select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to sponge skeleton data and passing 1.5C

A study published in Nature Climate Change looks at sponge skeleton data and global warming exceeding 1.5C.


Prof Joeri Rogelj, Director of Research at the Grantham Institute – Climate Change and Environment, Imperial College London, said:

“It’s exciting to see new research that allows us to peek centuries in the past and imagine how warm it was in the Caribbean when Napoleon was conquering Europe. However, the suggestion that this research directly changes how we see the goals of the UN Paris Agreement is quite a stretch. Countries use global temperatures as a proxy for climate impacts, and the well-below 2°C and 1.5°C limits of the Paris Agreement define changes in climate impacts that countries aim to avoid. Relabelling the warming that has occurred until today by using a different starting point does not change the impacts we are seeing today, or the impacts we are aiming to avoid.”


Dr Robin Lamboll, Research Fellow at the Centre for Environmental Policy, Grantham Institute – Climate Change and Environment, Imperial College London, said:

“This paper adds to the considerable body of evidence that humans were causing slight warming before industrialisation, but at a much slower rate. However the definition of ‘pre-industrial’ used in estimates of climate damages by the IPCC and implicitly used in international agreements excludes this warming, being based on more recent and higher-confidence measurements. While interesting, the results of this paper do not have legal ramifications for the Paris Agreement.”


Dr Shaun Fitzgerald FREng, Director of the Centre for Climate Repair, University of Cambridge, said:

“The authors state ‘Thus, the opportunity to limit global warming to no more than 1.5 °C by emission reductions alone has now passed and at current emission rates, the 2 °C threshold for GMSTs will be reached by the late 2020s.’ However, the strict target is 1.5C by (or rather at) the end of the century. If we agree that this is indeed the target, then the opportunity to limit global warming to no more than 1.5 °C has in fact not necessarily passed. It looks like it has passed if the only lever being considered is emissions reduction, but there are other levers too such as greenhouse gas removal. Various approaches are being developed and if scaled up quickly could help keep the target alive. Furthermore, there are other climate intervention approaches which are still in relatively early stages of research but if these are developed and found to be effective then the degree of overshoot could also be reduced.”


Prof Yadvinder Malhi FRS, Professor of Ecosystem Science at the Environmental Change Institute, University of Oxford, said:

“The way these findings have been communicated is flawed, and has the potential to add unnecessary confusion to public debate on climate change. Despite the headline, the results of this paper do not show that we have already exceeded the Paris climate targets.

“This paper presents a detailed sclerosponge record from the Caribbean that shows that ocean surface warming (something they term industrial-era warming) in this region began in the 1860s, a time earlier than the existence of good ocean temperature records. This finding is used to argue that the pre-industrial reference baseline should be pushed back to the period from the 1700s to 1860, rather than the 1850-1900 period usually used by the IPCC and others. With this new baseline, an extra 0.5C is added to our estimates of industrial-era warming, and hence the studies infers we are already well past 1.5C of industrial era warming and approaching 2.0C.

“While the Paris climate targets set reference warming targets against pre-industrial baselines, there is an implicit assumption this is dominated by human-caused warming caused by the industrial era. It is unlikely that warming of 0.5C in the 1800s is human-caused. By 1900 cumulative human fossil fuel/cement carbon emissions since 1750 were a mere 2.5 % of our cumulative emissions by 2021 (Source: Our World in Data). If we include land use change (much more uncertain) another few percent might be added. This is unlikely to have caused substantial warming compared to the 1.4C of warming caused by the remaining 97.5% of cumulative emissions.

“Hence this early industrial era warming, if real, is almost certainly not human-caused, is not industrial-caused warming. Our models of climate warming impacts are based on warming relative to 1850-1900, and moving the baseline definition of preindustrial does not make these expected impacts worse. Human-caused climate warming to date is still around 1.4C. It is the date of the reference period that matters rather than whether it is labelled pre-industrial or not. The period 1850-1900 is a period of relatively reliable global data when industrial era human-caused climate change was likely negligible.”


Dr Duo Chan, Lecturer in Climate Sciences at the University of Southampton, said:

“McCulloch et al have presented intriguing research using sclerosponges from the Caribbean Sea to infer global sea-surface temperatures (SSTs) over the past 300 years. They argue that these proxies indicate ocean warming began in the 1860s, up to 0.6°C cooler than existing instrumental estimates for 1850. This leads to an implication of a more rapid increase in global warming, reaching 1.7°C above pre-industrial levels by 2020.

“Exceeding the 1.5°C target is certainly headline-worthy, emphasising the significant impact of even a 0.1°C change in global warming levels. As I am not a paleo-geochemist, I cannot offer specific insights into the Sr/Ca measurements. However, from my perspective as a physical climate scientist, I advise caution in interpreting these findings, primarily for two reasons. 

  1. This new warming estimate does not align with historical estimates of external forcing. According to Haustein et al. (2019), the external forcing trend from 1850 to 1900, excluding volcanic eruptions that has a cooling effect, is 0.03W/m^2 per decade, with an SST warming estimate of 0.09°C per decade when including volcanic eruptions. In contrast, from 1970 to the present, external forcing has increased at a rate of 0.48W/m^2 per decade, but the SST increase is estimated at only 0.15°C per decade. 
  2. If we assume that the Berkeley Earth temperature estimates, which rely on a limited number of land station records, are reliable for the period 1850–1900, then land warming during this time was about 0.05°C per decade (as inferred from Figure 1a). This suggests that, according to the sclerosponge-based SST, the ocean warmed almost twice as fast as the land, a puzzling observation given the ocean’s large heat capacity. Interestingly, the authors use similar reasoning to argue that ocean warming outpaced land warming in the early 20th century.  It is important to note that their SST estimates are based on two critical assumptions: (1) SSTs in the Caribbean Sea are a reliable indicator of global SSTs, and (2) the calibration of sclerosponge Sr/Ca against HadSST4 from 1960 to 2020 remains consistent over time. If either of these assumptions is incorrect, it could result in an overestimation of early warming.

“Although I approach the quantitative implications of the sclerosponge record with caution, I acknowledge its contribution in shedding light on the qualitative aspects of SST evolution. A key question is whether historical SSTs exhibited “roller coaster”-like variability on a global scale, in contrast to the steady and smooth warming depicted in instrumental land records, which aligns with the evolution of external forcing both qualitatively and quantitatively. The fluctuation of SSTs, particularly the cooling from 1850 to 1910 followed by rapid warming until 1945, remains a complex challenge in climate physics. In my own ongoing research, which is under review at Nature, we have conducted a meticulous correction of instrumental SST data. Our findings indicate that the corrected SSTs qualitatively mirror the Sr/Ca record, albeit with a more modest warming between 1850 and 1900. This reinforces my cautious stance on their warming quantification. However, integrating these sclerosponge findings with corrected instrumental data could offer a more comprehensive view of historical SST evolution, thus contributing to the harmonisation of data archives across land and ocean temperatures, alongside external forcing.

“Additionally, I’d like to address the definition of the pre-industrial period. The conceptual model of global warming often includes a historical period characterised by a stable climate that fluctuated only around a baseline, with human emissions subsequently triggering temperature rises. This leads to the political implication that humanity should strive to limit anthropogenic warming to below 1.5 or 2°C. However, the climate has never really existed in such an ideally stable state. For instance, the “Little Ice Age” preceded the Industrial Revolution, and the warming observed during the 19th century could partly reflect a recovery from the “Little Ice Age” in conjunction with human activities. It is important to clarify that I am not suggesting the ongoing warming is unrelated to human activities. Rather, it is essential to more accurately distinguish the anthropogenic component from other factors, particularly during the early industrial period. This distinction not only enhances our understanding of climate change but also has significant political implications, informing policy and target-setting in the context of global warming.

“To the question Should surpassing or close to surpassing the 1.5°C target discourage efforts to limit further warming? my answer is NO. The impacts of climate change, including intensified heatwaves, extreme rainfall, and sea-level rise, are tangible and will only exacerbate as temperatures continue to rise. Setting specific goals is beneficial for political collaboration and global action, but failing to meet a target should not lead to despair.  Just like in our daily lives, where a missed objective in our schedule doesn’t mean we abandon the rest of the day’s plans.  In the face of climate challenges, we adapt and persevere; we recalibrate and re-energise, because every step counts and every fraction of a degree matters.”


Prof Gabi Hegerl FRS, Professor of Climate System Science, University of Edinburgh, said:

“The paper shows a new record that tracks known anomalies such as the cool period following two severe eruptions. But a single location cannot substitute global data, as climate varies across the globe which is why the only way to measure global temperature is to get data from across the globe.

“The authors state that their data ‘demonstrate that industrial-era warming began in the mid-1860s, more than 80 years earlier than instrumental sea surface temperature records.’ The second part of that claim is not true, as multiple analyses have shown that greenhouse warming can be detected by 1900,  by which time the warming can be separated from the recovery from cold volcanic period following the Mount Tambora eruption.  This is based on proxy data, and attribution analyses of instrumental data confirm significant warming even in the early 20th century from greenhouse gas increases.

“So it’s a nice new record that illustrates how temperatures in the Caribbean started to rise over the industrial period, punctuated by volcanic episodes that caused temporary cooling and significant variability. However, the interpretation in terms of global warming goals overstretches it.”


Prof Daniela Schmidt, Professor of Earth Sciences, University of Bristol, said:

“Reconstructing baselines for past climates is fundamentally important so that we understand the uniqueness of our current climate crisis. Skeletons and shells of fossils are an important archive for these changes and form the basis of this analysis.

“Such studies are challenging, due to small sample numbers, regional settings, and the strong biological modification of the climate signal by organisms which result in uncertainties in proxy, in this study temperature reconstructions. The claim that we might have overshot 1.5C is being made rather confidently, when in fact there are several uncertainties and limitations in this study which must be acknowledged. 

“Such time series will provide a target for our climate models and challenge us to face the realities of climate change altering our world.

“The absolute degree of warming will always depend on the baseline and different groups have different definitions. The absolute number should not be the focus of the discussion, though.  While the Paris Agreement strongly focused on 1.5C, we know that impacts increase with every increment of warming. Missing a target should not say we lose all hope but what we need to increase our efforts. Again, if we miss these target will depend on definitions. One year above 1.5C will allow people and Nature to bounce back. The duration and geographic extend of the overshoot is challenging us as a society to implement the mitigation options we already have and reduce the risks via adaptation.”


Dr Gavin Schmidt, Director of the NASA Goddard Institute for Space Studies, said:

“Paleo-climate records are essential to extend the temperature record beyond the instrumental era, and these proxies are a useful addition to that database. People should be careful in assuming that proxies from one part of the Atlantic are always reflective of the global mean.

“Estimates of the global mean temperatures before 1850 require multiple proxies from as wide a regional variation as possible, thus claims that records from a single record can confidently define the global mean warming since the pre-industrial are probably overreaching.

“There is a real uncertainty in what the mid-19th Century temperatures were compared to the modern period, around 0.2ºC even in the instrumental record, and so that complicates our ability to make definitive statements about the crossing of the 1.5ºC level.”


Prof Kate Hendry, Polar Oceans Team, British Antarctic Survey, said:

“If we are to meet climate targets, we obviously must agree internationally on what these targets are to be. To do this, we need a robust and agreed definition of pre-industrial baseline that all temperature anomalies are to be measured against. This paper attempts to do just this by providing a high-resolution temperature record from sponge skeletons collected in a region that could be argued to reflect global change. These archives reveal that the current warming trend may have started earlier and may have already reached 2°C above pre-industrial levels – a potentially significant benchmark given UN climate targets.

“However, we must keep in mind some considerable complexities surrounding the use of sponge skeleton chemistry as archives of past ocean change. We need to know far more about how these animals make their skeletons, and exactly how their chemistry relates to ambient temperatures – something we don’t have a good handle on now. We need a better understanding of these proxies before we can use them to make important statements about the state of the climate with confidence.

“This study calibrates their temperature archive by comparing sponge chemistry to sea surface temperature (SST) records from the latter part of the twentieth century, and they find a much greater temperature sensitivity than previous studies. Whilst this sensitivity is potentially very useful when building SST anomaly archives, we really need to delve into this more to understand it better. The calibration here only spans a portion of the temperature anomalies since pre-industrial times – it’s very important to know how the chemical proxy ‘works’ across a larger range. Experiments with sponges grown under controlled temperature conditions could be very useful in this respect, if challenging given their slow growth rates, as well as more field studies.

“There’s clearly more we need to know about these sponge archives, and we need more high-resolution climate records from more locations to build a robust, global picture. However, the importance of this paper is that it makes us ask the question: what if the planet has already warmed more than we thought?”



Prof Richard Allan, Professor of Climate Science, University of Reading, said:

“This indirect, measure of ocean warming adds to the pile of evidence of warming since the pre-industrial era before substantial emissions of heat-trapping greenhouse gases were pumped into the atmosphere by human industrial activities. Using this regional estimate to recalibrate global warming estimates above the dangerous 1.5 degrees Celsius threshold is something of a red herring since rapid climate change is already fast approaching or already at this point regardless of the mix of evidence used. Rapid cuts in greenhouse gas emissions to net zero remain absolutely paramount in avoiding an even more dire future for societies and the ecosystems upon which we depend.”




300 years of sclerosponge thermometry shows global warming has exceeded 1.5 °C’ by Malcolm T. McCulloch et al. was published in Nature Climate Change at 4pm UK time on Monday 5 February.





Declared interests

Daniela Schmidt: no competing interests

Richard Allan: no competing interests

Kate Hendry: no interests to declare

For all other experts, no reply to our request for DOIs was received.

in this section

filter RoundUps by year

search by tag