CarbonBrief – how could they get it so wrong?

In a paper just published in Nature Climate Change entitled “300 years of sclerosponge thermometry shows global warming has exceeded 1.5 °C” (https://www.nature.com/articles/s41558-023-01919-7) Professor Malcolm McCulloch and co-workers (McC-NCC) found that “Global temperatures have already exceeded 1.5°C warming and may pass 2°C later this decade. The worrying findings, based on temperature records contained in the carbonate skeletons, of long-lived marine sponges suggest global climate change has progressed much further than previously thought.”  The paper generated immense public interest and extensive media reporting. With a notable exception, the quality and accuracy of the reporting was excellent with the vast majority of media correspondents having done their homework.The notable exception was misinformation in the CarbonBrief report entitled “Scientists challenge ‘flawed communication‘ of study claiming 1.5 ^oC warming” documented below.

Misinformation in the Carbonbrief report.

1. “Between 1700-90 and 1840-60, the proxy data shows ocean warming of around 0.9 ^oC, according to the study. In the intervening time, there was some cooling, largely caused by volcanic eruptions, the authors say”.    AND

2. The authors have applied a 0.9 ^oC “offset” to their proxy data to account for pre-industrial temperature increase”,  which is repeated in the CarbonBrief Fig. caption.          YET        Contradicting this they also write “The authors assume that the 0.9C offset “can be applied to land-air as well as the ocean mixed layer anomalies”, therefore concluding that GMST increased by 0.9C between 1700-1860 and 1961-90.

Summary As repeatedly stated by McC-NCC “Relative to this extended and now well-defined pre-industrial baseline, an offset of 0.9 ± 0.1 °C is applied (MCC-NCC Fig. 4) to account for the temperature increase from the pre-industrial baseline period (that is, mean temperatures from 1700 to 1790 and from 1840 to 1860) to the 1961–1990 anomaly reference period.”

3.  CarbonBrief state ““However, many experts have warned that the framing of the study is misleading, arguing that the finding has no bearing on the Paris Agreement 1.5^oC limit, because it (the Paris Agreement?) specifically describes temperature rise relative to the late 19th century”.                                 YET    contradicting this they also write “In 2015, countries agreed under the Paris Agreement to hold the increase in the global average temperature to well below 2C above pre-industrial levels and pursue efforts to limit warming to 1.5C . “Pre-industrial” was not clearly defined in the agreement , but it has generally been taken tomean the average temperature over 1850-1900.”

Summary.    Nowhere in the Paris Agreement does it define temperature rise ‘relative to the late 19^th century’ as claimed by CarbonBrief.  As stated, in McC-NCC the ‘pre-industrial period’ is well understood, as being “the stable mean climate state just before human activities started to demonstrably change the climate through combustion of fossil fuels’.                                                                                                  But as opposed to CarbonBrief ‘expert’ opinions, when and by how much global temperatures have risen since the pre-industrial period has significant uncertainties. This is the focus of McC-NCC. For example, it well documented that concentrations of atmospheric CO₂ started to increase in the early-1800s although surprisingly it is contested by a CarbonBrief expert as to whether this is anthropogenic. In McC-NCC we also state that  ‘most limiting’ in defining the IPCC ‘pre-industrial’ “is that instrumental ship-based records of global sea surface temperature (SST) only began in the 1850s and then with limited coverage. Thus, for pragmatic reasons, the earliest available instrumental records from 1850 to 1900 have been used to define the IPCC ‘pre-industrial’ period.”

4. IN a mistaken view of how the pre-industrial is defined and how climate models work,  The University of Oxford’s Prof Yadvinder Malhi, is quoted by CarbonBrief as saying : “Our models of climate warming impacts are based on warming relative to 1850-1900 and moving the baseline definition of pre-industrial does not make these expected impacts worse…“It is the date of the reference period that matters rather than whether it is labelled pre-industrial or not..”

This is an incorrect statement at odds with CarbonBrief’s own description of climate models (https://www.carbonbrief.org/qa-how-do-climate-models-work/.  “Climate models are run for the historical period, from around 1850 to near-present these historical runs are not “fit” to actual observed temperatures or rainfall, but rather emerge from the physics of the model. This means they allow scientists to compare model predictions (“hindcasts ”) of the past climate to recorded climate observations. If climate models are able to successfully hindcast past climate variables, such as surface temperature, this gives scientists more confidence in model forecasts.” hat is climate models are not based on temperature changes over a fixed time period as Professor Mahli incorrectly claims, but to an initial ‘natural’ pre-industrial condition to which various levels and types of anthropogenic forcing are then added. Thus, while climate models are used to assess impacts as a function of global warming, the commencement and magnitude of present-day industrial-warming is dependent on paleoclimate/historical data.

5. Most concerning, is that these serious errors were then repeated by some of the ‘expert’ scientists quoted in CarbonBrief. For example, Dr Friederike Otto, a senior lecturer in climate science at Imperial College London’s Grantham Institute is reported by CarbonBrief to have said without justification “That the 1.5 ^oC limit was established as the threshold of unacceptably dangerous warming and describes temperature rise relative to the late 19th century. If this study has indeed identified warming from before the mid-1800s, that doesn’t mean the planet is any closer to breaking the 1.5^oC limit as it is widely understood.”                                                                                                                                          Similarly Dr Andrew King a senior lecturer in climate science at the University of Melbourne is reported by  Carbon Brief  to have said “that the findings of the study do not have any implications for the Paris Agreement warming limits, because these were written in 2015 with a view to limiting further global warming from that point onwards”. King also says that “the authors have not demonstrated that pre-1850s warming is due to human activity”.

Both Dr Otto and Dr King have misinterpreted the Paris Agreement and seem to assume that the current IPCC definition of the 1850-1900 is an irrevocable fact. This is not the4 case. IPCC regularly updates e.g. industrial temperature estimates (see AR5 versus AR6) as new data becomes available and there is an improved physical understanding of the processes.

6. Dr Duo Chan, a lecturer in climate sciences at the University of Southampton, notes that, “according to ​​Berkeley Earthtemperature estimates, the land warmed by around 0.05^oC per decade over 1850-1900. The new proxy data from the sponges suggests that the ocean warmed almost twice as quickly as the land over this time – a “puzzling observation given the ocean warms more slowly than land”, (https://www.carbonbrief.org/guest-post-why-does-land-warm-up-faster-than-the-oceans/) he says.

The reason is (McC-NCC) straightforward; the 19^th (and 18^th) Berkeley Earth centuries of land warming has larger uncertainties back in time variable. From 1850-1900 are large ±0.5 ^oC changes corresponding to cooler land-air temperatures during La Ninas (e.g. 1862) and amplification of warm El Ninos (e.g. 1877-78).  Most importantly though, despite this variability Berkeley Land temperatures most closely follow the OML ocean record (see figure below). By contrast over the same 1850-1900 period, HadSST4 is significantly warmer (0.3 to 4^oC) than either Berkeley Land or the cooler OML global ocean record.

Additionally as pointed out by McC-NCC (Fig.1 a ,b) the forced IPCC renormalisation of land-air and HadSST4 to 1850-1900, produces an abrupt stepwise permanent warming of the land from the early 1900s, while the oceans cool. This cannot be reproduced in physically constrained climate models.

Conclusion

The extent of inaccuracies/misreporting by CarbonBrief is surprising as they are a well-resourced organisation who have as their byline to be: ‘clear on climate’ claiming to specialise ‘in clear, data-driven articles and graphics to help improve the understanding of climate change’. We look forward to a renewed effort by CarbonBrief to meet these admirable objectives, and where they have failed as described above, make appropriate corrections and clarifications.

Finally, Prof Yadvinder Malhi’s comment: “the way these findings have been communicated is flawed and has the potential to add unnecessary confusion to public debate on climate change” is not our experience nor evident from other commentaries. The public is well informed of the likely consequences of temperatures now exceeding 1.5 ^oC above pre-industrial levels and strongly supports the now even greater urgency to reduce anthropogenic CO₂ emissions to net zero.