the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 11 Sep 2024
Chempath 1.0: An open-source pathway analysis program for photochemical models
Abstract. We describe the development of Chempath: an open-source pathway analysis program for photochemical models. This algorithm can help understand the results of complex photochemical models by identifying the most important reaction chains (pathways) for the production and destruction of a species of interest in a reaction system. The algorithm can also quantify the contribution of the pathways to the production and destruction of a species.
We demonstrate how to apply Chempath to a one-dimensional photochemical model, using an example of a reaction system for Earth's present-day atmosphere. We validate that Chempath can identify well-known chemical mechanisms for O3 production and destruction in this model, suggesting that this algorithm can be applied to understand photochemical models of less well-known atmospheres, like past and exoplanet atmospheres.
- Preprint
(778 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on gmd-2024-163', Anonymous Referee #1, 25 Sep 2024
-
AC1: 'Reply on RC1', Daniel Garduno, 30 Oct 2024
Thank you so much for a thorough and detailed review. Please see the attached document.
-
EC1: 'Reply on AC1', Rolf Sander, 30 Oct 2024
Thanks for the very interesting discussion! There seems to be a major
discrepancy about the lifetime of CH3OO. The reviewer mentions a
chemical lifetime of < 1 min. This is consistent with the number given
by Wolfe et al. (doi:10.5194/acp-14-4715-2014) in their Fig. 9a.The authors, however, found a lifetime between 1 and 170 days in the
Photochem model. Apparently, Photochem is a very generic model which can
be used for any planetary atmosphere. So maybe some tropospheric
reactions which are specifically necessary for the current Earth's
atmosphere are missing? For example, I could not find CH3OO + HO2
anywhere in the reactions.txt files of the Photochem github repository.To investigate this issue further, it would be very helpful if the
authors can provide the chemical mechanism (reactions and rate
constants) that they used in Photochem.Citation: https://doi.org/10.5194/gmd-2024-163-EC1 -
AC2: 'Reply on EC1', Daniel Garduno, 30 Oct 2024
Thanks for your comment and the reference. The chemical mechanism we used can be consulted in the file `zahnle_earth.yaml` hosted here: https://github.com/Nicholaswogan/photochem_clima_data/tree/main/photochem_clima_data/data/reaction_mechanisms
In this mechanism, there are seven reactions involving CH3O2. As you mentioned, there is no CH3OO + HO2 reaction:
- equation: CH3 + O2 (+ M) <=> CH3O2 (+ M)type: fallofflow-P-rate-constant: {A: 1.494515e-25, b: -3.3, Ea: 0.0}high-P-rate-constant: {A: 2.6e-11, b: 0.0, Ea: 0.0}- equation: CH3O2 + H <=> CH3O + OHrate-constant: {A: 1.6e-10, b: 0.0, Ea: 0.0}- equation: CH3O2 + H <=> CH4 + O2rate-constant: {A: 4.0e-14, b: 1.0, Ea: 8350.0}- equation: CH3O2 + O <=> CH3O + O2rate-constant: {A: 6.0e-11, b: 0.0, Ea: 0.0}- equation: CH3O2 + OH (+ M) <=> CH3OH + O2 (+ M)type: fallofflow-P-rate-constant: {A: 0.0, b: 0.0, Ea: 0.0}high-P-rate-constant: {A: 0.0, b: 0.0, Ea: 0.0}- equation: CH3O2 + CH3 <=> CH3O + CH3Orate-constant: {A: 4.0e-11, b: 0.0, Ea: 0.0}- equation: CH3O2 + NO <=> CH3O + NO2rate-constant: {A: 2.8e-12, b: 0.0, Ea: -285.0}Citation: https://doi.org/10.5194/gmd-2024-163-AC2 -
EC2: 'Reply on AC2', Rolf Sander, 02 Nov 2024
Thanks for providing the complete list of CH3OO reactions. I noticed
that not only CH3OO + HO2 but also the self-reaction CH3OO + CH3OO is
missing. I expect that after adding both, the chemical lifetime of CH3OO
will be much closer to one minute. Still, I cannot be sure if these are
the only reactions missing in the chemical mechanism of Photochem.How to proceed?
On the one hand, the Photochem model is not the topic of the current
manuscript. It is certainly not the task of the current authors to
improve the applicability of Photochem to the current terrestrial
troposphere.On the other hand, the evaluation of Chempath is not convincing when it
results in questionable values for CH3OO (and maybe other species as
well).Would it be possible to use the results of a different model run?
Ideally, Chempath should be applied to model output which has been
evaluated in a peer-reviewed publication. For example, Wogan et al.
(2023) used Photochem for the Early Earth, and it could be interesting
to identify significant pathways in those results. I think any
quality-checked model output from Photochem or any other chemistry model
would be fine for testing Chempath. It could be for the Early Early,
modern Earth, or even exoplanets.Citation: https://doi.org/10.5194/gmd-2024-163-EC2 -
AC3: 'Reply on EC2', Daniel Garduno, 05 Nov 2024
The purpose of the example provided in section 4 of the manuscript is to show how to use Chempath in a 1D photochemical model, not to apply Chempath to a model in which all species match the observations. We disagree that the evaluation of Chempath is not convincing when it results in questionable values for CH3OO. The CH3OO high number density results from an incomplete representation of its chemistry in the reaction system we used, and is not a result artificially produced by Chempath. Chempath is a tool for evaluating photochemical models. The fact that our code was able to show the lack of important reactions for CH3O2 chemistry in the photochem model ModernEarth reaction system demonstrates that Chempath fulfills its purpose: to evaluate and understand the results and photochemical models. We will update the manuscript to discuss the inconsistency of CH3O2 in the photochem model and warn the reader to take our results with caution since the model we used is not intended to be a perfect representation of the modern Earth's atmosphere.
If there is a concern that our code is not working well or producing artificial results, this can be easily discarded using our input files and the reviewer’s implementation of the PAP algorithm. For example, the reviewer could use their algorithm implementation to corroborate that they get results similar to ours in the simple example we used to explain how the algorithm works. The Chempath code repository also includes an example of how to apply Chempath to a simple photochemical box model (https://github.com/DanyIvan/chempath/blob/main/examples/box_model_pathways/box_model_pathways_example.ipynb). These results could also be corroborated with the reviewer's implementation of the algorithm. Finally, our results for our the photochem model run can also be corroborated with the existing version of the PAP algorithm with the input files we used to obtain these results. These input files are available here: https://github.com/DanyIvan/chempath/tree/main/examples/photochem_modern_earth/chempath_input.
We are preparing another publication in which we will apply Chempath to the model output of Garduno et. al. 2023 and 2024. If the editor thinks changing our example is absolutely necessary, we can try to use the output from other publications that used photochem or a similar model. However, this task is not straightforward since we would likely have to rerun the simulation to get the complete model output we need to run Chempath. Alternatively, we can modify our example to analyze only the stratosphere in the photochem model, where the reaction system we used seems to be fine, or we can try to update the reaction system to make our simulation more realistic. We are also happy to use our code in model output provided by the reviewer or the editor. Finally, we note that we used the same chemical mechanism Wogan et. al (2023) used, but we have different boundary conditions in the model.
References:
Garduno Ruiz, D., Goldblatt, C., and Ahm, A.-S.: Climate shapes the oxygenation of Earth’s atmosphere across the Great Oxidation Event, Earth and Planetary Science Letters, 607, 118 071, https://doi.org/10.1016/j.epsl.2023.118071, 2023.
Garduno Ruiz, D., Goldblatt, C., and Ahm, A.: Climate Variability Leads to Multiple Oxygenation Episodes Across the Great Oxidation Event, Geophysical Research Letters, 51, https://doi.org/10.1029/2023gl106694, 2024.
Wogan, N. F., Catling, D. C., Zahnle, K. J., and Lupu, R.: Origin-of-life Molecules in the Atmosphere after Big Impacts on the Early Earth, The Planetary Science Journal, 4, 169, https://doi.org/10.3847/psj/aced83, 2023.
Citation: https://doi.org/10.5194/gmd-2024-163-AC3 -
EC3: 'Reply on AC3', Rolf Sander, 10 Nov 2024
I fully agree that the high number density of CH3OO originates from an incomplete representation of its chemistry in the reaction system of Photochem, and that it is not artificially produced by Chempath. The same applies to O atoms (D2,5), which were also mentioned by the reviewer. When Chempath detects transport of O in spite of its short chemical lifetime, this most probably points to unusual values of O in the output of Photochem, not to an error in Chempath.
Nevertheless, I do not find the evaluation of Chempath in its current form convincing. For example, the manuscript states:
"It is widely accepted that tropospheric ozone production is dominated by the NOx mediated photochemical oxidation of CO and hydrocarbons. Our pathway analysis algorithm captures this fact in the photochem model output."
When Chempath finds accepted pathways in (at least partially) incorrect model output, this does not validate the Chempath results at all!
My initial suggestion was to run Chempath on different model output, i.e., on data that have already been properly evaluated. I now understand that this may take too much time. As an alternative, it would be fine to keep the current results but rewrite the analysis. Summarizing my main points, I'd like to emphasize that:
- When Chempath detects unexpected pathways (or transport) in some model output, the model should be checked carefully for potential problems. Thus, Chempath can be a great tool for model developers.
- However, when Chempath finds widely accepted pathways in questionable model output, this does not prove that Chempath works correctly.
Citation: https://doi.org/10.5194/gmd-2024-163-EC3
-
EC3: 'Reply on AC3', Rolf Sander, 10 Nov 2024
-
AC3: 'Reply on EC2', Daniel Garduno, 05 Nov 2024
-
EC2: 'Reply on AC2', Rolf Sander, 02 Nov 2024
-
AC2: 'Reply on EC1', Daniel Garduno, 30 Oct 2024
-
EC1: 'Reply on AC1', Rolf Sander, 30 Oct 2024
-
AC1: 'Reply on RC1', Daniel Garduno, 30 Oct 2024
-
RC2: 'Comment on gmd-2024-163', Rolf Sander, 13 Nov 2024
Unfortunately, Referee #2 (who initially agreed to review the manuscript) is now unable to provide a review. Therefore, I have decided to continue with only one review.
I encourage the authors to submit a revised version of the manuscript, rewriting the analysis of the Photochem results as discussed, and also taking into account the suggestions from Anonymous Referee #1.
Citation: https://doi.org/10.5194/gmd-2024-163-RC2
Status: closed
-
RC1: 'Comment on gmd-2024-163', Anonymous Referee #1, 25 Sep 2024
-
AC1: 'Reply on RC1', Daniel Garduno, 30 Oct 2024
Thank you so much for a thorough and detailed review. Please see the attached document.
-
EC1: 'Reply on AC1', Rolf Sander, 30 Oct 2024
Thanks for the very interesting discussion! There seems to be a major
discrepancy about the lifetime of CH3OO. The reviewer mentions a
chemical lifetime of < 1 min. This is consistent with the number given
by Wolfe et al. (doi:10.5194/acp-14-4715-2014) in their Fig. 9a.The authors, however, found a lifetime between 1 and 170 days in the
Photochem model. Apparently, Photochem is a very generic model which can
be used for any planetary atmosphere. So maybe some tropospheric
reactions which are specifically necessary for the current Earth's
atmosphere are missing? For example, I could not find CH3OO + HO2
anywhere in the reactions.txt files of the Photochem github repository.To investigate this issue further, it would be very helpful if the
authors can provide the chemical mechanism (reactions and rate
constants) that they used in Photochem.Citation: https://doi.org/10.5194/gmd-2024-163-EC1 -
AC2: 'Reply on EC1', Daniel Garduno, 30 Oct 2024
Thanks for your comment and the reference. The chemical mechanism we used can be consulted in the file `zahnle_earth.yaml` hosted here: https://github.com/Nicholaswogan/photochem_clima_data/tree/main/photochem_clima_data/data/reaction_mechanisms
In this mechanism, there are seven reactions involving CH3O2. As you mentioned, there is no CH3OO + HO2 reaction:
- equation: CH3 + O2 (+ M) <=> CH3O2 (+ M)type: fallofflow-P-rate-constant: {A: 1.494515e-25, b: -3.3, Ea: 0.0}high-P-rate-constant: {A: 2.6e-11, b: 0.0, Ea: 0.0}- equation: CH3O2 + H <=> CH3O + OHrate-constant: {A: 1.6e-10, b: 0.0, Ea: 0.0}- equation: CH3O2 + H <=> CH4 + O2rate-constant: {A: 4.0e-14, b: 1.0, Ea: 8350.0}- equation: CH3O2 + O <=> CH3O + O2rate-constant: {A: 6.0e-11, b: 0.0, Ea: 0.0}- equation: CH3O2 + OH (+ M) <=> CH3OH + O2 (+ M)type: fallofflow-P-rate-constant: {A: 0.0, b: 0.0, Ea: 0.0}high-P-rate-constant: {A: 0.0, b: 0.0, Ea: 0.0}- equation: CH3O2 + CH3 <=> CH3O + CH3Orate-constant: {A: 4.0e-11, b: 0.0, Ea: 0.0}- equation: CH3O2 + NO <=> CH3O + NO2rate-constant: {A: 2.8e-12, b: 0.0, Ea: -285.0}Citation: https://doi.org/10.5194/gmd-2024-163-AC2 -
EC2: 'Reply on AC2', Rolf Sander, 02 Nov 2024
Thanks for providing the complete list of CH3OO reactions. I noticed
that not only CH3OO + HO2 but also the self-reaction CH3OO + CH3OO is
missing. I expect that after adding both, the chemical lifetime of CH3OO
will be much closer to one minute. Still, I cannot be sure if these are
the only reactions missing in the chemical mechanism of Photochem.How to proceed?
On the one hand, the Photochem model is not the topic of the current
manuscript. It is certainly not the task of the current authors to
improve the applicability of Photochem to the current terrestrial
troposphere.On the other hand, the evaluation of Chempath is not convincing when it
results in questionable values for CH3OO (and maybe other species as
well).Would it be possible to use the results of a different model run?
Ideally, Chempath should be applied to model output which has been
evaluated in a peer-reviewed publication. For example, Wogan et al.
(2023) used Photochem for the Early Earth, and it could be interesting
to identify significant pathways in those results. I think any
quality-checked model output from Photochem or any other chemistry model
would be fine for testing Chempath. It could be for the Early Early,
modern Earth, or even exoplanets.Citation: https://doi.org/10.5194/gmd-2024-163-EC2 -
AC3: 'Reply on EC2', Daniel Garduno, 05 Nov 2024
The purpose of the example provided in section 4 of the manuscript is to show how to use Chempath in a 1D photochemical model, not to apply Chempath to a model in which all species match the observations. We disagree that the evaluation of Chempath is not convincing when it results in questionable values for CH3OO. The CH3OO high number density results from an incomplete representation of its chemistry in the reaction system we used, and is not a result artificially produced by Chempath. Chempath is a tool for evaluating photochemical models. The fact that our code was able to show the lack of important reactions for CH3O2 chemistry in the photochem model ModernEarth reaction system demonstrates that Chempath fulfills its purpose: to evaluate and understand the results and photochemical models. We will update the manuscript to discuss the inconsistency of CH3O2 in the photochem model and warn the reader to take our results with caution since the model we used is not intended to be a perfect representation of the modern Earth's atmosphere.
If there is a concern that our code is not working well or producing artificial results, this can be easily discarded using our input files and the reviewer’s implementation of the PAP algorithm. For example, the reviewer could use their algorithm implementation to corroborate that they get results similar to ours in the simple example we used to explain how the algorithm works. The Chempath code repository also includes an example of how to apply Chempath to a simple photochemical box model (https://github.com/DanyIvan/chempath/blob/main/examples/box_model_pathways/box_model_pathways_example.ipynb). These results could also be corroborated with the reviewer's implementation of the algorithm. Finally, our results for our the photochem model run can also be corroborated with the existing version of the PAP algorithm with the input files we used to obtain these results. These input files are available here: https://github.com/DanyIvan/chempath/tree/main/examples/photochem_modern_earth/chempath_input.
We are preparing another publication in which we will apply Chempath to the model output of Garduno et. al. 2023 and 2024. If the editor thinks changing our example is absolutely necessary, we can try to use the output from other publications that used photochem or a similar model. However, this task is not straightforward since we would likely have to rerun the simulation to get the complete model output we need to run Chempath. Alternatively, we can modify our example to analyze only the stratosphere in the photochem model, where the reaction system we used seems to be fine, or we can try to update the reaction system to make our simulation more realistic. We are also happy to use our code in model output provided by the reviewer or the editor. Finally, we note that we used the same chemical mechanism Wogan et. al (2023) used, but we have different boundary conditions in the model.
References:
Garduno Ruiz, D., Goldblatt, C., and Ahm, A.-S.: Climate shapes the oxygenation of Earth’s atmosphere across the Great Oxidation Event, Earth and Planetary Science Letters, 607, 118 071, https://doi.org/10.1016/j.epsl.2023.118071, 2023.
Garduno Ruiz, D., Goldblatt, C., and Ahm, A.: Climate Variability Leads to Multiple Oxygenation Episodes Across the Great Oxidation Event, Geophysical Research Letters, 51, https://doi.org/10.1029/2023gl106694, 2024.
Wogan, N. F., Catling, D. C., Zahnle, K. J., and Lupu, R.: Origin-of-life Molecules in the Atmosphere after Big Impacts on the Early Earth, The Planetary Science Journal, 4, 169, https://doi.org/10.3847/psj/aced83, 2023.
Citation: https://doi.org/10.5194/gmd-2024-163-AC3 -
EC3: 'Reply on AC3', Rolf Sander, 10 Nov 2024
I fully agree that the high number density of CH3OO originates from an incomplete representation of its chemistry in the reaction system of Photochem, and that it is not artificially produced by Chempath. The same applies to O atoms (D2,5), which were also mentioned by the reviewer. When Chempath detects transport of O in spite of its short chemical lifetime, this most probably points to unusual values of O in the output of Photochem, not to an error in Chempath.
Nevertheless, I do not find the evaluation of Chempath in its current form convincing. For example, the manuscript states:
"It is widely accepted that tropospheric ozone production is dominated by the NOx mediated photochemical oxidation of CO and hydrocarbons. Our pathway analysis algorithm captures this fact in the photochem model output."
When Chempath finds accepted pathways in (at least partially) incorrect model output, this does not validate the Chempath results at all!
My initial suggestion was to run Chempath on different model output, i.e., on data that have already been properly evaluated. I now understand that this may take too much time. As an alternative, it would be fine to keep the current results but rewrite the analysis. Summarizing my main points, I'd like to emphasize that:
- When Chempath detects unexpected pathways (or transport) in some model output, the model should be checked carefully for potential problems. Thus, Chempath can be a great tool for model developers.
- However, when Chempath finds widely accepted pathways in questionable model output, this does not prove that Chempath works correctly.
Citation: https://doi.org/10.5194/gmd-2024-163-EC3
-
EC3: 'Reply on AC3', Rolf Sander, 10 Nov 2024
-
AC3: 'Reply on EC2', Daniel Garduno, 05 Nov 2024
-
EC2: 'Reply on AC2', Rolf Sander, 02 Nov 2024
-
AC2: 'Reply on EC1', Daniel Garduno, 30 Oct 2024
-
EC1: 'Reply on AC1', Rolf Sander, 30 Oct 2024
-
AC1: 'Reply on RC1', Daniel Garduno, 30 Oct 2024
-
RC2: 'Comment on gmd-2024-163', Rolf Sander, 13 Nov 2024
Unfortunately, Referee #2 (who initially agreed to review the manuscript) is now unable to provide a review. Therefore, I have decided to continue with only one review.
I encourage the authors to submit a revised version of the manuscript, rewriting the analysis of the Photochem results as discussed, and also taking into account the suggestions from Anonymous Referee #1.
Citation: https://doi.org/10.5194/gmd-2024-163-RC2
Model code and software
chempath Daniel Garduno Ruiz https://doi.org/10.5281/zenodo.13715328
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 424 | 136 | 738 | 1,298 | 19 | 25 |
- HTML: 424
- PDF: 136
- XML: 738
- Total: 1,298
- BibTeX: 19
- EndNote: 25
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 347 | 27 |
| Canada | 2 | 122 | 9 |
| Romania | 3 | 86 | 6 |
| Sweden | 4 | 56 | 4 |
| Germany | 5 | 54 | 4 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 347