in Re to nigelj, 7 Aug 2025 at 8:22 PM,
Hallo Nigel,
Thank you very much for your feedback.
Following your advice, I started a step-by-step check of the extent in which I missed easily available climate science information circulating in the public space, for what I am frequently criticized by many regular Real Climate commenters.
As the first step, I tried to check with the AI tool Perplexity how relevant can be my feeling that
– the possibility of equilibrium climate sensitivity (ECS) changes due to anthropogenic interferences with land hydrology
has not been studied yet.
I went through four consecutive question-answer rounds. The obtained output including references provided by the searching engine is attached below.
I would say that on the first sight, it appears that my above-mentioned feeling was wrong, because easily accessible public information suggests that
– studies showing that climate sensitivity is not a constant but depends on the initial state of the climate system do exist and are cited in the IPCC assessment reports,
– the idea that better fit of medium-sensitivity models with previous climate and of the “hot” models with present climate could be explained by changing climate sensitivity looks obvious, because there are hints that climate system state (and feedback processes that may influence climate sensitivity) may have changed, and
– land hydrology including water availability for evaporation belongs to parameters defining the climate system state.
Nevertheless, an inspection of the provided references revealed that none of documents suggested by the engine deals specifically with the relationship between water availability for evaporation from the land and climate sensitivity.
I would therefore conclude that although my question to moderators, regarding feasibility of a study directed to clarification of this relationship, seems to be obvious in view of the publicly accessible information, it appears that there is not easily accessible information in the public domain yet, suggesting that this is already a resolved problem or a subject of an ongoing scientific research or discussion.
Thus, my feeling that this relationship may be still an open scientific problem can be correct, and if so, my question to moderators about feasibility of a modelling study for clarifying the problem may be indeed relevant.
Greetings
Tomáš
————————–
The record of my search follows
Q1:
In climate science discussions, there often appear disputes about so called “equilibrium climate sensitivity” (ECS) that is usually defined as the increase “delta T” of the global mean surface temperature (GMST) which brings a sudden Earth energy imbalance (EEI) caused by a hypothetical abrupt change of a certain climate “forcing” (usually doubling of the atmospheric carbon dioxide concentration).
It appears that although there are publications asserting high climate sensitivity towards atmospheric CO2 concentration changes (more than 5 K for CO2 concentration doubling) also from paleoclimate data, in the mainstream climate science (represented e.g. by the director of the NASA Goddard Institute for Space Studies Gavin Schmidt) prevails a consensus that paleoclimate data suggest rather a medium climate sensitivity (about 3 K for CO2 concentration doubling) and it further appears that this consensus is in accordance with prevailing results of global climate models assessed in the Climate Model Intercomparison Project (CMIP).
Is this understanding right, or do you have a correction?
A:
Your understanding of equilibrium climate sensitivity (ECS) in climate science discussions is largely correct with some nuanced context.
ECS is defined as the long-term increase in global mean surface temperature (GMST) resulting from a doubling of atmospheric carbon dioxide (CO2) concentration. It reflects the climate system’s response once it reaches equilibrium after the forcing change.
There is indeed a range of estimates of ECS in the scientific community:
1. Mainstream consensus:
o Many climate scientists, including Gavin Schmidt, director of NASA’s Goddard Institute for Space Studies, support a medium-range ECS estimate around 3 K (degrees Celsius) for CO2 doubling.
o Schmidt and colleagues highlight that recent assessments have narrowed the uncertainty range roughly to 2.6 to 3.9 K.
o This consensus aligns with the frequent range reported by the Intergovernmental Panel on Climate Change (IPCC) and prevailing results from global climate models assessed in the Climate Model Intercomparison Project phases (CMIP5 and CMIP6).
o The broader and more detailed assessments recognize that while ECS is not extremely low, it is also not so high as to foreclose the possibility of mitigation or adaptation.[1][2][3]
2. Paleoclimate data and ECS:
o Paleoclimate records, which study past climates using proxies like ice cores and sediment records, generally support ECS estimates within the traditional range of about 2 to 5 K.
o Paleoclimate studies confirm medium sensitivities and provide valuable constraints that reinforce the validity of climate model estimates.
o Some individual studies have suggested higher ECS values (above 5 K), particularly when interpreting certain warm paleoclimates or specific proxy data.
o However, these higher estimates are not the prevailing consensus and are often more uncertain due to complexities in interpreting ancient climate states.
o Recent research highlights that the newest CMIP6 climate models produce a somewhat broader range of ECS values, including some higher values than previous generations, but many of these high sensitivities are still being evaluated against paleoclimate evidence and observational data to confirm their realism.[4][5][6][7][8]
3. Climate models and ECS:
o CMIP5 models tended to cluster around ECS estimates of 2 to 4.5 K, with an average near 3.2 K.
o CMIP6 models have introduced some that simulate notably higher climate sensitivities, sometimes exceeding 5 K.
o This increase in modeled sensitivities has prompted intense research to understand whether these higher values are realistic or reflect model biases, especially concerning cloud feedbacks.
o Overall, even with some high ECS values in CMIP6 models, the ensemble mean and best estimates remain close to the medium sensitivity range, consistent with paleoclimate and observational constraints.[9][10][11]
In summary, the prevailing mainstream scientific view, including that articulated by Gavin Schmidt and supported by major climate modeling intercomparison projects, holds that Earth’s equilibrium climate sensitivity likely lies near 3 K, with a range roughly between 2.5 to 4 K for doubled CO2. Paleoclimate data broadly support this medium range, though high ECS estimates do exist in some studies. CMIP6 models have introduced somewhat higher ECS values, which are under active investigation. This nuanced consensus balances multiple lines of evidence and acknowledges uncertainty, but does not embrace the highest ECS estimates as the dominant view at this time.
Hence, your summary is accurate: the mainstream scientific consensus supports a moderate climate sensitivity of about 3 K for CO2 doubling, consistent with paleoclimate data and the majority of climate model results, despite some research suggesting higher values in certain cases.
⁂
1. https://science.nasa.gov/earth/climate-change/making-sense-of-climate-sensitivity/
2. https://www.carbonbrief.org/the-carbon-brief-interview-dr-gavin-schmidt/
3. https://en.wikipedia.org/wiki/Climate_sensitivity
4. https://psa2022.dryfta.com/program-schedule/program/detail/141/climate-sensitivity-paleoclimate-data-amp-the-end-of-model-democracy
5. https://link.springer.com/article/10.1007/BF00143708
6. https://www.nature.com/articles/s41612-025-00948-7
7. https://www.nature.com/articles/s41558-020-0764-6
8. https://www.nature.com/articles/s43247-024-01531-3
9. https://skepticalscience.com/gwpf-lewis-crock-climate-sensitivity-optimism-ill-founded.html
10. https://www.carbonbrief.org/cmip6-the-next-generation-of-climate-models-explained/
11. https://ukesm.ac.uk/climate-sensitivity-in-cmip6-causes-consequences-and-uses/
Q2:
In the above-mentioned climate science discussions, I have noticed a remark that the “hot” models of the CMIP6, suggesting high climate sensitivity, do appear to fit better (in comparison with the entire CMIP6 ensemble and/or with the CMIP5 models) the present climate change as observed in the last decades.
The mainstream scientists, however, are still reluctant to accept that the “hot” models should be preferred, because they do not fit the paleoclimate data as good as the “conservative” models with medium ECS.
Is it true?
If so, could I rightly conclude that in view of the present knowledge, climate models with medium climate sensitivity do fit paleoclimate data well but somewhat struggle with the present climate change, whereas the “hot” models with higher climate sensitivity seem to fit the present climate change better, however, they still seem to struggle with paleoclimate data?
A:
Your understanding is largely accurate and reflects the current nuanced discourse within climate science regarding the CMIP6 “hot” models and their climate sensitivities.
1. The “hot” models within the CMIP6 ensemble are those with relatively high equilibrium climate sensitivity (ECS), often above the upper end of the IPCC likely range of about 4 K, with some models showing ECS values exceeding 5 K. These models generally produce stronger warming projections for the present and future climate compared to the broader CMIP6 ensemble or the earlier CMIP5 models.
2. Recent studies and assessments indicate that these “hot” models tend to fit observed recent climate changes over the past decades somewhat better than models with medium ECS, particularly with respect to the magnitude of warming observed so far. This is likely why they attract attention for their apparent better agreement with the present-day warming trend.
3. However, mainstream climate scientists remain cautious about favoring the “hot” models due to their poorer fit to paleoclimate data. Multiple paleoclimate studies have found that models with very high ECS values simulate temperatures for past climate states (such as the Last Glacial Maximum or warmer greenhouse periods) that fall outside the ranges indicated by proxy data, suggesting those models may overestimate climate sensitivity.
4. Meanwhile, models with medium ECS (roughly 2.5 to 4 K) tend to perform well in reproducing paleoclimate conditions, providing a consistent and physically plausible picture across a wide range of past and present climate data.
5. The disagreement primarily arises because the constraints from paleoclimate and the constraints from recent instrumental observations and short-term variability sometimes pull the estimates towards slightly different sensitivities. The “hot” models, with stronger warming, appear more consistent with recent trends but less so with paleoclimate evidence, while the medium-sensitivity models are well aligned with paleoclimate but somewhat less aggressive in simulating recent warming magnitude.
6. This tension leads to approaches where the “hot” models are either down-weighted or combined with other models using sophisticated statistical methods (e.g., Bayesian model averaging) to produce balanced future projections that account for both sets of evidence, rather than outright rejecting either group.
In conclusion, yes, you can rightly sum up that:
• Climate models with medium climate sensitivity fit paleoclimate data well but may somewhat underestimate or struggle to fully capture the magnitude of observed recent climate change.
• The “hot” CMIP6 models with higher climate sensitivity tend to fit present climate change observations better but struggle more with paleoclimate constraints.
This ongoing evaluation reflects the complexity of climate system responses and the importance of integrating multiple lines of evidence to understand and project future climate.[1][2][3][4][5][6]
1. https://hess.copernicus.org/preprints/hess-2023-47/hess-2023-47.pdf
2. https://hess.copernicus.org/articles/27/4355/2023/
3. https://www.nature.com/articles/s43247-023-01009-8
4. http://su.diva-portal.org/smash/record.jsf?pid=diva2%3A1921490
5. https://www.nature.com/articles/s43247-024-01531-3
6. https://pmc.ncbi.nlm.nih.gov/articles/PMC11041668/
Q3:
Is there a scientific publication or another article or blog discussing the possibility that both “traditional” climate models fitting better the past climate and the “hot” models fitting better the present climate could be right if the present climate sensitivity is higher than the climate sensitivity in the past?
A:
Yes, there are scientific publications discussing the possibility that both “traditional” climate models, which better fit past climate, and “hot” climate models, which fit better the present climate, could be correct if the climate sensitivity has changed over time—specifically if the present climate sensitivity is higher than it was in the past.
Key points from the literature include:
• Climate sensitivity estimates from past climate changes and from modern observations may differ because feedback strengths and forcing timescales vary over time. For example, slow and fast feedback processes (like ocean heat uptake and ice sheet changes) may influence sensitivity differently in past versus present climates, leading to timescale-dependent or state-dependent climate sensitivity.[1]
• Models sometimes show that the present climate’s feedbacks (such as cloud behavior and water vapor feedback) and the base state of the climate system differ substantially from those in past climate states, potentially causing a higher effective climate sensitivity now than during earlier periods.[2]
• Some studies from paleoclimate and observational data suggest that the climate system response is non-stationary and dependent on background climate states and timescales, making it plausible for climate sensitivity to vary with time.[1]
• This understanding helps reconcile why some climate models aligned well with past climate changes but underpredict observed warming in recent decades whereas “hot” models with higher sensitivity better match current warming trends.[2][1]
• The IPCC and related research emphasize uncertainties in cloud feedbacks and other processes that contribute to different effective sensitivities in models over time, yet allow for models with different sensitivities to both be reasonable representations depending on timescale and epoch.[3][2]
Thus, the scientific community recognizes that climate sensitivity is not necessarily constant over time, and the difference between model fits to past versus present climate can be explained by this variability in sensitivity, making it plausible that both sets of models could be valid under different conditions.
References:
• Roe & Baker (2016) on lessons from past climate changes and time-dependent sensitivity[1]
• IPCC AR4 chapter on model evaluation and feedback differences[2]
• Discussions on climate model strengths, limitations, and feedback uncertainties[3]
⁂
1. https://link.springer.com/article/10.1007/s40641-016-0049-3
2. https://www.ipcc.ch/site/assets/uploads/2018/02/ar4-wg1-chapter8-1.pdf
3. https://www.wcrp-climate.org/images/modelling/WGCM/publications/sap3-climate-models.pdf
Q4:
Can you find any article suggesting or discussing possible climate sensitivity change due to a change in water availability for evaporation from the land?
A:
There are scientific articles discussing how changes in water availability for evaporation from the land can influence climate sensitivity or climate feedbacks.
Key points include:
• Climate change affects global water availability through changes in both precipitation and evaporation, which vary by region and season. Changes in evapotranspiration and soil moisture affect surface energy balances and can modulate climate responses. For example, increased evaporation in wetter regions can amplify climate impacts, while in drier regimes water limitation constrains evaporation changes. These changes influence the spatio-temporal distribution of water availability and thus the energy and moisture fluxes that drive climate processes.[1][2][3]
• Increased temperatures can raise evapotranspiration losses, reducing water availability and potentially modifying feedback mechanisms in regional climates. This can make land surface moisture a factor influencing climate sensitivity, as soil moisture and evaporation exert control over land-atmosphere interactions and surface temperature regulation.[4][5]
• Research indicates that models and observations of hydrology reveal that water availability changes due to climate forcing may be more sensitive than previously thought, suggesting a possible revision to how climate feedbacks related to evaporation and moisture are represented. This sensitivity in water availability could imply changes in effective climate sensitivity linked to land surface processes.[6]
While these articles do not always explicitly frame water availability changes as altering the “climate sensitivity” parameter per se, they discuss how changes in evaporation and soil moisture—key elements of land hydrology—influence climate feedbacks and thereby could contribute to variability in climate sensitivity over time or regions.
In summary, there is scientific discussion and evidence supporting the idea that changes in land water availability for evaporation can affect climate system feedbacks, thus potentially playing a role in changes in climate sensitivity.
References:
• Nature Communications article on changes in water availability through precipitation and evaporation under climate change.[2][1]
• Regional studies on evapotranspiration increases reducing water availability and influencing climate feedbacks.[5][4]
• Hydrological analyses suggesting higher sensitivity of water availability to climate parameters than previously accounted for.[6]
• Recent reviews of multifaceted water availability changes impacting climate dynamics.[3]
⁂
1. https://www.nature.com/articles/s41467-020-16757-w
2. https://www.sciencedirect.com/science/article/pii/S1674283424000862
3. https://www.nature.com/articles/s41612-025-00913-4
4. https://link.springer.com/article/10.1007/s10584-021-03122-z
5. https://hess.copernicus.org/articles/21/2107/2017/hess-21-2107-2017.pdf
6. https://smartwatermagazine.com/news/h2o-building-services/how-climate-change-impacts-water-availability-systematically
#direct #water #vapor #emissions #endangering
