the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 03 Jul 2023
Contrail formation on ambient aerosol particles for aircraft with hydrogen combustion: A box model trajectory study
Abstract. Future air traffic using (green) hydrogen (H2) promises zero carbon emissions but the effects of contrails from this new technology has hardly been investigated. We study contrail formation behind aircraft with H2 combustion by means of the particle-based Lagrangian Cloud Module (LCM) box model. Assuming the absence of soot and ultrafine volatile particle formation, contrail ice crystals form solely on atmospheric background particles mixed into the plume. While a recent study extended the original LCM with regard to the contrail formation on soot particles, we further advance the LCM to cover the contrail formation on ambient particles. For each simulation, we perform an ensemble of box model runs using the dilution along 1000 different plume trajectories that are based on 3D Large Eddy Simulations using the FLUDILES solver.
The formation threshold temperature of H2 contrails is by around 10 K higher than for conventional contrails (which form behind aircraft with kerosene combustion) due to a factor of 2.6 higher energy-specific water vapor emission. Therefore, contrail formation becomes primarily limited by the homogeneous freezing temperature of the water droplets formed on the ambient particles such that contrails can form at temperatures down to around 234 K.
The number of formed ice crystals varies strongly with ambient temperature even far away from the contrail formation threshold. The latter is because the water-supersaturation in the plume lasts longer for colder conditions and, hence, more of the entrained aerosol particles can form droplets and ice crystals. The contrail ice crystal number clearly increases for a higher ambient aerosol number concentration. The increase becomes weaker for higher number concentrations (>≈ 200 cm−3) and lower ambient temperatures (< 230 K). The ice crystal number decreases significantly for ambient particles with mean dry radii <≈ 10 nm due to the Kelvin effect.
Besides simulations with one aerosol particle ensemble, we analyze contrail formation scenarios with two co-existing aerosol particle ensembles that differ either in their mean dry size or hygroscopicity parameter. We compare them to scenarios with a single ensemble that is the average of the two ensembles. We find that the total ice crystal number can differ significantly between the two cases, in particular if nucleation mode particles are involved.
Due to the absence of soot particle emissions, the ice crystal number in H2 contrails is typically reduced by more than 80–90 % compared to conventional contrails. The contrail optical thickness is significantly reduced and H2 contrails either become later visible than kerosene contrails or are not visible at all for low ambient particle number concentrations. On the other hand, H2 contrails can form at lower flight altitudes where conventional contrails would not form.
-
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
(3672 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3672 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1321', Anonymous Referee #1, 21 Jul 2023
This is an excellent and timely analysis of the expected contrail behavior from an aviation gas turbine engine burning H2 fuel. There is increasing interest in this topic as H2 is being considered as a potential, longer-term solution to the problem of burning of fossil fuels by aviation. There are many questions and issues to adopting H2 fuels, but the changes in contrail properties are correctly identified by the authors as a key one, since contrails are such an important part of aviation's radiative impact.
I think the paper is very well written and the study very well executed and the paper can be published after addressing several small issues.
1. In section 2.1, the authors note that chemical ion nucleation is important source of particles in the UT. Some have, in the past, considered ion-nucleatiion in the aircraft exhaust plume. Since there are no soot particles and the current model is assuming nucleation on ambient particles, it is probably worth mentioning that ion nucleation from combustion chemi-iions is not being considered in this analysis. I don't think that is a shortcoming of the analysis, but it should be made clear that that may be another nucleation pathway, but is not being considered here (perhaps add in section 3.3?). If the authors have explicit reasons why they chose not to do so, that may be useful to add, but not necessary.
2. In section 4.2.2 line 433-434, the authors note "This behavior is similar to . . . kerosene combustion." I think it is notable and a bit remarkable that this is the case, despite the very different nuclei number density behavior (keerosene starts high and dilutes, while the H2 case has continual entrainment). I think a phrase should be added to emphasize that the dependence is similar DESPITE quite different nuclei number histories.
3. Fiigure 6 is an important result. However, I found the caption confusing. The legends do not have the T = 230 K line type (dashed) identified. And the caption only mentions this at the very end. I was trying to understand the first two panels and was reading that part of the caption and not finding out to what the dashed lines correspond. I would recommend putting the statement about the two temperatures at the beginning of the caption to avoid this problem, if they don't want to have the dashed lines in the figure legend itself.
4. Fiinally, section 5.1 is a bit inappropriate, I think. First, the analysis doesn't really do anything with oil particles. So this is sort of a side comment that this might need to be studies more. And the paper only cites one paper on oil, while there is signficant literature on aviation engine oil contributions to aviation PM emissions. And some of the other literature presents data significantly distinct from the one paper cited. The authors should either review the literature more broadly or minimize their discussion of the oil emission data. And the comment that "a hermetic and clean sealing of the engines from the oil system should be improved with regard to a complete jet oil recovery" indicates a lack of understanding of how the oil system on an aviation gas turbine engine functions. A comment along the lines of "reduction or elimination of oil emissions from the aircraft engines may be . . . a valuable mitigation effort for contrail formation" might be better. It should also be qualified ("may") since oil's role in contrails is still very poorly understood.
Citation: https://doi.org/10.5194/egusphere-2023-1321-RC1 - AC1: 'Comment on egusphere-2023-1321', Simon Unterstrasser, 19 Dec 2023
-
RC2: 'Comment on egusphere-2023-1321', Anonymous Referee #1, 21 Jul 2023
One more very minor comment from Reviewer 1:
line 454 "can easier form water droplets." might read better as "can form water droplets more easily."
The writing is so clear and clean, this somewhat awkward turn of phrase stands out. The meaning is still clear, so this is just a suggestion. Sorry I neglected to include this in my main review.
Citation: https://doi.org/10.5194/egusphere-2023-1321-RC2 - AC1: 'Comment on egusphere-2023-1321', Simon Unterstrasser, 19 Dec 2023
-
RC3: 'Comment on egusphere-2023-1321', Anonymous Referee #2, 20 Aug 2023
The manuscript explores how contrails might form and persist for zero-particle emissions engines with enhanced water vapor emissions, which are conditions relevant for hydrogen propulsion. Particle size distributions and hygroscopicities are assumed based on recent literature in order to parametrically understand the impacts on contrail ice crystal number throughout the near-field plume evolution. Interestingly, the results demonstrate that the increase in water vapor emissions and relatively low concentrations of available cloud condensation nuclei (CCN) allow contrails to form at warmer temperatures than for conventional kerosene burning engines, which could mean that a transition toward hydrogen propulsion results in an increase in climate-altering contrails. Simulations are carried out with both one and two lognormal size distribution modes, and the results are interesting for understanding the relative role of new particle formation versus Aitken and accumulation mode particles. These results hint that new particle formation from engine oil vapors may play a role in forming contrails; although, the choice of simulation parameters (e.g., forcing the concentrations of both modes to be the same) make it hard to understand how these particles may influence a contrail behind a hydrogen engine. Overall, the manuscript is well written and is an interesting and useful contribution to ACP and the contrail literature. I'd be inclined to recommend the paper for publication after the following comments are satisfactorily addressed:
1) The authors demonstrate awareness of recent work hinting that the venting of oil vapors may be a significant source of CCN for contrail formation in the "soot-poor" emissions regime, and the discussion in Section 5.1 is a good and important addition to the paper. However, I think not applying the model to directly explore the potential influence of these particles is a major gap in the present study. Could the 2-mode simulation be set up to create an additional figure to better understand how the contrail ice crystal number would change if a constant particle emissions source size mode from oil vapor with the following parameters were externally mixed with the ambient baseline particle number size distribution mode from Brock et al., 2021 and Section 4.3.1:
- Mode 1 (New Particle Formation From Continuously Vented Oil Vapor):
- r_d (nm) = 5, sigma = 1.2
- kappa = 0
- Emissions Index (number of particles emitted from 2 engines per second) = 10^11, 10^13, 10^15
- Mode 2 (Baseline ambient lognormal particle mode from Table 2):
- r_d (nm) = 15 nm, sigma = 1.6
- kappa = 0.5
- Concentration = 600 cm-3
The fundamental question is, do large numbers of small wetable, but insoluble particles dominate the contrail ice number in the presence of relatively few, large, sulfate particles from the ambient atmosophere?
2) Line 2: Change "has" to "have"
3) I find the use of the word "ensemble" to describe distinct size distribution modes to be confusing. I'd suggest replacing the word "ensemble" with "lognormal size mode" on Line 19 and throughout the manuscript.
4) Lines 33-34: Suggest "in particular on soot particles relative to co-emitted organic-sulfate particles"
5) Line 50: It would also be good to cite the field studies showing detection of oil signatures in engine particles for high-soot engines, albeit, with a focus on mass spectral detection methods that only see particles with diameters > 100 nm:
- Yu et al., 2010: https://doi.org/10.1021/es102145z
- Yu et al., 2012: https://doi.org/10.1021/es301692t
6) Line 74: Change "magnitudes" to "magnitude"
7) Line 78: large variability in atmospheric aerosol properties and co-emitted volatile particles from engine oil vapor
8) Line 130: It was not clear to me that the concentrations reported by Beer et al., 2020 in their supplement were for non-volatile particles. Concentrations of 200-300 cm^-3 seem high to me for dust and black carbon number. Please double check this.
9) Line 140: Suggest "higher" instead of "better"
10) Section 3.3.1, how would treating uncoated soot or dust as in soluble but able to adsorb water change things versus standard Köhler Theory, if at all? See, e.g., Kumar et al., 2009 (https://acp.copernicus.org/articles/9/2517/2009/).
11) Line 292: insert "is" to read "expression is the Kelvin term"
12) Line 325: Does this factor of 4 assume a 4-engine aircraft?
13) Line 364: is this because the soot particles are already well mixed with the plume water vapor?
14) Line 421: parameter values is hyphenated and misspelled
15) Lines 451-457: It's fascinating to think about this discussion in the context of low-kappa oil-nucleated particles that might be emitted by a hydrogen engine; although, it is not exactly the same set of parameters (as discussed above). I hope the authors will carry out the additional simulations to inform the interplay between new particle formation from co-emitted vapors and ambient aerosols
16) Lines 544: Strike "by"
17) Line 545: Replace "by" with "is"
18) Line 650: Strike "are"
Citation: https://doi.org/10.5194/egusphere-2023-1321-RC3 - AC1: 'Comment on egusphere-2023-1321', Simon Unterstrasser, 19 Dec 2023
- Mode 1 (New Particle Formation From Continuously Vented Oil Vapor):
- AC1: 'Comment on egusphere-2023-1321', Simon Unterstrasser, 19 Dec 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1321', Anonymous Referee #1, 21 Jul 2023
This is an excellent and timely analysis of the expected contrail behavior from an aviation gas turbine engine burning H2 fuel. There is increasing interest in this topic as H2 is being considered as a potential, longer-term solution to the problem of burning of fossil fuels by aviation. There are many questions and issues to adopting H2 fuels, but the changes in contrail properties are correctly identified by the authors as a key one, since contrails are such an important part of aviation's radiative impact.
I think the paper is very well written and the study very well executed and the paper can be published after addressing several small issues.
1. In section 2.1, the authors note that chemical ion nucleation is important source of particles in the UT. Some have, in the past, considered ion-nucleatiion in the aircraft exhaust plume. Since there are no soot particles and the current model is assuming nucleation on ambient particles, it is probably worth mentioning that ion nucleation from combustion chemi-iions is not being considered in this analysis. I don't think that is a shortcoming of the analysis, but it should be made clear that that may be another nucleation pathway, but is not being considered here (perhaps add in section 3.3?). If the authors have explicit reasons why they chose not to do so, that may be useful to add, but not necessary.
2. In section 4.2.2 line 433-434, the authors note "This behavior is similar to . . . kerosene combustion." I think it is notable and a bit remarkable that this is the case, despite the very different nuclei number density behavior (keerosene starts high and dilutes, while the H2 case has continual entrainment). I think a phrase should be added to emphasize that the dependence is similar DESPITE quite different nuclei number histories.
3. Fiigure 6 is an important result. However, I found the caption confusing. The legends do not have the T = 230 K line type (dashed) identified. And the caption only mentions this at the very end. I was trying to understand the first two panels and was reading that part of the caption and not finding out to what the dashed lines correspond. I would recommend putting the statement about the two temperatures at the beginning of the caption to avoid this problem, if they don't want to have the dashed lines in the figure legend itself.
4. Fiinally, section 5.1 is a bit inappropriate, I think. First, the analysis doesn't really do anything with oil particles. So this is sort of a side comment that this might need to be studies more. And the paper only cites one paper on oil, while there is signficant literature on aviation engine oil contributions to aviation PM emissions. And some of the other literature presents data significantly distinct from the one paper cited. The authors should either review the literature more broadly or minimize their discussion of the oil emission data. And the comment that "a hermetic and clean sealing of the engines from the oil system should be improved with regard to a complete jet oil recovery" indicates a lack of understanding of how the oil system on an aviation gas turbine engine functions. A comment along the lines of "reduction or elimination of oil emissions from the aircraft engines may be . . . a valuable mitigation effort for contrail formation" might be better. It should also be qualified ("may") since oil's role in contrails is still very poorly understood.
Citation: https://doi.org/10.5194/egusphere-2023-1321-RC1 - AC1: 'Comment on egusphere-2023-1321', Simon Unterstrasser, 19 Dec 2023
-
RC2: 'Comment on egusphere-2023-1321', Anonymous Referee #1, 21 Jul 2023
One more very minor comment from Reviewer 1:
line 454 "can easier form water droplets." might read better as "can form water droplets more easily."
The writing is so clear and clean, this somewhat awkward turn of phrase stands out. The meaning is still clear, so this is just a suggestion. Sorry I neglected to include this in my main review.
Citation: https://doi.org/10.5194/egusphere-2023-1321-RC2 - AC1: 'Comment on egusphere-2023-1321', Simon Unterstrasser, 19 Dec 2023
-
RC3: 'Comment on egusphere-2023-1321', Anonymous Referee #2, 20 Aug 2023
The manuscript explores how contrails might form and persist for zero-particle emissions engines with enhanced water vapor emissions, which are conditions relevant for hydrogen propulsion. Particle size distributions and hygroscopicities are assumed based on recent literature in order to parametrically understand the impacts on contrail ice crystal number throughout the near-field plume evolution. Interestingly, the results demonstrate that the increase in water vapor emissions and relatively low concentrations of available cloud condensation nuclei (CCN) allow contrails to form at warmer temperatures than for conventional kerosene burning engines, which could mean that a transition toward hydrogen propulsion results in an increase in climate-altering contrails. Simulations are carried out with both one and two lognormal size distribution modes, and the results are interesting for understanding the relative role of new particle formation versus Aitken and accumulation mode particles. These results hint that new particle formation from engine oil vapors may play a role in forming contrails; although, the choice of simulation parameters (e.g., forcing the concentrations of both modes to be the same) make it hard to understand how these particles may influence a contrail behind a hydrogen engine. Overall, the manuscript is well written and is an interesting and useful contribution to ACP and the contrail literature. I'd be inclined to recommend the paper for publication after the following comments are satisfactorily addressed:
1) The authors demonstrate awareness of recent work hinting that the venting of oil vapors may be a significant source of CCN for contrail formation in the "soot-poor" emissions regime, and the discussion in Section 5.1 is a good and important addition to the paper. However, I think not applying the model to directly explore the potential influence of these particles is a major gap in the present study. Could the 2-mode simulation be set up to create an additional figure to better understand how the contrail ice crystal number would change if a constant particle emissions source size mode from oil vapor with the following parameters were externally mixed with the ambient baseline particle number size distribution mode from Brock et al., 2021 and Section 4.3.1:
- Mode 1 (New Particle Formation From Continuously Vented Oil Vapor):
- r_d (nm) = 5, sigma = 1.2
- kappa = 0
- Emissions Index (number of particles emitted from 2 engines per second) = 10^11, 10^13, 10^15
- Mode 2 (Baseline ambient lognormal particle mode from Table 2):
- r_d (nm) = 15 nm, sigma = 1.6
- kappa = 0.5
- Concentration = 600 cm-3
The fundamental question is, do large numbers of small wetable, but insoluble particles dominate the contrail ice number in the presence of relatively few, large, sulfate particles from the ambient atmosophere?
2) Line 2: Change "has" to "have"
3) I find the use of the word "ensemble" to describe distinct size distribution modes to be confusing. I'd suggest replacing the word "ensemble" with "lognormal size mode" on Line 19 and throughout the manuscript.
4) Lines 33-34: Suggest "in particular on soot particles relative to co-emitted organic-sulfate particles"
5) Line 50: It would also be good to cite the field studies showing detection of oil signatures in engine particles for high-soot engines, albeit, with a focus on mass spectral detection methods that only see particles with diameters > 100 nm:
- Yu et al., 2010: https://doi.org/10.1021/es102145z
- Yu et al., 2012: https://doi.org/10.1021/es301692t
6) Line 74: Change "magnitudes" to "magnitude"
7) Line 78: large variability in atmospheric aerosol properties and co-emitted volatile particles from engine oil vapor
8) Line 130: It was not clear to me that the concentrations reported by Beer et al., 2020 in their supplement were for non-volatile particles. Concentrations of 200-300 cm^-3 seem high to me for dust and black carbon number. Please double check this.
9) Line 140: Suggest "higher" instead of "better"
10) Section 3.3.1, how would treating uncoated soot or dust as in soluble but able to adsorb water change things versus standard Köhler Theory, if at all? See, e.g., Kumar et al., 2009 (https://acp.copernicus.org/articles/9/2517/2009/).
11) Line 292: insert "is" to read "expression is the Kelvin term"
12) Line 325: Does this factor of 4 assume a 4-engine aircraft?
13) Line 364: is this because the soot particles are already well mixed with the plume water vapor?
14) Line 421: parameter values is hyphenated and misspelled
15) Lines 451-457: It's fascinating to think about this discussion in the context of low-kappa oil-nucleated particles that might be emitted by a hydrogen engine; although, it is not exactly the same set of parameters (as discussed above). I hope the authors will carry out the additional simulations to inform the interplay between new particle formation from co-emitted vapors and ambient aerosols
16) Lines 544: Strike "by"
17) Line 545: Replace "by" with "is"
18) Line 650: Strike "are"
Citation: https://doi.org/10.5194/egusphere-2023-1321-RC3 - AC1: 'Comment on egusphere-2023-1321', Simon Unterstrasser, 19 Dec 2023
- Mode 1 (New Particle Formation From Continuously Vented Oil Vapor):
- AC1: 'Comment on egusphere-2023-1321', Simon Unterstrasser, 19 Dec 2023
Peer review completion
Post-review adjustments
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 444 | 254 | 31 | 729 | 26 | 24 |
- HTML: 444
- PDF: 254
- XML: 31
- Total: 729
- BibTeX: 26
- EndNote: 24
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 278 | 37 |
| Germany | 2 | 148 | 20 |
| Canada | 3 | 55 | 7 |
| France | 4 | 39 | 5 |
| China | 5 | 34 | 4 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 278
Andreas Bier
Simon Unterstrasser
Josef Zink
Dennis Hillenbrand
Tina Jurkat-Witschas
Annemarie Lottermoser
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3672 KB) - Metadata XML