A Question about Calculation of rootdepth

[xxxxyyyyyzzzz]

Hi everyone,
Hope you all are doing well.
In landcovrerType.py, the calculation of "rootDepth[1][i]" uses "soildepth12". I think it would be more appropriate to use "soildepth[1]" instead. Because "rootDepth[1][i]" represents the root depth of the second layer of soil, while "soildepth12" is the sum of the depths of the second and third layers of soil. The calculation method of "np.minimum(self.var.soildepth12 - 0.05, h1)" cannot accurately represent the root depth of the second layer of soil.

Please help me this regard.
Thank you very much.
Jinyuan Liu

[Supsa]

Hi Jinyuan,
I think what you said would make sense. But from my understanding, the logic of CWatM is quite different.

To be honest it is quite tricky to figure it out, and I would really appreciate some review from CWatM team on that topic, but here is what (I think) I have understand of it:

• CWatM works with 3 soil layers. The top layer thickness is fixed at 5 cm. The bottom layer thickness is defined by StorDepth2 (in settings) minus the thickness of the top two layers. The thickness of the top two layers (the top layer (5 cm) plus the intermediate layer) is determined for each landcover by maxRootDepth inputs (in settings), but it cannot be any thiner than StorDepth1 (StorDepth1 is used as an inferior limit)
  • Except for grassland landcover type. For grasslands, grassland_maxRootDepth is ignored and the thickness of the top two layers directly equals StorDepth1.
• Soil parameters given in settings (like KSat1, alpha1... and forest_KSat1, forest_alpha1....) are therefore not used for the StorDepth1 and StorDepth2 thicknesses, but for the soil thicknesses computed by CWatM as above (and named rootDepth[0, 1 and 2] in CWatM scripts)
• Finally, the maximum plant transpiration (transpiration demand weighted by a transpiration reduction factor) is computed for each landcover. Then, it is distributed into each of the three soil layer, according to rootFraction1 inputs (in settings). For instance, if grassland_rootFraction1 = 0.7, it means that, for grasslands, 70% of the computed maximum transpiration will be demanded to the top two soil layers, and 30% to the bottom layer.
  • in theory, it is possible to deactivate this with rootFrac = False (in settings), in which case rootFraction is proportional to soil layer thicknesses
  • if I understand correctly, it means that maxRootDepth is not the maximum root depth (transpiration can occur below), but it rather plays two roles:
    • it is the soil depth above which "rootFraction1 part" of the roots are located
    • it is used as the thickness of the top two soil layers (as indicated in Burek et al. 2020), except for grasslands, and is therefore associated to the soil parameters defined for layers 1 and 2

But as I said, I am not sure I understood it correctly, and any official explanation would be much welcome! 🙏

Alexandre Coche