the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 25 Jun 2025
Modeling PFAS in the global atmosphere – The PRIEST extension for the ICON-ART modeling framework
Abstract. This study presents the ICON-ART PRIEST model extension, developed to simulate the transport and transformation of Per- and Polyfluorinated Substances (PFAS) in the atmosphere. While the ICON-ART framework was developed to simulate atmospheric physics and chemical composition, the newly developed PRIEST extension incorporates additional gas‐phase and aqueous physics, along with chemical reaction mechanisms, to model the transport, transformation, and deposition of PFCA precursors. Therefore, the model includes 22 aqueous-phase reactions that depend on liquid cloud water and temperature. The aqueous-phase processes represent the adsorption of precursors in water droplets, with variable absorption rates. The model follows a hierarchical initialization, starting with the emissions, followed by aerosols, chemistry, and finally removal. A simple parameterization of the OH radical is implemented to improve the simulation of PFCA precursors. The global model results (approx 105 km² grid resolution and 6 hours temporal resolution) show the capability of the model system to simulate regional and global variations of PFCA concentrations and their deposition processes. The results reveal an overestimation of observed atmospheric concentrations in Europe and an underestimation in East Asia. These differences are mainly related to the coarse spatial model resolution and the uncertainties arising from the underlying emissions model. In conclusion, ICON-ART PRIEST represents a significant step forward in simulating the atmospheric fate of PFCAs precursors and their transformation products by integrating an enhanced chemical mechanism into the ICON-ART framework that couples both gas-phase and aqueous-phase processes, and also with the incorporation of a detailed temporally resolved PFAS emission inventory.
- Preprint
(6951 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
CEC1: 'Comment on egusphere-2025-2289', Juan Antonio Añel, 24 Jul 2025
Dear authors,
Unfortunately, after checking your manuscript, it has come to our attention that it does not comply with our "Code and Data Policy".
https://www.geoscientific-model-development.net/policies/code_and_data_policy.htmlFirst, you have archived your code on a Git site. However, Git sites are not a suitable repositories for scientific publication; our policy is clear about it. Despite it, we have tried to access the code, and the link that you provide is empty, when trying to access the code through it, the message obtained is "page not found".
Also, we can not accept that the model output data is available upon request. You must share all the input files and output files used and produced in your work in a repository acceptable according to our policy.
Therefore, the current situation with your manuscript is irregular, as we can not accept in Discussions or send out for review manuscripts that do not comply with our policy. Please, publish your code and data in one of the appropriate repositories listed in our policy and reply to this comment with the relevant information (link and a permanent identifier for it (e.g. DOI)) as soon as possible. Also, you must include a modified 'Code and Data Availability' section in a potentially reviewed manuscript, containing the information of the new repositories.
I must note that if you do not fix this problem, we will not be able of proceeding with the review of your manuscript and publication in our journal.
Juan A. Añel
Geosci. Model Dev. Executive Editor
Citation: https://doi.org/10.5194/egusphere-2025-2289-CEC1 -
RC1: 'Comment on egusphere-2025-2289', Anonymous Referee #1, 19 Aug 2025
General Comments
Meza-Landero et al. reported a newly developed atmospheric PFAS modeling framework, ICON-ART (ICOsahedral Non-hydrostatic model framework with Aerosol and Reactive Tracers) PRIEST (PFCA Reactions In Earth System Transport). They described the implementation of multiphase chemistry, transport, and deposition of PFCA (Perfluoroalkyl Carboxylic Acids) in the model, and evaluated surface concentrations and wet deposition against in-situ observations. The validation demonstrated that the model could reasonably capture the wet deposition of PFOA (Perfluorooctanoic Acid) and PFNA (Perfluorononanoic Acid) compared with available observations. However, the model’s performance on surface concentrations of PFOA, PFNA, and FTOH (Fluorotelomer Alcohol) varied across regions. The research topic is novel and significant, since PFAS has been drawing attention as a new class of pollutants that raise public health concerns. The manuscript is suitable for publication in GMD. I have two major concerns that should be addressed before I can recommend it for publication.
- The setup of the model is quite concerning. The model reads the monthly output of NOx, OH, HO₂, and H₂O from CAM-chem, while organic peroxy radicals (RO₂) were calculated from the most abundant VOCs, including methane, ethane, and propane. First, CAM-chem was driven by its dynamic core, and it might be nudged by other reanalyses such as MERRA-2 (Modern-Era Retrospective analysis for Research and Applications, Version 2). However, the meteorological fields in the authors’ model were simulated by its own dynamic module, ICON. There is a mismatch between the meteorological fields driving the two models, which makes it unreasonable to directly match the chemical fields from CAM-chem to the ICON meteorology fields. Second, NOx, HO₂, OH, RO₂, and VOCs are chemically linked. The way the authors calculated RO₂ from methane, ethane, and propane is physically unreasonable, as it breaks the mass conservation between these species. Overall, the authors’ current approach of combining meteorology and chemical fields violates fundamental physical laws, including conservation of momentum and mass. The authors should adopt either a fully online or a fully offline approach, rather than the hybrid method they use now.
- Further analysis is required to better understand the model performance. Figures D1 and D2 demonstrate that the model cannot capture the variability of surface concentrations of PFNA and PFOA, and the simulation results appear overly consistent. I think this is reasonable, as it could represent an initial step for this type of modeling. The coarse resolution could be a potential explanation for the discrepancy, however, I believe that more effort is needed to identify the sources of uncertainty. For example, the meteorological fields should be examined before validating PFCA. In addition, emissions could also be a potential source of uncertainty, and sensitivity tests would be necessary to assess their contribution. All these analyses would be helpful for both the authors and the readers in understanding the underlying issues with the model system.
Specific Comments
Abstract: “ICON-ART PRIEST” should be written out in full the first time it appears.
Lines 60–73: This paragraph is confusing. It is unclear whether it refers to updates or to the previous version of the ICON-ART PRIEST model. The authors seem to describe ICON-ART PRIEST as a new model, but placing this paragraph here is confusing. The updates should instead be mentioned after line 76.
Section 2.2.1: As I mentioned in my major comments, this monthly OH estimation may be inappropriate, given the connections between OH and other species.
Line 240: The data link is not available.
Figure 6: Error bars are necessary to better understand the zonal variability.
Citation: https://doi.org/10.5194/egusphere-2025-2289-RC1 -
RC2: 'Comment on egusphere-2025-2289', Anonymous Referee #2, 10 Sep 2025
I find this a very challenging manuscript to review. Understanding global distributions of PFAS and associated atmospheric chemistry is an important problem and the authors present a significant advancement in the coupling of the simplified PFAS chemistry to the ICON model. On the other hand, it is apparent from Figures D1 and D2 that the model is not capable of reproducing the observed spatial variability of PFAO and PFNA. Given the simplified mechanism, uncertainties in global emission inventories, potential mismatches between chemical fields introduced by the use of the CAM-chem model for initialization of atmospheric oxidants, as well as limited observations for model validation, it is unclear what causes these mismatches and what steps could be undertaken to improve simulations. Given that this is a model development paper, I am not sure how to properly weigh these against each other or to suggest a more in-depth study to investigate the sources of uncertainty in model results.
However, specifically there are some suggestions outlined in specific comments
- In line with Reviewer 1's comment, the authors should explore ways to minimize the identified mismatch
- Whenever possible Figures should communicate uncertainties/ variation in the data beyond bar-plots (especially if n=2, which would merit just plotting point data). For higher n, boxplots or distribution plots would be appropriate- Figure 2: the different scales with change in color in Figure 2 are confusing and should be highlighted in caption (or otherwise changed)
- Code availability: It is inadvisable to solely rely on an institutional Git repository. The release should be tagged and submitted to an archive, such as zenodo to ensure preservation.
Citation: https://doi.org/10.5194/egusphere-2025-2289-RC2 -
EC1: 'Comment on egusphere-2025-2289', Lars Hoffmann, 16 Sep 2025
Dear Authors,
We have received feedback from a third reviewer of your paper, which was not available before the interactive discussion closed, unfortunately.
Please take these comments into account when revising your manuscript.
Thank you and best regards
Lars HoffmannAdditional Reviewer Comments- Overall, I think this work is an important step for atmospheric PFAS modeling.
- Explicit inclusion of aqueous/aerosol chemistry is an important advancement for PFAS specifically. Multiple aerosol modes with different behaviors is likely a reasonable representation, though there is little literature constraint on the PFAS side of this process.
- While I agree with Referee 1 that the treatment of non-PFAS reactants as semi-offline provided from a different model will introduce errors based on the mismatch of dynamics and chemistry, I believe these errors will be significantly smaller than the PFAS-specific uncertainties of the model. This echoes the point of both Referee 1 and Referee 2 that uncertainties and sensitivities are an important aspect of this stage of modeling and should be explored.
- A striking difference between the model and observations (e.g. in Table 1) is in the degree of variability. The model is significantly less variable at a given site. Exploring how much of this can be explained by the use of monthly non-PFAS reactant concentrations, possible unresolved time/space-variability of emissions, or other reasons would be informative.
- I wonder if the PFCA model concentrations compared to observations are the total gas+particle phase? My inference is that they must be, but I think it's unclear in the manuscript.
Citation: https://doi.org/10.5194/egusphere-2025-2289-EC1
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,051 | 91 | 13 | 1,155 | 16 | 28 |
- HTML: 1,051
- PDF: 91
- XML: 13
- Total: 1,155
- BibTeX: 16
- EndNote: 28
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1