search

geoschem / geos-chem

GEOS-Chem "Science Codebase" repository. Contains GEOS-Chem science routines, run directory generation scripts, and interface code. This repository is used as a submodule within the GCClassic and GCHP wrappers, as well as in other modeling contexts (external ESMs).
http://geos-chem.org
Other
167 stars 162 forks source link

[QUESTION]How to get the photolysis cross sections of a new species #994

Open ymfu2021 opened 2 years ago

ymfu2021 commented 2 years ago

Ask a question about GEOS-Chem:

I'd like to caculate some photolysis frequencies using Fast-JX v7.0a implemented in GEOS-Chem, and I found that I need to generate the cross-section using file 'UCI_fastJ_addX_73cx.tar.gz' first. When reading those codes, I am very confused about the 'subroutine for fast-JX Xsection generation', which seems interpolating X-section vs wavelength and outputting the cross-section used for photolysis frequency. Does that mean that I need to get the section data by some caculation first by some kind method? What I want to ask is if my understanding about those codes is right, and if so, how can I get the original section data for interpolation of specific species

UCI_fastJ_addX_73cx.tar.gz

Thanks.

yantosca commented 2 years ago

Thanks for writing @ymfu2021. The tar.tz file you listed is the proper package to compute optical properties for new species. Unfortunately, we cannot provide any assistance in this area, as this is a research issue rather than a coding issue. I would recommend that you reach out to the GEOS-Chem Aerosols Working Group for assistance.

ymfu2021 commented 2 years ago

Thanks. Also I wanna know if I have rewrited the code for my species, how to get the photolysis frequencies before I run the whole model. Can the 'dry-run' option give me such results? Besides, may you explain the data listed in FJX_spec.dat? Whether those figures behind the species' name are cross-section data corresponding to w-eff(nm) above or not. Are those numbers next to species names means temperature(such as 180 behind O2)?

O2 180 1.727E-21 1.989E-22 3.004E-23 9.833E-24 7.306E-24 6.827E-24 J10X 6.238E-24 5.748E-24 1.153E-25 5.030E-25 4.150E-25 0.000E+00 corr.bin8 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00

http://geoschemdata.wustl.edu/ExtData/CHEM_INPUTS/FAST_JX/v2020-02/FJX_spec.dat

yantosca commented 2 years ago

@ymfu2021, the dry-run package can only print out logs of data files to download.

I am tagging @sdeastham who will know the format of the FJX_spec.dat file. I think the number after the O2 may be temperature but I am not sure.

ymfu2021 commented 2 years ago

Thanks for helping @yantosca.

stale[bot] commented 2 years ago

This issue has been automatically marked as stale because it has not had recent activity. If there are no updates within 7 days it will be closed. You can add the "never stale" tag to prevent the Stale bot from closing this issue.

yantosca commented 2 years ago

Closing out this issue for now.

lizziel commented 2 years ago

I am reopening this issue to discuss further.

sdeastham commented 2 years ago

One detail I can add - interpretation of the FJX_spec.dat file should be done with great care. The wavelengths listed at the top are "representative" at best - in reality, the cross section for each "bin" is the weighted mean cross section over at least one contiguous wavelength bins. The package that was linked to gives you the code you need if you want to take a high-resolution set of cross section measurements and calculate the weighted averages which need to be put into FJX_spec.dat.

To get more information on how that code works, unfortunately all I really remember is that careful scrutiny of the code in that tarball will be needed - it has some examples on how to generate a new entry for FJX_spec.dat. There's slightly more info at https://kpp.readthedocs.io/en/latest/creating-mechanisms/reactions.html#photolysis-reactions which may help!

lizziel commented 2 years ago

I'm also bringing in @pratherUCI who I believe wrote the package.

pratherUCI commented 2 years ago

Ok. Who is involved and what is the question? Let us see if we can answer it Michael

⁣Michael Prather ---sent from phone, beware of android's Otto correct.

Get BlueMail for Android ​

On Dec 14, 2021, 16:56, at 16:56, Lizzie Lundgren @.***> wrote:

I'm also bringing in @pratherUCI who I believe wrote the package.

-- You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub: https://github.com/geoschem/geos-chem/issues/994#issuecomment-994116684

ymfu2021 commented 2 years ago

Thanks for your helping. After reading codes in 'UCI_fastJ_addX_73cx.tar.gz', I noticed that the sample subroutines for fast -JX Xsection generation are different for sample species. Does that just because of the format of original data used for caculating(for example, 'XCFCl3_JPL11X.dat' and 'XMeVK_JPL11X.dat')? What's more, my comprehension about file 'FJX_spec.dat' are: (1)Those 18 wavelengths listed at the top are "representative" at best - in reality as Sdeastham said. (2)The cross sections below are 18-bin, which all were the weighted mean outputed by those codes in 'UCI_fastJ_addX_73cx.tar.gz'. (3)the figures behind the species' name are corresponding temperatures setted in caculation.
Do I comprehend them rightly? Also I believe those 18 wavelengths and 18 cross sections are dependent instead of one to one correspondence, but I'm not sure. Can you explain it for me? Thanks.

stale[bot] commented 2 years ago

This issue has been automatically marked as stale because it has not had recent activity. If there are no updates within 7 days it will be closed. You can add the "never stale" tag to prevent the Stale bot from closing this issue.

stale[bot] commented 2 years ago

Closing due to inactivity

pratherUCI commented 2 years ago

If you want to understand how Fast-J integrates over wavelengths, then you must read the first 2 Fast-J & Fast-J2 papers: Wild, O., X. Zhu, and M.J. Prather (2000) Fast-J: Accurate simulation of in- and below-cloud photolysis in tropospheric chemical models, J. Atmos. Chem., 37, 245-282. Bian, H., and M.J. Prather (2002) Fast-J2: Accurate Simulation of stratospheric photolysis in global chemical models, J. Atmos. Chem., 41, 281-296.

Adding a new species is a scientific problem with scientific decisions that must be made. If it becomes 'push button' you will probably be making a mistake.

The user must gather the laboratory data on the cross sections and quantum yields and write a module that delivers these values at any wavelength, temperature and pressure. Then the two example of 'addX_' that are provided will do the integration. You must also make a decision as to how to interpolate across temperature or pressure and addX will generate cross-section x quantum-yield tables at the points you specify. once again, this is a scientific decision to be made.

There is a standalone version of Cloud-J that you can run to test your new cross sections, but this is provided as is.

Michael Prather