After AD SPINUP Simulation, NPP and NEP are 0

[zpc]

Dear all:

The AD SPINUP model was used to preheat clm5.0 for 100 years using self-produced atmospheric forcing data. However, after checking the history file, I find that the values of NNP and NEP becomes 0 after 10 years. Is this normal? Here is my code for running clm. Is there any error? ZIP file is my NEE and NPP of history file.
I have one more question. I will simulate the carbon dynamics from 2015 to 2100. Can I set RUN_TYPE=branch
RUN_REFCASE, RUN_REFDATE and other related information to run？

#! /bin/bash
cd /home/user/clm5.0.35/clm5new/cime/scripts
./create_newcase --case ~/cases/preheat/ACCESS.board1.preheat1206 --res CLM_USRDAT --compset SSP585_DATM%1PT_CLM50%BGC_SICE_SOCN_SROF_SGLC_SWAV --run-unsupported --compiler intel --mach myintel
cd /home/user/cases/preheat/ACCESS.board1.preheat1206
./xmlchange CLM_ACCELERATED_SPINUP="on"
./xmlchange CLM_USRDAT_NAME='tibet'
./xmlchange ATM_DOMAIN_PATH='/home/user/tibet/myout_board1/'
./xmlchange LND_DOMAIN_PATH='/home/user/tibet/myout_board1/'
./xmlchange ATM_DOMAIN_FILE='domain.lnd.fv0.47x0.63_gx1v7.180521_myout_board1_board1.nc'
./xmlchange LND_DOMAIN_FILE='domain.lnd.fv0.47x0.63_gx1v7.180521_myout_board1_board1.nc'
./xmlchange NTASKS=1
./xmlchange STOP_N=1
./xmlchange STOP_OPTION=nyears
./xmlchange RESUBMIT=99
./xmlchange DATM_CLMNCEP_YR_START=2015
./xmlchange DATM_CLMNCEP_YR_END=2015
./xmlchange DATM_CLMNCEP_YR_ALIGN=2015
./xmlchange RUN_REFDATE=0001-01-01
./xmlchange RUN_STARTDATE=0001-01-01
./case.setup
echo "fsurdat = '/home/user/tibet/myout_board1/surfdata_0.47x0.63_78pfts_CMIP6_simyr2000_c180508_myout_board1_board1.nc'
flanduse_timeseries = '/home/user/tibet/myin/landuse.timeseries_0.9x1.25_SSP5-8.5_78pfts_CMIP6_simyr1850-2100_c181209.nc'
finidat = '/home/user/cesm/inputdata/lnd/clm2/initdata_map/clmi.I2000Clm50BgcCrop.2011-01-01.1.9x2.5_gx1v7_gl4_simyr2000_c190312.nc'
! hist_empty_htapes = .true.
hist_fincl1='GPP:A','NPP:A','NEP:A','NEE:A','NBP:A','TOTECOSYSC:A','TOTSOMC:A','TOTVEGC:A','TLAI:A','TWS:A','H2OSNO:A'
hist_fincl2='GPP:X','NPP:X','NEP:X','NEE:X','NBP:X','TOTECOSYSC:X','TOTSOMC:X','TOTVEGC:X','TLAI:X','TWS:X','H2OSNO:X'
hist_nhtfrq= 0,0,0
hist_mfilt = 3600,3600,3600
use_init_interp = .true.
do_harvest = .false.
do_transient_crops = .false.
do_transient_pfts = .false." >> ./user_nl_clm
echo "<?xml version='1.0'?>
<file id='stream' version='1.0'>
<dataSource>
GENERIC
</dataSource>
<domainInfo>
<variableNames>
time time
xc lon
yc lat
area area
mask mask
</variableNames>
<filePath>
/home/user/tibet/myout_board1/
</filePath>
<fileNames>
domain.lnd.fv0.47x0.63_gx1v7.180521_myout_board1_board1.nc
</fileNames>
</domainInfo>
<fieldInfo>
<variableNames>
ZBOT z
TBOT tbot
QBOT shum
WIND wind
PRECTmms precn
FSDS swdn
PSRF pbot
FLDS lwdn
</variableNames>
<filePath>
/home/user/tibet/15pt_forcing
</filePath>
<fileNames>
ACC.a86.board1.nc
</fileNames>
<offset>
0
</offset>
</fieldInfo>
</file>" >> ./user_datm.streams.txt.CLM1PT.CLM_USRDAT
echo "<?xml version='1.0'?>
<file id='stream' version='1.0'>
<dataSource>
GENERIC
</dataSource>
<domainInfo>
<variableNames>
time time
lon lon
lat lat
area area
mask mask
</variableNames>
<filePath>
/home/user/cesm/inputdata/atm/cam/chem/trop_mozart_aero/aero
</filePath>
<fileNames>
aerodep_clm_SSP585_b.e21.BSSP585cmip6.f09_g17.CMIP6-SSP5-8.5.001_2014-2101_monthly_0.9x1.25_c190419.nc
</fileNames>
</domainInfo>
<fieldInfo>
<variableNames>
BCDEPWET bcphiwet
BCPHODRY bcphodry
BCPHIDRY bcphidry
OCDEPWET ocphiwet
OCPHIDRY ocphidry
OCPHODRY ocphodry
DSTX01WD dstwet1
DSTX01DD dstdry1
DSTX02WD dstwet2
DSTX02DD dstdry2
DSTX03WD dstwet3
DSTX03DD dstdry3
DSTX04WD dstwet4
DSTX04DD dstdry4
</variableNames>
<filePath>
/home/user/cesm/inputdata/atm/cam/chem/trop_mozart_aero/aero
</filePath>
<fileNames>
aerodep_clm_SSP585_b.e21.BSSP585cmip6.f09_g17.CMIP6-SSP5-8.5.001_2014-2101_monthly_0.9x1.25_c190419.nc
</fileNames>
<offset>
0
</offset>
</fieldInfo>
</file>" >> ./user_datm.streams.txt.presaero.SSP5-8.5
./case.build
./case.submit

 

[oleson]

It looks like vegetation is alive initially and then slowly dies over the course of the simulation. You could look at the trend in TLAI to confirm that.
Is this a single point or a regional simulation?
One thing to look at would be CO2 (PCO2) and the forcing fields (e.g., FSDS, FLDS, RAIN, SNOW, etc) to see if they look as you would expect.

 

[zpc]

It looks like vegetation is alive initially and then slowly dies over the course of the simulation. You could look at the trend in TLAI to confirm that.
Is this a single point or a regional simulation?
One thing to look at would be CO2 (PCO2) and the forcing fields (e.g., FSDS, FLDS, RAIN, SNOW, etc) to see if they look as you would expect.

I checked the TLAI and found that it also gradually became 0.
This a single point simulation.
My CO2 file is fco2_datm_globalSSP5-8.5__simyr_2014-2501_CMIP6_c190506.nc, and l checked my forcing fields in nc flies they look as my would expect. But the Variable "LATIXY" and "LONGXY" in my nc file is not set to my point.
The follow link is my nc file: ACC.a86.board1.nc

 

[oleson]

There may be other problems, but one thing I see in your forcing file is that ZBOT is 1774 meters, which seems unrealistic.

 

[zpc]

There may be other problems, but one thing I see in your forcing file is that ZBOT is 1774 meters, which seems unrealistic.

hahaha，that's right because my point is on the Tibetan Plateau.

 

[oleson]

ZBOT should be the height above the surface that the forcing temperature, humidity, etc are representative, not the elevation above sea level.
The global attributes on the file say:

site_location = "Latitude: 45.55980 Longitude: 275.28620 Elevation (masl): 50.

The elevation doesn't seem to correspond to the Tibetan Plateau.

 

[zpc]

ZBOT should be the height above the surface that the forcing temperature, humidity, etc are representative, not the elevation above sea level.
The global attributes on the file say:

site_location = "Latitude: 45.55980 Longitude: 275.28620 Elevation (masl): 50.

The elevation doesn't seem to correspond to the Tibetan Plateau.

By CLM existing single-point nc file to make my single-point nc file, I just changed the [' TBOT '] [' PRECTmms] [' WIND '] [' ZBOT] [' PSRF] [' FLDS] [' FSDS ']. 'RH' the eight variables, other attribute information has not changed. What's wrong with this process?

 

[oleson]

What existing CLM single-point file are you referring to?

 

[zpc]

What existing CLM single-point file are you referring to?

https://svn-ccsm-inputdata.cgd.ucar.edu/trunk/inputdata/atm/datm7/CLM1PT_data/1x1pt_US-UMB/

 

[oleson]

That first file (1999-01.nc) shows ZBOT = 50. And the last file (2006-12.nc) shows that as well. Maybe you had a bad download or something happened with your processing script?

 

[zpc]

That first file (1999-01.nc) shows ZBOT = 50. And the last file (2006-12.nc) shows that as well. Maybe you had a bad download or something happened with your processing script?

Thanks for your answer. Maybe there is something wrong with my ZBOT. I will try to change the value of ZBOT again, I mistook it for altitude.
Best wish!

 

[zpc]

That first file (1999-01.nc) shows ZBOT = 50. And the last file (2006-12.nc) shows that as well. Maybe you had a bad download or something happened with your processing script?

Even if I modified the ZBOT variable, TLAI had a rapid decline in the first five years. May I ask what other possible reasons you think？

 

[oleson]

It looks like you intend to spinup by repeating 2015 atmospheric forcing. I would expect TLAI to stabilize eventually after that type of forcing. It doesn't seem to do that, so there must be some variability induced by the forcing, perhaps the forcing is not being repeated exactly year to year.
Please attach all of your log files and your *_in files from your run directory. Also, your datm.streams.* files, and I can take a quick look.

You mentioned: "But the Variable "LATIXY" and "LONGXY" in my nc file is not set to my point."
I'm not sure what you mean by that. I see LATIXY=45.5598 and LONGXY=275.2862 in your datm forcing file, which is obviously not on the Tibetan Plateau. What latitude/longitude do you have on your surface dataset?

 

[zpc]

看起来你打算通过重复2015年的大气强迫来旋转。我预计TLAI在这种强迫之后最终会稳定下来。它似乎没有这样做，所以强迫一定有一些变化，也许强迫并没有逐年重复。
请附加运行目录中的所有日志文件和 *_in 文件。另外，你的datm.streams.*文件，我可以快速浏览一下。

你提到：“但是我的 nc 文件中的变量”LATIXY“和”LONGXY“没有设置为我的观点。
我不确定你的意思。我在你的datm强制文件中看到LATIXY=45.5598和LONGXY=275.2862，这显然不在青藏高原上。您的表面数据集上的纬度/经度是多少？

I was sick last week. I'm sorry to reply to your message so late. The attachment may be the file you need

 

[oleson]

I don't see any log files in that tar file. Can you attach those?

 

[zpc]

I don't see any log files in that tar file. Can you attach those?

 

[oleson]

Those are the bld logs. I would need to look at the actual run logs. They should be in your run directory or if the model completed successfully they should be in your archive directory.
What latitude/longitude do you have on your surface dataset?

 

[zpc]

Those are the bld logs. I would need to look at the actual run logs. They should be in your run directory or if the model completed successfully they should be in your archive directory.
What latitude/longitude do you have on your surface dataset?

My surface dataset latitude and longitude are: lon:95, lat:28.43342037.

 

[oleson]

Thanks for that information. If you look at your atm log file you will see that your atm forcing data is not being read in appropriately. It looks like the only time slice of forcing data that is being read in is:

(shr_dmodel_readstrm) open : /home/user/tibet/15pt_forcing/ACC.a86.board1.nc
(shr_dmodel_readstrm) file lb: /home/user/tibet/15pt_forcing/ACC.a86.board1.nc 251120
(shr_dmodel_readstrm) file ub: /home/user/tibet/15pt_forcing/ACC.a86.board1.nc 251120

No other time slices are being read in after that. So that time slice, which appears to be the last time slice in that file, is being used to force the model at every time step.
I see that you have the following set in your datm_in:

taxmode = "extend", "extend", "extend", "extend"

Generally, taxmode is set to cycle so that during a spinup (if you do intend to be spinning up here; I assume you might be because your model year indicated in the logs is 87, not a real year) the data at the beginning of the forcing file will be repeated once the end of the file is reached. If you don't intend a spinup here then maybe you want to change RUN_STARTDATE to 2015-01-01?

 

[zpc]

Thanks for that information. If you look at your atm log file you will see that your atm forcing data is not being read in appropriately. It looks like the only time slice of forcing data that is being read in is:

(shr_dmodel_readstrm) open : /home/user/tibet/15pt_forcing/ACC.a86.board1.nc
(shr_dmodel_readstrm) file lb: /home/user/tibet/15pt_forcing/ACC.a86.board1.nc 251120
(shr_dmodel_readstrm) file ub: /home/user/tibet/15pt_forcing/ACC.a86.board1.nc 251120

No other time slices are being read in after that. So that time slice, which appears to be the last time slice in that file, is being used to force the model at every time step.
I see that you have the following set in your datm_in:

taxmode = "extend", "extend", "extend", "extend"

Generally, taxmode is set to cycle so that during a spinup (if you do intend to be spinning up here; I assume you might be because your model year indicated in the logs is 87, not a real year) the data at the beginning of the forcing file will be repeated once the end of the file is reached. If you don't intend a spinup here then maybe you want to change RUN_STARTDATE to 2015-01-01?

Thank you for your patient reply! After modifying RUN_STARTDATE and taxmode my results are correct. I had mistakenly thought RUN_STARTDATE was just a date. Best wish!