fertilizer rate in CLM5BGC

[liliyao]

Hi all, I am new to CLM model, and I would like to ask a question about the fertilization rate in CLM5. I was doing simple test on one grid cell to see the impact of fertilization rate on hydrological variables (e.g., evaporation). I ran the offline CLM5 simulation with compset I2000Clm50BgcCropGs using CESM2.1.3 for 2 years, res=CLM_USRDAT, and this grid cell is fully coved by CFT45 (rainfed foddergrass). The constant fertilization rate for this crop is 4.1 gN/m^2/yr. To my understanding, fertilizer application in CLM continues for 20 days during the leaf emergence phase, and the fertilizer rate is same during these 20 days. I am confused by my output.

• In the first year’s output, there are 20 days with NFERTILIZATION=1.737E-07 gN/m^2/s
• While in the second year’s output, there are 19 days with NFERTILIZATION=1.737E-07 gN/m^2/s, 1 day with NFERTILIZATION=1.158E-07 gN/m^2/s, and 1 day with NFERTILIZATION=5.789E-08 gN/m^2/s (Note: 1.158E-07+5.789E-08=1.737E-07)

My questions are

• In each year’s output, the total added fertilizer is 0.3 gN/m^2 (1.737E-07 gN/m^2/s * 20 days * 86400 s/day), this number is much smaller than 4.1 gN/m^2/yr set in the surface data.
• For the second year’s output, I have no idea why there are 21 days having fertilizer added.

Any help will be greatly appreciated. Thank you very much.

 

[oleson]

It would be helpful if you could provide the information requested here so that we can try to replicate this behavior on our machines here:

Information to include in help requests

When you are posting to the forums because you want help with an issue, please include as much of the following information as possible. Before submitting a help request, please check to see if your question is already answered: Search the forums for similar issues Check the CIME...

bb.cgd.ucar.edu

CFT 45 is not an active (supported) crop. See this table for supported crops:

2.26. Crops and Irrigation — ctsm release-clm5.0 documentation

I believe what happens in this case is that the unsupported crop is mapped to another similar crop type given by the mergetoclmpft parameter on the parameter file, which would be 19 (rainfed spring wheat). I'm not quite sure what happens with fertilization in this case.

I'd like to look at the surface dataset (fsurdat in lnd_in) and the domain file (fatmlndfrc in lnd_in) if you can provide that.

 

[liliyao]

Hi Oleson, thanks for your quick response!
Yes, I know CFT 45 is not an active crop, and is remapped to rainfed spring wheat. But the fertilizer rate assigned to unsupported crop is still used by the model, right?
Surface dataset, domain file, and CLM version information have been attached this time. Thank you!

 

[oleson]

I tried CFT 45 at a single point site that I had already setup. The constant fertilization rate is 8.179 gN/m2/yr. I found that in the second year (I am using a cold start so crops don't start growing until the second year), I see 19 days of 5.890567e-06 gN/m2/s, 1 day of 2.699843e-06 gN/m2/s, and 1 day of 3.190724e-06 gN/m2/s. But the single days add up to 5.890567e-06 gN/m2/s, so there are effectively 20 days of 5.890567e-06 gN/m2/s.
This is a total of 5.890567e-06 X 20 days X 86400 seconds = 10.179 gN/m2/yr. If we subtract the background manure amount of 2 gN/m2, that is 8.179 gN/m2/yr, which agrees with the amount on the surface dataset. So I think that is ok, the fertilizer is spread out across 21 days, probably because it's application is begun during a day and ended during another day, however, the total is correct. You are correct that the fertilizer rate assigned to the unsupported crop is used.

I'm not sure why your totals don't agree, I'll take a look at your datasets next.

 

[oleson]

Your surface dataset shows:

PCT_NATVEG =
88.8749340004429 ;

PCT_CROP =
11.125065999557 ;

So the gridcell average NFERTILIZATION is going to be reduced from that of the crop value because natural vegetation makes up 89% of the grid cell. NFERTILIZATION will be zero for the natural vegetation landunit.

 

[liliyao]

Thank you so much! I did not realized PCT_CROP = 11.125065999557. and thanks for this information: the background manure amount of 2 gN/m2. Now the added fertilizer in my output agree with the amount on the surface dataset. 3.919e-7 x 20 days x 86400s / 11.1% - 2 = 4.1
(the daily added fertilizer in output is 3.919e-7gN/m2/s, not 1.737e-07, I put a wrong number by mistake in my original message)

 

[AnuNTNU]

@oleson , So If i change the crop distribution inside a gridcell, i also need to change the fertilization accordingly right. Is there a script or description on how to change fertilization in accordance with crop distribution?

Thanks in advance

 

[slevis]

The above conversation indicates that fertilization rate is per meter-squared, and that the model will adjust fertilization for you according to the crop distribution.