initial files for single site simulation

[jinmuluo]

Hi,

I have some questions on how to use the surface dataset and initial files.


What version of the code are you using?
ctsm5.3.080

Have you made any changes to files in the source tree?
None changes on the code

Describe every step you took leading up to the problem:
(1) I tried to run a single-site simulation with different PFTs (C3 crop, wheat, rice, and soybean)
(2) Use the CTSM/tools/site_and_regional/subset_data to create the surface dataset by specifying the --dompft 15 for C3 crop only for example

Describe your problem or question:
It looks like the GPP is always zero when I use the new surface dataset created by the "subset_data" tool, and the first year vegetation C is also zero.

Two questions
(1) The subset_data tool changes the PCT_NATVEG into 0 and PCT_CROP into 100, but I still see values under the PCT_NAT_PFT. Does it affect the surface dataset?
(2) The default surface dataset compiled by the subset_data at my target location is in 16 PFTs. Right now, I want to change it into 78 PFTs. Does that mean in the initial file, there isn't any biomass for the PFTs beyond the 16 PFTs?

Different PFT surface dataset here:
DEFAULT: /glade/work/jinmuluo/O3ProjectCases/CRUJRA_LAT42_LON266/surfdata_cornbelt_hist_2000_16pfts_c251022.nc
C3 CROP: /glade/work/jinmuluo/O3ProjectCases/CRUJRA_LAT42_LON266/surfdata_cornbelt_hist_2000_16pfts_dompft_c3crop.nc
WHEAT: /glade/work/jinmuluo/O3ProjectCases/CRUJRA_LAT42_LON266/surfdata_cornbelt_hist_2000_78pfts_dompft_spring_wheat.nc

case with zero vegetation C and GPP: /glade/work/jinmuluo/O3ProjectCases/C3_danica2015

 

[jinmuluo]

This issue arose from the unclear specifications of the output format; now the GPP is not zero after adding " hist_type1d_pertape(1) = ' ' " to the user_nl_clm.

 

[jinmuluo]

@oleson

Hi Keith,
Can you help on these two questions?

(1) The subset_data tool changes the PCT_NATVEG into 0 and PCT_CROP into 100, but I still see values under the PCT_NAT_PFT. Does it affect the surface dataset?
(2) The default surface dataset compiled by the subset_data at my target location is in 16 PFTs. Right now, I want to change it into 78 PFTs. Does that mean in the initial file, there isn't any biomass for the PFTs beyond the 16 PFTs?

 

[oleson]

(1) Those PCT_NAT_PFT values should be ignored in your 100% crop simulation.
(2) If PCT_CROP is 100% and PCT_CFT(1) is 100% then you won't be modeling anything other than that generic crop type #15.

 

[jinmuluo]

(1) Those PCT_NAT_PFT values should be ignored in your 100% crop simulation.
(2) If PCT_CROP is 100% and PCT_CFT(1) is 100% then you won't be modeling anything other than that generic crop type #15.

Hi Keith,

Thank you, I did find the model still simulates the non-zero pft types in PCT_NAT_PFT, even the PCT_NATVEG = 0; but the pfts1d_wtgcell indicates the natural vegetation pft weights are zero.

For example, in one of my outputs of study of the wheat only: /glade/derecho/scratch/jinmuluo/archive/o3cases/Springwheat_0damage/lnd/hist

GPP(time, pft)

GPP = 8.88333e-05, 1.204732e-05, 6.981624e-10, 4.973625e-06, 0.0001121794 ;

data: pfts1d_itype_veg = 1, 7, 13, 14, 19 ;

data: pfts1d_wtgcell = 0, 0, 0, 0, 1 ;

 

[jinmuluo]

(1) Those PCT_NAT_PFT values should be ignored in your 100% crop simulation.
(2) If PCT_CROP is 100% and PCT_CFT(1) is 100% then you won't be modeling anything other than that generic crop type #15.

One more question is, when I tried to run the one PFT-only simulation on the same location ( --lat 42.1834 --lon -95.0471), only the wheat and soybean simulations were successful (GPP were non-zero), but the C3crop and Rice were failed (GPP and TLAI were always zero), do you think that was because the initial file doesn't have any C3 crop and rice PFT at the location I selected?

 

[oleson]

Right, it seems as if the pfts that have non-zero PCT_NAT_PFT are still simulated but won't be included in the gridcell average because their pfts1d_wtgcell are zero. I imagine this is because they were active in the dataset that was subsetted and subset_data just changed the weights. @slevis may have more insight into this tool than I do.

 

[oleson]

Regarding the C3crop and Rice problem, is this your Rice simulation?

/glade/derecho/scratch/jinmuluo/archive/o3cases/Rice_0damage/lnd/hist

ncdump -v pfts1d_itype_veg Rice_0damage.clm2.h0a.2003-07.nc | more
data:

pfts1d_itype_veg = 1, 7, 13, 14, 61 ;
}

I think 61 is tropical corn, not rice?

So I guess you are saying that rice must not have been in the dataset that was subsetted? So even if you ask for Rice in subset data it doesn't give you that?
Again, @slevis may have some insight into this.

 

[jinmuluo]

Regarding the C3crop and Rice problem, is this your Rice simulation?

/glade/derecho/scratch/jinmuluo/archive/o3cases/Rice_0damage/lnd/hist

ncdump -v pfts1d_itype_veg Rice_0damage.clm2.h0a.2003-07.nc | more
data:

pfts1d_itype_veg = 1, 7, 13, 14, 61 ;
}

I think 61 is tropical corn, not rice?

So I guess you are saying that rice must not have been in the dataset that was subsetted? So even if you ask for Rice in subset data it doesn't give you that?
Again, @slevis may have some insight into this.

Hi Keith,

I reference the index of the PFT here: main/PatchType.F90, do you think it's the correct place to look at?
! 60 => irrigated_rapeseed
! 61 => rice
! 62 => irrigated_rice
! 63 => sorghum

 

[jinmuluo]

Regarding the C3crop and Rice problem, is this your Rice simulation?

/glade/derecho/scratch/jinmuluo/archive/o3cases/Rice_0damage/lnd/hist

ncdump -v pfts1d_itype_veg Rice_0damage.clm2.h0a.2003-07.nc | more
data:

pfts1d_itype_veg = 1, 7, 13, 14, 61 ;
}

I think 61 is tropical corn, not rice?

So I guess you are saying that rice must not have been in the dataset that was subsetted? So even if you ask for Rice in subset data it doesn't give you that?
Again, @slevis may have some insight into this.

Hi Keith,

My question is kind of like why the C3crop only PFT simulation has zero GPP, like the case I list below.

/glade/derecho/scratch/jinmuluo/archive/C3_0damage/lnd/hist


ncdump -v GPP C3_0damage.clm2.h0a.2000-08.nc

GPP = 9.029518e-05, 0.0001026504, 4.155958e-10, 7.812619e-05, 0 ;

ncdump -v pfts1d_itype_veg C3_0damage.clm2.h0a.2000-08.nc

pfts1d_itype_veg = 1, 7, 13, 14, 15 ;

 

[oleson]

Ah, yes, I think you are right. I was looking at the global attributes on the history files which shows:

:cft_tropical_corn = 61 ;

but I forgot that the crop types start their numbering at 15 as per the technical note:

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