the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 Apr 2025
Regional climate imprints of recent historical changes in anthropogenic Near Term Climate Forcers
Abstract. Near-Term Climate Forcers (NTCFs) play a crucial role in shaping Earth's climate, yet their effects are often overshadowed by long-lived greenhouse gases (GHGs) when addressing climate variability. This study explores the climatic impact of elevated non-methane NTCF concentrations from 1950 to 2014 using CMIP6-AerChemMIP simulations. We analyse data from four Earth System Models with interactive tropospheric chemistry and aerosol schemes, leveraging a twelve-member ensemble to ensure statistical robustness. Unlike single-species or idealised radiative forcing studies, our approach captures the combined effects of co-emitted NTCF species. Our results show that the negative radiative forcing of aerosols dominates the overall NTCF impact, offsetting the warming effects of absorbing aerosols and tropospheric ozone. Multi-model mean analyses reveal three key climate responses: (1) a global cooling, amplified in the Arctic, where autumn temperatures decrease by up to 5 °C, (2) a 38 % increase in Labrador Sea ocean convection, and (3) changes in tropical precipitation, including a 0.6° southward displacement of the Intertropical Convergence Zone (ITCZ). This research addresses the mechanisms driving these climatic changes and underscores the importance of incorporating interactive NTCFs in climate projections. As inferred from their historical impact, future NTCF reductions could amplify regional responses to increasing GHG concentrations, thus requiring more ambitious mitigation strategies.
- Preprint
(8969 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Review of “Regional climate imprints of recent historical changes in anthropogenic Near Term Climate Forcers”, by A. Santos-Espeso et al.', Anonymous Referee #1, 12 Jun 2025
This paper examines the climate effects of Near Term Climate Forcers (NTCFs) using two climate model experiments from CMIP6-AerChemMIP, “historical” and “hist-piNTCF”, which include time-varying and fixed pre-industrial NTCF forcings, respectively. Focus is given to three main climate responses to NTCFs: 1) Arctic-amplified global cooling, 2) increased Labrador Sea convection, and 3) changes in tropical precipitation, including a southward displacement of the ITCZ.
Overall, the paper is clear and well written, and the methodology is sound. However, there are a few occasions throughout where statements are made that are unclear and/or are unsupported by the authors’ results. I discuss these in my specific comments below. Once these comments are addressed, I believe that the paper should be acceptable for publication.
Specific comments:
1) Lines 52-54: Since this paragraph is focused on the ocean, presumably you are talking about ocean meridional circulation and ocean heat transport here?
2) Line 87: Should this be “key metrics”, not “key magnitudes”?
3) Table A1 header row: First model should be BCC-ESM1, not BSC-ESM1.
4) Lines 191-193 and Fig. 1c,d,g,h: It might be interesting to quantify how much of the variance change between historical and hist-piNTCF is due to different multidecadal trends in these two experiments versus different interannual variability. The impact of different trends on the variance change could be quantified by comparing Fig. 1d,h (which presumably include the effects of trend differences) with the analogous figures computed using detrended time series. Generally speaking, the effects of anthropogenic aerosols (which tend to dominate the NTCF response) counteract the effects of greenhouse gases, contributing to smaller trends in historical compared to hist-piNTCF. This is consistent with the overall decrease in variance shown in Fig. 1d,h.
5) Fig. B2: Should probably say something in the figure caption about why you don’t show the siconc from the EC-Earth model.
6) Lines 224-225: I would change “sea ice-albedo feedback” to “sea ice-related feedbacks” here. The albedo feedback over the Arctic mainly operates in summer, but you’re showing autumn siconc here. In the autumn, it is mainly the sea ice-insulation feedback that is acting to amplify temperature changes.
7) Line 230: “our results suggest” appears twice.
8) Lines 232-233: It’s unclear what is meant here by “regional radiative changes”.
9) Line 236: Should be “formation”.
10) Fig. 4: Figure title indicates that the period of focus is 1980-2014, while the caption indicates 1950-2014.
11) Lines 245-246: First of all, should say Fig. 4d, not 4f. Secondly, is it certain that these episodes of collapsed convection are purely stochastic? Could there be a state (and thus forcing) dependence to them? If so, the results in Fig. 4d might not change much if you had more ensemble members. All this is to say that it might be good to soften the language a bit here, e.g., say that the response to NTCFs “may be” underestimated, rather than “is likely” underestimated.
12) Line 251: Could be worth noting here that this model only shows a decline after ~1980, at which point global aerosol concentrations had stabilized.
13) Fig. 6 caption: The variable name for salinity seems to have been entered incorrectly, i.e., (b, e) salinity (textitso).
14) Lines 262-263: Missing parenthesis, i.e., (as observed in Fig. 4).
15) Lines 265-269: This part seems too speculative to me. Can you present any evidence that this recirculation of saltier subsurface water is actually happening in the models? Or at least some citation from the literature supporting the existence of this positive feedback in the Labrador Sea?
16) Fig. B5 caption: Should be “temperature’s contribution to density (sigmaT)”.
17) Lines 270-275: I think that Fig. B5 is useful for understanding the contributions of temperature and salinity to the simulated density anomalies. However, I don’t agree with the authors’ interpretation of this figure. Specifically, it is stated that “temperature initially triggers surface density increases, which are subsequently reinforced by a salinity-driven feedback” (echoing a similar statement on lines 265-269). However, in October, temperature and salinity contribute about equally to the surface density anomalies. So, I don’t see how Fig. B5 can be used to argue that temperature anomalies are the initial trigger of the density anomalies, and that salinity anomalies are a subsequent feedback. I think this section (and lines 265-269, which make similar statements) needs to be reworded a bit.
18) Line 310: Should be “rsut”, not “rust”.
19) Fig. 9 and Fig. B6 captions: I’m a bit confused here about the distinction between MRI-ESM2-0 and the other models in terms of representing the effects of major volcanic eruptions. Even if the models other than MRI-ESM2-0 don’t include interactive stratospheric chemistry, they should still prescribe the volcanic aerosols in their historical simulations, correct? If so, why isn’t this reflected in od550aer? Do these models simply exclude the stratosphere in their calculation of od550aer? Or, is there some other explanation?
20) Lines 329-330: I would change this to “supports the hypothesis that aerosols, through some combination of direct effects and aerosol-cloud interactions, force…”, or something similar. You haven’t actually quantified the relative impacts of aerosol direct and indirect effects on the net radiation.
21) Lines 330-331: “Notably, the clt magnitude…” I don’t understand this sentence. clt is the total cloud fraction/amount – how does it capture changes in other cloud properties besides that? And how can it be used to detect aerosol-cloud interactions? I would explain more clearly what you mean here, or just remove this sentence.
22) Lines 335-339: I would remove this paragraph as it does not fit well within the rest of the discussion. First of all, you have not actually quantified aerosol-cloud interactions in your model simulations, so it’s unclear how your results relate to those of Zhao and Suzuki (2021). Secondly, all of the previous discussion/analysis attempting to link aerosols to the ITCZ shift focused on the aerosol effect on the top-of-atmosphere radiation. Now, in this paragraph, you start to talk about aerosol effects on surface evaporation and the hemispheric atmospheric energy contrast. Again, I think this paragraph just doesn’t fit well, adds confusion, and is unnecessary.
23) Lines 344-346: The Byrne et al. (2018) paper is a review paper that does discuss “not only changes in ITCZ location but also in its width and strength”, in multiple contexts (e.g., observations, future climate projections). The role of aerosols is discussed some, but mainly (as far as I can tell) in terms of aerosol effects on ITCZ latitude. Please explain more clearly how your finding here of a negative correlation between netR_HD and equatorial rainfall amount is “consistent with” the Byrne et al. (2018) study. Or just remove this sentence.
24) Lines 370-372: As discussed in previous comments, I don’t believe that your results support these statements. The 38% increase in convection refers to the Feb.-Mar.-Apr. (FMA) season. During FMA, surface density (and thus convection) anomalies are driven primarily by salinity anomalies, not temperature anomalies (Fig. B5). And you’ve provided no evidence as far as I can tell to support the existence of the salinity feedback that is proposed here.
Citation: https://doi.org/10.5194/egusphere-2025-1286-RC1 -
RC2: 'Comment on egusphere-2025-1286', Matthew Kasoar, 01 Aug 2025
The authors present an analysis of the CMIP6 AerChemMIP hist-piNTCF experiment, using an ensemble of 4 earth system models each with 3 ensemble members. By comparing against the CMIP6 hist experiment they explore the coupled climate response to historical aerosol and reactive gas emissions, focusing on the period from 1950 onwards when aerosol emissions rapidly increased, and on the impacts to global temperature patterns, Arctic sea ice, North Atlantic ocean convection, and the ITCZ.
It is great to see the piNTCF experiment being further utilised, and insofar as this paper documents the response of the latest generation of earth system models to historical near-term climate forcers in combination, it is a useful and overdue addition to the literature.
I have some suggestions for areas that the manuscript could be improved, as well as some comments of a more technical nature which are mostly seeking clarification or further detail. I would note that my expertise is mainly around atmospheric composition and particularly aerosol-climate interactions, and so I did not feel able to assess the part of the manuscript that discusses ocean circulation responses.
My overarching comment on the manuscript is that while it adequately documents the climate response in the CMIP6 ensemble to historical NTCFs, which is valuable, the analysis mostly stops short of really attributing the responses, instead mainly relying on a review of existing literature and mechanisms that have been previously reported, to 'attribute' the responses to aerosol forcing, primarily on the basis that the response matches what would be expected and has previously been found for historical aerosol perturbations. In this respect, the results do not offer much new insight, although again in terms of documenting the CMIP6 response to piNTCF I still consider the contribution valuable.
It should be noted that the role of aerosols in driving the cooling trend between 1950-1980 in CMIP6 historical simulations has already been explored using AerChemMIP data by Zhang et al., 2021, (https://doi.org/10.5194/acp-21-18609-2021), and this paper ought be cited in appropriate places. The aerosol-driven cooling trend is one of the central results that the present study highlights, and the authors suggest that the observed pattern must be due to aerosols, but don't actually attribute this because they do not separate out the aerosols from reactive gas emissions. Comparison with the results of Zhang et al. based on the AerChemMIP hist-piAer experiment would therefore be a very useful comparison, and it's surprising that this study isn't referenced currently. Similarly the spatial pattern of historical aerosol cooling in CMIP6, with stronger cooling in northern high latitudes, is presented in AR6 WG1 Chapter 6 (Szopa et al. 2021) and this should probably also be cited.
As a result, a related limitation is that the rationale for using hist-piNTCF in the present study isn't very clear. There is minimal discussion and no quantitative analysis of the climate impacts of tropospheric ozone and other reactive gases, and the authors conclude that the overall NTCF response is probably dominated by aerosol forcing. So the value added from using piNTCF, rather than piAer which has already been analysed in Zhang et al, is unclear. Using the combination of the two experiments (i.e. piNTCF compared with piAer) would have been nice to separate out the contributions of aerosols and ozone precursors to the overall NTCF response, but as it is the focus is entirely on aerosols and an assumption that these drive all the response.
The authors loosely throughout the paper use phrases like 'we attribute this cooling to higher aerosol concentrations' (e.g. L204) - but, has it actually been attributed? 'Attribution' can have a specific meaning which is not really what the authors are doing here, I don't think. Although I completely agree this is the reason, nonetheless as currently presented this is really a hypothesis/assumption/interpretation, based on the fact that the cooling response is what we would expect to see, and what previous studies have shown, due to aerosols.
For example, in L230-231: "Overall, our results suggest that the primary driver of the Arctic response to NTCFs during 1950–1980 was high aerosol concentrations, with tropospheric ozone playing a secondary and opposing warming role" - I'm unclear how the results actually show this. The results show that the response to NTCFs (i.e. aerosols and ozone precursors combined) during 1950-1980 was a global cooling with a stronger response in the northern hemisphere. No results are shown which demonstrate that this coincided with high aerosol concentrations, or that tropospheric ozone contributed an opposing warming. While I completely agree that this is certainly the case, and is probably expected background knowledge for any reader working in this field, nonetheless in the present study it's an inference drawn from other literature and the expected behaviour, not something that is actually demonstrated. Comparing the piNTCF with the piAer experiment and/or the results of Zhang et al. would be one way to ascertain this, and would enable the contribution of aerosols and ozone precursors to be separated, which would then be a much more holistic analysis of the piNTCF experiment. Showing the distribution of aerosols and tropospheric ozone, and how their emissions and/or burden has changed over time and how this corresponds (or doesn't) with the timeseries of temperature response in Figure 3, could be another way to help motivate such a conclusion.
Indeed, it seems an omission that no timeseries of the emissions is included, given that the authors discuss how the cooling trend is predominately during the 1950-1980 time period which corresponds to when aerosol emissions were increasing, before they then plateau, whereas ozone precursor emissions continue to increase. But these emission timeseries are never actually shown, so the reader is unable to make a judgement on whether the temperature trends really do align with the aerosol emission trends or not. E.g. L208-209: "This trend change is particularly evident in the differences between ensembles (Fig. 3b,d), which closely follow historical aerosol concentration trends" - it would be nice if this was shown (i.e. the historical aerosol concentration trends) so that the reader can see this for themselves, rather than just asserting it.
Additional minor comments:
- As well as the Collins et al. (2017) AerChemMIP protocol paper, the recent AerChemMIP retrospective which summarises the outcomes of the initiative, by Griffiths, Wilcox, Allen et al., 2025, (https://doi.org/10.5194/acp-25-8289-2025) should probably also be cited in the introduction when introducing AerChemMIP.
- L39: "attributed to meridional forcing gradients in midlatitudes" - is it the gradient in the midlatitudes that matters, or the interhemispheric difference?
- Figures with maps, e.g. Figure 1, Figure B8 - please include global mean values for reference as well
- L215: "This aligns with previous studies (Wu et al., 2024)" - this is only one study, not multiple studies
- L221-225: "Indeed, the observed cooling in Fig. 1b aligns with an increase in sea ice concentration. Examining boreal autumn data—when sea ice retreat peaks reveals a consistent increase in sea ice extent in the historical ensemble relative to hist-piNTCF across multiple models (Fig. B2). The strongest sea ice expansion in the Barents Sea corresponds to the most pronounced temperature decreases (Fig.1b), reinforcing the role of sea ice-albedo feedback in amplifying Arctic cooling" - but, an expansion of the Arctic sea ice would also be an expected consequence of Arctic cooling, so on it's own I'm not sure this provides much evidence one way or the other for a dominant role of the sea ice feedback mechanism as the main factor amplifying the cooling. Even if other factors are actually more important for driving the amplification, you would still expect to see a sea ice response, and so the presence of such a sea ice expansion neither proves nor disproves the hypothesis that it plays the dominant role.
- L229: "A reverse mechanism — enhanced Arctic sea ice extent (Fig. B2) — could explain the observed tropical cooling" - enhanced tropical upper tropospheric warming is a well established temperature response to global warming, due to the lapse rate feedback. Vice versa, an enhanced cooling signal in the tropical upper troposphere will be the expected response to a global dimming caused e.g. by aerosols. Although Arctic sea ice expansion will also contribute as an additional feedback which adds to the global cooling and the tropical signature, I don't see any reason to think this is the main reason rather than the well-established lapse rate feedback pattern that would accompany any global cooling.
- L232-234: "The amplified cooling in the Arctic is largely mediated by sea ice feedbacks, though additional factors, such as regional radiative changes, as well as, variations in atmospheric, and oceanic energy transport, may also contribute to the observed temperature changes" - again, this is speculative based on mechanisms that have been reported in previous studies, rather than something which is demonstrated by any of the results presented here. There are multiple mechanisms that are known to contribute to Arctic Amplification, and based on the results shown here, the authors have no basis for concluding that sea ice feedbacks are the dominant driver, as far as I can see. I'm happy to be corrected if I've missed something which allows the contribution of different feedbacks to the overall Arctic cooling to be determined though.
- L330-331: "Notably, the clt magnitude captures both changes in cloud amount and cloud properties, allowing for detection of potential aerosol–cloud interactions" - does it? Surely clt (cloud area fraction) is just the 2D cloud amount, i.e. the fraction of the gridbox covered by cloud. It doesn't tell you about other cloud properties that are affected (possibly even more strongly so) by aerosol microphysics such as cloud albedo. If the authors were to calculate e.g. the change in cloud radiative forcing (difference between all-sky and clear-sky radiative fluxes), then this would capture both changes in cloud amount, and other properties like cloud albedo and optical thickness.
- Please double check figure numbering - for example, it looks like fig. B8 is referenced in the text before figure B7 (though apologies if I've missed an earlier instance of it).
- L341-343: "The analysis reveals a clear negative correlation, with higher netR_HD values associated with reduced equatorial rainfall. This suggests that an increased radiation imbalance between hemispheres, previously linked to a southward shift in the meridional circulation, also leads to a reduction in precipitation near the equator" - does it suggest this? Couldn't the reduced precipitation just be a consequence of globally cooler temperatures, rather than because of the hemispheric difference? Because the global cooling is caused by aerosols, which are preferentially located in the northern hemisphere, the hemispheric difference in netR will also be correlated with global temperature. So, this analysis can't distinguish between an effect caused by the hemispheric temperature difference, and an effect caused by globally cooler temperatures. The precipitation reduction could therefore just be due to the overall temperature reduction and not the interhemispheric difference; you will find a correlation either way. If the authors wish to argue that it is indeed the *gradient* of radiative forcing and not just the overall magnitude that drives reduced ITCZ precipitation, this needs to be backed up with additional analysis elucidating the mechanism for this, I think.
Citation: https://doi.org/10.5194/egusphere-2025-1286-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 489 | 59 | 14 | 562 | 18 | 37 |
- HTML: 489
- PDF: 59
- XML: 14
- Total: 562
- BibTeX: 18
- EndNote: 37
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1