Atmospheric boundary layer simulation

[Ritambhara Dubey]

Hello,

I am trying to simulate an atmospheric boundary layer in a domain of size 3200 m X 3200 x X 4800 m. The initial temperature stratification is neutral up to a height of 800 m with a stable layer of 1K/100m aloft. The boundary layer is heated by imposing a uniform heat flux at the bottom surface. Currently, I am using a mesh size of 320 X 320 X 240. I have a questions related to this problem:

If I use 'ABL' for this case and provide a value for Roughness length and Obukhov length, my initial temperature profile (neutral up to 800 m and then 1K/100 m aloft) is overwritten. Is there a way how, I can provide both? 

Thanks in advance for the help.

[Kevin McGrattan]

If you specify an Obukhov length and similar M-O parameters, FDS automatically generates a velocity and temperature profile for the atmosphere. 

[Ritambhara Dubey]

Thanks Kevin. 

Is there a way I can generate this specific profile using ABL parameters?

[Kevin McGrattan]

Show me you WIND and SURF lines.

[Ritambhara Dubey]

Sure, here it is.

&WIND U0 = 0, RAMP_TMP0_Z='T profile' /

&VENT DB='ZMIN', SURF_ID='GROUND' /

&VENT DB='ZMAX', SURF_ID='TOP' /

&VENT DB='XMIN', SURF_ID='PERIODIC' /

&VENT DB='XMAX', SURF_ID='PERIODIC' /

&VENT DB='YMIN', SURF_ID='PERIODIC' /

&VENT DB='YMAX', SURF_ID='PERIODIC' /

&SURF ID ='GROUND', HEAT_TRANSFER_MODEL='ABL', CONVECTIVE_HEAT_FLUX=0.125 /

&SURF ID ='TOP', FREE_SLIP=.TRUE. /

[Kevin McGrattan]

When you use these lines, do you get the desired temperature profile?

[Ritambhara Dubey]

Yes. Here the T profile is as shown below:

&RAMP ID='T profile', Z= 800 , F= 1 /

&RAMP ID='T profile', Z= 900 , F= 1.0033 /

&RAMP ID='T profile', Z= 1000 , F= 1.0066 /

&RAMP ID='T profile', Z= 1100 , F= 1.01 /

&RAMP ID='T profile', Z= 1200 , F= 1.0133 /

&RAMP ID='T profile', Z= 1300 , F= 1.0166 /

&RAMP ID='T profile', Z= 1400 , F= 1.02 /

&RAMP ID='T profile', Z= 1500 , F= 1.0233 /

&RAMP ID='T profile', Z= 1600 , F= 1.0266 /

&RAMP ID='T profile', Z= 1700 , F= 1.03 /

&RAMP ID='T profile', Z= 1800 , F= 1.0333 /

&RAMP ID='T profile', Z= 1900 , F= 1.0366 /

&RAMP ID='T profile', Z= 2000 , F= 1.04 /

&RAMP ID='T profile', Z= 2100 , F= 1.0433 /

&RAMP ID='T profile', Z= 2200 , F= 1.0466 /

&RAMP ID='T profile', Z= 2300 , F= 1.05 /

&RAMP ID='T profile', Z= 2400 , F= 1.0533 /

&RAMP ID='T profile', Z= 2500 , F= 1.0566 /

&RAMP ID='T profile', Z= 2600 , F= 1.06 /

&RAMP ID='T profile', Z= 2700 , F= 1.0633 /

&RAMP ID='T profile', Z= 2800 , F= 1.0666 /

&RAMP ID='T profile', Z= 2900 , F= 1.07 /

&RAMP ID='T profile', Z= 3000 , F= 1.0733 /

&RAMP ID='T profile', Z= 3100 , F= 1.0766 /

&RAMP ID='T profile', Z= 3200 , F= 1.08 /

&RAMP ID='T profile', Z= 3300 , F= 1.0833 /

&RAMP ID='T profile', Z= 3400 , F= 1.0866 /

&RAMP ID='T profile', Z= 3500 , F= 1.09 /

&RAMP ID='T profile', Z= 3600 , F= 1.0933 /

&RAMP ID='T profile', Z= 3700 , F= 1.0966 /

&RAMP ID='T profile', Z= 3800 , F= 1.1 /

&RAMP ID='T profile', Z= 3900 , F= 1.1033 /

&RAMP ID='T profile', Z= 4000 , F= 1.1066 /

&RAMP ID='T profile', Z= 4100 , F= 1.11 /

&RAMP ID='T profile', Z= 4200 , F= 1.1133 /

&RAMP ID='T profile', Z= 4300 , F= 1.1166 /

&RAMP ID='T profile', Z= 4400 , F= 1.12 /

&RAMP ID='T profile', Z= 4500 , F= 1.1233 /

&RAMP ID='T profile', Z= 4600 , F= 1.1266 /

&RAMP ID='T profile', Z= 4700 , F= 1.13 /

&RAMP ID='T profile', Z= 4800 , F= 1.1333 /

[Kevin McGrattan]

OK, so is your problem solved?

[Ritambhara Dubey]

The problem of providing the input is solved.

I am actually trying to validate a paper in FDS. The paper is ' Raasch, S., & Etling, D. (1991). Numerical simulation of rotating turbulent thermal convection. Contributions to atmospheric physics, 64(3), 185-199.'  When I simulate the problem, the inversion height reduces quickly from 800 m to ~400 m. I am not sure where why this is happening. Any help from your side would be appreciated. 

[Kevin McGrattan]

If you do not provide a heat flux at the ground, and just create the thermal profile with no velocity or wind, does the temperature profile persist? That is, is your initial temperature profile stable?

[Ritambhara Dubey]

Hey Kevin,

I tried a case with no velocity(wind) and no heat flux for 1000 seconds. My initial temperature profile does not persists. Please find attached the plot of it.

[Kevin McGrattan]

Could you plot the actual temperature profiles that FDS exhibits, not the potential temperature, but the actual temperature in degrees C.

[Kevin McGrattan]

OK, good. So is there a problem?