the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 15 Aug 2023
Global and Indian precipitation responses to anthropogenic aerosol and carbon dioxide forcings from PDRMIP experiments
Abstract. Global precipitation change in response to climate change is closely related to surface temperature, the forcing agent, and the atmospheric dry energy budget, but regional precipitation change is more complex. In this study, we use experiments from the Precipitation Driver and Response Model Intercomparison Project (PDRMIP) wherein carbon dioxide, sulfate aerosols, and black carbon aerosols are perturbed to study the global precipitation response in contrast with the regional response over India. The response to global warming from carbon dioxide increases precipitation both globally and regionally, whereas the cooling response to sulfate aerosol leads to a reduction in precipitation in both cases. The response to black carbon aerosols, however, is a global decrease but a regional increase of precipitation over India. The mechanism is increased atmospheric heating driving a stronger monsoon circulation and stronger low-level winds. This intensification of the Indian monsoon is, somewhat surprisingly, stronger for global black carbon emissions than when the emissions are limited to those from the Asian region. Overall, our study presents heterogeneity in precipitation responses at both global and regional levels and the potential underlying physical processes under a variety of climate forcings that would be useful in designing further model experiments with higher spatial resolution.
- Preprint
(8258 KB) - Metadata XML
-
Supplement
(3332 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2023-1605', Anonymous Referee #2, 22 Aug 2023
General Comments
The paper uses a promising framework to study the ESM-estimated effect of different climate forcers on the Indian Monsoon using the PDRMIP dataset. However, the current paper unfortunately spends far too long describing general effects which are already well-known, devoting insufficient space to what could be a useful investigation into the specific response of the Indian Monsoon.
In addition, the framework for analysing the precipitation response is incomplete, and doesn’t utilise previous work on the topic, again preventing a deep dive into novel grounds.
The paper could, with some substantial changes and additions, present a useful analysis of the response of a key climate phenomenon to anthropogenic emissions changes. These should be carried out before a further review can be done. In particular, the work has to be situated amongst the existing literature and present a new avenue of research.
The first few pages of the results present the global and spatial impacts of CO2, BC, and Sulfate, which are already well-documented elsewhere including using this dataset. Some standard, intuitive effects – such as the effect of global perturbations being larger than that of regional ones – are noted without discussing that these are consistent. This is reflected in the conclusion section, in which the first 2 points represent well-known effects, without noting this.
The approach to modelling the energy budget is convoluted. The initial focus on dry energy leads to conclusions on the linearity between precipitation and dry energy on the global scale, without noting that this has been documented in prior work (see specific references below). Local discrepancies from linearity over India are then explored without explicit reference to the change in horizontal flow that they represent.
I would suggest that this paper be re-oriented to focus specifically on the Indian Monsoon, with little need for global analysis. With this, and a much-reduced focus on previously-documented results, more space would then be available for devotion to a deeper dive into, and discussion of, the effects of these forcing agents on the Indian Monsoon. This should include further analysis of the ITCZ and horizontal energy transfer, and more extensive situation of the results within the wider literature.
Focus should be given to error analysis; currently only multi-model means are presented in most cases, with no quantification of R squared values for linear fits, and no discussion of internal variability in the annual cycles in Figure 6. One issue which needs exploration is the differing sets of models which undertook each experiment. Many conclusions are drawn from Sulfur reduction experiments with 1 or 2 models; these merit explicit comparison to the base experiments from those same model sets, and analysis of inter-annual variability when only 1 or 2 models are available.
Specific Comments
L62 “The GHGs have significantly warmed the climate by 1.5°C” – this is the level that would have occurred in the absence of Aerosols, right? Present-day warming is closer to 1C.
L190: what type of interpolation did you apply to each variable?
This part warrants expansion, with reference to other studies which have looked at this. PDRMIP was used by Xie et al 2022 (in the reference list but not referred to in the text https://www.nature.com/articles/s43247-022-00660-x), and in https://acp.copernicus.org/articles/20/11823/2020/ using a different energy budget analysis, and you need to be clearer on the novelty of your approach versus this. The motivation for focusing only on the coupled response should be made clear.
L192: what is the motivation for focusing only on the dry component of the energy budget? Will this not miss the latent heat changes?
L225: the relationship between northern hemisphere cooling and ITCZ shifting warrants explanation, with references.
L231: “Reducing the sulfate aerosols enhances the surface warming as noticed in the Figure 2h, which can alter the climate sensitivity leading to various feedbacks” – need to be more careful with terminology; it reads like you’re suggesting the Equilibrium Climate Sensitivity is altered by the sulfate response.
L234: it should be noted here that this consistency between sulred and sulx5 isn’t surprising
L241: I don’t follow the logic of “global BC aerosols contribute more to precipitation increase than the Asian emitted BC aerosols”. The global BC experiment contains the Asia perturbation, and additionally increases BC outside Asia; it’s not surprising that this overall effect is larger. It would be surprising if the Indian precipitation change per change in BC emissions was larger for Global than Asian emissions, but I don’t think this is what you’re suggesting here.
L263 “surface temperature (Figure 1g).” – Figure 2g
L267: with the important exception of BC perturbations! It is well worth quantifying these by displaying the gradients and R squareds of linear fits for the Global and Indian regions to back up your point
L272: you should give exact values for these numbers rather than approximations, to a consistent number of significant figures
L278: this is consistent with the effects of BC as you’ve described – this should be noted here
L281: what do you mean by “synchronous”?
L287: it’s interesting but not surprising that the precipitation change is linear in the dry energy change on the global scale; using the formulation in e.g. Muller and O’Gorman 2011 (DOI: 10.1038/NCLIMATE1169), and used by Liu et al 2018 that you refer to, that latent heat gain equals losses to radiative cooling, sensible heat, and horizontal dry static energy flux, i.e. Lc*delP = delQ + delH; the dry heat is delQ, and delH is zero when averaged globally so you have delP/delQ = 1/Lc; a linear relation. You need to contextualise this section with prior results such as these, noting that Figure 4a is consistent with this. This does raise the question of what’s happening in the GISS 10xBC experiment, as this is some distance from the linear fit.
Given this, Figure 4b’s lack of linearity can be assumed to be due to a non-zero delH over the Indian region. E.g. on L307 it is noted that under sulx10asia precipitation decreases (negative Lc delP), and dry heat decreases (which represents an increase in the losses, i.e. positive delQ); this therefore must be compensated by a strongly negative delH i.e. horizontal energy flow into the region. See e.g. https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2019GL083479 and https://acp.copernicus.org/articles/21/10179/2021/ for more discussion on this kind of effect. Generally the balance between delQ and delH depends on the latitude, with a greater role for delH near the equator and delQ more so in the extratropics; so a mix seems sensible over this 10N-50N region you’re using.
L317: the global 5a plot should probably be in the supplement; you note it should be zero in x anyway, and you can show it on the regional plot. The near-zero locations of different models aren’t key to the results. I would use a different colour/style arrow on the regional plot to point to the global line, as they’re pretty similar currently.
Figure 5 seems to be exploring the non-dry energy changes i.e. delH, though not framed in this way currently. Perhaps the vertical velocity change is related to the horizontal energy flow i.e. delH?
L330: I’m not sure what this sentence means with “higher than that of global perturbation experiments” – 3 of the 4 experiments listed are global experiments.
Figure 6: this is interesting but there are some issues with the analysis:
There aren’t any error bars so it’s not possible to know if the differences are due to noise or a real effect. This is complicated by the fact that sulasiared only has 1 model and sulred has 2. This reflects a broader issue; we aren’t really comparing like-with-like when we compare different sets of models performing different experiments. Obviously you would throw a lot of information away if you used only the one model which has done sulasiared, but the claimed effects arise from only the models which ran each the related experiment, so there has to be a direct comparison between the experiments within just these smaller model subsets, at least in the supplement to verify the effect manifests in this case.
It seems odd that the precip cycle in sulasiared increases from May, whereas the sulred experiment features increases from July only, while both see similar responses in the temperature gradient to which these changes are attributed. This warrants exploration, and might suggest that there is a large role for internal variability, or other drivers, here. Similarly, it is interesting that the sulx10Asia and sulx5 experiments see little response in the temperature gradient, but large reductions in precip. This strong nonlinearity needs explaining and again suggests perhaps the temperature gradient isn’t the only driver here. You discuss this possibility in relation to other experiments which is useful but overall this needs expansion.
L342-343: not clear what this sentence is saying.
L350: you should clarify this is averaged zonally across Asia in this sentence
Figure 7: this is interesting, but there is no scale for the circulation changes. The apparent warming regions under increased SO2 are odd – maybe there’s essentially zero change here but the mean happens to be >0? Would suggest making the areas white if the magnitude of the change is below some threshold say 0.1K as the smallest contour is currently 0.4K.
L377: it’s good to note the ITCZ but this effect isn’t explored further in this way. This could be a good avenue for digging further into the response.
L801 “Table 1: Description of the 11 models used from the Precipitation Driver Model” – missing “Response”
In the abstract you say you present “potential underlying physical processes under a variety of climate forcings that would be useful in designing further model experiments with higher spatial resolution” – what do you mean by this? It isn’t elaborated further in the main text.
Citation: https://doi.org/10.5194/egusphere-2023-1605-RC1 - AC1: 'Reply on RC1', Sushant Das, 19 Jan 2024
-
RC2: 'Comment on egusphere-2023-1605', Anonymous Referee #1, 25 Aug 2023
The authors investigated the responses of global and India precipitation to CO2 and aerosol forcing with a set of LUMIP model simulations, and found the different responses to CO2 and aerosols on both global and regional scales. The main mechanisms driving the responses are dynamical responses. The results are robust and well presented. However, this paper needs substantial revisions before considerations for publication.
Major comments:
- The main concern regarding this study is that the results are not new. Many literatures have documented the impact of greenhouses and aerosols on India monsoon. The authors just qualitatively analyzed some dynamical responses, contributing limited new insights.
- The direct comparison between different simulations throughout this paper does not have meaning. For example, the authors compared the precipitation response in BCx10 experiment with that in the CO2x2 experiment (Line 221). The global effective radiative forcing (ERF) in the CO2x2 is roughly 3.7 W m-2 while the forcing in the BCx10 experiment is around 1.4 W m-2. The forcing in the BCasiax10 would be even smaller. It is not surprising that the response in the CO2x2 experiment is larger than BC in most regions due to its larger forcing and larger temperature change, which does not contribute any new insights. I suggest the authors normalize the results by warming level or TOA forcing. In that case, the results shown would be precipitation change per K (e.g., Figure 4 of Myhre et al., 2017) or change per unit forcing (e.g., Tang et al., 2018), following earlier PDRMIP studies. An advantage of normalized results is that you can find how sensitive the precipitation is in India to different forcing agent, and if scaled by historical forcing, you can then compare their historical contributions.
- The linear relationship between global precipitation change and temperature or energy change has been discussed by previous work, including using PDRMIP. The results presented here are not new. What the authors conclude from these analyses (Figure 3-5)?
Minor comments:
- Line 57-59, we have so many precipitation data (obs. Station, satellite, radar, balloon). Why quantifying precipitation change is challenging?
- Line 62, a comma after 1.5C.
- Line 83, “Zhao and Suzuki (2019) using…” or “used”?
- Line 88, “decreases…”
- Line 89-90, precipitation decreases over NH caused southward migration of ITCZ or the latter caused the former?
- Line 92, uniform?
- Line 94, play?
- line 109-110, attributed…due to or to?
- Line 113, comma after ‘seas’,
- Line 122, Ganguly et al., (2012), using or used
- Line 129-130, “due to computational constraints”, any literature on this statement?
- Line 131-134, any literature supporting this claim? There are many studies using coupled model simulations.
- Line 152, due to or to?
- Line 185, grammar error
- Line 244-247, grammar error.
- Line 196-298, any evidence to support this claim?
- Line 306 to 308, why these models are inconsistent? Any explanations?
References
Myhre, G., and Coauthors, 2017: PDRMIP: A Precipitation Driver and Response Model Intercomparison Project—Protocol and Preliminary Results. Bull. Amer. Meteor. Soc., 98, 1185–1198, https://doi.org/10.1175/BAMS-D-16-0019.1.
Tang, T., Shindell, D., Samset, B. H., Boucher, O., Forster, P. M., Hodnebrog, Ø., Myhre, G., Sillmann, J., Voulgarakis, A., Andrews, T., Faluvegi, G., Fläschner, D., Iversen, T., Kasoar, M., Kharin, V., Kirkevåg, A., Lamarque, J.-F., Olivié, D., Richardson, T., Stjern, C. W., and Takemura, T.: Dynamical response of Mediterranean precipitation to greenhouse gases and aerosols, Atmos. Chem. Phys., 18, 8439–8452, https://doi.org/10.5194/acp-18-8439-2018, 2018.
Citation: https://doi.org/10.5194/egusphere-2023-1605-RC2 - AC3: 'Reply on RC2', Sushant Das, 19 Jan 2024
-
RC3: 'Comment on egusphere-2023-1605', Anonymous Referee #3, 01 Sep 2023
The authors utilized the perturbation experiments to investigate the global and Indian precipitation responses to external forcings, including CO2, black carbon (BC), and sulfate aerosol. Based on the analysis discussed in the manuscript, the authors demonstrated that the precipitation over the Indian subcontinent shows different responses to aerosols compared to the global precipitation changes, which can be explained by aerosol-induced circulation changes. The regional precipitation responses induced by the circulation shifts are interesting and worth investigating. However, a large piece of the conclusion discussed and summarized in the manuscript, unfortunately, is well documented by multiple previous studies. In addition, the current analysis of the circulation part stopped at a very preliminary step, further weakening the robustness of the current conclusions. Overall, the motivation of this manuscript is interesting, but the current manuscript needs careful revision and deeper analysis. It would be more valuable to the community if the authors could focus on the regional responses due to the circulation shift.
Major comments:
- The first two conclusions in section 4 are well-documented in multiple previous studies, especially the global precipitation responses to CO2 and anthropogenic aerosol. The authors spent too many words discussing these well-known results, which leads to a redundant section 3. It would be better if the authors acknowledged the existing literature in the introduction and included more regional changes and dynamics analyses in section 3.
- The major part of the study focuses on the annual total precipitation, which is strange. The authors already mentioned ITCZ and the monsoon activities may be important to the regional precipitation changes over India, and it is less investigated. If so, why not focus on the monsoon season? The circulation background near India varies greatly, comparing the monsoon seasons and the other months.
- All the discussion based on the sulaasiared and sulred is unconvincing because of the limited ensemble sizes (actually, they should not be called an ensemble), meaning the internal variability heavily impacts the related patterns. I suggest removing any discussions on these two experiments.
- The magnitude comparisons between experiments in the first few paragraphs of section 3 are less meaningful because of the totally different global energy budgets between the models. For example, the authors argued that the magnitude of temperature/precipitation changes over the Tibetan Plateau in bcx10 is greater than that in bcx10asia (lines 204–244), but this does not indicate any useful information because the bc forcing in bcx10 is much greater than that in bcx10asia. Magnitude comparisons between sulx5 and sulx10asia are also problematic because of the same issue. The authors should only look at the spatial patterns or relative changes in magnitude.
- Some of the descriptions of the spatial patterns are less convincing. For example, the “increase in precipitation over most of the continental land region in co2x2” is inconsistent with what is shown in Fig. 1a. If I read it correctly, most of the increases in Fig1a occur over the ocean, while land regions show negligible changes based on the current colormap. Similar issues occur repeatedly in this paragraph.
- The authors kept describing the potential importance of ITCZ in forcing regional precipitation changes, which is certainly important to better understand the dynamical responses. However, no figures/analyses are included in the current manuscript. I suggest putting more effort into the analysis and discussion about ITCZ.
Minor comments:
Line 61: should remove ‘its’
Line 250–253: It might be better to move the literature summaries to the introduction section.
Line 255: The precipitation changes during monsoon over India are not shown in the figures.
Line 270–280: Most of the discussions have been covered in previous paragraphs about Fig. 1.
Line 363–364: The argument about heavier air seems wrong. It is mainly due to the lower water vapor capacity and weaker upwelling.
Line 366–368: The temperature changes in Fig 7f are rather small, which means all the discussions here are unconvincing before the significance test.
Line 367: The downdrafts should be more related to circulation shifts than tropospheric cooling. Please double check
Line 462–464: Unclear. Please rewrite the sentence.
Citation: https://doi.org/10.5194/egusphere-2023-1605-RC3 - AC2: 'Reply on RC3', Sushant Das, 19 Jan 2024
Status: closed
-
RC1: 'Comment on egusphere-2023-1605', Anonymous Referee #2, 22 Aug 2023
General Comments
The paper uses a promising framework to study the ESM-estimated effect of different climate forcers on the Indian Monsoon using the PDRMIP dataset. However, the current paper unfortunately spends far too long describing general effects which are already well-known, devoting insufficient space to what could be a useful investigation into the specific response of the Indian Monsoon.
In addition, the framework for analysing the precipitation response is incomplete, and doesn’t utilise previous work on the topic, again preventing a deep dive into novel grounds.
The paper could, with some substantial changes and additions, present a useful analysis of the response of a key climate phenomenon to anthropogenic emissions changes. These should be carried out before a further review can be done. In particular, the work has to be situated amongst the existing literature and present a new avenue of research.
The first few pages of the results present the global and spatial impacts of CO2, BC, and Sulfate, which are already well-documented elsewhere including using this dataset. Some standard, intuitive effects – such as the effect of global perturbations being larger than that of regional ones – are noted without discussing that these are consistent. This is reflected in the conclusion section, in which the first 2 points represent well-known effects, without noting this.
The approach to modelling the energy budget is convoluted. The initial focus on dry energy leads to conclusions on the linearity between precipitation and dry energy on the global scale, without noting that this has been documented in prior work (see specific references below). Local discrepancies from linearity over India are then explored without explicit reference to the change in horizontal flow that they represent.
I would suggest that this paper be re-oriented to focus specifically on the Indian Monsoon, with little need for global analysis. With this, and a much-reduced focus on previously-documented results, more space would then be available for devotion to a deeper dive into, and discussion of, the effects of these forcing agents on the Indian Monsoon. This should include further analysis of the ITCZ and horizontal energy transfer, and more extensive situation of the results within the wider literature.
Focus should be given to error analysis; currently only multi-model means are presented in most cases, with no quantification of R squared values for linear fits, and no discussion of internal variability in the annual cycles in Figure 6. One issue which needs exploration is the differing sets of models which undertook each experiment. Many conclusions are drawn from Sulfur reduction experiments with 1 or 2 models; these merit explicit comparison to the base experiments from those same model sets, and analysis of inter-annual variability when only 1 or 2 models are available.
Specific Comments
L62 “The GHGs have significantly warmed the climate by 1.5°C” – this is the level that would have occurred in the absence of Aerosols, right? Present-day warming is closer to 1C.
L190: what type of interpolation did you apply to each variable?
This part warrants expansion, with reference to other studies which have looked at this. PDRMIP was used by Xie et al 2022 (in the reference list but not referred to in the text https://www.nature.com/articles/s43247-022-00660-x), and in https://acp.copernicus.org/articles/20/11823/2020/ using a different energy budget analysis, and you need to be clearer on the novelty of your approach versus this. The motivation for focusing only on the coupled response should be made clear.
L192: what is the motivation for focusing only on the dry component of the energy budget? Will this not miss the latent heat changes?
L225: the relationship between northern hemisphere cooling and ITCZ shifting warrants explanation, with references.
L231: “Reducing the sulfate aerosols enhances the surface warming as noticed in the Figure 2h, which can alter the climate sensitivity leading to various feedbacks” – need to be more careful with terminology; it reads like you’re suggesting the Equilibrium Climate Sensitivity is altered by the sulfate response.
L234: it should be noted here that this consistency between sulred and sulx5 isn’t surprising
L241: I don’t follow the logic of “global BC aerosols contribute more to precipitation increase than the Asian emitted BC aerosols”. The global BC experiment contains the Asia perturbation, and additionally increases BC outside Asia; it’s not surprising that this overall effect is larger. It would be surprising if the Indian precipitation change per change in BC emissions was larger for Global than Asian emissions, but I don’t think this is what you’re suggesting here.
L263 “surface temperature (Figure 1g).” – Figure 2g
L267: with the important exception of BC perturbations! It is well worth quantifying these by displaying the gradients and R squareds of linear fits for the Global and Indian regions to back up your point
L272: you should give exact values for these numbers rather than approximations, to a consistent number of significant figures
L278: this is consistent with the effects of BC as you’ve described – this should be noted here
L281: what do you mean by “synchronous”?
L287: it’s interesting but not surprising that the precipitation change is linear in the dry energy change on the global scale; using the formulation in e.g. Muller and O’Gorman 2011 (DOI: 10.1038/NCLIMATE1169), and used by Liu et al 2018 that you refer to, that latent heat gain equals losses to radiative cooling, sensible heat, and horizontal dry static energy flux, i.e. Lc*delP = delQ + delH; the dry heat is delQ, and delH is zero when averaged globally so you have delP/delQ = 1/Lc; a linear relation. You need to contextualise this section with prior results such as these, noting that Figure 4a is consistent with this. This does raise the question of what’s happening in the GISS 10xBC experiment, as this is some distance from the linear fit.
Given this, Figure 4b’s lack of linearity can be assumed to be due to a non-zero delH over the Indian region. E.g. on L307 it is noted that under sulx10asia precipitation decreases (negative Lc delP), and dry heat decreases (which represents an increase in the losses, i.e. positive delQ); this therefore must be compensated by a strongly negative delH i.e. horizontal energy flow into the region. See e.g. https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2019GL083479 and https://acp.copernicus.org/articles/21/10179/2021/ for more discussion on this kind of effect. Generally the balance between delQ and delH depends on the latitude, with a greater role for delH near the equator and delQ more so in the extratropics; so a mix seems sensible over this 10N-50N region you’re using.
L317: the global 5a plot should probably be in the supplement; you note it should be zero in x anyway, and you can show it on the regional plot. The near-zero locations of different models aren’t key to the results. I would use a different colour/style arrow on the regional plot to point to the global line, as they’re pretty similar currently.
Figure 5 seems to be exploring the non-dry energy changes i.e. delH, though not framed in this way currently. Perhaps the vertical velocity change is related to the horizontal energy flow i.e. delH?
L330: I’m not sure what this sentence means with “higher than that of global perturbation experiments” – 3 of the 4 experiments listed are global experiments.
Figure 6: this is interesting but there are some issues with the analysis:
There aren’t any error bars so it’s not possible to know if the differences are due to noise or a real effect. This is complicated by the fact that sulasiared only has 1 model and sulred has 2. This reflects a broader issue; we aren’t really comparing like-with-like when we compare different sets of models performing different experiments. Obviously you would throw a lot of information away if you used only the one model which has done sulasiared, but the claimed effects arise from only the models which ran each the related experiment, so there has to be a direct comparison between the experiments within just these smaller model subsets, at least in the supplement to verify the effect manifests in this case.
It seems odd that the precip cycle in sulasiared increases from May, whereas the sulred experiment features increases from July only, while both see similar responses in the temperature gradient to which these changes are attributed. This warrants exploration, and might suggest that there is a large role for internal variability, or other drivers, here. Similarly, it is interesting that the sulx10Asia and sulx5 experiments see little response in the temperature gradient, but large reductions in precip. This strong nonlinearity needs explaining and again suggests perhaps the temperature gradient isn’t the only driver here. You discuss this possibility in relation to other experiments which is useful but overall this needs expansion.
L342-343: not clear what this sentence is saying.
L350: you should clarify this is averaged zonally across Asia in this sentence
Figure 7: this is interesting, but there is no scale for the circulation changes. The apparent warming regions under increased SO2 are odd – maybe there’s essentially zero change here but the mean happens to be >0? Would suggest making the areas white if the magnitude of the change is below some threshold say 0.1K as the smallest contour is currently 0.4K.
L377: it’s good to note the ITCZ but this effect isn’t explored further in this way. This could be a good avenue for digging further into the response.
L801 “Table 1: Description of the 11 models used from the Precipitation Driver Model” – missing “Response”
In the abstract you say you present “potential underlying physical processes under a variety of climate forcings that would be useful in designing further model experiments with higher spatial resolution” – what do you mean by this? It isn’t elaborated further in the main text.
Citation: https://doi.org/10.5194/egusphere-2023-1605-RC1 - AC1: 'Reply on RC1', Sushant Das, 19 Jan 2024
-
RC2: 'Comment on egusphere-2023-1605', Anonymous Referee #1, 25 Aug 2023
The authors investigated the responses of global and India precipitation to CO2 and aerosol forcing with a set of LUMIP model simulations, and found the different responses to CO2 and aerosols on both global and regional scales. The main mechanisms driving the responses are dynamical responses. The results are robust and well presented. However, this paper needs substantial revisions before considerations for publication.
Major comments:
- The main concern regarding this study is that the results are not new. Many literatures have documented the impact of greenhouses and aerosols on India monsoon. The authors just qualitatively analyzed some dynamical responses, contributing limited new insights.
- The direct comparison between different simulations throughout this paper does not have meaning. For example, the authors compared the precipitation response in BCx10 experiment with that in the CO2x2 experiment (Line 221). The global effective radiative forcing (ERF) in the CO2x2 is roughly 3.7 W m-2 while the forcing in the BCx10 experiment is around 1.4 W m-2. The forcing in the BCasiax10 would be even smaller. It is not surprising that the response in the CO2x2 experiment is larger than BC in most regions due to its larger forcing and larger temperature change, which does not contribute any new insights. I suggest the authors normalize the results by warming level or TOA forcing. In that case, the results shown would be precipitation change per K (e.g., Figure 4 of Myhre et al., 2017) or change per unit forcing (e.g., Tang et al., 2018), following earlier PDRMIP studies. An advantage of normalized results is that you can find how sensitive the precipitation is in India to different forcing agent, and if scaled by historical forcing, you can then compare their historical contributions.
- The linear relationship between global precipitation change and temperature or energy change has been discussed by previous work, including using PDRMIP. The results presented here are not new. What the authors conclude from these analyses (Figure 3-5)?
Minor comments:
- Line 57-59, we have so many precipitation data (obs. Station, satellite, radar, balloon). Why quantifying precipitation change is challenging?
- Line 62, a comma after 1.5C.
- Line 83, “Zhao and Suzuki (2019) using…” or “used”?
- Line 88, “decreases…”
- Line 89-90, precipitation decreases over NH caused southward migration of ITCZ or the latter caused the former?
- Line 92, uniform?
- Line 94, play?
- line 109-110, attributed…due to or to?
- Line 113, comma after ‘seas’,
- Line 122, Ganguly et al., (2012), using or used
- Line 129-130, “due to computational constraints”, any literature on this statement?
- Line 131-134, any literature supporting this claim? There are many studies using coupled model simulations.
- Line 152, due to or to?
- Line 185, grammar error
- Line 244-247, grammar error.
- Line 196-298, any evidence to support this claim?
- Line 306 to 308, why these models are inconsistent? Any explanations?
References
Myhre, G., and Coauthors, 2017: PDRMIP: A Precipitation Driver and Response Model Intercomparison Project—Protocol and Preliminary Results. Bull. Amer. Meteor. Soc., 98, 1185–1198, https://doi.org/10.1175/BAMS-D-16-0019.1.
Tang, T., Shindell, D., Samset, B. H., Boucher, O., Forster, P. M., Hodnebrog, Ø., Myhre, G., Sillmann, J., Voulgarakis, A., Andrews, T., Faluvegi, G., Fläschner, D., Iversen, T., Kasoar, M., Kharin, V., Kirkevåg, A., Lamarque, J.-F., Olivié, D., Richardson, T., Stjern, C. W., and Takemura, T.: Dynamical response of Mediterranean precipitation to greenhouse gases and aerosols, Atmos. Chem. Phys., 18, 8439–8452, https://doi.org/10.5194/acp-18-8439-2018, 2018.
Citation: https://doi.org/10.5194/egusphere-2023-1605-RC2 - AC3: 'Reply on RC2', Sushant Das, 19 Jan 2024
-
RC3: 'Comment on egusphere-2023-1605', Anonymous Referee #3, 01 Sep 2023
The authors utilized the perturbation experiments to investigate the global and Indian precipitation responses to external forcings, including CO2, black carbon (BC), and sulfate aerosol. Based on the analysis discussed in the manuscript, the authors demonstrated that the precipitation over the Indian subcontinent shows different responses to aerosols compared to the global precipitation changes, which can be explained by aerosol-induced circulation changes. The regional precipitation responses induced by the circulation shifts are interesting and worth investigating. However, a large piece of the conclusion discussed and summarized in the manuscript, unfortunately, is well documented by multiple previous studies. In addition, the current analysis of the circulation part stopped at a very preliminary step, further weakening the robustness of the current conclusions. Overall, the motivation of this manuscript is interesting, but the current manuscript needs careful revision and deeper analysis. It would be more valuable to the community if the authors could focus on the regional responses due to the circulation shift.
Major comments:
- The first two conclusions in section 4 are well-documented in multiple previous studies, especially the global precipitation responses to CO2 and anthropogenic aerosol. The authors spent too many words discussing these well-known results, which leads to a redundant section 3. It would be better if the authors acknowledged the existing literature in the introduction and included more regional changes and dynamics analyses in section 3.
- The major part of the study focuses on the annual total precipitation, which is strange. The authors already mentioned ITCZ and the monsoon activities may be important to the regional precipitation changes over India, and it is less investigated. If so, why not focus on the monsoon season? The circulation background near India varies greatly, comparing the monsoon seasons and the other months.
- All the discussion based on the sulaasiared and sulred is unconvincing because of the limited ensemble sizes (actually, they should not be called an ensemble), meaning the internal variability heavily impacts the related patterns. I suggest removing any discussions on these two experiments.
- The magnitude comparisons between experiments in the first few paragraphs of section 3 are less meaningful because of the totally different global energy budgets between the models. For example, the authors argued that the magnitude of temperature/precipitation changes over the Tibetan Plateau in bcx10 is greater than that in bcx10asia (lines 204–244), but this does not indicate any useful information because the bc forcing in bcx10 is much greater than that in bcx10asia. Magnitude comparisons between sulx5 and sulx10asia are also problematic because of the same issue. The authors should only look at the spatial patterns or relative changes in magnitude.
- Some of the descriptions of the spatial patterns are less convincing. For example, the “increase in precipitation over most of the continental land region in co2x2” is inconsistent with what is shown in Fig. 1a. If I read it correctly, most of the increases in Fig1a occur over the ocean, while land regions show negligible changes based on the current colormap. Similar issues occur repeatedly in this paragraph.
- The authors kept describing the potential importance of ITCZ in forcing regional precipitation changes, which is certainly important to better understand the dynamical responses. However, no figures/analyses are included in the current manuscript. I suggest putting more effort into the analysis and discussion about ITCZ.
Minor comments:
Line 61: should remove ‘its’
Line 250–253: It might be better to move the literature summaries to the introduction section.
Line 255: The precipitation changes during monsoon over India are not shown in the figures.
Line 270–280: Most of the discussions have been covered in previous paragraphs about Fig. 1.
Line 363–364: The argument about heavier air seems wrong. It is mainly due to the lower water vapor capacity and weaker upwelling.
Line 366–368: The temperature changes in Fig 7f are rather small, which means all the discussions here are unconvincing before the significance test.
Line 367: The downdrafts should be more related to circulation shifts than tropospheric cooling. Please double check
Line 462–464: Unclear. Please rewrite the sentence.
Citation: https://doi.org/10.5194/egusphere-2023-1605-RC3 - AC2: 'Reply on RC3', Sushant Das, 19 Jan 2024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 472 | 181 | 54 | 707 | 72 | 36 | 36 |
- HTML: 472
- PDF: 181
- XML: 54
- Total: 707
- Supplement: 72
- BibTeX: 36
- EndNote: 36
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 219 | 30 |
| India | 2 | 99 | 14 |
| China | 3 | 76 | 10 |
| Germany | 4 | 45 | 6 |
| United Kingdom | 5 | 29 | 4 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 219