the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 28 Feb 2024
AERA-MIP: Emission pathways, remaining budgets and carbon cycle dynamics compatible with 1.5 ºC and 2 ºC global warming stabilization
Abstract. While international climate policies now focus on limiting global warming well below 2 °C, or pursuing 1.5 °C, the climate modeling community has not provided an experimental design in which all Earth System Models (ESMs) converge and stabilize at the same prescribed global warming levels. This gap hampers accurate estimations based on comprehensive ESMs of the carbon emission pathways needed to meet such agreed warming levels, and of the associated climate impacts under temperature stabilization. Here, we apply the Adaptive Emission Reduction Approach (AERA) with ESMs to provide such simulations in which all models converge at 1.5 °C and 2.0 °C warming levels by iteratively adjusting their emissions. These emission-driven simulations provide a wide range of emission pathways and resulting atmospheric CO2 projections for a given warming level, uncovering uncertainty ranges that were previously missing in the traditional CMIP scenarios with prescribed greenhouse gas concentration pathways. Meeting the 1.5°C warming level necessitates a 40 % (model full range: 7 to 76 %) reduction in multi-model mean CO2-forcing equivalent (CO2-fe) emissions from 2025 to 2030, a 98 % (57 to 127 %) reduction from 2025 to 2050, and a stabilization at 1.0 (-1.7 to 2.9) PgC yr-1 from 2100 onward after the 1.5 °C target is reached. For the 2.0 °C warming level, CO2-fe emissions require a 47 % (8 to 92 %) reduction until 2050 and a stabilization at 1.7 (-1.5 to 2.7) PgC yr-1 from 2100 onward. The on-average positive emissions under stabilized global temperatures are the result of a decreasing transient climate response to cumulative CO2-fe emissions. This evolution is consistent with a slightly negative zero emissions commitment – initially assumed zero – and leads to an increase in the post-2025 CO2-fe emission budget by a factor 2.2 (-0.8 to 6.9) by 2150 for the 1.5 °C warming level and a factor 1.4 (0.9 to 2.4) for the 2.0 °C warming level compared to its first estimate in 2025. Our simulations highlight shifts in carbon uptake dynamics under stabilized temperature, such as a cessation of the carbon sinks in the North Atlantic and in tropical forests. On the other hand, the Southern Ocean and the northern high-latitude land remain carbon sinks over centuries after temperatures stabilize. Overall, this new type of target-based emission-driven simulations offers a more coherent assessment across climate models and opens up a wide range of possibilities for studying both the carbon cycle and climate impacts, such as extreme events, under climate stabilization.
- Preprint
(10986 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-488', Andrew MacDougall, 13 Mar 2024
Review of: AERA-MIP: Emission pathways, remaining budgets and carbon cycle dynamics compatible with 1.5 ºC and 2 ºC global warming stabilization
Overall Evaluation:
The paper presents the results of AERA-MIP, an effort to use Earth system models to quantify the emission pathway needed to reach global temperature stabilization. I commend the authors for the substantial effort needed to put together a model intercomparison of this complexity, and for this comprehensive manuscript. I have some relativity minor suggestions to clarify some of the methods and results.General Comments:
(1) I recommend that you add either a box or an appendix to explain CO2 forcing equivalent emissions to a more general Climate Science audience. Model intercomparison papers such of these are often read by audiences far beyond those immediately involved in the field, and without explaining CO2 forcing equivalent emissions that audience is going to be very confused. It is also important to explain if there is an assumed airborne fraction built into CO2 forcing equivalent emissions.(2) There are three explanations in the paper as to why many of the models project that continued fossil fuel CO2 emissions could be compatible with temperature stabilization:
I) Negative ZEC.
II) Net-negative forcing from non-CO2 forcing being compensated with fossil fuel CO2 emissions.
III) Net-negative carbon emissions from land use change being compensated with fossil fuel CO2 emissions.The paper shows that all three mechanisms are at work at different times for at least some models. It would be good to collect these explanations together in the discussion to clarify what are the dominant mechanisms on what time-scale and how mechanism may translate to the natural world.
(3) Related to comment 2, I suggest adding a section on anthropogenic carbon emissions similar to section 3.3. That is, I think it is also important to highlight what the sum of fossil fuel and land use change emissions to distinguish between true continued net CO2 emissions and fossil fuel emissions being compensated by net-negative land use change emissions.
(4) When examining the model's carbon cycle responses you should examine whether models with a terrestrial nitrogen cycle have a different response.
Specific Comments:
Line 19: I suspect that only a handful of your models had a representation of the permafrost carbon feedback (UVic, CESM2, maybe Bern?), so this statement needs a caveat.Line 41: Delete 'Damon'.
Figure 1: Delete 'or Concentration'. This panel of the figure is clearly only represents an emissions pathway. The 'or Concentration' is just confusing.
Line 53: 'climate-carbon feedbacks' are either part of TCRE if they are included in the ESM or are part of the 'unrepresented feedbacks' if not represented in the ESM. Delete 'climate-carbon feedbacks'.
Section 2.2. Should note here that the two models with the largest positive ZEC values in ZECMIP, UKESM and CNRM, did not participate in AERA-MIP.
Line 128 to 129: Fix grammar.
Line 132 to 136: Is this due to the way non-CO2 is simulated in the radiative transfer code of ACCESS, or due to something like dynamic CH4 emissions?
Line 180: Fix grammar.
Line 205: I suggest not using abbreviations as variables. For consistency with your other variables C$_{AF}$, C$_{OF}$, and C$_{LF}$ would be preferable.
Line 210: Operationally these are actually sums not integrals (just adding up the yearly or monthly values from the model output). Would be sensible just to use sum notation to not make this seem more complicated than it really is.
Line 218 to 220: Fix grammar.
Line 317 to 324: Be clear which numbers are calendar years in this paragraph.
Line 334 to 335: 'already' should go before 'exhibits'.
Line 379 to 380: Be clear that you are referring to either effective-TCRE or TCRE-fe. By definition true-TCRE only includes the effect of CO2 emissions.
Line 392: Cannot make a strong statement of significance for your collection of ESMs. Climate models are not independent, and mathematically closer to phylogenetically related (Knutti et al. 2013). Independence is a foundational assumption underlying all tests of significance. Can replace 'significant' with 'strong' or 'substantial'.
Figure 6: Would be good to include a panel with the land-use change emissions used. Maybe as its own figure as land-use change is a forcing not an output of the model experiment.
Line 443 to 445: Re-write this sentence for clarity.
References:
Knutti R, Masson D, Gettelman A. Climate model genealogy: Generation CMIP5 and how we got there. Geophysical Research Letters. 2013 Mar 28;40(6):1194-9.Sincerely:
-Andrew MacDougallCitation: https://doi.org/10.5194/egusphere-2024-488-RC1 -
RC2: 'Comment on egusphere-2024-488', Ric Williams, 27 Mar 2024
The study estimates emissions pathways that are adjusted to be compatible with temperature targets, referred to as Adaptive Emission Reduction Approach (AERA) using AERA-MIP, made up of 13 full Earth system models, 2 intermediate complexity Earth system models and 1 ocean general circulation model coupled to a carbon cycle emulator. The analyses of AERA-MIP are reported here and provide compatible emission pathways and remaining carbon budgets, together with ocean and land carbon responses. The responses are shown for model means and the model ranges, as well as including ensemble means and ensemble ranges to reveal internal variability where appropriate. The study is very comprehensive, thorough and impressive.
I only have minor comments, but given the importance of the study I recommend that these comments are taken on board to aid communication:
- There is a statement (L177) that the future CO2 emissions curve compatible with a temperature target is largely insensitive to the non-CO2 radiative forcing and land-use changes. As long as this statement is representative of the study (as this seems slightly surprising to me) , then I recommend making more of this statement and discussing the implications in the Conclusion and perhaps including in the Abstract.
- The methodology is clearly described and novel, but does overlap with a prior study of Goodwin et al. (2018a) introducing “Adjusting Mitigation Pathways”. The present approach does extend the prior study of Terhaar et al. (2022a). The text states the limitations of the Goodwin et al. (2018a) approach, but does not outline the similarity in the two approaches. Further explanation would be helpful to the wider community.
- Please include any caveats about the set of Earth system models that are included in the study, so that there is clarity about any limitations to the approach. For example, there is probably a large range in the climate feedbacks in your model set and limitations in the closures of some carbon cycles particularly for the land, which might then affect the estimates of the remaining carbon budget.
- While the manuscript is very comprehensive, the text is cryptic in places and sometimes not easy to follow due to the large number of acronyms and the choices made of those acronyms. I recommend making variable names more internally consistent, rather than using a range of different symbols to represent variables with the same units. For example, the carbon inventory in (4) includes variables written as E, G and S with subscripts, but when those variables are referred to later in isolation their meaning requires the reader to go back and find their definitions (such as L319). Likewise Table 2 uses REM and EB_2026 and both are in PgC, so unsure as to why the change in symbols used. Sometimes repetition in their definitions would also be helpful to the reader, such as in the final Discussion or Conclusions. I also recommend including a Table to list those variables.
In summary, this study is very impressive and substantial, and will make an important contribution to discussions about the remaining carbon budget, and provide key information for the next IPCC report. Minor editorial work can help in the readability of the study for a wider audience.
Minor details
Line 41 Add Goodwin et al. (2018b)
Equation (6) and (7) add the dt in the integrals to make more explicit.
Line 243 Overlong sentence including repeated use of but.
Line 506. Try to avoid using ”it”, you know what you mean, but better to be explicit to the reader.
Citation: https://doi.org/10.5194/egusphere-2024-488-RC2 -
RC3: 'Comment on egusphere-2024-488', Charles Koven, 30 Mar 2024
Comments on "AERA-MIP: Emission pathways, remaining budgets and carbon cycle dynamics compatible with 1.5oC and 2oC global warming stabilization" by Silvy et al.
Overall, I think this is a really nice paper and an important contribution that shows a new way of using ESMs to understand remaining emissions budgets at different reference warming levels. The authors have done a great job of describing a complex set of experimental results. I do have a couple concerns that I describe below, but overall I think it is a very worthwhile publication.
My first concern is semantic but I think it is important. I object to the usage of the word "target" in the sense of "global warming target" starting in the very first sentence of manuscript, and its many instances thereafter. The Paris agreement (English-language version) uses the word "target" only in article 4 to describe emissions reduction targets, it never refers to warming levels as "targets". Nor do any of the IPCC AR6 SPMs. So the first sentence is not correct: the Paris agreement does not primarily focus on global warming "targets". And it is a mistake to think of warming levels like 1.5C or 2C as targets; just because we do not wish to exceed them does not mean that these warming levels are where we wish the climate to arrive at or remain, either as a peak temperature or after a peak-and-decline. So in this paper, there are basically two distinct usages of the word "target": in the control-theory sense of "setpoint", and in the policy sense of reference warming level, and I advocate that the authors do not conflate them. I.e., there is a 1.5 degree global warming level as specified in the Paris agreement, and because one is interested in both the impacts at that or another policy-relevant global warming level, as well as the emissions pathway that is compatible with that warming level and how it might evolve over time, one uses the AERA-MIP protocol with a setpoint at that level to drive the simulations to that setpoint. I think that it is very important *not* to use the word "target" for the setpoint value, because doing so implies a value judgement that the setpoint warming level is in fact a desired goal. So I would strongly encourage the authors to replace every single instance in the entire manuscript of the word "target" for either of the more neutral words "level" or "setpoint", or other similar wording where appropriate, throughout (including when used within figures such as figure 1).
My second major concern is about how this manuscript treats ACCESS as an outlier. The overall hypothesis of the manuscript seems to be that it is possible to apply the AERA in a set of full-complexity ESMs, but then the one model that does not work as intended is not fully included in the analysis. The implications of this would seem to be pretty important, as it implies that certain types of physical uncertainties might lead to a failure of climate policies that are structured like the AERA approach. The explanation on lines 132-136 makes sense, but what is the evidence for it? How confident are we that this mismatch in non-CO2 forcing is not pointing to a real uncertainty that we need to consider in remaining emissions budgets? So I would advocate keeping ACCESS within the ensemble statistics unless a stronger case is made that the non-CO2 forcing estimate difference represents an unphysical model artifact.
I am also a bit confused about the implications of removing the historical warming differences between the models by lining them up so that they all pass through 1.22 degrees at year 2020. How was this handled for the models with multiple ensemble members -- were the ensemble members each lined up at their own year 2020 temperature, and thus they have different absolute temperature setpoints? If so, is there any correlation between the REBs and the absolute temperature at 2020 across the ensemble members? Or was the ensemble-mean global temperature used for the year 2020 value so that all members have the same absolute temeprature setpoint? Given that there is real uncertainty in the current level of global warming due to internal variability (e.g. as pointed out on line 218), what are the implications of this choice on the uncertainty in REBs? Some further discussion of this would be helpful. Lastly, I recommend revising the schematic in figure 1 bottom-right panel to show that the minimum in temperature spread is at present-day, and then increases prior to that, rather than what is shown where the temperature spread increases over time starting from preindustrial conditions.
Why is the time evolution of the ensemble spread of CO2-fe in figure 3b and 3c so different between GFDL-ESM2M and EC-Earth?
Line 172-175: The E_LUC won't be quite the same as in the references, since the temperature and CO2 pathways will differ, and thus the loss of additional sink capacity and weakening of carbon-climate feedbacks will be different. Can you quantify that effect, and does it matter?
Lines 216-217. Is the +/- 10% correspond to the light red and blue shading in fig. 1? Or is it the +/- 0.2C referred to on line 217? Either way, please note that in the figure caption.
In fig. B3, why is the difference between the AERA-derived and "true" estimate of TCRE so systematic over the early years of the experiment? I.e. the multi-model ensemble looks roughly to have about the same difference as individual models. Is that due to some systematic path dependence in aerosol, land-use, and other SLCF dynamics?
Citation: https://doi.org/10.5194/egusphere-2024-488-RC3
Data sets
Model data for "AERA-MIP: Emission pathways, remaining budgets and carbon cycle dynamics compatible with 1.5 ºC and 2 ºC global warming stabilization" Yona Silvy, Thomas L. Frölicher, Jens Terhaar, Fortunat Joos, Friedrich A. Burger, Fabrice Lacroix, Myles Allen, Raffaele Bernadello, Laurent Bopp, Victor Brovkin, Jonathan R. Buzan, Patricia Cadule, Martin Dix, John Dunne, Pierre Friedlingstein, Goran Georgievski, Tomohiro Hajima, Stuart Jenkins, Michio Kawamiya, Nancy Y. Kiang, Vladimir Lapin, Donghyun Lee, Paul Lerner, Nadine Mengis, Estela A. Monteiro, David Paynter, Glen P. Peters, Anastasia Romanou, Jörg Schwinger, Sarah Sparrow, Eric Stofferahn, Jerry Tjiputra, Etienne Tourigny, and Tilo Ziehn https://doi.org/10.5281/zenodo.10715168
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 654 | 150 | 17 | 821 | 10 | 11 |
- HTML: 654
- PDF: 150
- XML: 17
- Total: 821
- BibTeX: 10
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1