Several questions about soil layer structure

[Zh Chen]

Dear scientists,
I have several questions about soil layer structure in CLM4.5 and CLM5.0
(1) Why soil layer structure is non-linear in CLM4.5 and CLM5.0? Why is the difference between the two version?
(2) Running CLM4.5 or CLM5.0 over a single-point, observations at 5, 10, 20, 40cm were compared against CLM4.5 at node depth 6.23,11.89, 21.22, 36.61cm or CLM5.0 at node depth 4, 9, 16, 40cm, it's right?
(3) I read CLM5 technical notes: The soil column can be discretized into an arbitrary number of layers. Whether soil structure can be modified in CLM4.5 and 5.0, is it recommended to modify it in two versions?
Thanks

 

[oleson]

(1) The non-linear structure is mainly to better resolve water and heat dynamics closer to the surface. CLM5 has increased soil vertical resolution plus bedrock layers, was well as spatially varying soil depth per observations.
(2) That seems like the right node depths for comparison. You could also interpolate the CLM data to the observed depths.
(3) I'm not aware of any studies in which this has actually been done.