the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 25 Mar 2025
Constraining a Radiative Transfer Model with Satellite Retrievals: Implications for Cirrus Cloud Thinning
Abstract. The complex mechanisms governing the formation of cirrus clouds pose significant challenges in the accurate simulation of cirrus clouds within climate models, leading to uncertainties in predicting the cirrus cloud response to aerosols and efficacy of cirrus cloud thinning (CCT), a climate intervention method. One issue is related to the relative contributions of homogeneous and heterogeneous ice nucleation. Recent satellite observations from the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) suggest that cirrus clouds strongly affected by homogeneous ice nucleation (i.e., homogeneous cirrus) play a more important role than previously assumed. We employ a radiative transfer model to quantify the cloud radiative effect for homogeneous and heterogeneous cirrus clouds at the top of atmosphere (TOA), Earth's surface, and within the atmosphere. The experiments are conducted using cirrus ice water content and effective diameter vertical profiles from CALIPSO retrievals for homogeneous and heterogeneous cirrus clouds across different regions (Arctic, Antarctic, and midlatitude) and surface types (ocean and land). Results indicate that homogeneous cirrus clouds exhibit stronger radiative effects than heterogeneous cirrus, implying that transitioning from homogeneous to heterogeneous cirrus, as an indicator of CCT efficacy, could induce substantial surface cooling, particularly in polar regions during winter. Estimated instantaneous surface cooling effects range from -0.7 to -1.0 W m-2, with the TOA cooling reaching up to -1.6 W m⁻2. This study highlights the need for improved representation of homogeneous cirrus in models to better predict the climatic impacts of cirrus clouds and to assess the CCT viability.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(2519 KB)
-
Supplement
(784 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2519 KB) - Metadata XML
-
Supplement
(784 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-1165', Anonymous Referee #1, 05 May 2025
- AC1: 'Reply on RC1', Ehsan Erfani, 31 Jul 2025
-
RC2: 'Comment on egusphere-2025-1165', Anonymous Referee #2, 09 May 2025
Review of “Constraining a Radiative Transfer Model with Satellite Retrievals: Implications for Cirrus Cloud Thinning”
This work by Erfani and Mitchell focuses on the quantitative analysis of radiative effects of two cirrus types – homogeneously formed versus heterogeneous formed cirrus. The methodology of the work is very straightforward and the contrast between the two scenarios is easy to follow. Overall, the manuscript is well written. The reviewer has some main comments regarding the interpretation of the contrast between two types of cirrus as the effects of cirrus thinning as described more in detail below. The reviewer thinks that the manuscript needs some major revisions, but this work is appropriate to be considered for publication in ACP after making these changes.
Main comments:
[1] The reviewer can see the merits of using satellite-based observations to constrain the radiative transfer models for testing the radiative effects between two different types of cirrus. But there are some main concerns about using the contrast between the two cirrus types to represent what would happen if the homogeneously formed cirrus were to be transformed or replaced by heterogeneously formed cirrus. Below the reviewer will list a few reasons for this concern.
The thermodynamic and dynamic conditions that form homogeneous cirrus based on satellite observations may be quite different from those supporting the formation of heterogeneous cirrus. Cirrus formed via homogeneous freezing often experienced higher relative humidity (RH) (such as under more turbulent conditions with more orographic waves). Thus, it is highly likely that the heterogeneous cirrus observed by satellite is not going to be the same as the modified cirrus formed via thinning method since they would form under different thermodynamic/dynamic conditions and at different locations.
Assuming that the dynamic conditions (turbulence or waves) would generate the same amount of excess water vapor mass concentrations over ice saturation, then the ice water content (IWC) of the cirrus clouds could be similar whether forming heterogeneous or homogeneous cirrus, but the two cirrus types may have significantly different lifetime, since larger ice particles formed heterogeneous generally will sediment faster than small ice. In addition, the sedimentation of ice can further lead to higher RH in the lower levels, which used to be drier. Such increasing water vapor in the lower altitudes also can increase LW trapping due to the water vapor’s greenhouse effect.
[2] It is not clear if the CCT here considers forming heterogeneous cirrus from clear-sky ice supersaturation (ISS) or transforming the existing homogeneous cirrus into heterogeneous cirrus. The latter one would need to consider competition between heterogeneous and homogeneous nucleation, which is quite a complex process as former cloud modeling studies showed. Some examples include:
Kärcher et al., 2022, Studies on the Competition Between Homogeneous and Heterogeneous Ice Nucleation in Cirrus Formation https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2021JD035805
Spichtinger and Cziczo, 2010, Impact of heterogeneous ice nuclei on homogeneous freezing events in cirrus clouds, https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2009JD012168
Spichtinger, P., and K. Gierens, 2009, Modelling of cirrus clouds. Part 2: Competition of different nucleation mechanisms, Atmos. Chem. Phys., 9, 2319–2334.
Barahona and A. Nenes, 2009, Parameterizing the competition between homogeneous and heterogeneous freezing in cirrus cloud formation – monodisperse ice nuclei, https://acp.copernicus.org/articles/9/369/2009/
Because the existing heterogeneous cirrus is formed directly from clear sky ISS, not going through this extra step of ISS to homogeneous cirrus and then adding new INPs, it is quite uncertain how the modified cirrus microphysical properties would be.
The reviewer’s suggestion includes two possible approaches for the revisions. For the first approach, because the work is anchored with satellite observations, which provide realistic estimates of radiative effects of two cirrus types, the title would be more appropriate if it emphasizes this contrast, such as: “Constraining a Radiative Transfer Model with Satellite Retrievals: Contrasts between Cirrus Formed via Homogeneous and Heterogeneous Freezing”. The reviewer disagrees with the other reviewer’s comments of making the CCT an emphasis in the title because of the caveats mentioned above. The introduction section should also be modified accordingly as well to tune down the emphasis solely on CCT, since the contrast of heterogeneous and homogeneous cirrus would not reflect the real CCT result. It is better to discuss more implications for CCT in the discussion section.
For the second approach of revision, if the authors prefer to keep focusing on CCT, the reviewer suggests that the study to include cloud-resolving model simulations, using models that can represent the more realistic aerosol-cloud interactions, especially the effects of INPs. One simulation can focus on adding INP to clear sky ISS, while another case can focus on adding INP to homogeneously formed cirrus. The authors can use the observed conditions that form homogeneous cirrus clouds to initialize the cloud model for various regions, which will make the results more realistic.
The current results in this work show that the radiative effects are highly sensitive to IWC, which is a useful finding. But this also points to the caveat that the fact that heterogeneous cirrus tend to have lower IWC than homogeneous cirrus based on observations, is not necessarily true if CCT creates new heterogeneous cirrus in conditions that used to favor homogeneous cirrus.
[3] Regarding the experimental design of the RTM tests, the reviewer has a main comment regarding the locations and seasons selected. Since this study is based on global-scale satellite observations, why not take advantage of global-scale satellite observations and provide a contrast between two cirrus types for all locations, four seasons, as well as annual average? The analysis does not need to limit itself to only Arctic, Antarctic, and NH midlatitudes in selected seasons, but can be applied to the global scale, such as every 10 degree by 10 degree or 30 degrees by 30 degrees in terms of latitudinal * longitudinal boxes, and compared between various seasons.
[4] The section 5 “Suggestions for improving cirrus cloud modeling” seems out of place. Most of the discussions in this section focuses on using previous work on various climate models to speculate how the model may lack the ability to do certain things. This section does not fit as a result section of a research article, and it is more like part of an introduction or a review article on the status of the current modeling work. In addition, the Section 5.1 uses a parameterization by Sun and Rikus (1999) and Sun (2001), which seems quite obsolete and may not be relevant to the current climate modeling field. But if this parameterization is still actively used in some models, can the authors provide the model’s name and version so the readers are aware of how impactful this parameterization is to the current climate modeling field? In addition, more up-to-date parameterizations are recommended, if the authors would like to keep this section 5.1 for comparison with model parameterizations. For instance, Gettelman and Morrison (2015, https://journals.ametsoc.org/view/journals/clim/28/3/jcli-d-14-00102.1.xml) or P3 scheme (Morrison, H., & Milbrandt, J. A., 2015, https://doi.org/10.1175/jas-d-14-0065.1) would be more relevant to the model users. Because of the caveats of this Section 5, the reviewer recommends that the entire section be removed, and the discussion of modeling implications can be summarized in a succinct way, like a paragraph, in the last conclusions and discussions section.
[5] The discussions for CCT of this work emphasizes one scenario of the CCT outcome, which is homogeneous cirrus replaced by heterogeneous cirrus, but it is possible that cloud seeding (adding INP) can also induce new cirrus formation from clear-sky ice supersaturated regions, where used to have no cirrus. For the radiative effects of forming cirrus from clear sky, a former study using RTM has shown large radiative impacts (X. Tan et al. 2016, https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2016GL071144). One remaining question is when adding INPs, how frequently this clear-sky ISS to cirrus scenario may happen, compared with the scenario of homogeneous cirrus changing to heterogeneous cirrus. Would the two scenarios cancel out each other or one would likely dominate. Some discussions regarding this uncertainty and their scope of impact would be good, to help the readers understand the complexity and uncertainty of the CCT method, since the current discussion seems to only focus on a favorable cooling effect through CCT.
Minor comments:
Line 128, the two references of Cziczo et al. 2013 and Froyd et al. 2022 did not sample the tropical tropopause layer, where homogeneous cirrus very likely dominates. So please revise the text of “heterogeneous cirrus is considered to be dominant globally”.
Line 139, about the Arctic amplification, some other studies also show 4 times faster warming rate in the Arctic than the rest of the global, such as Rantanen, M., et al., 2022. https://doi.org/10.1038/s43247-022-00498-3
Line 148, about the model’s biases for representing aerosol indirect effects (AIE) on cirrus, there are several newer papers showing that climate model simulations significantly underestimate the AIE compared with aircraft observations, such as Patnaude et al., 2021, https://acp.copernicus.org/articles/21/1835/2021/, and Maciel et al., 2023, https://acp.copernicus.org/articles/23/1103/2023/
Line 337, the authors mentioned here that the difference between two cirrus types are consistent between satellite and in-situ observations, such as Krämer et al. 2016, 2020. It would also be very helpful and important to know for the exact ranges of distributions for IWC, Ni and Di as a function of temperature, how the overall average values of satellite observed cirrus compare with in-situ observations for various regions. This is also related to the main comment #3 about producing a global map of radiative effects using the mean state of cirrus of satellite observations in each location. It would be very helpful if the authors can provide a short paragraph of discussion on how the microphysical properties of cirrus compare between satellite and aircraft observations at various geographical locations. Besides the study of Krämer et al., 2016 and 2020, which focused over European and African regions, other US-funded studies provided in-situ observed cirrus microphysical properties in complementary geographical locations, such as around the N. America, Pacific, and Southern Ocean regions, for example, Patnaude et al., 2020 (https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2019gl086550), Patnaude et al. 2021, https://acp.copernicus.org/articles/21/1835/2021/), and Ngo et al. (2025, https://egusphere.copernicus.org/preprints/2024/egusphere-2024-2122/)
This additional discussion on the similarity or differences between different observation techniques can provide the readers a deeper understanding of the uncertainties related to cirrus cloud measurements, which also demonstrates the importance of providing more accurate observations in order to narrow down the estimates of CCT outcomes for the general audience.
Overall, the reviewer thinks that the work provides fundamental quantifications of radiative effects of cirrus clouds, which is a major component in the Earth’s climate system. The reviewer would be happy to review the revised manuscript after these recommended changes have been accounted for by the authors.
Citation: https://doi.org/10.5194/egusphere-2025-1165-RC2 - AC1: 'Reply on RC1', Ehsan Erfani, 31 Jul 2025
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-1165', Anonymous Referee #1, 05 May 2025
Please find my comments in form of the uploaded supplement.
- AC1: 'Reply on RC1', Ehsan Erfani, 31 Jul 2025
-
RC2: 'Comment on egusphere-2025-1165', Anonymous Referee #2, 09 May 2025
Review of “Constraining a Radiative Transfer Model with Satellite Retrievals: Implications for Cirrus Cloud Thinning”
This work by Erfani and Mitchell focuses on the quantitative analysis of radiative effects of two cirrus types – homogeneously formed versus heterogeneous formed cirrus. The methodology of the work is very straightforward and the contrast between the two scenarios is easy to follow. Overall, the manuscript is well written. The reviewer has some main comments regarding the interpretation of the contrast between two types of cirrus as the effects of cirrus thinning as described more in detail below. The reviewer thinks that the manuscript needs some major revisions, but this work is appropriate to be considered for publication in ACP after making these changes.
Main comments:
[1] The reviewer can see the merits of using satellite-based observations to constrain the radiative transfer models for testing the radiative effects between two different types of cirrus. But there are some main concerns about using the contrast between the two cirrus types to represent what would happen if the homogeneously formed cirrus were to be transformed or replaced by heterogeneously formed cirrus. Below the reviewer will list a few reasons for this concern.
The thermodynamic and dynamic conditions that form homogeneous cirrus based on satellite observations may be quite different from those supporting the formation of heterogeneous cirrus. Cirrus formed via homogeneous freezing often experienced higher relative humidity (RH) (such as under more turbulent conditions with more orographic waves). Thus, it is highly likely that the heterogeneous cirrus observed by satellite is not going to be the same as the modified cirrus formed via thinning method since they would form under different thermodynamic/dynamic conditions and at different locations.
Assuming that the dynamic conditions (turbulence or waves) would generate the same amount of excess water vapor mass concentrations over ice saturation, then the ice water content (IWC) of the cirrus clouds could be similar whether forming heterogeneous or homogeneous cirrus, but the two cirrus types may have significantly different lifetime, since larger ice particles formed heterogeneous generally will sediment faster than small ice. In addition, the sedimentation of ice can further lead to higher RH in the lower levels, which used to be drier. Such increasing water vapor in the lower altitudes also can increase LW trapping due to the water vapor’s greenhouse effect.
[2] It is not clear if the CCT here considers forming heterogeneous cirrus from clear-sky ice supersaturation (ISS) or transforming the existing homogeneous cirrus into heterogeneous cirrus. The latter one would need to consider competition between heterogeneous and homogeneous nucleation, which is quite a complex process as former cloud modeling studies showed. Some examples include:
Kärcher et al., 2022, Studies on the Competition Between Homogeneous and Heterogeneous Ice Nucleation in Cirrus Formation https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2021JD035805
Spichtinger and Cziczo, 2010, Impact of heterogeneous ice nuclei on homogeneous freezing events in cirrus clouds, https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2009JD012168
Spichtinger, P., and K. Gierens, 2009, Modelling of cirrus clouds. Part 2: Competition of different nucleation mechanisms, Atmos. Chem. Phys., 9, 2319–2334.
Barahona and A. Nenes, 2009, Parameterizing the competition between homogeneous and heterogeneous freezing in cirrus cloud formation – monodisperse ice nuclei, https://acp.copernicus.org/articles/9/369/2009/
Because the existing heterogeneous cirrus is formed directly from clear sky ISS, not going through this extra step of ISS to homogeneous cirrus and then adding new INPs, it is quite uncertain how the modified cirrus microphysical properties would be.
The reviewer’s suggestion includes two possible approaches for the revisions. For the first approach, because the work is anchored with satellite observations, which provide realistic estimates of radiative effects of two cirrus types, the title would be more appropriate if it emphasizes this contrast, such as: “Constraining a Radiative Transfer Model with Satellite Retrievals: Contrasts between Cirrus Formed via Homogeneous and Heterogeneous Freezing”. The reviewer disagrees with the other reviewer’s comments of making the CCT an emphasis in the title because of the caveats mentioned above. The introduction section should also be modified accordingly as well to tune down the emphasis solely on CCT, since the contrast of heterogeneous and homogeneous cirrus would not reflect the real CCT result. It is better to discuss more implications for CCT in the discussion section.
For the second approach of revision, if the authors prefer to keep focusing on CCT, the reviewer suggests that the study to include cloud-resolving model simulations, using models that can represent the more realistic aerosol-cloud interactions, especially the effects of INPs. One simulation can focus on adding INP to clear sky ISS, while another case can focus on adding INP to homogeneously formed cirrus. The authors can use the observed conditions that form homogeneous cirrus clouds to initialize the cloud model for various regions, which will make the results more realistic.
The current results in this work show that the radiative effects are highly sensitive to IWC, which is a useful finding. But this also points to the caveat that the fact that heterogeneous cirrus tend to have lower IWC than homogeneous cirrus based on observations, is not necessarily true if CCT creates new heterogeneous cirrus in conditions that used to favor homogeneous cirrus.
[3] Regarding the experimental design of the RTM tests, the reviewer has a main comment regarding the locations and seasons selected. Since this study is based on global-scale satellite observations, why not take advantage of global-scale satellite observations and provide a contrast between two cirrus types for all locations, four seasons, as well as annual average? The analysis does not need to limit itself to only Arctic, Antarctic, and NH midlatitudes in selected seasons, but can be applied to the global scale, such as every 10 degree by 10 degree or 30 degrees by 30 degrees in terms of latitudinal * longitudinal boxes, and compared between various seasons.
[4] The section 5 “Suggestions for improving cirrus cloud modeling” seems out of place. Most of the discussions in this section focuses on using previous work on various climate models to speculate how the model may lack the ability to do certain things. This section does not fit as a result section of a research article, and it is more like part of an introduction or a review article on the status of the current modeling work. In addition, the Section 5.1 uses a parameterization by Sun and Rikus (1999) and Sun (2001), which seems quite obsolete and may not be relevant to the current climate modeling field. But if this parameterization is still actively used in some models, can the authors provide the model’s name and version so the readers are aware of how impactful this parameterization is to the current climate modeling field? In addition, more up-to-date parameterizations are recommended, if the authors would like to keep this section 5.1 for comparison with model parameterizations. For instance, Gettelman and Morrison (2015, https://journals.ametsoc.org/view/journals/clim/28/3/jcli-d-14-00102.1.xml) or P3 scheme (Morrison, H., & Milbrandt, J. A., 2015, https://doi.org/10.1175/jas-d-14-0065.1) would be more relevant to the model users. Because of the caveats of this Section 5, the reviewer recommends that the entire section be removed, and the discussion of modeling implications can be summarized in a succinct way, like a paragraph, in the last conclusions and discussions section.
[5] The discussions for CCT of this work emphasizes one scenario of the CCT outcome, which is homogeneous cirrus replaced by heterogeneous cirrus, but it is possible that cloud seeding (adding INP) can also induce new cirrus formation from clear-sky ice supersaturated regions, where used to have no cirrus. For the radiative effects of forming cirrus from clear sky, a former study using RTM has shown large radiative impacts (X. Tan et al. 2016, https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2016GL071144). One remaining question is when adding INPs, how frequently this clear-sky ISS to cirrus scenario may happen, compared with the scenario of homogeneous cirrus changing to heterogeneous cirrus. Would the two scenarios cancel out each other or one would likely dominate. Some discussions regarding this uncertainty and their scope of impact would be good, to help the readers understand the complexity and uncertainty of the CCT method, since the current discussion seems to only focus on a favorable cooling effect through CCT.
Minor comments:
Line 128, the two references of Cziczo et al. 2013 and Froyd et al. 2022 did not sample the tropical tropopause layer, where homogeneous cirrus very likely dominates. So please revise the text of “heterogeneous cirrus is considered to be dominant globally”.
Line 139, about the Arctic amplification, some other studies also show 4 times faster warming rate in the Arctic than the rest of the global, such as Rantanen, M., et al., 2022. https://doi.org/10.1038/s43247-022-00498-3
Line 148, about the model’s biases for representing aerosol indirect effects (AIE) on cirrus, there are several newer papers showing that climate model simulations significantly underestimate the AIE compared with aircraft observations, such as Patnaude et al., 2021, https://acp.copernicus.org/articles/21/1835/2021/, and Maciel et al., 2023, https://acp.copernicus.org/articles/23/1103/2023/
Line 337, the authors mentioned here that the difference between two cirrus types are consistent between satellite and in-situ observations, such as Krämer et al. 2016, 2020. It would also be very helpful and important to know for the exact ranges of distributions for IWC, Ni and Di as a function of temperature, how the overall average values of satellite observed cirrus compare with in-situ observations for various regions. This is also related to the main comment #3 about producing a global map of radiative effects using the mean state of cirrus of satellite observations in each location. It would be very helpful if the authors can provide a short paragraph of discussion on how the microphysical properties of cirrus compare between satellite and aircraft observations at various geographical locations. Besides the study of Krämer et al., 2016 and 2020, which focused over European and African regions, other US-funded studies provided in-situ observed cirrus microphysical properties in complementary geographical locations, such as around the N. America, Pacific, and Southern Ocean regions, for example, Patnaude et al., 2020 (https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2019gl086550), Patnaude et al. 2021, https://acp.copernicus.org/articles/21/1835/2021/), and Ngo et al. (2025, https://egusphere.copernicus.org/preprints/2024/egusphere-2024-2122/)
This additional discussion on the similarity or differences between different observation techniques can provide the readers a deeper understanding of the uncertainties related to cirrus cloud measurements, which also demonstrates the importance of providing more accurate observations in order to narrow down the estimates of CCT outcomes for the general audience.
Overall, the reviewer thinks that the work provides fundamental quantifications of radiative effects of cirrus clouds, which is a major component in the Earth’s climate system. The reviewer would be happy to review the revised manuscript after these recommended changes have been accounted for by the authors.
Citation: https://doi.org/10.5194/egusphere-2025-1165-RC2 - AC1: 'Reply on RC1', Ehsan Erfani, 31 Jul 2025
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 956 | 158 | 28 | 1,142 | 50 | 27 | 44 |
- HTML: 956
- PDF: 158
- XML: 28
- Total: 1,142
- Supplement: 50
- BibTeX: 27
- EndNote: 44
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
David L. Mitchell
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2519 KB) - Metadata XML
-
Supplement
(784 KB) - BibTeX
- EndNote
- Final revised paper
Please find my comments in form of the uploaded supplement.