Contrarian Manipulation of Paleoclimate Data

 

Now that 2023 is in the books, it's pretty clear that all global temperature datasets will show 2023 to be the warmest year on record by a significant margin. And since 21st century has exceeded the warmest temperatures of the Holocene (the current interglacial), it's also fair to say that 2023 has likely been the warmest year at least since the last interglacial ended 125K years ago. I'm going to wait until all the major datasets are updated to 2023 before saying more about 2023, but Patrick Moore posted a tweet with the above graph and this:

The claim is being made that “2023 was the hottest year in 125,000 years”. This is an outright lie. The Holocene Climatic Optimum from about 10,000-5,000 years ago, when the Sahara was green, was warmer than this Modern Warm Period.

If the graph is intended to demonstrate that 2023 isn't the warmest of the Holocene, it fails. In fact, the graph is a convoluted mess. The CH4 time series is attributed to Kobashi et al 2007, but it looks like it actually comes from either from Grachev et al 2007[1] or Monnin et al 2004[2]. I'm guessing both the CO2 and CH4 time series come from the latter. The "model temperatures" are strangely put on a different temperature scale, but they actually do come from Liu et al 2014, which was a very important paper on the "Holocene Temperature Conundrum." The paper discusses the discrepancy between the models (which generally show slowly increasing temperatures throughout the Holocene with increasing CO2) and Marcott 2013, which shows slowly cooling temperatures from a warm HTM about 6500 years ago.

Liu et al 2014

The conundrum is likely caused by seasonal biases affecting the proxy reconstructions, and this has been understood for some time and discussed in the literature. Recently, Samantha Bova developed a technique to correct for these biases in reconstructions[4], and newer reconstructions like the reconstruction by Osman et al 2020[5] shows the Holocene temperatures to be closer to the those expected by models and CO2 proxies.

Patrick Moore doesn't tell you any of this, though. Instead he shows a time series that he attributed to Marcott et al 2013. The problem, though, is that this time series doesn't come from Marcott. It's actually a manipulation of Marcott's proxy data performed by Javier Vinós. Vinós took Marcott's proxies and manipulated them to make the HTM warmer. In the Vinós' manipulation, GMST changes by about 1.4 C. In Marcott's reconstruction (see below) warmest and coolest portions of the preindustrial Holocene differ by about half that amount. And of course, Moore doesn't put the instrumental record on the graph either. The graph ends in 1950, so it's missing the last 73 years. It's amazing to me that he thinks he can show that it's a "lie" that 2023 is the warmest of the last 125K years while displaying a graph that does not show 2023.

Marcott 2013 with the Instrumental Record

In fairness to Moore, I've seen several versions of this graph in various places, including Andy May's blog, Judith Curry's blog and in a manuscript by Nicola Scafetta. Perhaps the dishonesty is not on his part but on the part of others. But I don't think that lets him off the hook. Moore (and other contrarians) want us to think of them as skeptics, but at best, Moore wasn't skeptical here. He took a dishonest graph and used it to make an invalid point. And anyone can see that the graph doesn't include modern temperatures (it ends in 1950) and does not place 2023 on the graph. In point of fact, Marcott et al 2013 probably makes the HTM too warm, but even if we use Marcott instead of Osman, temperatures have increased above the warmest temperatures of the Holocene.

References:

[1] Grachev, A. M., E. J. Brook, and J. P. Severinghaus (2007), Abrupt changes in atmospheric methane at the MIS 5b–5a transition, Geophys. Res. Lett., 34, L20703, doi:10.1029/2007GL029799.

[2] Eric Monnin, Eric J Steig, Urs Siegenthaler, Kenji Kawamura, Jakob Schwander, Bernhard Stauffer, Thomas F Stocker, David L Morse, Jean-Marc Barnola, Blandine Bellier, Dominique Raynaud, Hubertus Fischer. Evidence for substantial accumulation rate variability in Antarctica during the Holocene, through synchronization of CO2 in the Taylor Dome, Dome C and DML ice cores. Earth and Planetary Science Letters, Volume 224, Issues 1–2 (2004): 45-54. https://doi.org/10.1016/j.epsl.2004.05.007.

[3] Liu, Z., Zhu, J., Rosenthal, Y., Zhang, X., Otto-Bliesner, B. L., Timmermann, A., … Elison Timm, O. (2014). The Holocene temperature conundrum. Proceedings of the National Academy of Sciences, 111(34), E3501–E3505. doi:10.1073/pnas.1407229111.

[4] Bova, S., Rosenthal, Y., Liu, Z. et al. Seasonal origin of the thermal maxima at the Holocene and the last interglacial. Nature 589, 548–553 (2021).
https://doi.org/10.1038/s41586-020-03155-x

[5] Osman, M.B., Tierney, J.E., Zhu, J. et al. Globally resolved surface temperatures since the Last Glacial Maximum. Nature 599, 239–244 (2021). https://doi.org/10.1038/s41586-021-03984-4