Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

Generation of Tracer Emissions Data Files

K

kwyang

New member
Hi everyone, I am hoping to generate a methane plume in WRF_chem. My understanding based on the provided WRF chem v4.4 User Guide and Emission Guide 3.9.1.1. is that emission files are separate from WRF_Chem, one that we have to generate own its own as a intermediate file and be converted via convert_emiss.F program.

I am trying to follow the following example routine: ftp://aftp.fsl.noaa.gov/divisions/taq/emissions_data_2005/emiss_v03_tracer.F (From 6.1 of Emission guide). However, I was bit confused on where to run this Fortran program. Is it to be stored in "chem" directory in my WRF_Chem and be compiled? The emission guide states "Once the sample program has been modified, compiled and run, a binary intermediate file containing the trader emissions will be produced." I was not sure where to compile and run since if I compile wrf_chem with this emiss_v03_tracer.F stored in "chem" directory nothing really seems to happening. Thank you for the support.

Emission guide: https://ruc.noaa.gov/wrf/wrf-chem/Emission_guide.pdf
 
Hi kwyang, I am a beginner of WRF, Do you know how to generate a plume with a specified emission rate? Is this Fortran file use to set Emission rate?：）
 
richard_n413 said:
This may be of interest to you: Brian Blaylock, UofU. It explains how to add passive tracers to your WRF setup.
Click to expand...
Thanks:), I successfully created the plume by referring to this blog, but I am confused about what the 1 in
Code: 
tracer(i,1,j,P_S_SLV)= tracer(i,1,j,P_S_SLV)+1
represents. Most papers link the plume simulated by wrf to the emission rate and concentration(ppb), Many places say this is dimensionless, which is very confusing to me. I wonder what the values in the 3D matrix of the obtained plume represent.
 
I'm using this procedure as well, and my understanding is that it is dimensionless, but can be mapped to an emission rate. I've been using an arbitrary emissions rate scale factor to convert the dimensionless value to a meaningful value in post-processing.
 
richard-newton said:
I'm using this procedure as well, and my understanding is that it is dimensionless, but can be mapped to an emission rate. I've been using an arbitrary emissions rate scale factor to convert the dimensionless value to a meaningful value in post-processing.
Click to expand...
Thanks~ Really helpful. And is there a certain formula for vertical integration? Or just a simple integration with the 'z' in wrfout?
 
I can't provide a definitive answer to that. Perhaps there's something in WRF-Python that may be able to do that accurately, but I think a simple integration would work.
 
hi richard-newton, I'm also working with passive tracer simulations and also I have an issue to interpret the units of the passive tracer and how to link them to a emission rate. Can you explain me how are you using the scale factors for the emission rate you mention earlier? I'm trying to simulate CH4 plumes but I don't know if I should transform the tracer values into concentration (how? ppm? ppb?) or directly to a flux (which units? kg/h?). If you have some insights on this I would greatly appreciate, thanks
 
For my experiment, it would be sufficient to just convert directly into a concentration, and then the overall methane concentrations would be the plume concentrations + background concentrations. I'm using the same plume to represent different concentrations, so I'll have a scale factor of, e.g. 10 ppm, 50 ppm, and 100 ppm to represent different plume sizes. However, if you are simulating a plume of known emission rate, using that as the scale factor, and then somehow converting that to concentrations would be needed. There are most likely better ways of doing this though.

I'm finding it hard to get the background concentrations of methane into the WRF model, however, which is preventing me from testing this at the moment.
 
richard-newton said:
For my experiment, it would be sufficient to just convert directly into a concentration, and then the overall methane concentrations would be the plume concentrations + background concentrations. I'm using the same plume to represent different concentrations, so I'll have a scale factor of, e.g. 10 ppm, 50 ppm, and 100 ppm to represent different plume sizes. However, if you are simulating a plume of known emission rate, using that as the scale factor, and then somehow converting that to concentrations would be needed. There are most likely better ways of doing this though.

I'm finding it hard to get the background concentrations of methane into the WRF model, however, which is preventing me from testing this at the moment.
Click to expand...
Thank you for your ideas. I did the following in Python to get vertical integration：
Code: 
plume = getvar(ncfile,'plume')
z=getvar(ncfile,'z')
dz=np.diff(z.data,axis=0)
plume_i=(plume[:-1]*dz).sum(axis=0)

Can I regard the unit of each grid as kg/m^2?
Then according to(IME):

1720528932274.png
Where Q [kg s−1], ΔΩ[kg m−2], Ueff[m s−1], A[m^2] represents the area of each small grid and L[m] is the plume length.
I calculated L and Ueff, set Q to 10000/3600 kg s−1, A is 400m^2, and then calculated the total ΔΩ.
Is the scale factor = total ΔΩ/sum of plume values after vertical integration?
After this i can use the scale factor * plume values after vertical integration to connect with real concentration?
I apologize if this is a very basic question, thanks for your reading.
 
Hi, sorry for interruption. Can convert_emiss run properly in the WRFv4.4? I have ran then failed many times. It always shows the 'memory corruption'. Compiling an older version is also difficult since it's too old and many libraries couldn't support it. I have worked on it for several weeks.
 
You must log in or register to reply here.
Top