Question on distributed_ahe_opt Implementation in WRF v4.6/4.7

[ZHOU Wan]

Dear WRF Developers and Users,

I'm currently working with WRF versions 4.6 and 4.7 and exploring the newly added features related to urban modeling, specifically the Anthropogenic Heat Emission (AHE) and SLUCM-related urban parameters.

According to the release notes and the README.namelist file, the distributed_ahe_opt namelist parameter provides three options:



However, when inspecting the source code, I only found logic related to distributed_ahe_opt = 1 inside phys/module_pbl_driver.F (around line 2268). GitHub codes can be found in WRF/phys /module_pbl_driver.F.



I could not locate any implementation or conditional logic for distributed_ahe_opt = 2. My questions are:
1、Is the implementation for distributed_ahe_opt = 2 located in another module or file?
2、Or is this option not yet fully implemented despite being documented in the namelist?
3、If available, could you point me to the relevant code sections or usage examples for options 1 vs. 2?

Any clarification or guidance would be greatly appreciated. Thank you for your efforts in continuously improving the WRF system.

 

[ZHOU Wan]

1、Is the implementation for distributed_ahe_opt = 2 located in another module or file?

Additionally, I noticed that the implementation for `distributed_ahe_opt = 2` appears in `module_surface_driver.F`, where `HFX(i,j)` is directly modified by adding `ahe(i,ihour,j)`. However, this differs from the documented or expected location in the original task submission (see in New option for SLUCM to use global distributed urban aerodynamic parameters (#1986)), where AHE was added elsewhere. Could this difference in the insertion point affect the simulation results due to the order of module calls or data flow across physics components?



When I run with `distributed_ahe_opt = 2` (3 cases with AHE, 5 AHE, and 10 AHE to see the impacts of AHE), the ground and air temperatures remain nearly unchanged, but the sensible heat flux (HFX) shows a significant increase, which aligns with the added anthropogenic heat.




Could this discrepancy be caused by AHE being added too late (or too early) in the calling sequence, such that it doesn't feed back into surface temperature calculations?
Any insights into the correct location or timing for injecting AHE in the WRF physics flow would be greatly appreciated.

 

[Ming Chen]

I have forwarded your question to our expert. We will get back to you soon.

 

[cenlinhe]

The `distributed_ahe_opt = 2` option adds the AHE to the total surface sensible heat flux to the atmosphere after land and urban model computes the 2m temperature for the current timestep, then the added AHE will further affect next step's atmospheric sensible heat flux from the surface and potentially further affects surface temperature in later steps indirectly through the land-atmosphere interaction. From what I can tell, the code place of the added `distributed_ahe_opt = 2` in module_surface_driver.F looks correct. The very small difference in T2 might be because the temperature is already very high (>38degC), so additional AHE shows relatively small impact. Could you please plot the difference maps for T2 and HFX between 10xAHE and AHE?

 

[ZHOU Wan]

Could you please plot the difference maps for T2 and HFX between 10xAHE and AHE?

Hi cenlinhe, thank you for your detailed explanation.
As suggested, I have plotted the difference maps between the 10AHE and AHE experiments for T2, TSK, and HFX at 10 UTC on August 2, 2022.



Under the configuration of distributed_ahe_opt = 0 and AHOPTION = 1 (consider fixed AHE based on URBPARAM.TBL), the increased anthropogenic heat primarily heats the surface (TSK), leading to a significant rise in surface temperature. However, the impact on 2-meter air temperature (T2) is minimal. Does this imply that AHE, in this setup, is not directly involved in sensible heat flux (HFX) exchange with the atmosphere and thus has limited influence on the near-surface atmospheric temperature?