isoprene is too low in WRF-Chem with MEGAN

[Suqian Chu]

Dear all,

I am using WRF-Chem 3.7.1 with biogenic emissions from MEGAN v2.04 and anthropogenic emissions from NEI 2011. "isop" and "terp" from the 2-week simulation are very low (<0.1 ppb) in the entire domain.

I compared the results with MEGAN (bio_emiss_opt=3) to the simulation with simple Guenther approach (bio_emiss_opt=1). isop and terp are different and the other variables remain the same.

Attached are the maps of isop on 20220424 (daily average) averaged for the first three levels from these two simulations, and the wrfbiochemi_d01 that I used to run the model. I was expecting the model simulation with MEGAN would produce higher isop so the results do not make sense to me. Has anyone had this issue?

Any help would be greatly appreciated!

Suqian

[Gabriele Pfister]

Hi Suqian

I looked at your wrfbiochemi file and that seems to cover Europe while your plots show the Eastern US. Did you make sure you are using the correct wrfbiochemi file in your simulation?

Gabi

--

================================
Gabriele Pfister

Atmospheric Chemistry Observations & Modeling (ACOM) Lab

NSF National Center for Atmospheric Research (NCAR)

Email: pfi...@ucar.edu

Work Phone: +1 303 497 2915

Web: https://staff.ucar.edu/users/pfister

[Suqian Chu]

Hi Gabi,

Thanks for your quick response! The maps from wrfchem output are due to the plotting area. The domain is the entire US but I just plotted the Eastern US. 

I think I was using the correct wrfinput (same domain and time info) to generate the wrfbiochemi. Here I attached the map of MSEBIO_ISOP from wrfbiochemi file. I am a bit confused because this MSEBIO_ISOP map does not conver Europe. Do you have the plot of my wrfbiochemi file?

Suqian

[Gabriele Pfister]

Hi Suqian

sorry, then something must have gotten mixed up when I downloaded your file to my computer. 

Have you compared your isoprene and terpene concentrations to observations to determine they are indeed too low? Also, the plots you had sent seemed to represent daily averages and I assume the daytime values are quite a bit higher?

Gabi

[Suqian Chu]

Hi Gabi,

Thanks for your reply. I do not have isoprene observations at SGP site, but at least the isoprene should not be always below 0.05ppb over Southeastern US . I did check the maps at 12Z and they are still all purple (<0.05ppb) over the entire domain. 

I am using chem_opt=132. I am wondering if you could have a quick test by running WRF-Chem with my wrfbiochemi_d01 file from MEGAN v2.04 to see if the isoprene is also such low?

Thanks,

Suqian

[Gabriele Pfister]

Dear Suqian

it would take quite some time for me to set up a run and I am afraid I just do not have the bandwidth for that. Please also consider that we are not familiar with chemistry options other than the by us developed suite of MOZART schemes and therefore cannot even say whether for chem_opt = 132  the biogenic emissions are indeed correctly implemented. 

I attach, however, an output file from our experimental CONUS forecast system (https://www.acom.ucar.edu/firex-aq/forecast.shtml) for 220422 18 UTC. It included EBIO_ISO and isopr fields and you might be able to compare whether your emissions are in a reasonable range. 

Gabi

 wrfout_d01_2022-04-22_18:00:00

[Suqian Chu]

Hi Gabi,

Thank you so much for providing this wrfout file! Is it generated using MEGAN biogenic emission? What anthropogenic emission was used?

Thanks,

Suqian

[Gabriele Pfister]

Yes, this simulation is using MEGAN and chem_opt=114. We used trend adjusted NEI2017v2.

Gabi

[Suqian Chu]

Thanks for the information! I am wondering if you could attach the wrfbiochemi file for your wrfout? I think it would be helpful if I could also compare the emission files. 

Thanks,

Suqian

[Gabriele Pfister]

Attached. 

Gabi

[Suqian Chu]

Thank you very much! The plots from wrfbiochemi files are the same as mine. It looks my issue are from a certain chem option in WRF-Chem. 

Best,

Suqian 

[Mary Barth]

Hi Suqian,

I would be surprised if chem_opt=132 does not permit MEGAN biogenic emissions. You could check this in the code in chem/emissions_driver.F. 

I would also suggest writing EBIO_ISO from your simulation to see if the isoprene emissions are generated. 

Good luck,

Mary

^--^--^--^--^--^--^--^--^--^--^--^--^--^--^-^--^--^--^-^--^

Mary Barth 

Senior Scientist   

phone: 303-497-8186    email: bar...@ucar.edu

NSF - National Center for Atmospheric Research

P.O. Box 3000

Boulder, CO 80307

Staff web page

SIMA Web Site

MUSICA Web Site

DC3 Web Site 

^--^--^--^--^--^--^--^--^--^--^--^--^--^--^-^--^--^--^-^--^

[Suqian Chu]

Hi Mary,

Based on the emission_driver code, I think chem_opt132 is supported for MEGAN by calling bio_emissions_megan2. Attached is the daily average EBIO_ISO on 20220424 at the first level from my wrfout and EBIO_ISO from Gabi`s wrfout. It looks isoprene emissions are generated but very low.  

The extra settings for MEGAN in my namelist.input are:

io_form_auxinput6       = 2,

auxinput6_inname        = "wrfbiochemi_d<domain>",

auxinput6_interval_d    = 30,     ! simulation length is a week

 bio_emiss_opt           = 3,
 ne_area                 = 500,

If the wrfbiochemi and namelist.input look good to you, it should be from the chemical mechanism that I am using.

Thanks,

Suqian

[Suqian Chu]

Hi Mary,

I just found the attached EBIO_ISO from my output in my previous reply is daily average. Here I attached the map at 20220422 18Z which is consistent with the other figure. This looks more reseaonable.

Suqian