Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

Questions about anthropogenic heat flux in CLM5

KeerZ

Member
Hi all,

I am a little puzzled by my anthropogenic heat flux result from an I2000Clm50SpGs run. It is much smaller than I expected (Figures attached). I am using CLM5 and in the user_nl_clm I specified:
urban_hac='ON_WASTEHEAT'
building_temp_method=1

I calculated Qanthro as (HEAT_FROM_AC+WASTEHEAT)/Total PCT_URBAN. The Qanthro should be the anthropogenic heat flux with respect to the urban area. However, I found that only U.S. cities have Qanthro of 2-20W/m2, and cities from other counties have near-zero Qanthro. I know that CLM does not consider heat from the traffic system and will underestimate anthropogenic heat. But the Qanthro I got is surprisingly low for the non-US cities (I felt that they should not be 0...). Is this a normal behavior of CLM5?
Also, I wonder if there are any setups I can use to get higher Qanthro. For example, try building_temp_method=0?

Thank you very much!!
 

Attachments

  • Picture1.png
    Picture1.png
    214.9 KB · Views: 16

oleson

Keith Oleson
CSEG and Liaisons
Staff member
The amount of AHF going into the climate system is URBAN_HEAT+WASTEHEAT (see equation 13 in Oleson&Feddema, 2019). You are using HEAT_FROM_AC+WASTEHEAT, which are the terms that enter the energy balance. The difference is that another term in the energy balance (FGR) already incorporates the energy applied for space heating and cooling.
Your figure shows AHF in summer. In the northern hemisphere most of the AHF will come from AC, however, the model does not apply much AC except in regions where AC is available in nearly all urban areas (e.g., in the U.S.). In winter, you should see more AHF because all urban areas are kept warm.
There are several ways to increase AHF. You could increase the wasteheat factors, which are currently set as 0.2 and 0.6 for heating and cooling in clm_varcon, respectively.
Or you can increase/decrease the thermostat settings. The thermostat settings for heating are in the surface dataset (T_BUILDING_MIN) and for cooling are in the "urbantv" stream defined in lnd_in.
 

KeerZ

Member
The amount of AHF going into the climate system is URBAN_HEAT+WASTEHEAT (see equation 13 in Oleson&Feddema, 2019). You are using HEAT_FROM_AC+WASTEHEAT, which are the terms that enter the energy balance. The difference is that another term in the energy balance (FGR) already incorporates the energy applied for space heating and cooling.
Your figure shows AHF in summer. In the northern hemisphere most of the AHF will come from AC, however, the model does not apply much AC except in regions where AC is available in nearly all urban areas (e.g., in the U.S.). In winter, you should see more AHF because all urban areas are kept warm.
There are several ways to increase AHF. You could increase the wasteheat factors, which are currently set as 0.2 and 0.6 for heating and cooling in clm_varcon, respectively.
Or you can increase/decrease the thermostat settings. The thermostat settings for heating are in the surface dataset (T_BUILDING_MIN) and for cooling are in the "urbantv" stream defined in lnd_in.
I see. Is it correct to say that the FGR contains two parts? One part is URBAN_HEAT which is due to space heating and another part is (FGR-URBAN_HEAT) which is not relevant to human activities.

Also, I think maybe the total energy flux due to human activities can be calculated as the sum of HEAT_FROM_AC, WASTEHEAT, and URBAN_HEAT. Here HEAT_FROM_AC is not additional energy into the climate system but it is moved from buildings into the canyons by AC.

Thanks for your help!
 

oleson

Keith Oleson
CSEG and Liaisons
Staff member
FGR includes both heating and cooling fluxes.
The way I think of the energy balance for an urban area is:
:
energy_balance_u = fsa_u - fira_u - eflx_lh_tot_u - fsh_u - fgr_u + wasteheat + heat_from_ac

or, since urban_ac = heat_from_ac:

energy_balance_u = fsa_u - fira_u - eflx_lh_tot_u - fsh_u - fgr_u + wasteheat + urban_ac

This can be re-written such that heating and cooling are removed from fgr_u (note that urban_ac and urban_heat have different signs because they
act in opposite directions with respect to the urban surface):

energy_balance_u = fsa_u - fira_u - eflx_lh_tot_u - fsh_u - (fgr_u - urban_ac + urban_heat) + wasteheat + urban_heat

You are right that HEAT_FROM_AC is not additional energy into the climate system because heat is removed from the interior of the building into the urban canyon. The only part that is additional energy is the wasteheat associated with inefficiencies in the AC system and the power systems that generate the electricity that runs the AC.
I guess you could consider HEAT_FROM_AC or URBAN_AC to be due to human activities from a philosophical viewpoint.
 

KeerZ

Member
FGR includes both heating and cooling fluxes.
The way I think of the energy balance for an urban area is:
:
energy_balance_u = fsa_u - fira_u - eflx_lh_tot_u - fsh_u - fgr_u + wasteheat + heat_from_ac

or, since urban_ac = heat_from_ac:

energy_balance_u = fsa_u - fira_u - eflx_lh_tot_u - fsh_u - fgr_u + wasteheat + urban_ac

This can be re-written such that heating and cooling are removed from fgr_u (note that urban_ac and urban_heat have different signs because they
act in opposite directions with respect to the urban surface):

energy_balance_u = fsa_u - fira_u - eflx_lh_tot_u - fsh_u - (fgr_u - urban_ac + urban_heat) + wasteheat + urban_heat

You are right that HEAT_FROM_AC is not additional energy into the climate system because heat is removed from the interior of the building into the urban canyon. The only part that is additional energy is the wasteheat associated with inefficiencies in the AC system and the power systems that generate the electricity that runs the AC.
I guess you could consider HEAT_FROM_AC or URBAN_AC to be due to human activities from a philosophical viewpoint.
Thanks, Keith! This is really helpful.
 

KeerZ

Member
FGR includes both heating and cooling fluxes.
The way I think of the energy balance for an urban area is:
:
energy_balance_u = fsa_u - fira_u - eflx_lh_tot_u - fsh_u - fgr_u + wasteheat + heat_from_ac

or, since urban_ac = heat_from_ac:

energy_balance_u = fsa_u - fira_u - eflx_lh_tot_u - fsh_u - fgr_u + wasteheat + urban_ac

This can be re-written such that heating and cooling are removed from fgr_u (note that urban_ac and urban_heat have different signs because they
act in opposite directions with respect to the urban surface):

energy_balance_u = fsa_u - fira_u - eflx_lh_tot_u - fsh_u - (fgr_u - urban_ac + urban_heat) + wasteheat + urban_heat

You are right that HEAT_FROM_AC is not additional energy into the climate system because heat is removed from the interior of the building into the urban canyon. The only part that is additional energy is the wasteheat associated with inefficiencies in the AC system and the power systems that generate the electricity that runs the AC.
I guess you could consider HEAT_FROM_AC or URBAN_AC to be due to human activities from a philosophical viewpoint.
Hi Keith, I have a follow-up discussion. I have completed two simulations, one uses BSSP585 compset and another uses I2000Clm50SpGs compset. Both of them used urban_hac='ON_WASTEHEAT' and building_temp_method=1.

However, I did not output URABN_HEAT because I thought HEAT_FROM_AC + WASTEHEAT is the total anthropogenic heat flux. Now if I want to use URABN_HEAT+WASTEHEAT or URABN_HEAT+WASTEHEAT+HEAT_FROM_AC as my anthropogenic heat. I think I don't need to rerun the simulations and output URBAN_HEAT again. I can calculate the URABN_HEAT using this relationship WASTEHEAT=0.2*HEAT_FROM_AC+0.6*URABN_HEAT. I've run a land-only simulation with the URABN_HEAT, WASTEHEAT, and HEAT_FROM_AC output to verify this simple linear relationship. I found that my calculated URABN_HEAT(URABN_HEAT=5*(WASTEHEAT-0.6*HEAT_FROM_AC)) is almost the same as the URABN_HEAT output.

I don't think it is risky to calculate URABN_HEAT offline using WASTEHEAT and HEAT_FROM_AC. Do you think so?
 

oleson

Keith Oleson
CSEG and Liaisons
Staff member
I think that would work fine. Your first equation should be WASTEHEAT = 0.6*HEAT_FROM_AC + 0.2*URBAN_HEAT, but your second equation is correct.
 

KeerZ

Member
I think that would work fine. Your first equation should be WASTEHEAT = 0.6*HEAT_FROM_AC + 0.2*URBAN_HEAT, but your second equation is correct.
Hi Keith, now I calculate the anthropogenic heat flux as '(URBAN_HEAT+WASTEHEAT)/Total PCT_URBAN. I might be overthinking about this, but I still feel my summer anthropogenic heat flux is too small even for US cities (0-7.8W/m2).

Thus, I am trying to confirm the magnitudes of AFH based on other people’s simulations. However, I haven't found any published studies which show the summer anthropogenic heat flux results from CLM5. Do you know whether there are any available sources/data that I can use to double-check the magnitude of my summer AFH? I looked at Land Diagnostics Package (lnd_diag) but there are no AFH results.

Or I wonder if you (or other CLM5 users) have one-summer AFH data and are willing to share them with me just to help me confirm my AFH results?

Thanks a lot!
 

oleson

Keith Oleson
CSEG and Liaisons
Staff member
The numbers you are getting are probably about right. There will likely be very little space heating in CONUS in JJA so URBAN_HEAT will be zero. WASTEHEAT will include the amount of air conditioning but multiplied by a factor of 0.6. The amounts in W/m2 might seem small but remember that they are occurring every model time step on average throughout the entire summer. Presumably, the AC would be close to zero during late nighttime and early morning and then ramp up in the afternoon. So there should be a diurnal cycle in the AC applied.
If you want to check to see if your results are similar to another simulation, the data for the CLM5 simulation reported on in Oleson and Feddema (2019) is at:

/glade/p/cgd/tss/people/oleson/CLM_LAND_ONLY_RELEASE/CLM5/clm50sp_r270_1deg_cplh_Uv2_hist/lnd/hist
 

KeerZ

Member
The numbers you are getting are probably about right. There will likely be very little space heating in CONUS in JJA so URBAN_HEAT will be zero. WASTEHEAT will include the amount of air conditioning but multiplied by a factor of 0.6. The amounts in W/m2 might seem small but remember that they are occurring every model time step on average throughout the entire summer. Presumably, the AC would be close to zero during late nighttime and early morning and then ramp up in the afternoon. So there should be a diurnal cycle in the AC applied.
If you want to check to see if your results are similar to another simulation, the data for the CLM5 simulation reported on in Oleson and Feddema (2019) is at:

/glade/p/cgd/tss/people/oleson/CLM_LAND_ONLY_RELEASE/CLM5/clm50sp_r270_1deg_cplh_Uv2_hist/lnd/hist
Thank you so much! I confirmed that the summer AFH in our two simulations has similar values. The AFH with respect to the urban areas is about 0-8W/m2 in summer. Really appreciate your help.
 

acgvar

Alvin C.G. Varquez
Member
The amount of AHF going into the climate system is
Code:
URBAN_HEAT+WASTEHEAT
(see equation 13 in Oleson&Feddema, 2019). You are using HEAT_FROM_AC+WASTEHEAT, which are the terms that enter the energy balance. The difference is that another term in the energy balance (FGR) already incorporates the energy applied for space heating and cooling.
Your figure shows AHF in summer. In the northern hemisphere most of the AHF will come from AC, however, the model does not apply much AC except in regions where AC is available in nearly all urban areas (e.g., in the U.S.). In winter, you should see more AHF because all urban areas are kept warm.
There are several ways to increase AHF. You could increase the wasteheat factors, which are currently set as 0.2 and 0.6 for heating and cooling in clm_varcon, respectively.
Or you can increase/decrease the thermostat settings. The thermostat settings for heating are in the surface dataset (T_BUILDING_MIN) and for cooling are in the "urbantv" stream defined in lnd_in.
I really appreciate this discussion. It clarifies a lot about the treatment of anthropogenic heat in CLM. Thanks a lot!

I am curious to know about the source codes where

Code:
URBAN_HEAT+WASTEHEAT

is passed into the climate model.
 

oleson

Keith Oleson
CSEG and Liaisons
Staff member
You could start by grepping the variables "eflx_wasteheat" (or more generally "wasteheat") and eflx_urban_heat in src/biogeophys.
There are several urban specific modules in /src/biogeophys also:
UrbanAlbedoMod.F90
UrbanFluxesMod.F90
UrbanParamsType.F90
UrbanRadiationMod.F90
UrbanTimeVarType.F90
UrbBuildTempOleson2015Mod.F90

A description of the building energy model in CLM5 can be found here:

Oleson, K.W., and J. Feddema, 2019: Parameterization and surface data improvements and new capabilities for the Community Land Model Urban (CLMU), JAMES, 11, doi:10.1029/2018MS001586.
 

acgvar

Alvin C.G. Varquez
Member
You could start by grepping the variables "eflx_wasteheat" (or more generally "wasteheat") and eflx_urban_heat in src/biogeophys.
There are several urban specific modules in /src/biogeophys also:
UrbanAlbedoMod.F90
UrbanFluxesMod.F90
UrbanParamsType.F90
UrbanRadiationMod.F90
UrbanTimeVarType.F90
UrbBuildTempOleson2015Mod.F90

A description of the building energy model in CLM5 can be found here:

Oleson, K.W., and J. Feddema, 2019: Parameterization and surface data improvements and new capabilities for the Community Land Model Urban (CLMU), JAMES, 11, doi:10.1029/2018MS001586.
Thanks for the quick reply! I've browsed through these modules in the CLM and the article. I was wondering about a specific script/location of the script that links the outputs of the CLM5 as surface boundaries to the CAM.
 

oleson

Keith Oleson
CSEG and Liaisons
Staff member
The module clm/src/main/lnd2atmMod.F90 creates the gridcell level fields to be passed to the atmosphere and the module clm/src/cpl/lnd_import_export.F90 converts the data to be sent from the clm model to the coupler (subroutine lnd_export). And then I believe these fields are passed to CAM in cam/src/cpl/atm_import_export.F90 (subroutine atm_import).
 

acgvar

Alvin C.G. Varquez
Member
The module clm/src/main/lnd2atmMod.F90 creates the gridcell level fields to be passed to the atmosphere and the module clm/src/cpl/lnd_import_export.F90 converts the data to be sent from the clm model to the coupler (subroutine lnd_export). And then I believe these fields are passed to CAM in cam/src/cpl/atm_import_export.F90 (subroutine atm_import).
Wow, thanks a lot for these specific scripts! I was about to inspect lnd2atmMod.F90 in the clm. Will investigate further.
 

acgvar

Alvin C.G. Varquez
Member
Good day Keith!

This is embarrassing. I cannot seem to find (via grep of wasteheat) the source code that mentions the anthropogenic heat flux (summing of "eflx_wasteheat" and "eflx_urban_heat") that will eventually be passed to the CAM. In the "lnd2atmMod.F90" module, the closest relevant lines would be as follows

Code:
230     call p2g( bounds, &
231          energyflux_inst%eflx_sh_tot_patch (bounds%begp:bounds%endp), &
232          lnd2atm_inst%eflx_sh_tot_grc      (bounds%begg:bounds%endg), &
233          p2c_scale_type='unity',c2l_scale_type='urbanf',l2g_scale_type='unity')

but neither "eflx_wasteheat" nor "eflx_urban_heat" are directly influencing it.
 

oleson

Keith Oleson
CSEG and Liaisons
Staff member
Sorry, I wasn't very clear on this, I thought you were looking for where the turbulent (e.g., sensible) fluxes were passed to the atmosphere. The anthropogenic heat flux is released within the urban canyon, more specifically, as sensible heat flux applied to the canyon floor. So the anthropogenic heat is already incorporated into the surface fluxes passed to the atmosphere.
For more details, see equation 4.26 of the urban technical note:

 

acgvar

Alvin C.G. Varquez
Member
Sorry for taking time to respond. I've been reflecting and trying to understand deeply the methodology. I appreciate the very clear response. I'm trying to conduct a test of utilizing an hourly global AHF dataset into CESM. Now, strategizing to on its implementation.
 
Top