Question about field capacity/porosity and wilting point simulation in CLM

[yan Chen]

When I was looking at the soil moisture data published by CESM in CMIP6 (mrsol_Emon_CESM2_amip_r1i1p1f1_gn_195001-201412.nc), I failed to find the corresponding field capacity,ilting point data porosity, and w.

I only found soil texture data on CESM's website (mksrf_soitex.10level.c010119.nc).

I used the data of the percent of clay and percent of sandy to estimate their values but failed.

I would like to know where I can find the corresponding field capacity/porosity and wilting point data. If not, is there any recommended calculation method?

 

[nusbaume]

Hi Yan Chen,

I am moving this post to the land model forum, as they are the most likely group to know the answer to your question.

Also pinging @strandwg, as he helps manage the CMIP6 data, and may have some insights as well.

Hope that helps, and have a great day!

Jesse

 

[oleson]

Is there any of the following information on your file(s):

float WATSAT(levgrnd, lat, lon) ;
WATSAT:_FillValue = 1.e+36f ;
WATSAT:long_name = "saturated soil water content (porosity)" ;
WATSAT:units = "mm3/mm3" ;
WATSAT:missing_value = 1.e+36f ;
float SUCSAT(levgrnd, lat, lon) ;
SUCSAT:_FillValue = 1.e+36f ;
SUCSAT:long_name = "saturated soil matric potential" ;
SUCSAT:units = "mm" ;
SUCSAT:missing_value = 1.e+36f ;
float BSW(levgrnd, lat, lon) ;
BSW:_FillValue = 1.e+36f ;
BSW:long_name = "slope of soil water retention curve" ;
BSW:units = "unitless" ;
BSW:missing_value = 1.e+36f ;
float HKSAT(levgrnd, lat, lon) ;
HKSAT:_FillValue = 1.e+36f ;
HKSAT:long_name = "saturated hydraulic conductivity" ;
HKSAT:units = "mm s-1" ;
HKSAT:missing_value = 1.e+36f ;

WATSAT is porosity. We don't normally output wilting point/field capacity by default in our model. But there are a couple of instances of these concepts in CLM. One calculation of field capacity is watfc in SoilStateInitTimeConstMod.F90, but that history field is off by default. The wilting point concept is used in IrrigationMod.F90. So maybe these quantities could be calculated from soil texture and other soil properties. For soil texture, it would be useful to know what the simulation case was so that we could provide the corresponding surface dataset.

 

[yan Chen]

Is there any of the following information on your file(s):

float WATSAT(levgrnd, lat, lon) ;
WATSAT:_FillValue = 1.e+36f ;
WATSAT:long_name = "saturated soil water content (porosity)" ;
WATSAT:units = "mm3/mm3" ;
WATSAT:missing_value = 1.e+36f ;
float SUCSAT(levgrnd, lat, lon) ;
SUCSAT:_FillValue = 1.e+36f ;
SUCSAT:long_name = "saturated soil matric potential" ;
SUCSAT:units = "mm" ;
SUCSAT:missing_value = 1.e+36f ;
float BSW(levgrnd, lat, lon) ;
BSW:_FillValue = 1.e+36f ;
BSW:long_name = "slope of soil water retention curve" ;
BSW:units = "unitless" ;
BSW:missing_value = 1.e+36f ;
float HKSAT(levgrnd, lat, lon) ;
HKSAT:_FillValue = 1.e+36f ;
HKSAT:long_name = "saturated hydraulic conductivity" ;
HKSAT:units = "mm s-1" ;
HKSAT:missing_value = 1.e+36f ;

WATSAT is porosity. We don't normally output wilting point/field capacity by default in our model. But there are a couple of instances of these concepts in CLM. One calculation of field capacity is watfc in SoilStateInitTimeConstMod.F90, but that history field is off by default. The wilting point concept is used in IrrigationMod.F90. So maybe these quantities could be calculated from soil texture and other soil properties. For soil texture, it would be useful to know what the simulation case was so that we could provide the corresponding surface dataset.

Hi Keith Oleson,

Thank you for the clarification.
I downloaded soil moisture from CMIP6 website: https://esgf-node.llnl.gov/search/cmip6/
Experiment ID: amip; Variant Label: r1i1p1f1; Source ID: CESM; variation: mrsol.
The name of the soil moisture file is mrsol_Emon_CESM2_amip_r1i1p1f1_gn_195001-201412.nc

Could you provide corresponding soil texture dataset?
I am not only looking for surface dataset, I am interested soil texture in top 1 meter and top 2 meter.

BTW. Previously, I downloaded a soil texture file from: https://svn-ccsm-inputdata.cgd.ucar.edu/trunk/inputdata/lnd/clm2/rawdata/, the file name is mksrf_soitex.10level.c010119.nc
However, after I estimated the field capacity via: Soil Hydraulic Properties, I found many values of field capacity are below soil moisture. So, I guess I download a wrong file. That is why I am looking for a right file of the corresponding soil texture.

Thank you for the help.

 

[oleson]

Is there a case name in the attributes of that soil moisture file? E.g., ncdump -h mrsol_Emon_CESM2_amip_r1i1p1f1_gn_195001-201412.nc.
That would help me find the right case and surface dataset that was used.
Or maybe there is even a surface dataset identified in the attributes?

 

[yan Chen]

Is there a case name in the attributes of that soil moisture file? E.g., ncdump -h mrsol_Emon_CESM2_amip_r1i1p1f1_gn_195001-201412.nc.
That would help me find the right case and surface dataset that was used.
Or maybe there is even a surface dataset identified in the attributes?

Thank you for the quick reply, the case name is :
:cesm_casename = "f.e21.FHIST_BGC.f09_f09_mg17.CMIP6-AMIP.001" ;

 

[oleson]

The surface dataset for that case is:
inputdata/lnd/clm2/surfdata_map/surfdata_0.9x1.25_78pfts_CMIP6_simyr1850_c170824.nc

 

[yan Chen]

The surface dataset for that case is:
inputdata/lnd/clm2/surfdata_map/surfdata_0.9x1.25_78pfts_CMIP6_simyr1850_c170824.nc

Thank you for the reply. I noticed percent_clay and percent_sand in this file have 10 layers. However, the output of mrsol (soil moisture) I downloaded from CMIP6 website has a total of 20 layers (up to 8.03 m). I'm curious if the thickness of each layer for both soil texture and soil moisture is consistent, and then the soil texture after the tenth layer remains the same and used to simulate the soil moisture of layers 11-20?

 

[oleson]

Yes, the pct_sand and pct_clay from the 10th layer are used for layers 11-20. See biogeophys/SoilStateInitTimeConstMod.F90 in the code.