Extreme negative values of ground heat flux (G_URB) (GRDFLX)

Would you please clarify your question? Your namelist.input indicates that is a triply-nested case with very high resolution of 0.388km in your finest domain. When you look at GRDFLX, did you find extremely large negative values over all the 3 domains? What did you mean by ' simulate 100% PV'?

I also have a few concerns about your namelist options:
(1) cumulus scheme should be turned off for all the 3 domains
(2) The basis assumption of PBL scheme becomes invalid when the resolution is smaller than 1km. This will cause some trouble in your D03.
(3) why do you set gwd_opt = 1, 0, 1 ? I would suggest that you turn off GWD over all domains.
 
By 100% PV i mean turning on PV on roof =1 (PV_FRAC_ROOF=1) in Urbparm.tbl

Should I increase D03 domain greater than 1 km to have better PBL Scheme working ?

Yes GRDFLX is negative also for other domains
 
It is worth trying to increase grid interval to 1km for your D03 domain. Please keep me updated about the result. Thanks.
 
You are right that negative GRDFLX implies that the heat is transferred from the surface to underground.
 
hamzanisar470 said:
I changed the version, now i am getting around -80w/m2 of GRDLX during day. Why this value is negative ?
Click to expand...
Hello, may I ask which versions of WRF you were using before and which version you are using now? I am very interested in the issue you encountered.


In addition, I was wondering whether you have also found that, in WRF simulations, the energy budget over non-urban areas is nearly balanced, whereas over urban areas it appears to be unbalanced. If so, did changing the WRF version improve this issue?


Wishing you all the best with your research.
 
Would you please upload a map showing the energy budget? It is better to denote urban areas in the map, too.

Please also clarify the details of your case (resolution, input data, physics option, etc.)

Thanks.
 
Ming Chen said:
Would you please upload a map showing the energy budget? It is better to denote urban areas in the map, too.

Please also clarify the details of your case (resolution, input data, physics option, etc.)

Thanks.
Click to expand...
Hello Dr. Chen, I’m really sorry for missing your message.

My WRF version is 4.3.3. The innermost domain resolution is 500m, and the physics schemes used are as shown in the document.

My questions include:

1. The energy balance issue with Roof Photovoltaics (RPV).
The GRDFLX output for RPV is generally a massive negative value, similar to the case reported by hamzanisar470. As long as RPV is deployed, GRDFLX will show abnormal negative values, regardless of the proportion. The size of the negative values changes according to the proportion. I think this may be due to a bug somewhere in the code. Please refer to the provided simulation results for details.
It should be noted that I tried two different physics schemes (see the corresponding excerpted documents), but neither resolved the GRDFLX issue under the PV1 scenario. I also tried using WRF version 4.4 to solve this problem, but it also failed. May I ask which version of WRF should be used to resolve this issue?

2. The energy imbalance issue in urban areas.

After performing the calculations, it can be seen that the energy in non-urban areas is relatively balanced according to the equation, whereas there is an energy imbalance in urban areas. The corresponding land use map and residual calculation results are attached.

My energy balance calculation equations are as follows:

Equation: SWnet + LWnet + AH = LH + SH + G

LH: Calculated using variable LH

SH: Calculated using variable HFX

G: Calculated using variable GRDFLX

SWnet = (1-albedo)*swdown

LWnet = emissivity*(GLW-5.67*10^-8*TSK^4)

For details, please refer to the compressed file. Please let me know if you need any further information. Thank you very much.

RPV result Link: PVResult.zip

WRF input result link: WRFinput.zip
 
cenlinhe said:
which urban physics are you using?
@tslin2 @andreazonato can you take a look at this urban energy balance issue?
Click to expand...
Hi Dr. He,

I'm using the BEP-BEM scheme (sf_urban_physics = 3) coupled with LCZ (landcat = 41).

My main concern right now is the first question: the energy balance issue with Roof Photovoltaics (RPV).
Could you advise on how to fix this? I am currently on WRF 4.3.3 and 4.4. If this is a known bug that has been fixed in the latest WRF release, I can just upgrade. If not, is there any method I can apply to my current versions?

Thanks so much for your help!
 
tslin2 said:
Which land surface model option was used? In version WRFv4.6, there was a fix of ground heat flux sign for NoahMP with urban schemes Bug fix for Noah-MP snow, vegetation and urban by cenlinhe · Pull Request #1929 · wrf-model/WRF
Click to expand...
Hi Dr. Lin,

Thank you for the information! I used the Noah LSM (sf_surface_physics = 2, not Noah-MP sf_surface_physics = 4) for the simulation, so I think the results may not be affected by this bug.

I will also try running the simulation with WRF v4.6 to see see if the newer version resolve this problem.

Thanks again for your help!
 
You must log in or register to reply here.
Back
Top