the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 May 2026
Predicting Aviation Contrail Occurrence Using Bayesian Population Statistics From Reanalysis Data
Abstract. Despite the ongoing climate crisis and recent pandemic-induced disruption, the aviation sector is expected to experience 5% annual growth over the next decade. While the industry moves towards decarbonisation through use of sustainable fuels and improved operating practices, the contribution by non-CO2 effects become ever more apparent. Contrails and contrail-induced cirrus clouds contribute an estimated 57% to the sector's total effective radiative forcing (ERF). Contrail avoidance methods are gaining ground as tools to strategically reroute flights to reduce their ERF by predicting contrail forming regions in advance.
The task of prediction remains a challenge however, with typical methodologies employing either highly parametrised models that suffer from uncertainties, or machine learning methods that are heavily abstracted away from the background physics. We propose a novel, robust method for contrail prediction that leverages large-scale population behaviours. Using ERA-5 reanalysis and the OpenContrails dataset for over 50,000 confirmed contrails between 2019 and 2020 over North America, we train an informed contrail predictor using Bayesian methods which we verify on unseen data. Results and statistical evaluation of this model are presented, providing a scalable but interpretable contrail predictor with good skill (F1 = 0.801) that could be run using output from numerical weather prediction models, or time-slice outputs from high-resolution climate models.
- Preprint
(4979 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-2490', Anonymous Referee #1, 09 Jun 2026
-
RC2: 'Comment on egusphere-2026-2490', Anonymous Referee #2, 22 Jul 2026
GENERAL REMARKS
The manuscript reports ion the development of an approach for predicting contrails, based on population-level statistics and Bayesian methods. To train the model for meteorological conditions associated with contrail formation, the authors used contrail observations from the OpenContrails dataset, along with ERA-5 reanalysis data. Then, the authors applied Bayesian methods to create a predictive tool for contrail occurrence at specific humidity and pressure levels, based on conditions associated with contrails confirmed from remote sensing observations. The performance of the prediction tool was evaluated by means of binary classification metrics.
To evaluate their tool, the authors matched maps of predicted contrail occurrence against satellite observations. This matching exercise demonstrated that the developed method could reproduce regions of increased likelihood of contrail formation.
The authors propose that this approach may be useful in the development of future contrail avoidance strategies by offering a predictable and expandable prediction framework.
The manuscript is clearly structured and well written, and the statistical methods are carefully introduced and described. However, there are four main limitations which need to be clarified before the manuscript might be ready for publication. After consideration of these major issues and the minor issues and typos listed below, the study is most likely suitable for publication in the best-suited journal. The recommendation is to withdraw the article, revise accordingly and submit to the best-suited journal. If the authors are still convinced that ACP is the best-suited journal, withdrawing is of course not necessary.
SPECIFIC COMMENTS
1| The manuscript outlines the development and evaluation of a promising tool for predicting contrail occurrence from the output of numerical weather prediction models, or of high-resolution climate models. The method is promising and the study surely deserves publication. My question is whether Atmospheric Chemistry and Physics is the right Copernicus journal. Since the manuscript is mainly focusing on the description of the method and its evaluation, there are no conclusions drawn of relevance for the state of the atmosphere, or atmospheric sciences in general. This, however, is necessary to meet the ACP objective of having important and clearly argued implications for our understanding of the state and behaviour of the atmosphere and climate or present substantial new insights into the atmosphere's role in other parts of the Earth system.
This statement is not intended as criticism, since the purpose of the manuscript is different, but it could be used as a stimulus for reconsidering the target journal. Looking at the portfolio of Open Access Copernicus journals, either Geoscientific Model Development, or – even more suitable – the recently launched Journal of Environmentally Compatible Air Transport System(https://www.jecats.net/home.html) may be more appropriate. It particular, JECATS is probably reaching out for a much more targeted audience than the broad audience of Atmospheric Chemistry and Physics.
2| More methodologically oriented, the manuscript uses the term “relative humidity” in a not very well-defined manner. From the study topic, it should always be the relative humidity with respect to ice. For the formation of contrails, independent of their lifetime, the exceedance of the relative humidity with respect to liquid water is necessary within the cooling and expanding plume. This condition is handled by the Schmidt-Appleman criterium (SAC; Schumann, 1996). For the establishment of long-lived and climate-impactful contrail-cirrus the embedding air masses need to be above 90% relative humidity with respect to ice (Li et al., 2023). Clarification is needed of the RH to which the authors are referring in a certain context.
Additionally, the authors use ERA5 humidity fields but there is a well-known misrepresentation of ice-supersaturation in ERA5. This drawback is mentioned the manuscript several times but never extensively discussed in one paragraph. As this is an important feature of ERA5 humidity fields, it should be discussed prominently in the Methods section, how this limitation of ERA5 is handled in the context of the present method development.
3| The OpenContrails Data set and similar approaches focus on the identification of linear-shaped contrails. These young contrails are an easier target since the linear shape is kind of unique for ice clouds. However, contrails become only climate-effective if they live long enough and spread out to contrail-cirrus in an atmosphere impacted by strong shear winds. A discussion is recommended which specifies how the taken approach considers this difference between detected linear-shaped contrails and climate-effective contrail-cirrus which have lost their clear linear shape.
4| The Conclusions section is not in line with the current requirements of ACP. The guidelines for authors (https://www.atmospheric-chemistry-and-physics.net/policies/guidelines_for_authors.html) state that the concluding section needs to contain a summary of the main results, synthesis and interpretation, comparison and context, caveats and limitations and finally a section on implications where the authors should discuss what the results mean for our understanding of the state and/or behaviour of the atmosphere and climate, which is the main requirement for publication in ACP.
Please rework the conclusions section accordingly or adjust to another journal’s requirements, if appropriate.
MINOR ISSUES
1| Abstract, line 4: The estimated contribution of 57% to the total aviation ERF depends on present-day knowledge, since the proportion depends entirely on recent growth rates of aviation. It is recommended to add “(present-day knowledge) after 57%”. Generally speaking, Lee et al. (2021) reports that contrail-cirrus contributes 57.4 mW m2 to the total aviation ERF of 100.9 mW m2. It is probably better to list these numbers instead of calculating 57%. If readers look for the fractional number in Lee et al. (2021), they will not find it.
2| Line 32: The sentence starting with “the contribution from which …” is hard to understand, rephrasing is suggested.
3| Line 43: Contrail formation requires exceedance of saturation with respect to liquid water. This should be specified.
4| Line 48: The authors state that during daytime the solar and thermal radiative effects are opposite but of comparable magnitude. This statement is too general since it depends on time of day, solar zenith angle etc. Some clarification is suggested.
5| Line 60: The authors state that “the conditions required for persistent contrails are commonly found in extratropical regions”. It might be worthwhile to specify that Ice-supersaturated regions occur mostly below but close to the tropopause, and in extratropical regions, this altitude range corresponds to the cruising altitude of commercial aviation (Petzold et al., 2020).
6| Line 71: The reference to the IAGOS dataset on relative humidity observations by commercial aircraft should be added as (Petzold et al., 2020).
7| Line 89: The adjusted humidity distribution used by Arriolabengoa et al. (2025) was adjusted to IAGOS-based humidity observations. This information might be added here because it demonstrates that Arriolabengoa and co-workes used in-situ observations to construct their adjusted humidity distributions.
8| In the introductory section to the Methods chapter, a discussion of the misrepresentation of ice-supersaturation and how this tackled in the context of the presented study needs to be discussed.
9| Figure 3: The nomenclature of Relative Humidity should be harmonised and specified; it should be RHi (relative humidty with respect to ice). In Panel 3, there is no region shown, I assume the authors show solid and dashed line as the bounding lines at the start and end of the identified contrail. If so, it should be clarified.
10| Line 213: The explanation of the “area under the curve (AUC)” is difficult to follow. Here, an additional figure might help readers without in-depth knowledge of Bayesian methods. If this paragraph is somehow connected to Figure 8, it should me mentioned.
11| Figure 5 shows RH probability distribution functions across the entire atmospheric column from ground to 150 hPa. For clarity, it might be appropriate to only show those PDFs relevant for the contrail existence levels.
TYPOS
Line 19: “;” should be replaced by “,”.
Line 30: SAF stands for Sustainable Aviation Fuel” not aircraft fuels.
Line 324: Shouldn’t it read “…, however this should be less important …” ?
REFERENCES
Arriolabengoa, S., Crispel, P., Jaron, O., Bouteloup, Y., Vié, B., Li, Y., Petzold, A., and Plu, M.: Modeling and verifying ice supersaturated regions in the ARPEGE model for persistent contrail forecast, Atmospheric Chemistry and Physics, 25, 18051-18076, https://doi.org/10.5194/acp-25-18051-2025, 2025.
Lee, D. S., Fahey, D. W., Skowron, A., Allen, M. R., Burkhardt, U., Chen, Q., Doherty, S. J., Freeman, S., Forster, P. M., Fuglestvedt, J., Gettelman, A., De Leon, R. R., Lim, L. L., Lund, M. T., Millar, R. J., Owen, B., Penner, J. E., Pitari, G., Prather, M. J., Sausen, R., and Wilcox, L. J.: The contribution of global aviation to anthropogenic climate forcing for 2000 to 2018, Atmospheric Environment, 244, 29, 10.1016/j.atmosenv.2020.117834, 2021.
Li, Y., Mahnke, C., Rohs, S., Bundke, U., Spelten, N., Dekoutsidis, G., Groß, S., Voigt, C., Schumann, U., Petzold, A., and Krämer, M.: Upper-tropospheric slightly ice-subsaturated regions: frequency of occurrence and statistical evidence for the appearance of contrail cirrus, Atmospheric Chemistry and Physics, 23, 2251-2271, https://doi.org/10.5194/acp-23-2251-2023, 2023.
Petzold, A., Neis, P., Rütimann, M., Rohs, S., Berkes, F., Smit, H. G. J., Krämer, M., Spelten, N., Spichtinger, P., Nédélec, P., and Wahner, A.: Ice-supersaturated air masses in the northern mid-latitudes from regular in situ observations by passenger aircraft: vertical distribution, seasonality and tropospheric fingerprint, Atmospheric Chemistry and Physics, 20, 8157-8179, https://doi.org/10.5194/acp-20-8157-2020, 2020.
Schumann, U.: On conditions for contrail formation from aircraft exhausts, Meteorologische Zeitschrift, N.F.5, 4-23, https://doi.org/10.1127/metz/5/1996/4, 1996.
Citation: https://doi.org/10.5194/egusphere-2026-2490-RC2 - RC3: 'Comment on egusphere-2026-2490', Anonymous Referee #3, 28 Jul 2026
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 249 | 94 | 29 | 372 | 23 | 26 |
- HTML: 249
- PDF: 94
- XML: 29
- Total: 372
- BibTeX: 23
- EndNote: 26
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
You can find my comments in the PDF document.