CMAQ v.5.3.1 IPR cloud process problem

Hi, all!

I have a question of cloud process in CMAQ IPR program.

I tried to calculate contribution of wet scavenging using process analysis IPR program.
So, I set all of values in cloud chemistry routine from aqchem.F to zero and did run of IPR program.
And I expected to get some results with all of hourly negative values in cloud process.
But I got some positive values from cloud process.

I think if there is zero concentration from cloud chemistry (e.g. sulfate formation from droplets), cloud process of IPR program only reflect wet scavenging process, so, the results would be negative.

So, the result with positive values from cloud process with zero cloud chemistry is a error or bug of cloud process in IPR program ??

I’m looking forward to getting any responses from you all!

Thank you

HyeonYeong Park

Please give more description of what changes you made to aqchem.F, specifically which values you set to 0. Is it aerosol species that have positive changes, or gas-phase species?

Hi, nolte!
Thank you for your response :slight_smile:
I attached my revised aqchem.F file with text format below.

I revised line 1260 to 1321 in aqchem.F, which means sulfate formation from aqueous-phase chemistry reaction (DSIVDT(1~5)) set to zero.

I thought if I set values of these reactions to zero, cloud process of IPR program would not reflect sulfate formation from cloud chemistry.
So, only wet scavenging would be reflected to calculate cloud process of IPR program and value of cloud process would be negative.

If you available, please confirm my attached file.
And my assumption is wrong, please let me know how to revise aqchem.F.

I wish I get any responses from you!

Thank you.
aqchem.F.txt (76.7 KB)

Without aqueous-phase sulfate oxidation reactions, I also would have expected changes due to clouds to be negative from aerosol scavenging.

Please give more information. What does your process analysis control file contain? Are you looking at variables beginning with “CLDS_” in the CCTM_PA_1 file? What species have positive changes, and how large are those changes?

Hi, nolte!
I attached some files from my process analysis results and process analysis directory. And I used the ‘CLDS’ values from CCTM_PA_1 file too.

First, you may find some weird points from excel file named ‘CMAS FORUM’, that means the values of CLDS_ASO4 without sulfate formation from chemical reaction (DSIVDT(0~5) = 0.0 from aqchem.F) are bigger than those values with sulfate formation from chemical reaction (DSIVDT(0~5) NOT 0.0).

Also, I think the values of ‘CLDS_ASO4 (DSIVDT = 0.0 CASE) + WETDEP_ASO4’ should be similar to 0.0. But some cases the values are bigger than those values with DSIVDT NOT 0.0 case.

And also, I found that gaseous H2SO4 concentrations were included to load aqueous sulfate concentrations at 690 line from aqchem.F. So, I revised that line and compared the results with two cases.

  1. LOADING(LSO4,ACC) = (AEROSOL(LSO4,ACC) + GAS(LH2SO4)) (‘with H2SO4’ at ‘dsivdt=0.0’ sheet)
  2. LOADING(LSO4,ACC) = AEROSOL(LSO4,ACC) (‘without H2SO4’ at ‘dsivdt=0.0’ sheet)

Then, you can find that if DSIVDT set to 0.0, the wet scavenging of sulfate occurred more in without H2SO4 case compared to with H2SO4 case.

In other words, if there is no sulfate formation from aqueous chemical reaction, most of sulfate concentration from ‘LOADING(LSO4,ACC) =…’ would be assigned H2SO4 concentration.

I think this is the reason why I get positive value of CLDS_ASO4 when DSIVDT set to 0.0.

Please confirm attached files, and let me get your answers.

Thank you.

CMAS_FORUM_dsivdt=0.0.csv|attachment (6.7 KB)
CMAS_FORUM_example.csv (10.8 KB)
pa_cb6r3_ae7_aq_mod.txt (2 KB)

1 Like

Do the values in your spreadsheets represent sums over the entire modeling domain, or are they values in specific grid cells?

Regardless, I think you have answered your own original question. Aerosol sulfate comes from aqueous phase oxidation of SO2 (actually S(IV)) and from scavenging of sulfuric acid vapor, SULF. If you set the S(IV) oxidation rates to 0, you still have scavenging of SULF, so can get positive values for CLDS_ASO4J.

1 Like

Thank you for your response!

I made the sums over the entire modeling domain.
And I’m appreciate to you to recognize me to get answer of my question.

But I have also a question.
CMAQ is assuming that all of sulfuric acid vapor (H2SO4) converted into the aerosol like sulfate.
And also, wet deposition values of sulfuric acid vapor are set to zero.
But why scavenging of sulfuric acid vapor (CLDS_SULF) is not similar to zero ??

I think if all of sulfuric acid vapor are converted into the aerosol, they should be considered as sulfate because CMAQ is assuming all of sulfuric acid vapor would be scavenged and be sulfate.
So, if there may be a little of sulfuric acid vapor in the atmosphere, their concentration would be similar to 0.0.
But my results represented that the sum of CLDS_SULF over the entire domain was not similar to zero.

If you have any answer about this question, please don’t hesitate and give me your answer!

Thank you.

phy,
I do not understand what it is that is confusing you.

SULF is gas-phase sulfuric acid. CLDS_SULF is the change in gas-phase sulfuric acid that occurred during the cloud processing step. If gas-phase sulfuric acid is scavenged it becomes either aerosol sulfate or is wet deposited. Either way, CLDS_SULF should be a negative quantity equal to the amount by which SULF has decreased.

Hi,
I think it’s possible that the value of H2SO4 gas is changing in the aerosol routine. Before CMAQv5.3.2, we had slight sulfur conservation issues with the algorithm used to calculate the pseudo-steady-state concentration of H2SO4 gas and apply it to condensation on particles.

A hypothesis for what might be happening here is that when you turn off SIV conversion to SVI and reduce the amount of ASO4 that is generated by clouds in every time step, you then reduce the size of particles in those cells and increase the PSSA concentration of H2SO4 when the aerosol partitioning is calculated. This could potentially then feed back to the cloud chemistry routine and appear as larger values of H2SO4 being remove every time through.

We have corrected the sulfur conservation issue in CMAQv5.3.2. It could be worth trying that version if sulfur conservation is critical for your work.

Best wishes,
Ben

The modification Ben mentions above did not make it into the v5.3.2 release. It will be included in the next release.

Hi, Dr. nolte and murphy!
I read your responses.

I think I made you confused because I didn’t explain what I was wondering specifically.
I’m so sorry about that.

I was confused because I thought the default sum of CLDS_SULF and CLDS_ASO4 values when I turned on SIV conversion to SVI and the sum of revised CLDS_SULF and CLDS_ASO4 values when I turned off SIV conversion to SVI were quite similar to each other.

Because I thought conversion amount of CLDS_SULF to CLDS_ASO4 would be same regardless of turning off SIV conversion to SVI, so, if I turned off the conversion, the same amount of CLDS_SULF would be remained in CLDS_SULF.

But the results said that turning off the conversion had more H2SO4 and more scavenging of H2SO4 concentration.
I think this is the sulfur conservation you mentioned.

So, I want you to explain more detailed about sulfur conservation you’ve already mentioned!

Thank you.

HyeonYeong Park

Hi HyeonYeong,

Try the following two updates in the file VOLINORG.F:

  1. Somewhere near line 1102, comment the line:

    PRECURSOR_CONC( SULF_IDX ) = REAL( XH2SO4, 8 )
    
  2. Somewhere near lines 1676-1677, add the lines:

        PRECURSOR_CONC( SULF_IDX ) = PRECURSOR_CONC( SULF_IDX )
    &                            +SUM( -DELSO4(:) * H2SO4RAT )
    

    These lines for (2) should be added after the comment reading “Update gas-phase concentrations from Aitken and Accumulation Mode Partitioning” and before the the lines:

        PRECURSOR_CONC( NH3_IDX )  = GNH3R8 + sum( CINORG( KNH4,1:2 )
    &                            -CFINAL( KNH4,1:2 ) ) * NH3RAT 
    

Good luck!
Ben

1 Like

Hi, Dr. Murphy!

Thank you to let me know these updates :slight_smile:

I’m going to try the run with these updates soon!

Thank you so much again.

HyeonYeong Park

Hi, Dr. Murphy!

Finally, I got the results that I want from cloud process of process analysis with your revision code.
I’m really appreciate to you to help me solve this difficult problem.

Then, Dr. nolte said that the revision of code have not been updated in CMAQ v5.3.2. too.

So, the revised code of sulfuric acid you mentioned will be applied to next CMAQ version in the future as Dr. nolte mentioned??

HyeonYeong Park