How to output tracer tendency due to vertical transport by convective processes in CAM6/WACCM6

[czheng]

Greetings,

I am trying to set up model runs in CAM6/WACCM6 to understand tracer (e.g. CO or pollutants) transport by atmosphere. As far as I know (I have seen papers and have run experiments on my own), in CAM4/CAM5, the models can output the tendency of tracer concentration at each grid point due to different transport processes, including transport due to resolved dynamics (mainly advection), vertical diffusion, as well as transport due to deep and shallow convection.
In order to output/save these tendency terms in CAM4/CAM5, I can just include TAxx (for resolved dynamics), VDxx (for vertical diffusion), TCOND_xx and TCONS_xx (for deep and shallow convection) in usr_cam_nl file, where "xx" is the name of the tracer.
However, when running CAM6/WACCM6, TAxx and VDxx still work (I could still save the tendency of resolved dynamics and vertical diffusion), but I couldn't find out what variable names I should put in usr_cam_nl file to save the tendency terms due to convective processes (TCOND_xx and TCONS_xx don't work). So I wonder how could I output/save these tendency terms due to convection? Or what variable names I should put in usr_cam_nl file? I have checked the name list for output variables in CAM6 and I couldn't find the way to do it.

Thanks in advance!
--Cheng

 

[peverley]

Hi Cheng,

It looks like you should use DTCOND

Here's the master field list for future reference: CAM History Fields

 

[czheng]

Hi Cheng,

It looks like you should use DTCOND

Here's the master field list for future reference: CAM History Fields

Hi Courtney,

I don't think that works...

DTCOND outputs temperature tendency due to moist processes, which is not exactly what I need (tracer [e.g. carbon monoxide ] tendency due to moist processes). I also checked the history fields link you provided, and couldn't find the variable I need (that is the reason I posted this new thread).

Do you suggest I use DxxCOND (xx for the tracer tendency)？

--Cheng

 

[peverley]

Hi Cheng,

Apologies, I answered too quickly! Right - that is not the correct tracer to use. I'll ask around on Tuesday and see who knows what the correct field is. In the meantime, if you look at your atm.log file in your case (after you run it), you should see a "MASTER FIELD LIST". See if anything in there looks like it would what you're looking for.

 

[czheng]

Hi Cheng,

Apologies, I answered too quickly! Right - that is not the correct tracer to use. I'll ask around on Tuesday and see who knows what the correct field is. In the meantime, if you look at your atm.log file in your case (after you run it), you should see a "MASTER FIELD LIST". See if anything in there looks like it would what you're looking for.

Hi Courtney,

Thanks a lot for your reply!!
Also, I just checked the atm.log file, and I don't think I could find the fields I am looking for...

--Cheng

 

[brianpm]

@czheng -- I'm not sure those tendencies are included as outputs in the current version of the code. I just took a look, and it seems like they could be added by putting in addfld/outfld calls. The output fields that you want, I think, are in ptend%q that come out of the call to subroutine convtran inside the subroutine zm_conv_tend_2 (in the file zm_conv_intr.F90). That data structure, ptend%q, contains all the advected constituents/tracers, so you'll need to figure out which index you want, and I am guessing you can figure that out by looking at similar pbuf_get_index calls that access those tracers elsewhere in the code.

 

[czheng]

@czheng -- I'm not sure those tendencies are included as outputs in the current version of the code. I just took a look, and it seems like they could be added by putting in addfld/outfld calls. The output fields that you want, I think, are in ptend%q that come out of the call to subroutine convtran inside the subroutine zm_conv_tend_2 (in the file zm_conv_intr.F90). That data structure, ptend%q, contains all the advected constituents/tracers, so you'll need to figure out which index you want, and I am guessing you can figure that out by looking at similar pbuf_get_index calls that access those tracers elsewhere in the code.

@brainpm

Thanks for your reply!!
I am able to output the tendencies in zm_conv_intr.F90 following your instructions.

And, for the shallow convection part (actually the CLUBB scheme in CAM6/WACCM6), similar modification is needed in clubb_intr.F90

--Cheng