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

How to calculate Net radiation in WRF

W

wonseok ko

New member
Hello,

WRF version: WRF-4.3WPS version: WPS-4.3

I am trying to calculate net radiation with the equation Rn = SWD + GLW - SWU − LWU.
  1. I think that SWD is SWDOWN and GLW is GLW. Is that correct?
  2. I can't find SWU (Upward shortwave flux at the ground surface) and LWU (Upward longwave flux at the ground surface). How can I obtain these values?
  3. What is the difference between SWDOWN, SWDNT, and SWDNB?
  4. What is the difference between LWDNT and LWDNB?
  5. Is there any method to calculate net radiation with radiation flux? I have already calculated it with HFX, LFX, and GFX.
Any help is highly appreciated!
 
Please see my answers below:
wonseok ko said:
Hello,

WRF version: WRF-4.3WPS version: WPS-4.3

I am trying to calculate net radiation with the equation Rn = SWD + GLW - SWU − LWU.
  1. I think that SWD is SWDOWN and GLW is GLW. Is that correct?
Click to expand...
You are correct.
wonseok ko said:
  1. I can't find SWU (Upward shortwave flux at the ground surface) and LWU (Upward longwave flux at the ground surface). How can I obtain these values?
Click to expand...
WRF doesn't output SWU and LWU. However, GSW is "NET SHORT WAVE FLUX AT GROUND SURFACE", which is equivalent to SWD-SWU.

wonseok ko said:
  1. What is the difference between SWDOWN, SWDNT, and SWDNB?
Click to expand...

SWDOWN is "DOWNWARD SHORT WAVE FLUX AT GROUND SURFACE"
SWDNT is "INSTANTANEOUS DOWNWELLING SHORTWAVE FLUX AT TOP"
SWDNB is "INSTANTANEOUS DOWNWELLING SHORTWAVE FLUX AT BOTTOM"

wonseok ko said:
  1. What is the difference between LWDNT and LWDNB?
Click to expand...
LWDNT is "INSTANTANEOUS DOWNWELLING LONGWAVE FLUX AT TOP"
LWDNB is "INSTANTANEOUS DOWNWELLING LONGWAVE FLUX AT BOTTOM"
wonseok ko said:
  1. Is there any method to calculate net radiation with radiation flux? I have already calculated it with HFX, LFX, and GFX.
Click to expand...
net radiation is equal to GSW+netLW, netLW is net longwave radiation in the ground.
wonseok ko said:
Any help is highly appreciated!
Click to expand...
 
Ming Chen said:
Please see my answers below:

You are correct.

WRF doesn't output SWU and LWU. However, GSW is "NET SHORT WAVE FLUX AT GROUND SURFACE", which is equivalent to SWD-SWU.



SWDOWN is "DOWNWARD SHORT WAVE FLUX AT GROUND SURFACE"
SWDNT is "INSTANTANEOUS DOWNWELLING SHORTWAVE FLUX AT TOP"
SWDNB is "INSTANTANEOUS DOWNWELLING SHORTWAVE FLUX AT BOTTOM"


LWDNT is "INSTANTANEOUS DOWNWELLING LONGWAVE FLUX AT TOP"
LWDNB is "INSTANTANEOUS DOWNWELLING LONGWAVE FLUX AT BOTTOM"

net radiation is equal to GSW+netLW, netLW is net longwave radiation in the ground.
Click to expand...
Hello, Ming:
I'm new to WRF, and I'm currently using Noah-MP as the surface physics scheme. I noticed that the variable GSW is not available in the output. However, I found SWDOWN, FSA, and FIRA. What's the difference between SWDOWN and FSA? Does FSA represent netSW, and FIRA represent netLW? Does Rn = FSA + FIRA?
I’d greatly appreciate your detailed feedback.
 
Last edited:
SWDOWN is total downward solar radiation on the surface.
FSA, and FIRA are variables from NoahMP. They represent total absorbed solar radiation and total net longwave radiation.
Positive FSA means downward, and positive FIRA means upward (from land to the atmosphere), so net Rn = FSA - FIRA.
 
Ming Chen said:
SWDOWN is total downward solar radiation on the surface.
FSA, and FIRA are variables from NoahMP. They represent total absorbed solar radiation and total net longwave radiation.
Positive FSA means downward, and positive FIRA means upward (from land to the atmosphere), so net Rn = FSA - FIRA.
Click to expand...
Thanks for your answer! I have another question. Suppose my WRF output is every three hours — are the radiation variables (eg., FSA and FIRA) three-hour averages or instantaneous values at that time step? Is there any official documentation that provides detailed descriptions of the WRF output variables?
 
All variables in wrfout are instantaneous. A brief description of these variables can be found in WRF/Registry/Regsitry.EM_COMMON
 
Hi @husiyang22

GLW is DOWNWARD LONG WAVE FLUX AT GROUND SURFACE. It is not equal to netLW.

netLW should be the sum of downward LW and upward LW emitted from the surface.
 
Ming Chen said:
Please see my answers below:

You are correct.

WRF doesn't output SWU and LWU. However, GSW is "NET SHORT WAVE FLUX AT GROUND SURFACE", which is equivalent to SWD-SWU.



SWDOWN is "DOWNWARD SHORT WAVE FLUX AT GROUND SURFACE"
SWDNT is "INSTANTANEOUS DOWNWELLING SHORTWAVE FLUX AT TOP"
SWDNB is "INSTANTANEOUS DOWNWELLING SHORTWAVE FLUX AT BOTTOM"


LWDNT is "INSTANTANEOUS DOWNWELLING LONGWAVE FLUX AT TOP"
LWDNB is "INSTANTANEOUS DOWNWELLING LONGWAVE FLUX AT BOTTOM"

net radiation is equal to GSW+netLW, netLW is net longwave radiation in the ground.
Click to expand...
Hello Dr.Chen @Ming Chen ,
I’m currently using the WRF BEP+BEM urban scheme, and the WRF version is 4.3.3. I am attempting to calculate and study the surface energy balance, but I have encountered several issues.

1. Issues with _URB Variables and Energy Imbalance

I previously attempted to calculate the energy balance using the following formula:

(Reference: Question about CM_AC_URB3D calculation)

Rn_URB = SH_URB + LH_URB + G_URB

I tested this using the variables SH_URB2D (SENSIBLE HEAT FLUX FROM URBAN SFC), LH_URB2D (LATENT HEAT FLUX FROM URBAN SFC), G_URB2D (GROUND HEAT FLUX INTO URBAN), and RN_URB2D (NET RADIATION ON URBAN SFC).
There is a clear energy imbalance. The difference between RN_URB2D and the sum of the fluxes reaches nearly 1000 W/m².

Additionally, enabling the output of these four variables in BEM mode seems to introduce significant errors in the temperature and humidity simulation. To output RN_URB2D (and related URB2D variables), I modified the Registry and recompiled WRF. After this Registry change + recompilation, the simulation results changed significantly: for the same domain/case (with identical namelist/physics settings), the urban temperature RMSE increased by nearly 1°C compared to the first run result before changing the registry (activating URB2D variables) and complied the WRF. Although I am unsure of the root cause, I hope this observation might be helpful to others.

I have a few questions regarding this:
  • Does this energy imbalance suggest that the standard variables HFX, LH, and GRDFLX are more reliable than the SH_URB2D, LH_URB2D, and G_URB2D variables?
  • Why does such a massive energy imbalance occur? Even when accounting for anthropogenic heat, RN_URB2D consistently reaches values as high as 1400+ W/m² seems abnormal. If you need detailed results, I can provide them.

2. Calculation of NetLW and Anthropogenic Heat

I subsequently switched to the following formula: SWDOWN (SWDOWN) - SWUP + LWDOWN (GLW) - LWUP + Qanthro = SH (HFX) + LH (LH) + G (GRDFLX)
(i.e., NetRad + Qanthro = SH + LH + G).

According to previous discussions in this post, GSW = SWDOWN - SWUP , and NetRad = GSW + NetLW. However, I have checked the registry and variable tables but could not find a specific variable for NetLW.
  • Could you please advise on how to calculate NetLW?
  • Additionally, is there a specific variable or calculation method in WRF to determine Qanthro (anthropogenic heat)?

Thanks for your time and consideration. Wishing you a Happy New Year and a Happy Lunar New Year!

Best regards,

Zhikai Wang
 
You must log in or register to reply here.
Top