 years resolution) shows a dip of 6 ppmv around 1600, thus hardly a lag at the onset of the LIA.

    Further, the ocean mixed layer has a limited capacity for CO2 sequestering / release: the carbon content is about 1000 GtC (800 GtC in the atmosphere), but a change of 100% in the atmosphere only gives a 10% change of CO2 in the ocean mixed layer, due to changes in the equilibrium reactions between CO2, bicarbonate and carbonate. The exchanges with the deep oceans are limited and probably less climate related.

  1. No trackbacks yet.
Comments are closed.