Added a new inorganic nucleation subroutine to tomas_mod.F90. by samuelod-atmos · Pull Request #2528 · geoschem/geos-chem

samuelod-atmos · 2024-10-17T20:08:02Z

The nucleation scheme is based on this paper:
https://www.science.org/doi/10.1126/science.aaf2649.

Name and Institution (Required)

Name: Samuel O'Donnell
Institution: Colorado State University

Describe the update

We have added a new inorganic nucleation mechanism to the TOMAS aerosol package. The nucleation mechanisms are based on Dunne et al. (2016; https://www.science.org/doi/10.1126/science.aaf2649), and was originally written by Jeff Pierce at CSU. We have not included the organic nucleation mechanism outlined in Riccobono et al. (2014) which is part of the Dunne et al. (2016) paper.

A new switch has been added (dunn_nuc; 0/1) in order to turn the mechanism off/on, consistent with the existing inorganic nucleation mechanisms. The updated scheme uses sulfuric acid (molecules/cm^3) and ammonia (molecules/cm^3) in a simple power-law function to give the formation rate of small particles. In order to better match observations of new particle formation events at the Southern Great Plains (SGP) observatory, we had to scale up the nucleation mechanism by 1000x for the 15 bin simulation. This updated is relatively untested with TOMAS40, so the 1000x multiplier should be tested with a TOMAS40 simulation.

Expected changes

Overall, nucleation and particle growth is a buffered system, so the nucleation rates should result in little changes to bulk aerosol mass. Ideally, the updated mechanisms should lead to better agreement with observations in terms of new particle formation event frequency, duration, and strength throughout the globe, but we have only tested this at the SGP site so far.

Reference(s)

The work with updates to TOMAS are not published yet.

Related Github Issue

Please link to the corresponding Github issue(s) here. If fixing a bug, there should be an issue describing it with steps to reproduce.

The nucleation scheme is based on this paper: https://www.science.org/doi/10.1126/science.aaf2649. We added a 1000x multiplier to the total inorganic nucleation rate in order to match observations better.

yantosca · 2024-12-02T22:09:45Z

Also tagging @BettyCroft who has worked with TOMAS more recently

yantosca · 2025-01-27T20:32:09Z


-  INTEGER  :: bin_nuc = 1, tern_nuc = 1  ! Switches for nucleation type.
+  INTEGER  :: bin_nuc = 0, tern_nuc = 0  ! Switches for nucleation type.
+  INTEGER  :: dunn_nuc = 1


If these are intended to be constant values, consider declaring these with the PARAMETER attribute (i.e. INTEGER, PARAMETER :: bin_nuc =0, etc. That will provide some extra protection in case someone else tries to modify the value elsewhere.

yantosca · 2025-01-27T20:36:43Z

    nh3ppt = Gci(srtnh4)/17.e+0_fp/(boxmass/29.e+0_fp)*1e+12_fp* &
             PRES/101325.*273./TEMPTMS ! corrected for pressure (because this should be concentration)
+
+    nh3moleccm3 = Gci(srtnh4)/boxvol*1000.e+0_fp/17e+0_fp*6.022e+23_fp ! Changed by SamO


For consistency, please use the MW of 17.04 for NH3 as defined in the species_database.yml file here:

https://github.com/samuelod-atmos/geos-chem/blob/aedbeb6b4fb2384828131425d831bcd95364d6d0/run/shared/species_database.yml#L3399-L3418

Also please use the AIRMW parameter from Headers/physconstants.F90 for the MW of air. This is defined as 28.9644 g/mol.

https://github.com/samuelod-atmos/geos-chem/blob/main/Headers/physconstants.F90

yantosca · 2025-01-27T20:40:05Z

+
+    nh3moleccm3 = Gci(srtnh4)/boxvol*1000.e+0_fp/17e+0_fp*6.022e+23_fp ! Changed by SamO

+    Mair = 2.69E19*273.15/TEMPTMS*PRES/101325.


Use _fp to prevent loss of precision:

Mair = 2.69E19_fp * 273.15_fp / TEMPTMS * PRES / 101325.0_fp

yantosca · 2025-01-27T20:41:19Z


    errorswitch = .false.

+    Mair = 2.69E19*273.15/TEMPTMS*PRES/101325.


Use _fp to prevent loss of precision:

Mair = 2.69E19_fp * 273.15_fp / TEMPTMS * PRES / 101325.0_fp

yantosca · 2025-01-27T20:41:36Z

    h2so4 = Gci(srtso4)/boxvol*1000.e+0_fp/98.e+0_fp*6.022e+23_fp
    nh3ppt = Gci(srtnh4)/17.e+0_fp/(boxmass/29.e+0_fp)*1e+12_fp* &
             PRES/101325.*273./TEMPTMS ! corrected for pressure (because this should be concentration)
+    nh3moleccm3 = Gci(srtnh4)/boxvol*1000.e+0_fp/17e+0_fp*6.022e+23_fp ! Changed by SamO


Use 17.04 for the MW of NH3 (see above comment)

yantosca · 2025-01-27T20:43:03Z

+      parameter(pbn=3.95, ubn=9.70, vbn=12.6, wbn=-0.00707, ptn=2.89)
+      parameter(utn=182., vtn=1.20, wtn=-4.19, pAn=8.00, an=1.6d-6)
+      parameter(pbi=3.37, ubi=-11.5, vbi=25.5, wbi=0.181, pti=3.14)
+      parameter(uti=-23.8, vti=37.0, wti=0.227, pAi=3.07, ai=0.00485)


Each of the parameters should be followed by d0 to prevent a loss of precision

yantosca · 2025-01-27T20:43:29Z

+      temp=tempi   !SamO
+      !temp=278.0
+      nh3=nh3i*1E-6 ! I think they want it in units of 1E6 molec cm-3
+      cna=cnai*1E-6


Use 1d-6 to prevent loss of precision

yantosca · 2025-01-27T20:43:58Z

+      !fion = 75.0 !SamO
+
+! CALCULATE ION CONCENTRATION
+      alpha_ion = 6d-8*sqrt(300./temp) + 6d-26*Mair*(300./temp)**4 ! Need Mair in air molec per cm3


Use 300.0d0 to prevent a loss of precision.

yantosca · 2025-01-27T20:44:22Z

+      kbn = exp(ubn - exp(vbn*(temp/1000. - wbn)))
+      ktn = exp(utn - exp(vtn*(temp/1000. - wtn)))
+      kbi = exp(ubi - exp(vbi*(temp/1000. - wbi)))
+      kti = exp(uti - exp(vti*(temp/1000. - wti)))


Use 1000.0d0 to prevent a loss of precision.

yantosca

Thanks for this PR @samuelod-atmos. I've noted several minor issues.

Also would add a note to the CHANGELOG.md file under the Added section Something

- Added new inorganic nucleation subroutine to `tomas_mod.F90` (cf. 10.1126/science.aaf2649)

samuelod-atmos · 2025-07-10T21:26:31Z

Hi @yantosca, just curious about the next steps that I need to take to get this merged. Thanks!

yantosca · 2025-07-11T15:13:30Z

Thanks @samuelod-atmos. Can you resolve the issues that I've mentioned and push updates to this branch?

samuelod-atmos · 2025-07-15T19:21:28Z

Thanks, @yantosca! I just want to make sure I'm doing this correctly. I thought I pushed most of the updates to tomas_mod.F90 to this branch, but I may be mistaken. Should I resolve the issues I've addressed here on github? Sorry, I'm still learning the details of contributing.

yantosca · 2025-07-15T19:40:34Z

Thanks @samuelod-atmos, maybe this slipped thru the cracks on my part. I'll take a look.

samuelod-atmos · 2025-07-15T19:43:46Z

Thanks, @yantosca! I definitely haven't updated everything (e.g., AIRMW), so I'll try to get to that ASAP.

Added a new inorganic nucleation subroutine to tomas_mod.F90.

aedbeb6

The nucleation scheme is based on this paper: https://www.science.org/doi/10.1126/science.aaf2649. We added a 1000x multiplier to the total inorganic nucleation rate in order to match observations better.

yantosca added category: Feature Request New feature or request topic: Aerosols Related to aerosol species and/or microphysics in GEOS-Chem labels Oct 29, 2024

yantosca self-requested a review January 27, 2025 20:27

yantosca self-assigned this Jan 27, 2025

yantosca reviewed Jan 27, 2025

View reviewed changes

yantosca self-requested a review January 27, 2025 20:49

yantosca requested changes Jan 27, 2025

View reviewed changes

samuelod-atmos and others added 2 commits February 20, 2025 16:03

Addressed the review comments for the pull request.

1be8d12

Merge branch 'geoschem:main' into main

90e4fe5


		nh3moleccm3 = Gci(srtnh4)/boxvol1000.e+0_fp/17e+0_fp6.022e+23_fp ! Changed by SamO

		Mair = 2.69E19273.15/TEMPTMSPRES/101325.


		errorswitch = .false.

		Mair = 2.69E19273.15/TEMPTMSPRES/101325.

Conversation

samuelod-atmos commented Oct 17, 2024

Name and Institution (Required)

Describe the update

Expected changes

Reference(s)

Related Github Issue

Uh oh!

yantosca commented Dec 2, 2024

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

yantosca left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

samuelod-atmos commented Jul 10, 2025

Uh oh!

yantosca commented Jul 11, 2025

Uh oh!

samuelod-atmos commented Jul 15, 2025

Uh oh!

yantosca commented Jul 15, 2025

Uh oh!

samuelod-atmos commented Jul 15, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

yantosca left a comment •

edited

Loading