the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Hydrogen combustion mitigates contrail-induced climate forcing relative to kerosene for 2035-technology A320 aircraft
Abstract. Global-scale estimates of hydrogen combustion's contrail impact have relied on general circulation models with simplified contrail parameterizations or on gridded lifecycle simulations with generic aircraft, and no study has evaluated hydrogen against kerosene using real-world flight trajectories for a defined aircraft under directly comparable conditions. This study presents a trajectory-based comparison of contrail climate forcing from kerosene and hydrogen combustion for 2035-technology A320 aircraft, applying the Contrail Cirrus Prediction model (CoCiP) to 10,298 flights sampled from 2024 global air traffic, with 2024 ERA5 reanalysis meteorology and fuel-specific baseline aircraft designs simulated along identical routes. Hydrogen's higher water vapor emission expands the contrail formation envelope, more than doubling the persistent contrail fraction from 7.31% to 15.7% of fleet distance. Yet under baseline nucleation assumptions, fleet-level energy forcing per kilometer decreases by 86.8%, because eliminating soot lowers ice crystal numbers by 1 order of magnitude, yielding larger, faster-sedimenting crystals that form shorter-lived, narrower and optically thinner contrails. This reduction depends mainly on the assumed nucleation floor for soot-free hydrogen and on kerosene non-volatile particulate matter (nvPM) emissions: across the plausible hydrogen floor (10¹²–10¹⁴ ice crystals per kilogram of fuel burned) the reduction spans 99.9% to 86.8%; even the most conservative hydrogen floor remains 28.7% below the lowest-nvPM kerosene scenario (10% nvPM). A complementary surrogate model reproduces the same monotonic decline in forcing with decreasing kerosene nvPM, corroborating the trend. Hydrogen combustion thus offers considerable contrail mitigation potential exceeding that of soot reduction alone.
- Preprint
(4966 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 06 Dec 2026)
- CC1: 'Comment on egusphere-2026-4555', Michael Ponater, 10 Nov 2026 reply
-
RC1: 'Comment on egusphere-2026-4555', Anonymous Referee #1, 10 Nov 2026
reply
Review ofHydrogen combustion mitigates contrail-induced climate forcing relative to kerosene for 2035-technology A320 aircraftby Conti, Wiegand, Kiendl, and Balderas-XicohtencatlGeneral comments:The authors assess the climate benefit that an introduction of LH2 as aircraft fuel would give to climate in comparison to the use of kerosene. For that purpose, a fleet of A320 aircraft with assumed 2035 technology is considered, including more than 10000 realistic flights, once driven with kerosene of different grades of EI(nvPM), representing more of less technical progress, and once driven with LH2, assuming different levels of AEIs in this no-soot regime (to cover currect uncertainties). CoCiP, fed with ERA5 reanalysis data is applied to compute all the contrails that arise from these to fleets. It turns out, that on the one hand LH2 combustion produces much more (persistent) contrails than kerosene combustion, but on the other hand LH2 contrails have in all considered scenarios much less ice crystals than kerosene contrails. LH2 contrail crystals are thus larger and sediment faster; the contrails are optically thinner, they have shorter lifetime and are narrower than kerosene contrails. Thus the overall radiative impact (measured as energy forcing and derivatives of it) of contrails of a LH2 fleet is considerably smaller than the corresponding impact of kerosene contrails.The paper is well written and mostly easily comprehensible. The methods used are plausible and the results convincing. I have no major critique. Nevertheless, below I list a number of minor questions etc. that should be considered for further improvement of the paper, before the paper will be finally accepted.Minor issues:Language:"Nucleation floor": after a while one gets a feeling of what you mean, but initially this unusual expression is puzzling. Please define or explain it on first appearance or use different wording.Line 39 "...volatile aerosols can become the dominant nucleation pathway...". Aerosols are not a pathway. This is bad language and needs reformulation.Line 240: Please avoid expressions like "contrail-forming waypoints". This means a waypoint that forms a contrail, which is obviously nonsense.Line 82: It is not quite clear whether "this study" means the present one or the one by Petterson et al., to which you refer in the text just before.LL 131ff: What you describe here, is simple physics. But you write it as if this were a special effect of Unterstrasser's parameterization. Please reformulate.LL 144, 145 "Crystal size and number influence contrail optical depth both directly and indirectly, since crystal size also determines the sedimentation rate that governs contrail lifetime." I suggest to split this sentence into two and to delete "directly and indirectly". As it stands, it seems to state that liftime affects optical depth.L 146: "absorbing OLR" is a bit wrong, since OLR is the outgoing LW radiation at the top of the atmosphere. What you mean here, is rather upwelling LW radiation.Eq. 2: I am surprised that the width of a contrail appears in the formula for optical thickness. Please explain.L 150: Please define "number line concentration".LL 153-155: I think, two things are brought together that shouldn't. First, for large optical thickness the short-wave effcts begins to dominate, generally. Second, there are cooling contrails during the day. I think this has more to do with the special radiation geometry (low sun) than with large optical thickness. Large optical thickness needed to let the SW effect dominate is very rare in contrails.L 186: Is "duration-weighted mean" the duration-averaged mean?L 188: HOPE-BL, what is it?L 192: "mass penalty applied to the dry wing" is unclear for a reader not familiar with special aviation design slang.L 197: spell out ICAO EDB.L 205: "half the value derived by Schumann". This is unclear. A vertical resolution (which I think you speak of) is a matter of choice, not a matter of derivation. I don't understand what you want to say here.Eq. 5: The sense of this correction factor is not clear to me. The text before does not really help. My understanding is that the emission index is number of ice crystals (in this case) per unit mass of fuel burnt. But here it seems to become number of ice crystals per unit energy consumed (ignoring the overall propulsion efficiency). I think, it is necessary to give a derivation of this correction, and why it is necessary.Sect. 4.2: Good description of results.Sect. 4.3: This raises a critical question: Isn't your comparison a bit biased if the kerosene-fleet you consider is quite often at the lower edge of the range where contrail formation is possible? How would the results presumably change if you would consider a fleet with higher ceiling altitudes close to the tropopause (e.g. FL 350-400)?Sect. 5.3: Please explain "LW and SW enhancement factors", "sedimentation impact factor", "wind shear enhancement exponent", "emission index enhancement factor". I am not familiar with the method that is employed here, but I suppose that these enhancements are notions that refer to this method and not factors from CoCiP. This and the meaning of these factors must be clarified.L 479: I suppose, k=6 is the number of uncertain quantities that are varied. But what is N=256?L 481: I suppose, coefficient of variation is the ratio of standard deviation and mean value, correct? Could you please write this down, as this notion may be less known.Last par of 5.3: This paragraph needs more explanation. Unfortunaltely the reviewer and probably many readers are not familiar with the statistical method that you used for variance decomposition. It seems to me that this par. contains much technical slang, so that I have only a vague imagination of what the results mean. For instance, the meaning of "This behavior reflects that the RF enhancement factors are applied as multiplicative output-level scalars whose uncertainty ranges are independent of contrail microphysics" is not clear to me, neither mathematically nor physically. More explanation and more detail would be highly welcomed.Sect. 5.4: The question that I raised in reference to sect. 4.3 could be discussed here as an additional critical point.LL 520 ff: "Hydrogen combustion more than doubles the persistent contrail fraction (from 7.31% to 15.7% of fleet distance) due to the steeper mixing line caused by its higher water vapor emission index..." Please delete the words "due to the steeper mixing line". Currently, the statement mixes real physics (higher vapour emission) with the REPRESENTATION of the physics in a phase diagram. The latter has no causal effects and thus the sentence is bad in its current form.L 534: What does the unit pp mean?Citation: https://doi.org/
10.5194/egusphere-2026-4555-RC1 -
RC2: 'Comment on egusphere-2026-4555', Anonymous Referee #2, 10 Nov 2026
reply
The study compares the climate forcing and related properties of contrail cirrus produced by a potential hydrogen-powered aircraft and a kerosene-powered aircraft with 2035 technology. The authors focus on a specific aircraft type, the A320, and use flight paths of this aircraft from 2024. Flight performance for both kerosene and hydrogen aircraft is modelled with the Bauhaus Luftfahrt Aircraft Design Environment (BLADE), and CoCiP is then used to simulate the properties and climate forcing of the resulting contrail cirrus.
For the kerosene aircraft, non-volatile particulate matter (nvPM) is assumed to be the main source of ice-nucleating particles. Possible future reductions in nvPM emissions are approximated as fixed percentages of the emission profiles in the International Civil Aviation Organization (ICAO) Engine Emissions Databank (EDB). The possible contribution of volatile particulate matter (vPM) at low nvPM emission levels is addressed in a sensitivity analysis in which the background apparent ice particle emission index is varied. For the hydrogen aircraft, uncertainty in ice particle emission index is represented by varying the background apparent ice particle emission index, scaled by the ratio of fuel energy content.
The manuscript is well written, well structured, and provides solid background. The choice of methods and parameter values is well motivated by related studies, and the results are presented clearly. The study makes a valuable contribution to the growing literature on the climate effects of hydrogen contrail cirrus, placing its results in the context of previous work. My comments are minor and mainly concern clarifications and some additional explanation in places.
Comments
- L11: The term “nucleation floor” is not standard and should either be explained or reformulated.
- The comparison between the surrogate model and the authors’ results is fine and can remain in the manuscript. However, it seems to me to be more a comparison between CoCiP and the surrogate model than something that addresses the main questions of the study. I would therefore suggest removing it from the abstract (L13–14), particularly since, for the informed reader, it raises questions about the role of vPM at low nvPM emissions, which is not discussed in the abstract.
- L20: I suggest changing “Alongside these strategies, …” to “Alongside these strategies for contrail cirrus mitigation, …”.
- L188: Please briefly explain what HOPE-BL is.
- Fig. 2b: Is there a simple explanation for the double peak in the volume mean ice crystal radius for the hydrogen aircraft? A brief comment in the text would be helpful.
- L408: A sentence or two explaining why the EF per contrail km varies over the year would be helpful here.
- Sect. 5.3: The sensitivity analysis of the CoCiP model parameters is much appreciated. To make this section even more informative, the authors could include actual numbers, or a plot, showing how much the EF/km changes when the enhancement factors are varied. Alternatively, since these are multiplicative factors, it would be useful to at least state their range and briefly explain how it was chosen (e.g., whether it is based on a fit to observations or other data).
- L481: Please clarify what is meant by “the coefficient of variation” here.
Citation: https://doi.org/10.5194/egusphere-2026-4555-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 353 | 126 | 25 | 504 | 26 | 28 |
- HTML: 353
- PDF: 126
- XML: 25
- Total: 504
- BibTeX: 26
- EndNote: 28
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Thanks for your timely work on the potential contrail climate impact reduction gained by use of alternative, hydrogen based, fuel. I fully acknowledge your kind mentioning of the pioneering papers contributing to the subject by Marquart et al. (2005) and Ponater et al. (2006). Yet, I think your way of citing them is not fully to the point, writing that in those papers the contrail contribution was lumped with other effects rather than isolated. Actually, Tables 7 and 8 in Ponater et al. (2006) quantify dedicated contributions of contrails to the total effect. According to those numbers the respective gain of liquid hydrogen use, relative to kerosene use, range about 25% for radiative forcing reduction, and somewhat less than 10% for surface temperature increase reduction. The numbers, of course, depend on the details of the assumed technology transition, but they give a qualitative impression of the order of magnitude for the potential contrail impact reduction on the respective time scales. They could, thus, be used for quantitative comparison with your new contrail cirrus results.
This said, I do not hesitate to concede the superiority of the more recent results on the subject, as the classical work was limited to the contribution of line-shaped contrails (rather than contrail cirrus) and was based on contrail parametrisation schemes considerably more simplistic than those developed over the past 20 years.