Difference in the Number of Soil classes between LDT generated netcdf file (16 classes) and Noah-MP parameter table (19 classes) #1718
-
| Hello, I was working with Noah-MP land surface model within LIS. While doing some changes in soil parameters I found that in the soil parameter table of Noah-MP 4.0.1 uses 19 soil classes whereas in LDT as per (LIS-modified) STATSGO+FAO blended soil texture map it considers 16 soil classes as per the image from python where I tried to read the LDT generated LIS input and parameter file (netcdf). It would be really helpful to know how this mismatch is actually getting handled inside the LIS source code and how the parameter values from the SOILPARM values are getting assigned to the pixels of that specific soil class. | 
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 3 replies
-
| The SOILPARM.TBL contains 3 extra classes that are only used for specialized high-resolution regions. The input STATSGO+FAO soil texture map in LDT only contains the first 16 of the full list of texture classes in the SOILPARM file. That's because at the resolution of the data, types 17-19 are not the most dominant texture at any location. A README and other info on the input STATSGO-FAO data can be found here: | 
Beta Was this translation helpful? Give feedback.

Hi @Arijit22IITDelhi
Yes, I believe that it will, as it currently reads the first 19 texture types from SOILPARM file.
Be sure to change this line from 16 to 17 when you manually edit the LDT-generated netCDF file:
soiltypes = 17 ;